[Federal Register Volume 85, Number 84 (Thursday, April 30, 2020)]
[Rules and Regulations]
[Pages 24174-25278]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2020-06967]
[[Page 24173]]
Vol. 85
Thursday,
No. 84
April 30, 2020
Part IV
Book 2 of 3 Books
Pages 24173-25278
Environmental Protection Agency
-----------------------------------------------------------------------
40 CFR Parts 86 and 600
Department of Transportation
-----------------------------------------------------------------------
National Highway Traffic Safety Administration
49 CFR Parts 523, 531, 533, et al.
-----------------------------------------------------------------------
The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model
Years 2021-2026 Passenger Cars and Light Trucks; Final Rule
Federal Register / Vol. 85 , No. 84 / Thursday, April 30, 2020 /
Rules and Regulations
[[Page 24174]]
-----------------------------------------------------------------------
ENVIRONMENTAL PROTECTION AGENCY
40 CFR Parts 86 and 600
DEPARTMENT OF TRANSPORTATION
National Highway Traffic Safety Administration
49 CFR Parts 523, 531, 533, 536, and 537
[NHTSA-2018-0067; EPA-HQ-OAR-2018-0283; FRL 10000-45-OAR]
RIN 2127-AL76; RIN 2060-AU09
The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for
Model Years 2021-2026 Passenger Cars and Light Trucks
AGENCY: Environmental Protection Agency and National Highway Traffic
Safety Administration.
ACTION: Final rule.
-----------------------------------------------------------------------
SUMMARY: EPA and NHTSA, on behalf of the Department of Transportation,
are issuing final rules to amend and establish carbon dioxide and fuel
economy standards. Specifically, EPA is amending carbon dioxide
standards for model years 2021 and later, and NHTSA is amending fuel
economy standards for model year 2021 and setting new fuel economy
standards for model years 2022-2026. The standards set by this action
apply to passenger cars and light trucks, and will continue our
nation's progress toward energy independence and carbon dioxide
reduction, while recognizing the realities of the marketplace and
consumers' interest in purchasing vehicles that meet all of their
diverse needs. These final rules represent the second part of the
Administration's action related to the August 24, 2018 proposed Safer
Affordable Fuel-Efficient (SAFE) Vehicles Rule. These final rules
follow the agencies' actions, taken September 19, 2019, to ensure One
National Program for automobile fuel economy and carbon dioxide
emissions standards, by finalizing regulatory text related to
preemption under the Energy Policy and Conservation Act and withdrawing
a waiver previously provided to California under the Clean Air Act.
DATES: This final rule is effective on June 29, 2020.
Judicial Review: NHTSA and EPA undertake this joint action under
their respective authorities pursuant to the Energy Policy and
Conservation Act and the Clean Air Act. Pursuant to CAA section 307(b),
42 U.S.C. 7607(b), any petitions for judicial review of this action
must be filed in the United States Court of Appeals for the D.C.
Circuit. Given the inherent relationship between the agencies' action,
any challenges to NHTSA's regulation under 49 U.S.C. 32909 should also
be filed in the United States Court of Appeals for the D.C. Circuit.
ADDRESSES: EPA and NHTSA have established dockets for this action under
Docket ID Nos. EPA-HQ-OAR-2018-0283 and NHTSA-2018-0067, respectively.
All documents in the docket are listed in the http://www.regulations.gov index. Although listed in the index, some
information is not publicly available, e.g., confidential business
information (CBI) or other information whose disclosure is restricted
by statute. Certain other material, such as copyrighted material, will
be publicly available in hard copy in EPA's docket, and electronically
in NHTSA's online docket. Publicly available docket materials can be
found either electronically in www.regulations.gov by searching for the
dockets using the Docket ID numbers above, or in hard copy at the
following locations:
EPA: EPA Docket Center, EPA/DC, EPA West, Room 3334, 1301
Constitution Ave. NW, Washington, DC. The Public Reading Room is open
from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal
holidays. The telephone number for the Public Reading Room is (202)
566-1744.
NHTSA: Docket Management Facility, M-30, U.S. Department of
Transportation (DOT), West Building, Ground Floor, Rm. W12-140, 1200
New Jersey Ave. SE, Washington, DC 20590. The DOT Docket Management
Facility is open between 9 a.m. and 5 p.m. Eastern Time, Monday through
Friday, except Federal holidays.
FOR FURTHER INFORMATION CONTACT: EPA: Christopher Lieske, Office of
Transportation and Air Quality, Assessment and Standards Division,
Environmental Protection Agency, 2000 Traverwood Drive, Ann Arbor, MI
48105; telephone number: (734) 214-4584; fax number: (734) 214-4816;
email address: [email protected], or contact the Assessment
and Standards Division, email address: [email protected]. NHTSA: James Tamm,
Office of Rulemaking, Fuel Economy Division, National Highway Traffic
Safety Administration, 1200 New Jersey Avenue SE, Washington, DC 20590;
telephone number: (202) 493-0515.
SUPPLEMENTARY INFORMATION:
Does this action apply to me?
This action affects companies that manufacture or sell new light-
duty vehicles, light-duty trucks, and medium-duty passenger vehicles,
as defined under EPA's CAA regulations,\1\ and passenger automobiles
(passenger cars) and non-passenger automobiles (light trucks) as
defined under NHTSA's CAFE regulations.\2\ Regulated categories and
entities include:
---------------------------------------------------------------------------
\1\ ``Light-duty vehicle,'' ``light-duty truck,'' and ``medium-
duty passenger vehicle'' are defined in 40 CFR 86.1803-01. Generally
speaking, a ``light-duty vehicle'' is a passenger car, a ``light-
duty truck'' is a pick-up truck, sport-utility vehicle, or minivan
up to 8,500 lbs. gross vehicle weight rating, and a ``medium-duty
passenger vehicle'' is a sport-utility vehicle or passenger van from
8,500 to 10,000 lbs. gross vehicle weight rating.
\2\ ``Passenger car'' and ``light truck'' are defined in 49 CFR
part 523.
---------------------------------------------------------------------------
[[Page 24175]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.000
This list is not intended to be exhaustive, but rather provides a
guide regarding entities likely to be regulated by this action. To
determine whether particular activities may be regulated by this
action, you should carefully examine the regulations. You may direct
questions regarding the applicability of this action to the person
listed in FOR FURTHER INFORMATION CONTACT.
I. Executive Summary
II. Overview of Final Rule
III. Purpose of the Rule
IV. Purpose of Analytical Approach Considered as Part of Decision-
Making
V. Regulatory Alternatives Considered
VI. Analytical Approach as Applied to Regulatory Alternatives
VII. What does the analysis show, and what does it mean?
VIII. How do the final standards fulfill the agencies' statutory
obligations?
IX. Compliance and Enforcement
X. Regulatory Notices and Analyses
I. Executive Summary
NHTSA (on behalf of the Department of Transportation) and EPA are
issuing final rules to adopt and modify standards regulating corporate
average fuel economy and tailpipe carbon dioxide (CO2)
emissions and use/leakage of other air conditioning refrigerants for
passenger cars and light trucks for MYs 2021-2026.\3\ These final rules
follow the proposal issued in August 2018 and respond to each agency's
legal obligation to set standards based on the factors Congress
directed them to consider, as well as the direction of the United
States Supreme Court in Massachusetts v. EPA, which stated that ``there
is no reason to think the two agencies cannot both administer their
obligations and yet avoid inconsistency.'' \4\ These standards are the
product of significant and ongoing work by both agencies to craft
regulatory requirements for the same group of vehicles and vehicle
manufacturers. This work aims to facilitate, to the extent possible
within the statutory directives issued to each agency, the ability of
automobile manufacturers to meet all requirements under both programs
with a single national fleet under one national program of fuel economy
and tailpipe CO2 emission regulation.
---------------------------------------------------------------------------
\3\ Throughout this document and the accompanying FRIA, the
agencies will often use the term ``CO2'' or ``tailpipe
CO2'' to refer broadly to EPA's suite of light duty
vehicle GHG standards.
\4\ 549 U.S. 497, 532 (2007).
---------------------------------------------------------------------------
The CAFE and CO2 emissions standards established by
these final rules will increase in stringency at 1.5 percent per year
from MY 2020 levels over MYs 2021-2026. The ``1.5 percent'' regulatory
alternative is new for the final rule and was not expressly analyzed in
the NPRM, but it is a logical outgrowth of the NPRM analysis, being
well within the range of alternatives then considered and consistent
with discussions by both the agencies and commenters that there are
benefits to having standards that increase at the same rate for all
fleets. These standards apply to light-duty vehicles, which NHTSA
divides for purposes of regulation into passenger cars and light
trucks, and EPA divides into passenger cars, light-duty trucks, and
medium-duty passenger vehicles (i.e., sport utility vehicles, cross-
over utility vehicles, and light trucks). Both the CAFE and
CO2 standards are vehicle-footprint-based, as are the
standards currently in effect. These standards will become more
stringent for each model year from 2021 to 2026, relative to the MY
2020 standards. Generally, the larger the vehicle footprint, the less
numerically stringent the corresponding vehicle CO2 and
miles-per-gallon (mpg) targets. As a result of the footprint-based
standards, the burden of compliance is distributed across all vehicle
footprints and across all manufacturers. Each manufacturer is subject
to individualized standards for passenger cars and light trucks, in
each model year, based on the vehicles it produces. When standards are
carefully crafted, both in terms of the footprint curves and the rate
of increase in stringency of those curves, manufacturers are not
[[Page 24176]]
compelled to build vehicles of any particular size or type.
Knowing that many readers are accustomed to considering CAFE and
CO2 emissions standards in terms of the mpg and grams-per-
mile (g/mi) values that the standards are projected to eventually
require, the agencies include those projections here. EPA's standards
are projected to require, on an average industry fleet-wide basis, 201
grams per mile (g/mi) of CO2 in model year 2030, while
NHTSA's standards are projected to require, on an average industry
fleet-wide basis, 40.5 miles per gallon (mpg) in model year 2030. The
agencies note that real-world CO2 is typically 25 percent
higher and real-world fuel economy is typically 20 percent lower than
the CO2 and CAFE compliance values discussed here, and also
note that a portion of EPA's expected ``CO2'' improvements
will in fact be made through improvements in minimizing air
conditioning leakage and through use of alternative refrigerants, which
will not contribute to fuel economy but will contribute toward
reductions of climate-related emissions.
In these final rules, NHTSA and EPA are reaching similar
conclusions on similar grounds: even though each agency has its own
distinct statutory authority and factors, the relevant considerations
overlap in many ways. Both agencies recognize that they are balancing
the relevant considerations in somewhat different ways from how they
may have been balanced previously, as in the 2012 final rule and in
EPA's Initial Determination, but the current balancing is called for in
light of the facts before the agencies. The balancing in these final
rules is also somewhat different from how the agencies balanced their
respective considerations in the proposal, in part because of updates
to analytical inputs and methodologies, previewed in the NPRM and made
in response to public comments, that collectively resulted in changes
to the analytical outputs. For example, between the notice and final
rule, the agencies updated fuel price projections to somewhat greater
values, updated the analysis fleet to MY 2017, updated estimates of the
efficacy and cost of fuel-saving technologies, revised procedures for
calculating impacts on vehicle sales and scrappage, updated models for
estimating highway safety impacts, updated estimates of highway
congestion costs, and updated estimates of annual mileage accumulation,
holding VMT (before applying the rebound effect) constant between
regulatory alternative. Moreover, the cost-benefit analysis conducted
for these final rules has even been overtaken by events in many ways
over recent weeks. Based upon current events, and for additional
reasons discussed in Section VI.D.1 the benefits of saving additional
fuel through more stringent standards are potentially even smaller than
estimated in this rulemaking analysis.
The standards finalized today fit the pattern of gradual, tough,
but feasible stringency increases that take into account real world
performance, shifts in fuel prices, and changes in consumer behavior
toward crossovers and SUVs and away from more efficient sedans. This
approach ensures that manufacturers are provided with sufficient lead
time to achieve standards, considering the cost of compliance. The
costs to both industry and automotive consumers would have been too
high under the standards set forth in 2012, and by lowering the auto
industry's costs to comply with the program, with a commensurate
reduction in per-vehicle costs to consumers, the standards enhance the
ability of the fleet to turn over to newer, cleaner and safer vehicles.
More stringent standards also have the potential for overly
aggressive penetration rates for advanced technologies relative to the
penetration rates seen in the final standards, especially in the face
of an unknown degree of consumer acceptance of both the increased costs
and of the technologies themselves--particularly given current
projections of relatively low fuel prices during that timeframe. As a
kind of insurance policy against future fuel price volatility,
standards that increase at 1.5 percent per year for cars and trucks
will help to keep fleet fuel economy higher than they would be
otherwise when fuel prices are low, which is not improbable over the
next several years.\5\ At the same time, the standards help to address
these issues by maintaining incentives to promote broader deployment of
advanced technologies, and so provides a means of encouraging their
further penetration while leaving manufacturers alternative technology
choices. Steady, gradual increases in stringency ensure that the
benefits of reduced GHG emissions and fuel consumption are achieved
without the potential for disruption to automakers or consumers.
---------------------------------------------------------------------------
\5\ For example, EIA currently expects U.S. retail gasoline
prices to average $2.14/gallon in 2020, compared to $2.69/gallon in
2019 (see https://www.eia.gov/outlooks/steo/archives/mar20.pdf), and
$3.68/gallon in 2012 (see https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=EMM_EPM0_PTE_NUS_DPG&f=A). While gasoline
prices may foreseeably rise over the rulemaking time frame, it is
also very foreseeable that they will not rise to the $4-5/gallon
that many Americans saw over the 2008-2009 time frame, that caused
the largest shift seen toward smaller and higher-fuel-economy
vehicles. See, e.g., Figure VIII-2 below.
---------------------------------------------------------------------------
Standards that increase at 1.5 percent per year represent a
reasonable balance of additional technology and required per-vehicle
costs, consumer demand for fuel economy, fuel savings and emissions
avoided given the foreseeable state of the global oil market and the
minimal effect on climate between finalizing 1.5 percent standards
versus more stringent standards. The final standards will also result
in year-over-year improvements in fleetwide fuel economy, resulting in
energy conservation that helps address environmental concerns,
including criteria pollutant, air toxic pollutant, and carbon
emissions.
The agencies project that under these final standards, required
technology costs would be reduced by $86 to $126 billion over the
lifetimes of vehicles through MY 2029. Equally important, purchase
prices costs to U.S. consumers for new vehicles would be $977 to $1,083
lower, on average, than they would have been if the agencies had
retained the standards set forth in the 2012 final rule and originally
upheld by EPA in January 2017. While these final standards are
estimated to result in 1.9 to 2.0 additional billion barrels of fuel
consumed and from 867 to 923 additional million metric tons of
CO2 as compared to current estimates of what the standards
set forth in 2012 would require, the agencies explain at length below
why the overall benefits of the final standards outweigh these
additional costs.\6\
---------------------------------------------------------------------------
\6\ 1.9 to 2.0 barrels of fuel is approximately 78 to 84 gallons
of fuel.
---------------------------------------------------------------------------
For the CAFE program, overall (fleetwide) net benefits vary from
$16.1 billion at a 7 percent discount rate to -$13.1 billion at a 3
percent discount rate. For the CO2 program, overall
(fleetwide) societal net benefits vary from $6.4 billion at a 7 percent
discount rate to -$22.0 billion at a 3 percent discount rate. The net
benefits straddle zero, and are very small relative to the scale of
reduced required technology costs, which range from $86.3 billion to
$126.0 billion for the CAFE and CO2 programs across 7
percent and 3 percent discount rates. Likewise, net benefits are very
small relative to the scale of reduced retail fuel savings over the
full life of all vehicles manufactured during the 2021 through 2029
model years, which range from $108.6 billion to $185.1 billion for the
CAFE and CO2 programs across 7 percent and 3 percent
discount rates. Similarly, all of the alternatives have small net
benefits, ranging from $18.4 billion to -$31.1
[[Page 24177]]
billion for the CAFE and CO2 programs across 7 percent and 3
percent discount rates.\7\
---------------------------------------------------------------------------
\7\ See Table II-12 to Table II-15 for costs, benefits and net
benefits.
---------------------------------------------------------------------------
NHTSA and EPA believe their analysis of the final rule represents
the best available science, evidence, and methodologies for assessing
the impacts of changes in CAFE and CO2 emission standards.
In fact, the agencies note that today's analysis represents a marked
improvement over prior rulemakings. Previously, the agencies were
unable to model the impact of the standards on new vehicle sales or the
retirement of older vehicles in the fleet, and, instead, were forced to
assume, contrary to economic theory and empirical evidence, that the
number of new vehicles sold and older vehicles scrapped remained static
across regulatory alternatives. Today's analysis--as commenters to
previous rulemakings and EPA's Science Advisory Board have argued is
necessary \8\--quantifies the sales and scrappage impacts of the
standards, including the associated safety benefits, and represents a
significant step forward in agencies' ability to comprehensively
analyze the impacts of CAFE and CO2 emission standards.
---------------------------------------------------------------------------
\8\ Science Advisory Board, U.S. EPA. Review of EPA's Proposed
SAFE rule at 4 (Feb. 27, 2020), available at https://
yosemite.epa.gov/sab/sabproduct.nsf/LookupWebProjectsCurrentBOARD/
1FACEE5C03725F268525851F006319BB/$File/EPA-SAB-20-003+.pdf
[hereinafter ``SAB Report''].
---------------------------------------------------------------------------
However, the agencies also believe it is important to be
transparent about analytical limitations. For example, EPA's Science
Advisory Board stressed that the agencies account for ``evolving
consumer preferences for performance and other vehicle attributes,''
\9\ yet due to limitations on the agencies' current ability to model
buyers' choices among combinations of various attributes and their
costs, the primary analysis does not account for the consumer benefits
of other vehicle features that may be sacrificed for costly
technologies that improve fuel economy. The agencies' analysis assumes
that under these final standards, attributes of new cars and light
trucks other than fuel economy would remain identical to those under
the baseline standards, so that changes in sales prices and fuel
economy would be the only sources of benefits or costs to new car and
light truck buyers. In other words, the agencies' primary analysis does
not consider that producers will likely respond to buyers' demands by
reallocating some their savings in production costs due to lower
technology costs to add or improve other attributes that consumers
value more highly than the increases in fuel economy the augural
standards would have required. The agencies have long debated whether
and how best to model the consumer benefits of other vehicle
attributes, and note that they have made considerable progress.\10\
However, despite these potential analytical shortcomings, the agencies
reaffirm that today's analysis represents the most complete and
rigorous examination of CAFE and CO2 emission standards to
date, and provide decision-makers a powerful analytical tool--
especially since the limitations are known, do not bias the central
analysis' results, and are afforded due consideration.
---------------------------------------------------------------------------
\9\ SAB at 10.
\10\ In their evaluations of previous CAFE and CO2
rules, the agencies attempted to account for this possibility by
conducting sensitivity analyses that reduced the fuel savings and
other benefits to vehicle buyers by a significant fraction. For
example, NHTSA's analysis supporting the Final Rule establishing
CAFE standards for model year 2012-16 cars and light trucks tested
the sensitivity of their central estimates of social costs and
benefits to the assumptions that 25 percent and 50 percent of
benefits to buyers were offset by opportunity costs of foregone
improvements in attributes other than fuel economy; see NHTSA, Final
Regulatory Impact Analysis: Corporate Average Fuel Economy for Model
year 2012-16 Passenger Cars and Light Trucks, March 2010, at 563-565
and Table X-9, at 566-56; see also, NHTSA, Final Regulatory Impact
Analysis: Corporate Average Fuel Economy for Model year 2017-25
Passenger Cars and Light Trucks, August 2012, at 1087 and Tables X-
18a, X-18b, and X-18c, at 1099-1104. The agencies acknowledged that
this was not a completely satisfactory way to represent the
sacrifices in vehicles' other attributes that car and light truck
manufacturers might find it necessary to make in order to comply
with the increasingly stringent standards those previous rules
established. At the time, however, the agencies were unable to
identify specific attributes that manufacturers were most likely to
sacrifice, measure the tradeoffs between increased fuel economy and
improvements in those attributes, or assess the potential losses in
utility to car and light truck buyers. In an effort to improve on
their previous treatment of this issue, the agencies' evaluation of
this final rule includes a sensitivity case that assumes
manufacturers redirect their technology cost savings from complying
with less stringent standards to instead improve a combination of
cars' and light trucks' other attributes that offers benefits to
their buyers significantly exceeding those costs. The magnitude of
these (net) benefits is interpreted as the opportunity cost of the
improvements in vehicles' other attributes that would have been
sacrificed if the augural standards had been enacted. The method the
agencies use to approximate these benefits, together with its effect
on the rule's overall benefits and costs, is discussed in detail in
Section VI.D.1.b)(8). Briefly, the results of this sensitivity
analysis suggest the Final Rule would generate net benefits for the
CAFE and CO2 programs ranging from $34.9 to $55.4 billion
at 3% and 7% discount rates.
---------------------------------------------------------------------------
In terms of the agencies' respective statutory authorities, EPA is
setting national tailpipe CO2 emissions standards for
passenger cars and light trucks under section 202(a) of the Clean Air
Act (CAA),\11\ and taking other actions under its authority to
establish metrics and measure passenger car and light truck fleet fuel
economy pursuant to the Energy Policy and Conservation Act (EPCA),\12\
while NHTSA is setting national corporate average fuel economy (CAFE)
standards under EPCA, as amended by the Energy Independence and
Security Act (EISA) of 2007.\13\ As summarized above and as discussed
in much greater detail below, the agencies believe that these represent
appropriate levels of CO2 emissions standards and maximum
feasible CAFE standards for MYs 2021-2026, pursuant to their respective
statutory authorities. Sections III and VIII below contain detailed
discussions of both agencies' statutory obligations and authorities.
---------------------------------------------------------------------------
\11\ 42 U.S.C. 7521(a).
\12\ 49 U.S.C. 32904(c).
\13\ 49 U.S.C. 32902.
---------------------------------------------------------------------------
Section 202(a) of the CAA requires EPA to establish standards for
emissions of pollutants from new motor vehicles that cause or
contribute to air pollution that may reasonably be anticipated to
endanger public health or welfare. Standards under section 202(a) thus
take effect only ``after providing such period as the Administrator
finds necessary to permit the development and application of the
requisite technology, giving appropriate consideration to the cost of
compliance within such period.'' \14\ In establishing such standards,
EPA must consider issues of technical feasibility, cost, and available
lead time, among other things.
---------------------------------------------------------------------------
\14\ CAA Sec. 202(a); 42 U.S.C. 7512(a)(2).
---------------------------------------------------------------------------
EPCA, as amended by EISA, contains a number of provisions governing
how NHTSA must set CAFE standards. EPCA requires that the Department of
Transportation establish separate passenger car and light truck
standards \15\ at ``the maximum feasible average fuel economy level
that the Secretary decides the manufacturers can achieve in that model
year,'' \16\ based on the agency's consideration of four statutory
factors: technological feasibility, economic practicability, the effect
of other standards of the Government on fuel economy, and the need of
the United States to conserve energy.\17\ EPCA does not define these
terms or specify what weight to give each concern in balancing them--
such considerations are left within the discretion of the Secretary of
Transportation (delegated to NHTSA) based upon current information.
Accordingly, NHTSA interprets these factors and determines the
appropriate weighting that leads to the maximum
[[Page 24178]]
feasible standards given the circumstances present at the time of
promulgating each CAFE standard rulemaking. While EISA, for MYs 2011-
2020, additionally required that standards increase ``ratably'' and be
set at levels to ensure that the CAFE of the industry-wide combined
fleet of new passenger cars and light trucks reach at least 35 mpg by
MY 2020,\18\ EISA requires that standards for MYs 2021-2030 simply be
set at the maximum feasible level as determined by the Secretary (and
by delegation, NHTSA).\19\
---------------------------------------------------------------------------
\15\ 49 U.S.C. 32902(b)(1).
\16\ 49 U.S.C. 32902(a).
\17\ 49 U.S.C. 32902(f).
\18\ 49 U.S.C. 32902(b)(2)(A) and (C).
\19\ 49 U.S.C. 32902(b)(2)(B).
---------------------------------------------------------------------------
In the NPRM, the agencies sought comment on a variety of possible
changes to existing compliance flexibilities that have been created
over the past several years. The vast majority of the existing
compliance flexibilities are not being changed, but a small number of
flexibilities related to real-world fuel efficiency improvements are
being finalized. In addition, EPA will continue to allow manufacturers
to make improvements relating to air conditioning refrigerants and
leakage and will credit those improvements toward CO2
compliance, and EPA is making no changes in the amounts of credits
available. EPA is also not making any changes to the existing
CH4 and N2O standards. EPA is also extending the
``0 g/mi upstream'' incentive for electric vehicles beyond its current
sunset of MY 2021, through MY 2026. EPA is also establishing a credit
multiplier for natural gas vehicles through the 2026 model year.
Otherwise, compliance flexibilities in the two programs do not change
significantly for the final rule. These changes should help to
streamline manufacturer use of those flexibilities in certain respects.
While manufacturers and suppliers sought a number of other additional
compliance flexibilities, the agencies have concluded that the
aforementioned existing flexibilities are reasonable and appropriate,
and that additional flexibilities are not justified.
Table I-1 and Table I-2 present the total costs, benefits, and net
benefits for the 2021-2026 preferred alternative CAFE and
CO2 levels, relative to the MY 2022-2025 existing/augural
standards (with the MY 2025 standards repeated for MY 2026) and current
MY 2021 standard. The preferred alternative exhibits a stringency rate
increase of 1.5 percent per year for both passenger cars and light
trucks. The values in Table I-1 and Table I-2 display (in total and
annualized forms) costs for all MYs 1978-2029 vehicles, and the
benefits and net benefits represent the impacts of the standards over
the full lifetimes of the vehicles sold or projected to be sold during
model years 1978-2029.
For this analysis, negative signs are used for changes in costs or
benefits that decrease from those that would have resulted from the
existing/augural standards. Any changes that would increase either
costs or benefits are shown as positive changes. Thus, an alternative
that decreases both costs and benefits, will show declines (i.e., a
negative sign) in both categories. From Table I-1 and Table I-2, the
preferred alternative (Alternative 3) is estimated to decrease costs
relative to the baseline by $182 to $280 billion over the lifetime of
MYs 1978-2029 passenger vehicles (range determined by discount rate
across both CAFE and CO2 programs). It will also decrease
benefits from $175 to $294 billion over the life of these MY fleets.
The net impact will be a decrease from $22 billion to an increase of
$16 billion in total net benefits to society over this roughly 52-year
timeframe. Annualized, this amounts to roughly -$0.8 to 1.2 billion in
net benefits per year.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.001
[GRAPHIC] [TIFF OMITTED] TR30AP20.002
[[Page 24179]]
Table I-3 and Table I-4 lists costs, benefits, and net benefits for
all seven alternatives that were examined.
[GRAPHIC] [TIFF OMITTED] TR30AP20.003
[GRAPHIC] [TIFF OMITTED] TR30AP20.004
Table I-5 and Table I-6 show a summary of various impacts of the
preferred alternative for CAFE and CO2 standards. Impacts
are presented in monetized and non-monetized values, as well as from
the perspective of society and the consumer.
[[Page 24180]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.005
[[Page 24181]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.006
BILLING CODE 4910-59-C
The agencies note that the NPRM drew more public comments (and,
particularly, more pages of substantive comments) than any rulemaking
in the history of the CAFE or CO2 tailpipe emissions
programs--exceeding 750,000 comments. The agencies recognized in the
NPRM that the proposal was significantly different from the final rules
set forth in 2012, and explained at length the reasons for those
differences--namely, that new information and considerations, along
with an expanded and updated analysis, had led to different tentative
conclusions. Today's final rules represent a further evolution of the
work that supported the proposal, based on improved quantitative
methodology and in careful consideration of the hundreds of thousands
of public comments and deep reflection on the serious issues before the
agencies. Simply put, the agencies have heard the comments, and today's
analysis and decision reflect the agencies' grappling with the issues
commenters raised, as well as all of the other information before the
agencies. These programs and issues are weighty, and the agencies
believe that a reasonable balance has been struck in these final rules
between the many competing national needs that these regulatory
programs collectively address.
II. Overview of Final Rule
A. Summary of Proposal
In the NPRM, the National Highway Traffic Safety Administration
(NHTSA) and the Environmental Protection Agency (EPA) (collectively,
``the
[[Page 24182]]
agencies'') proposed the ``Safer Affordable Fuel-Efficient (SAFE)
Vehicles Rule for Model Years 2021-2026 Passenger Cars and Light
Trucks'' (SAFE Vehicles Rule). The proposed SAFE Vehicles Rule would
set Corporate Average Fuel Economy (CAFE) and carbon dioxide
(CO2) emissions standards, respectively, for passenger cars
and light trucks manufactured for sale in the United States in model
years (MYs) 2021 through 2026.\20\
---------------------------------------------------------------------------
\20\ NHTSA sets CAFE standards under the Energy Policy and
Conservation Act of 1975 (EPCA), as amended by the Energy
Independence and Security Act of 2007 (EISA). EPA sets
CO2 standards under the Clean Air Act (CAA).
---------------------------------------------------------------------------
The agencies explained that they must act to propose and finalize
these standards and do not have discretion to decline to regulate.
Congress requires NHTSA to set CAFE standards for each model year.\21\
Congress also requires EPA to set emissions standards for light-duty
vehicles if EPA has made an ``endangerment finding'' that the pollutant
in question--in this case, CO2--``cause[s] or contribute[s]
to air pollution which may reasonably be anticipated to endanger public
health or welfare.'' \22\ NHTSA and EPA proposed the standards
concurrently because tailpipe CO2 emissions standards are
directly and inherently related to fuel economy standards,\23\ and, if
finalized, the rules would apply concurrently to the same fleet of
vehicles. By working together to develop the proposals, the agencies
aimed to reduce regulatory burden on industry and improve
administrative efficiency.
---------------------------------------------------------------------------
\21\ 49 U.S.C. 32902.
\22\ 42 U.S.C. 7521; see also 74 FR 66495 (Dec. 15, 2009)
(``Endangerment and Cause or Contribute Findings for Greenhouse
Gases under Section 202(a) of the Clean Air Act'').
\23\ See, e.g., 75 FR 25324, at 25327 (May 7, 2010) (``The
National Program is both needed and possible because the
relationship between improving fuel economy and reducing tailpipe
CO2 emissions is a very direct and close one. The amount
of those CO2 emissions is essentially constant per gallon
combusted of a given type of fuel. Thus, the more fuel efficient a
vehicle is, the less fuel it burns to travel a given distance. The
less fuel it burns, the less CO2 it emits in traveling
that distance. [citation omitted] While there are emission control
technologies that reduce the pollutants (e.g., carbon monoxide)
produced by imperfect combustion of fuel by capturing or converting
them to other compounds, there is no such technology for
CO2. Further, while some of those pollutants can also be
reduced by achieving a more complete combustion of fuel, doing so
only increases the tailpipe emissions of CO2. Thus, there
is a single pool of technologies for addressing these twin problems,
i.e., those that reduce fuel consumption and thereby reduce
CO2 emissions as well.'').
---------------------------------------------------------------------------
The agencies discussed some of the history leading to the proposal,
including the 2012 final rule, the expectations regarding a mid-term
evaluation as required by EPA regulation, and the rapid process over
2016 and early 2017 by which EPA issued its first Final Determination
that the CO2 standards set in 2012 for MYs 2022-2025
remained appropriate based on the information then before the EPA
Administrator.\24\ The agencies also discussed President Trump's
direction in March 2017 to restore the original mid-term evaluation
timeline, and EPA's subsequent information-gathering process and
announcement that it would reconsider the January 2017
Determination.\25\ EPA ultimately concluded that the standards set in
2012 for MYs 2022-2025 were no longer appropriate.\26\ For NHTSA, in
turn, the ``augural'' CAFE standards for MYs 2022-2025 were never
final, and as explained in the 2012 final rule, NHTSA was obligated
from the beginning to undertake a new rulemaking to set CAFE standards
for MYs 2022-2025.
---------------------------------------------------------------------------
\24\ See 83 FR at 42987 (Aug.24, 2018).
\25\ Id.
\26\ 83 FR 16077 (Apr. 2, 2018).
---------------------------------------------------------------------------
The NPRM thus began the rulemaking process for both agencies to
establish new standards for MYs 2022-2025 passenger cars and light
trucks. Standards were concurrently proposed for MY 2026 in order to
provide regulatory stability for as many years as is legally
permissible for both agencies together. The NPRM also included revised
standards for MY 2021 passenger cars and light trucks, because the
agencies tentatively concluded, based on the information and analysis
then before them, that the CAFE standards previously set for MY 2021
were no longer maximum feasible, and the CO2 standards
previously set for MY 2021 were no longer appropriate. Agencies always
have authority under the Administrative Procedure Act to revisit
previous decisions in light of new facts, as long as they provide
notice and an opportunity for comment, and the agencies stated that it
is plainly the best practice to do so when changed circumstances so
warrant.\27\
---------------------------------------------------------------------------
\27\ See FCC v. Fox Television, 556 U.S. 502 (2009).
---------------------------------------------------------------------------
The NPRM proposed to maintain the CAFE and CO2 standards
applicable in MY 2020 for MYs 2021-2026, and took comment on a wide
range of alternatives, including different stringencies and retaining
existing CO2 standards and the augural CAFE standards.\28\
Table II-1, Table II-2, and Table II-3 show the estimates, under the
NPRM analysis, of what the MY 2020 CAFE and CO2 curves would
translate to, in terms of miles per gallon (mpg) and grams per mile (g/
mi), in MYs 2021-2026, as well as the regulatory alternatives
considered in the NPRM. In addition to retaining the MY 2020
CO2 standards through MY 2026, EPA proposed and sought
comment on excluding air conditioning refrigerants and leakage, and
nitrous oxide and methane emissions for compliance with CO2
standards after model year 2020, in order to improve harmonization with
the CAFE program. EPA also sought comment on whether to change existing
methane and nitrous oxide standards that were finalized in the 2012
rule. The proposal was accompanied by a 1,600 page Preliminary
Regulatory Impact Analysis (PRIA) and, for NHTSA, a 500 page Draft
Environmental Impact Statement (DEIS), with more than 800 pages of
appendices and the entire CAFE model, including the software source
code and documentation, all of which were also subject to comment in
their entirety and all of which received significant comments.
---------------------------------------------------------------------------
\28\ The agencies noted that this did not mean that the miles
per gallon and grams per mile levels that were estimated for the MY
2020 fleet in 2012 would be the ``standards'' going forward into MYs
2021-2026. Both NHTSA and EPA set CAFE and CO2 standards,
respectively, as mathematical functions based on vehicle footprint.
These mathematical functions that are the actual standards are
defined as ``curves'' that are separate for passenger cars and light
trucks, under which each vehicle manufacturer's compliance
obligation varies depending on the footprints of the cars and trucks
that it ultimately produces for sale in a given model year. It was
the MY 2020 CAFE and CO2 curves that the agencies
proposed would continue to apply to the passenger car and light
truck fleets for MYs 2021-2026. The mpg and g/mi values which those
curves would eventually require of the fleets in those model years
would be known for certain only at the ends of each of those model
years. While it is convenient to discuss CAFE and CO2
standards as a set ``mpg,'' ``g/mi,'' or ``mpg-e'' number,
attempting to define those values based on the information then
before the agency would necessarily end up being inaccurate.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24183]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.007
[[Page 24184]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.008
---------------------------------------------------------------------------
\29\ The carbon dioxide equivalents of air conditioning
refrigerant leakage, nitrous oxide emissions, and methane emissions
were included for compliance with the EPA standards for all MYs
under the baseline/no action alternative in the NPRM. Carbon dioxide
equivalent is calculated using the Global Warming Potential (GWP) of
each of the emissions.
\30\ Beginning in MY 2021, the proposal provided that the GWP
equivalents of air conditioning refrigerant leakage, nitrous oxide
emissions, and methane emissions would no longer be able to be
included with the tailpipe CO2 for compliance with
tailpipe CO2 standards.
---------------------------------------------------------------------------
[[Page 24185]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.009
BILLING CODE 4910-59-C
The agencies explained in the NPRM that new information had been
gathered and new analysis performed since publication of the 2012 final
rule establishing CAFE and CO2 standards for MYs 2017 and
beyond and since issuance of the 2016 Draft TAR and EPA's 2016 and
early 2017 ``mid-term evaluation'' process. This new information and
analysis helped lead the agencies to the tentative conclusion that
holding standards constant at MY 2020 levels through MY 2026 was
maximum feasible, for CAFE purposes, and appropriate, for
CO2 purposes.
The agencies further explained that technologies had played out
differently in the fleet from what the agencies previously assumed:
That while there remain a wide variety of technologies available to
improve fuel economy and reduce CO2 emissions, it had become
clear that there were reasons to temper previous optimism about the
costs, effectiveness, and consumer acceptance of a number of
technologies. In addition, over the years between the previous analyses
and the NPRM, automakers had added considerable amounts of technologies
to their new vehicle fleets, meaning that the agencies were no longer
free to make certain assumptions about how some of those technologies
could be used going forward. For example, some technologies that could
be used to improve fuel economy and reduce emissions had not been used
entirely for that purpose, and some of the benefit of these
technologies had gone instead toward improving other vehicle
attributes. Other technologies had been tried, and had been met with
significant customer acceptance issues. The agencies underscored the
importance of reflecting the fleet as it stands today, with the
technology it has and as that technology has been used, and considering
what technology remains on the table at this point, whether and when it
can realistically be available for widespread use in production, and
how much it would cost to implement.
The agencies also acknowledged the math of diminishing returns: As
CAFE and CO2 emissions standards increase in stringency, the
benefit of continuing to increase in stringency decreases. In mpg
terms, a vehicle owner who drives a light vehicle 15,000 miles per year
(a typical assumption for analytical purposes) \31\ and trades in a
vehicle with fuel economy of 15 mpg for one with fuel economy of 20
mpg, will reduce their annual fuel consumption from 1,000 gallons to
750 gallons--saving 250 gallons annually. If, however, that owner were
to trade in a vehicle with fuel economy of 30 mpg for one with fuel
economy of 40 mpg, the owner's annual gasoline consumption would drop
from 500 gallons/year to 375 gallons/year--only 125 gallons even though
the mpg improvement is twice as large. Going from 40 to 50 mpg would
save only 75 gallons/year. Yet each additional fuel economy improvement
becomes much more expensive as the easiest to achieve low-cost
technological improvement options are chosen. In CO2 terms,
if a vehicle emits 300 g/mi CO2,
[[Page 24186]]
a 20 percent improvement is 60 g/mi, so the vehicle would emit 240 g/
mi; but if the vehicle emits 180 g/mi, a 20 percent improvement is only
36 g/mi, so the vehicle would get 144 g/mi. In order to continue
achieving similarly large (on an absolute basis) emissions reductions,
the percentage reduction must also continue to increase.
---------------------------------------------------------------------------
\31\ A different vehicle-miles-traveled (VMT) assumption would
change the absolute numbers in the example, but would not change the
mathematical principles.
---------------------------------------------------------------------------
Related, average real-world fuel economy is lower than average fuel
economy required under CAFE and CO2 standards. The 2012
Federal Register notice announcing augural CAFE and CO2
standards extending through MY 2025 indicated that, if met entirely
through the application of fuel-saving technology, the MY 2025
CO2 standards would result in an average requirement
equivalent to 54.5 mpg. However, because the CO2 standards
provide credit for reducing leakage of AC refrigerants and/or switching
to lower-GWP refrigerants, and these actions do not affect fuel
economy, the notice explained that the corresponding fuel economy
requirement (under the CAFE program) would be 49.7 mpg. These estimates
were based on a market forecast grounded in the MY 2008 fleet. The
notice also presented analysis using a market forecast grounded in the
MY 2010 fleet, showing a 48.7 mpg average CAFE requirement.
In the real world, fuel economy is, on average, about 20% lower
than as measured under regulatory test procedures. In the real world,
then, these new standards were estimated to require 39.0-39.8 mpg.
Today's analysis indicates that the requirements under the
baseline/augural CAFE standards would average 46.6 mpg in MY 2029. The
lower value results from changes in the fleet forecast which reflects
consumer preference for larger vehicles than was forecast for the 2012
rulemaking. In the real world, the requirements average about 37.1 mpg.
Under the final standards issued today, the regulatory test procedure
requirements average 40.5 mpg, corresponding to 33.2 mpg in the real
world. Buyers of new vehicles experience real-world fuel economy, with
levels varying among drivers (due to a wide range of factors). Vehicle
fuel economy labels provide average real-world fuel economy information
to buyers.
[GRAPHIC] [TIFF OMITTED] TR30AP20.010
Vehicle owners also face fuel prices at the pump. The agencies
noted in the NPRM that when fuel prices are high, the value of fuel
saved may be enough to offset the cost of further fuel economy/
emissions reduction improvements, but the agencies recognized that
then-current projections of fuel prices by the Energy Information
Administration did not indicate particularly high fuel prices in the
foreseeable future. The agencies explained that fundamental structural
shifts had occurred in global oil markets since the 2012 final rule,
largely due to the rise of U.S. production and export of shale oil. The
consequence over time of diminishing returns from more stringent fuel
economy/emissions reduction standards, especially when combined with
relatively low fuel prices, is greater difficulty for automakers to
find a market of consumers willing to buy vehicles that meet the
increasingly stringent standards. American consumers have long
demonstrated that in times of relatively low fuel prices, fuel economy
is not a top priority for the majority of them, even when highly fuel
efficient vehicle models are available.
The NPRM analysis sought to improve how the agencies captured the
effects of higher new vehicle prices on fleet composition as a whole by
including an improved model for vehicle scrappage rates. As new vehicle
prices increase, consumers tend to continue using older vehicles for
longer, slowing fleet turnover and thus slowing improvements in fleet-
wide fuel economy, reductions in CO2 emissions, reductions
in criteria pollutant emissions, and advances in safety. That aspect of
the analysis was also driven by the agencies' updated estimates of
average per-vehicle cost increases due to
[[Page 24187]]
higher standards, which were several hundred dollars higher than
previously estimated. The agencies cited growing concerns about
affordability and negative equity for many consumers under these
circumstances, as loan amounts grow and loan terms extend.
For all of the above reasons, the agencies proposed to maintain the
MY 2020 fuel economy and CO2 emissions standards for MYs
2021-2026. The agencies explained that they estimated, relative to the
standards for MYs 2021-2026 put forth in 2012, that an additional 0.5
million barrels of oil would be consumed per day (about 2 to 3 percent
of projected U.S. consumption) if that proposal were finalized, but
that they also expected the additional fuel costs to be outweighed by
the cost savings from new vehicle purchases; that more than 12,700 on-
road fatalities and significantly more injuries would be prevented over
the lifetimes of vehicles through MY 2029 as compared to the standards
set forth in the 2012 final rule over the lifetimes of vehicles as more
new and safer vehicles are purchased than the current (and augural)
standards; and that environmental impacts, on net, would be relatively
minor, with criteria and toxic air pollutants not changing noticeably,
and with estimated atmospheric CO2 concentrations increasing
by 0.65 ppm (a 0.08 percent increase), which the agencies estimated
would translate to 0.003 degrees Celsius of additional temperature
increase relative to the standards finalized in 2012.
Under the NPRM analysis, the agencies tentatively concluded that
maintaining the MY 2020 curves for MYs 2021-2026 would save American
auto consumers, the auto industry, and the public a considerable amount
of money as compared to EPA retaining the previously-set CO2
standards and NHTSA finalizing the augural standards. The agencies
explained that this had been identified as the preferred alternative,
in part, because it appeared to maximize net benefits compared to the
other alternatives analyzed, and recognizing the statutory
considerations for both agencies. Relative to the standards issued in
2012, under CAFE standards, the NPRM analysis estimated that costs
would decrease by $502 billion overall at a three-percent discount rate
($335 billion at a seven-percent discount rate) and benefits were
estimated to decrease by $326 billion at a three-percent discount rate
($204 billion at a seven-percent discount rate). Thus, net benefits
were estimated to increase by $176 billion at a three-percent discount
rate and $132 billion at a seven-percent discount rate. The estimated
impacts under CO2 standards were estimated to be similar,
with net benefits estimated to increase by $201 billion at a three-
percent discount rate and $141 billion at a seven-percent discount
rate.
The NPRM also sought comment on a variety of potential changes to
NHTSA's and EPA's compliance programs for CAFE and CO2 as
well as related programs, including questions about automaker requests
for additional flexibilities and agency interest in reducing market-
distorting incentives and improving transparency; and on a proposal to
withdraw California's CAA preemption waiver for its ``Advanced Clean
Car'' regulations, with an accompanying discussion of preemption of
State standards under EPCA.\32\ The agencies sought comment broadly on
all aspects of the proposal.
---------------------------------------------------------------------------
\32\ Agency actions relating to California's CAA waiver and EPCA
preemption have since been finalized, see 84 FR 51310 (Sept. 27,
2019), and will not be discussed in great detail as part of this
final rule.
---------------------------------------------------------------------------
B. Public Participation Opportunities and Summary of Comments
The NPRM was published on NHTSA's and EPA's websites on August 2,
2018, and published in the Federal Register on August 24, 2018,
beginning a 60-day comment period. The agencies subsequently extended
the official comment period for an additional three days, and left the
dockets open for more than a year after the start of the comment
period, considering late comments to the extent practicable. A separate
Federal Register notice also published on August 24, 2018, which
announced the locations, dates, and times of three public hearings to
be held on the proposal: One in Fresno, California, on September 24,
2018; one in Dearborn, Michigan, on September 25, 2018; and one in
Pittsburgh, Pennsylvania, on September 26, 2018. Each hearing started
at 10 a.m. local time; the Fresno hearing ended at 5:10 p.m. and
resulted in a 235 page transcript; the Dearborn hearing ran until 5:26
p.m. and resulted in a 330 page transcript; and the Pittsburgh hearing
ran until 5:06 p.m. and also resulted in a 330 page transcript. Each
hearing also collected several hundred pages of comments from
participants, in addition to the hearing transcripts.
Besides the comments submitted as part of the public hearings,
NHTSA's docket received a total of 173,359 public comments in response
to the proposal as of September 18, 2019, and EPA's docket a total of
618,647 public comments, for an overall total of 792,006. NHTSA also
received several hundred comments on its DEIS to the separate DEIS
docket. While the majority of individual comments were form letters,
the agencies received over 6,000 pages of substantive comments on the
proposal.
Many commenters generally supported the proposal and many
commenters opposed it. Commenters supporting the proposal tended to
cite concerns about the cost of new vehicles, while commenters opposing
the proposal tended to cite concerns about additional fuel expenditures
and the impact on climate change. Many comments addressed the modeling
used for the analysis, and specifically the inclusion, operation, and
results of the sales and scrappage modules that were part of the NPRM's
analysis, while many addressed the NPRM's safety findings and the role
that those findings played in the proposal's justification. Many other
comments addressed California's standards and role in Federal decision-
making; as discussed above, those comments are further summarized and
responded to in the separate Federal Register notice published in
September 2019. Nearly every aspect of the NPRM's analysis and
discussion received some level of comment by at least one commenter.
The comments received, as a whole, were both broad and deep, and the
agencies appreciate the level of engagement of commenters in the public
comment process and the information and opinions provided.
C. Changes in Light of Public Comments and New Information
The agencies made a number of changes to the analysis between the
NPRM and the final rule in response to public comments and new
information that was received in those comments or otherwise became
available to the agencies. While these changes, their rationales, and
their effects are discussed in detail in the sections below, the
following represents a high-level list of some of the most significant
changes:
Some regulatory alternatives were dropped from
consideration, and one was added;
updated analysis fleet, and changes to technologies on
``baseline'' vehicles within the fleet to reflect better their current
properties and improve modeling precision;
no civil penalties assumed to be paid after MY 2020 under
CAFE program;
updates and expansions in accounting for certain over-
compliance
[[Page 24188]]
credits, including early credits earned in EPA's program;
updates and expansions to CAFE Model's technology paths;
updates to inputs defining the range of manufacturer-,
technology-, and product-specific constraints;
updates to allow the model to adopt a more advanced
technology if it is more cost-effective than an earlier technology on
the path;
precision improvements to the modeling of A/C efficiency
and off-cycle credits;
updates to model's ``effective cost'' metric;
extended explicit simulation of technology application
through MY 2050;
expanded presentation of the results to include ``calendar
year'' analysis;
quantifying different types of health impacts from changes
in air pollution, rather than only accounting for such impacts in
aggregate estimates of the social costs of air pollution;
updated costs to 2018 dollars;
updated fuel costs based on the AEO 2019 version of NEMS;
a variety of technology updates in response to comments
and new information;
updated accounting of rebound VMT between regulatory
alternatives;
updated estimates of the macroeconomic cost of petroleum
dependence;
updated response of total new vehicle sales to increases
in fuel efficiency and price; and
updated response of vehicle retirement rates to changes in
new vehicle fuel efficiency and transaction price.
Sections IV and VI below discuss these updates in significant
detail.
D. Final Standards--Stringency
As explained above, the agencies have chosen to set CAFE and
CO2 standards that increase in stringency by 1.5 percent
year over year for MYs 2021-2026. Separately, EPA has decided to retain
the A/C refrigerant and leakage and CH4 and N2O
standards set forth in 2012 for MYs 2021 and beyond, and the stringency
of the CO2 standards in this final rule reflect the
``offset'' also established in 2012 based on assumptions made at that
time about anticipated HFC emissions reductions.
When the agencies state that stringency will increase at 1.5
percent per year, that means that the footprint curves which actually
define the standards for CAFE and CO2 emissions will become
more stringent at 1.5 percent per year. Consistent with Congress's
direction in EISA to set CAFE standards based on a mathematical
formula, which EPA harmonized with for the CO2 emissions
standards, the standard curves are equations, which are slightly
different for CAFE and CO2, and within each program,
slightly different for passenger cars and light trucks. Each program
has a basic equation for a fleet standard, and then values that change
to cause the stringency changes are the coefficients within the
equations. For passenger cars, consistent with prior rulemakings, NHTSA
is defining fuel economy targets as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.011
where:
TARGETFE is the fuel economy target (in mpg) applicable to a
specific vehicle model type with a unique footprint combination,
a is a minimum fuel economy target (in mpg),
b is a maximum fuel economy target (in mpg),
c is the slope (in gallons per mile per square foot, or gpm, per
square foot) of a line relating fuel consumption (the inverse of
fuel economy) to footprint, and
d is an intercept (in gpm) of the same line.
Here, MIN and MAX are functions that take the minimum and maximum
values, respectively, of the set of included values. For example,
MIN[40,35] = 35 and MAX(40, 25) = 40, such that MIN[MAX(40, 25), 35] =
35.
For light trucks, also consistent with prior rulemakings, NHTSA is
defining fuel economy targets as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.012
where:
TARGETFE is the fuel economy target (in mpg) applicable to a
specific vehicle model type with a unique footprint combination,
a, b, c, and d are as for passenger cars, but taking values specific
to light trucks,
e is a second minimum fuel economy target (in mpg),
f is a second maximum fuel economy target (in mpg),
g is the slope (in gpm per square foot) of a second line relating
fuel consumption (the inverse of fuel economy) to footprint, and
h is an intercept (in gpm) of the same second line.
The final CAFE standards (described in terms of their footprint-
based curves) are as follows, with the values for the coefficients
changing over time:
[[Page 24189]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.013
These equations are presented graphically below, where the x-axis
represents vehicle footprint and the y-axis represents fuel economy,
showing that in the CAFE context, targets are higher (fuel economy) for
smaller footprint vehicles and lower for larger footprint vehicles:
BILLING CODE 4910-59-C
[[Page 24190]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.014
BILLING CODE 4910-59-P
[[Page 24191]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.015
BILLING CODE 4910-59-C
EPCA, as amended by EISA, requires that any manufacturer's
domestically-manufactured passenger car fleet must meet the greater of
either 27.5 mpg on average, or 92 percent of the average fuel economy
projected by the Secretary for the combined domestic and non-domestic
passenger automobile fleets manufactured for sale in the U.S. by all
manufacturers in the model year, which projection shall be published in
the Federal Register when the standard for that model year is
promulgated in accordance with 49 U.S.C. 32902(b).\33\ Any time NHTSA
establishes or changes a passenger car standard for a model year, the
MDPCS for that model year must also be evaluated or re-evaluated and
established accordingly. Thus, this final rule establishes the
applicable MDPCS for MYs 2021-2026. Table II-8 lists the minimum
domestic passenger car standards.
---------------------------------------------------------------------------
\33\ 49 U.S.C. 32902(b)(4).
[GRAPHIC] [TIFF OMITTED] TR30AP20.016
EPA CO2 standards are as follows. Rather than expressing
these standards as linear functions with accompanying minima and
maxima, similar to the approach NHTSA has followed since 2005 in
specifying attribute-based standards, the following tables specify flat
standards that apply below and above specified footprints, and a linear
function that applies between those footprints. The two approaches are
mathematically identical. For passenger cars with a footprint of less
than or equal to 41 square feet, the gram/mile CO2 target
value is selected for the appropriate model year from Table II-9:
[[Page 24192]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.017
For passenger cars with a footprint of greater than 56 square feet,
the gram/mile CO2 target value is selected for the
appropriate model year from Table II-10:
[GRAPHIC] [TIFF OMITTED] TR30AP20.018
For passenger cars with a footprint that is greater than 41 square
feet and less than or equal to 56 square feet, the gram/mile
CO2 target value is calculated using the following equation
and rounded to the nearest 0.1 grams/mile.
[[Page 24193]]
Target CO2 = [a x f] + b
Where f is the vehicle footprint and a and b are selected from Table
II-11 for the appropriate model year:
[GRAPHIC] [TIFF OMITTED] TR30AP20.019
For light trucks with a footprint of less than or equal to 41
square feet, the gram/mile CO2 target value is selected for
the appropriate model year from Table II-12:
[[Page 24194]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.020
For light trucks with a footprint greater than the minimum value
specified in the table below for each model year, the gram/mile
CO2 target value is selected for the appropriate model year
from Table II-13:
[GRAPHIC] [TIFF OMITTED] TR30AP20.021
[[Page 24195]]
For light trucks with a footprint that is greater than 41 square
feet and less than or equal to the maximum footprint value specified in
Table II-14 below for each model year, the gram/mile CO2
target value is calculated using the following equation and rounded to
the nearest 0.1 grams/mile.
Target CO2 = (a x f) + b
Where f is the footprint and a and b are selected from Table II-14
below for the appropriate model year:
[GRAPHIC] [TIFF OMITTED] TR30AP20.022
These equations are presented graphically below, where the x-axis
represents vehicle footprint and the y-axis represents the
CO2 target. The targets are lower for smaller footprint
vehicles and higher for larger footprint vehicles:
BILLING CODE 4910-59-P
[[Page 24196]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.023
[[Page 24197]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.024
BILLING CODE 4910-59-C
Except that EPA elected to apply a slightly different slope when
defining passenger car targets, CO2 targets may be expressed
as direct conversion of fuel economy targets, as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.025
where 8887 g/gal relates grams of CO2 emitted to gallons
of fuel consumed, and OFFSET reflects the fact that that HFC
emissions from lower-GWP A/C refrigerants and less leak-prone A/C
systems are counted toward average CO2 emissions, but
EPCA provides no basis to count reduced HFC emissions toward CAFE
levels.
For the reader's benefit, Table II-15, Table II-16, and Table II-17
show the estimates, under the final rule analysis, of what the MYs
2021-2026 CAFE and CO2 curves would translate to, in terms
of miles per gallon (mpg) and grams per mile (g/mi).
BILLING CODE 4910-59-P
[[Page 24198]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.026
BILLING CODE 4910-59-C
[[Page 24199]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.027
As the following tables demonstrate, averages of manufacturers'
estimated requirements are more stringent (i.e., for CAFE, higher, and
for CO2, lower) under the final standards than under the
proposed standards:
[GRAPHIC] [TIFF OMITTED] TR30AP20.028
[[Page 24200]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.029
E. Final Standards--Impacts
This section summarizes the estimated costs and benefits of the MYs
2021-2026 CAFE and CO2 emissions standards for passenger
cars and light trucks, as compared to the regulatory alternatives
considered. These estimates helped inform the agencies' choices among
the regulatory alternatives considered and provide further confirmation
that the final standards are maximum feasible, for NHTSA, and
appropriate, for EPA. The costs and benefits estimated to result from
the CAFE standards are presented first, followed by those estimated to
result from the CO2 standards. For several reasons, the
estimates for costs and benefits presented for the different programs,
while consistent, are not identical. NHTSA's and EPA's standards are
projected to result in slightly different fuel efficiency improvements.
EPA's CO2 standard is nominally more stringent in part due
to its assumptions about manufacturers' use of air conditioning
leakage/refrigerant replacement credits, which are expected to result
in reduced emissions of HFCs. NHTSA's final standards are based solely
on assumptions about fuel economy improvements, and do not account for
emissions reductions that do not relate to fuel economy. In addition,
the CAFE and CO2 programs offer somewhat different program
flexibilities and provisions, primarily because NHTSA is statutorily
prohibited from considering some flexibilities when establishing CAFE
standards, while EPA is not.\34\ The analysis underlying this final
rule reflects many of those additional EPA flexibilities, which
contributes to differences in how the agencies estimate manufacturers
could comply with the respective sets of standards, which in turn
contributes to differences in estimated impacts of the standards. These
differences in compliance flexibilities are discussed in more detail in
Section IX below.
---------------------------------------------------------------------------
\34\ See 49 U.S.C. 32902(h); CAA Sec. 202(a).
---------------------------------------------------------------------------
Table II-20 to Table II-23 present all subcategories of costs and
benefits of this final rule for all seven alternatives proposed. Costs
include application of fuel economy technology to new vehicles,
consumer surplus, crash costs due to changes in VMT, as well as, noise
and congestion. Benefits include fuel savings, consumer surplus,
refueling time, and clean air.
BILLING CODE 4910-59-P
[[Page 24201]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.030
[[Page 24202]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.031
[[Page 24203]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.032
[[Page 24204]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.033
[[Page 24205]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.034
[[Page 24206]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.035
[[Page 24207]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.036
[[Page 24208]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.037
[[Page 24209]]
BILLING CODE 4910-59-C
F. Other Programmatic Elements
1. Compliance and Flexibilities
Automakers seeking to comply with the CAFE and CO2
standards are generally expected to add fuel economy-improving
technologies to their new vehicles to boost their overall fleet fuel
economy levels. Readers will remember that improving fuel economy
directly reduces CO2 emissions, because CO2 is a
natural and inevitable byproduct of fossil fuel combustion to power
vehicles. The CAFE and CO2 programs contain a variety of
compliance provisions and flexibilities to accommodate better
automakers' production cycles, to reward real-world fuel economy
improvements that cannot be reflected in the 1975-developed test
procedures, and to incentivize the production of certain types of
vehicles. While the agencies sought comment on a broad variety of
changes and potential expansions of the programs' compliance
flexibilities in the NPRM, the agencies determined, after considering
the comments, to make a few changes to the flexibilities proposed in
the NPRM in this final rule. The most noteworthy change is the
retention, in the CO2 program, of the flexibilities that
allow automakers to continue to use HFC reductions toward their
CO2 compliance, and that extend the ``0 grams/mile''
assumption for electric vehicles through MY 2026 (i.e., recognizing
only the tailpipe emissions of full battery-electric vehicles and not
recognizing the upstream emissions caused by the electricity usage of
those vehicles). In the NPRM, EPA had proposed to remove and sought
comment on removing those flexibilities from the CO2
program, but determined not to remove them in this final rule. EPA and
NHTSA are also removing from the programs, starting in MY 2022, the
credit/FCIV for full-size pickup trucks that are either hybrids or
over-performing by a certain amount relative to their targets, and
allowing technology suppliers to begin the petition process for off-
cycle credits/adjustments.
Table II-24, Table II-25, Table II-26, and Table II-27 provide a
summary of the various compliance provisions in the two programs; their
authorities; and any changes included as part of this final rule:
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.038
[[Page 24210]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.039
BILLING CODE 4910-59-P
[[Page 24211]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.040
---------------------------------------------------------------------------
\35\ The CAFE program uses an energy efficiency metric and
standards that are expressed in miles per gallon. For PHEVs and
BEVs, to determine gasoline the equivalent fuel economy for
operation on electricity, a Petroleum Equivalency Factor (PEF) is
applied to the measured electrical consumption. The PEF for
electricity was established by the Department of Energy, as required
by statute, and includes an accounting for upstream energy
associated with the production and distribution for electricity
relative to gasoline. Therefore, the CAFE program includes upstream
accounting based on the metric that is consistent with the fuel
economy metric. The PEF for electricity also includes an incentive
that effectively counts only 15 percent of the electrical energy
consumed.
---------------------------------------------------------------------------
Providing a technology neutral basis by which manufacturers meet
fuel economy and CO2 emissions standards encourages an
efficient and level playing field. The agencies continue to have a
desire to minimize incentives that disproportionately favor one
technology over another. Some of this may involve regulations
established by other Federal agencies. In the near future, NHTSA and
EPA intend to work with other relevant Federal agencies to pursue
regulatory means by which we can further ensure technology neutrality
in this field.
2. Preemption/Waiver
As discussed above, the issues of Clean Air Act waivers of
preemption under Section 209 and EPCA/EISA preemption under 49 U.S.C.
32919 are not addressed in today's final rule, as
[[Page 24212]]
they were the subject of a separate final rulemaking action by the
agencies in September 2019. While many comments were received in
response to the NPRM discussion of those issues, those comments have
been addressed and responded to as part of that separate rulemaking
action.
III. Purpose of the Rule
The Administrative Procedure Act (APA) requires agencies to
incorporate in their final rules a ``concise general statement of their
basis and purpose.'' \36\ While the entire preamble document represents
the agencies' overall explanation of the basis and purpose for this
regulatory action, this section within the preamble is intended as a
direct response to that APA (and related CAA) requirements. Executive
Order 12866 further states that ``Federal agencies should promulgate
only such regulations as are required by law, are necessary to
interpret the law, or are made necessary by compelling public need,
such as material failures of private markets to protect or improve the
health and safety of the public, the environment, or the well-being of
the American people.'' \37\ Section III.C of the FRIA accompanying this
rulemaking discusses at greater length the question of whether a market
failure exists that these final rules may address.
---------------------------------------------------------------------------
\36\ 5 U.S.C. 553(c); see also Clean Air Act section
307(d)(6)(A), 42 U.S.C. 7607(d)(6)(A).
\37\ E.O. 12866, Section 1(a).
---------------------------------------------------------------------------
NHTSA and EPA are legally obligated to set CAFE and GHG standards,
respectively, and do not have the authority to decline to regulate.\38\
The agencies are issuing these final rules to fulfill their respective
statutory obligations to provide maximum feasible fuel economy
standards and limit emissions of pollutants from new motor vehicles
which have been found to endanger public health and welfare (in this
case, specifically carbon dioxide (CO2); EPA has already set
standards for methane (CH4), nitrous oxide (N2O),
and hydrofluorocarbons (HFCs) and is not revising them in this rule).
Continued progress in meeting these statutory obligations is both
legally necessary and good for America--greater energy security and
reduced emissions protect the American public. The final standards
continue that progress, albeit at a slower rate than the standards
finalized in 2012.
---------------------------------------------------------------------------
\38\ For CAFE, see 49 U.S.C. 32902; for CO2, see 42
U.S.C. 7521(a).
---------------------------------------------------------------------------
National annual gasoline consumption and CO2 emissions
currently total about 140 billion gallons and 5,300 million metric
tons, respectively. The majority of this gasoline (about 130 billion
gallons) is used to fuel passenger cars and light trucks, such as will
be covered by the CAFE and CO2 standards issued today.
Accounting for both tailpipe emissions and emissions from ``upstream''
processes (e.g., domestic refining) involved in producing and
delivering fuel, passenger cars and light trucks account for about
1,500 million metric tons (mmt) of current annual CO2
emissions. The agencies estimate that under the standards issued in
2012, passenger car and light truck annual gasoline consumption would
steadily decline, reaching about 80 billion gallons by 2050. The
agencies further estimate that, because of this decrease in gasoline
consumption under the standards issued in 2012, passenger car and light
truck annual CO2 emissions would also steadily decline,
reaching about 1,000 mmt by 2050. Under the standards issued today, the
agencies estimate that, instead of declining from about 140 billion
gallons annually today to about 80 billion gallons annually in 2050,
passenger car and light truck gasoline consumption would decline to
about 95 billion gallons. The agencies correspondingly estimate that
instead of declining from about 1,500 mmt annually today to about 1,000
mmt annually in 2050, passenger car and light truck CO2
emissions would decline to about 1,100 mmt. In short, the agencies
estimate that under the standards issued today, annual passenger car
and light truck gasoline consumption and CO2 emissions will
continue to steadily decline over the next three decades, even if not
quite as rapidly as under the previously-issued standards.
The agencies also estimate that these impacts on passenger car and
light truck gasoline consumption and CO2 emissions will be
accompanied by a range of other energy- and climate-related impacts,
such as reduced electricity consumption (because today's standards
reduce the estimated rate at which the market might shift toward
electric vehicles) and increased CH4 and N2O
emissions. These estimated impacts, discussed below and in the FEIS
accompanying today's notice, are dwarfed by estimated impacts on
gasoline consumption and CO2 emissions.
As explained above, these final rules set or amend fuel economy and
carbon dioxide standards for model years 2021-2026. Many commenters
argued that it was not appropriate to amend previously-established
CO2 and CAFE standards, generally because those commenters
believed that the administrative record established for the 2012 final
rule and EPA's January 2017 Final Determination was superior to the
record that informed the NPRM, and that that prior record led
necessarily to the policy conclusion that the previously-established
standards should remain in place.\39\ Some commenters similarly argued
that EPA's Revised Final Determination--which, for EPA, preceded this
regulatory action--was invalid because, they allege, it did not follow
the procedures established for the mid-term evaluation that EPA
codified into regulation,\40\ and also because the Revised Final
Determination was not based on the prior record.\41\
---------------------------------------------------------------------------
\39\ Comments arguing that the prior record was superior to the
current record, and thus a better basis for decision-making, will be
addressed throughout the balance of this preamble.
\40\ 40 CFR 86.1818-12(h).
\41\ See, e.g., comments from the States and Cities, Attachment
1, Docket No. NHTSA-2018-0067-11735, at 40-42; CARB, Detailed
Comments, Docket No. NHTSA-2018-0067-11873, at 71-72; CBD et. al,
Appendix A, Docket No. NHTSA-2018-0067-12000, at 214-228.
---------------------------------------------------------------------------
The agencies considered a range of alternatives in the proposal,
including the baseline/no action alternative of retaining the existing
EPA carbon dioxide standards. As the agencies explained in the
proposal, the proposal was entirely de novo, based on an entirely new
analysis reflecting the best and most up-to-date information available
to the agencies.\42\ This rulemaking action is separate and distinct
from EPA's Revised Final Determination, which itself was neither a
proposed nor a final decision that the standards ``must'' be revised.
EPA retained full discretion in this rulemaking to revise the standards
or not revise them. In any event, the case law is clear that agencies
are free to reconsider their prior decisions.\43\ With that legal
principle in mind, the agencies agree with commenters that the amended
(and new) CO2 and CAFE standards must be consistent with the
[[Page 24213]]
CAA and EPCA/EISA, respectively, and this preamble and the accompanying
FRIA explain in detail why the agencies believe they are consistent.
The section below discusses briefly the authority given to the agencies
by their respective governing statutes, and the factors that Congress
directed the agencies to consider as they exercise that authority in
pursuit of fulfilling their statutory obligations.
---------------------------------------------------------------------------
\42\ 83 FR 42968, 42987 (Aug. 24, 2018).
\43\ See, e.g., Encino Motorcars, LLC v. Navarro, 136 S. Ct.
2117, 2125 (2016) (``Agencies are free to change their existing
policies as long as they provide a reasoned explanation for the
change.''); FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009) (When an agency changes its existing position, it ``need not
always provide a more detailed justification than what would suffice
for a new policy created on a blank slate. Sometimes it must--when,
for example, its new policy rests on factual findings that
contradict those which underlay its prior policy; or when its prior
policy has engendered serious reliance interests that must be taken
into account . . . . In such cases it is not that further
justification is demanded by the mere fact of policy change, but
that a reasoned explanation is needed for disregarding facts and
circumstances that underlay or were engendered by the prior
policy.'')
---------------------------------------------------------------------------
A. EPA's Statutory Requirements
EPA is setting national CO2 standards for passenger cars
and light trucks under Section 202(a) of the Clean Air Act (CAA).\44\
Section 202(a) of the CAA requires EPA to establish standards for
emissions of pollutants from new motor vehicles which cause or
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare.\45\ In establishing such standards,
EPA considers issues of technical feasibility, cost, available lead
time, and other factors. Standards under section 202(a) thus take
effect only ``after providing such period as the Administrator finds
necessary to permit the development and application of the requisite
technology, giving appropriate consideration to the cost of compliance
within such period.'' \46\ EPA's statutory requirements are further
discussed in Section VIII.A.
---------------------------------------------------------------------------
\44\ 42 U.S.C. 7521(a).
\45\ See Coalition for Responsible Regulation v. EPA, 684 F.3d
102, 114-115 (D.C. Cir. 2012) (`` `If EPA makes a finding of
endangerment, the Clean Air Act requires the [a]gency to regulate
emissions of the deleterious pollutant from new motor vehicles . . .
. Given the non-discretionary duty in Section 202(a)(1) and the
limited flexibility available under Section 202(a)(2), which this
court has held related only to the motor vehicle industry, . . . EPA
had no statutory basis on which it could ground [any] reasons for
further inaction' '') (quoting Massachusetts v. EPA, 549 U.S. 497,
533-35 (2007).
\46\ 42 U.S.C. 7521(a)(2).
---------------------------------------------------------------------------
B. NHTSA's Statutory Requirements
NHTSA is setting national Corporate Average Fuel Economy (CAFE)
standards for passenger cars and light trucks for each model year as
required under EPCA, as amended by EISA.\47\ EPCA mandates a motor
vehicle fuel economy regulatory program that balances statutory factors
in setting minimum fuel economy standards to facilitate energy
conservation. EPCA allocates the responsibility for implementing the
program between NHTSA and EPA as follows: NHTSA sets CAFE standards for
passenger cars and light trucks; EPA establishes the procedures for
testing, tests vehicles, collects and analyzes manufacturers' data, and
calculates the individual and average fuel economy of each
manufacturer's passenger cars and light trucks; and NHTSA enforces the
standards based on EPA's calculations.
---------------------------------------------------------------------------
\47\ EPCA and EISA direct the Secretary of Transportation to
develop, implement, and enforce fuel economy standards (see 49
U.S.C. 32901 et. seq.), which authority the Secretary has delegated
to NHTSA at 49 CFR 1.94(c).
---------------------------------------------------------------------------
The following sections enumerate specific statutory requirements
for NHTSA in setting CAFE standards and NHTSA's interpretations of
them, where applicable. Many comments were received on these
requirements and interpretations. Because this is intended as an
overview section, those comments will be addressed below in Section
VIII rather than here, and the agencies refer readers to that part of
the document for more information.
For each future model year, EPCA (as amended by EISA) requires that
DOT (by delegation, NHTSA) establish separate passenger car and light
truck standards at ``the maximum feasible average fuel economy level
that the Secretary decides the manufacturers can achieve in that model
year,'' \48\ based on the agency's consideration of four statutory
factors: ``technological feasibility, economic practicability, the
effect of other motor vehicle standards of the Government on fuel
economy, and the need of the United States to conserve energy.'' \49\
The law also allows NHTSA to amend standards that are already in place,
as long as doing so meets these requirements.\50\ EPCA does not define
these terms or specify what weight to give each concern in balancing
them; thus, NHTSA defines them and determines the appropriate weighting
that leads to the maximum feasible standards given the circumstances in
each CAFE standard rulemaking.\51\
---------------------------------------------------------------------------
\48\ 49 U.S.C. 32902(a) and (b).
\49\ 49 U.S.C. 32902(f).
\50\ 49 U.S.C. 32902(g).
\51\ See Center for Biological Diversity v. NHTSA, 538 F.3d
1172, 1195 (9th Cir. 2008) (hereafter ``CBD v. NHTSA'') (``The EPCA
clearly requires the agency to consider these four factors, but it
gives NHTSA discretion to decide how to balance the statutory
factors--as long as NHTSA's balancing does not undermine the
fundamental purpose of the EPCA: Energy conservation.'')
---------------------------------------------------------------------------
EISA added several other requirements to the setting of separate
passenger car and light truck standards. Standards must be ``based on 1
or more vehicle attributes related to fuel economy and express[ed] . .
. in the form of a mathematical function.'' \52\ New standards must
also be set at least 18 months before the model year in question, as
would amendments to increase standards previously set.\53\ NHTSA must
regulations prescribing average fuel economy standards for at least 1,
but not more than 5, model years at a time.\54\ A number of comments
addressed these requirements; for the reader's reference, those
comments will be summarized and responded to in Section VIII. EISA also
added the requirement that NHTSA set a minimum standard for
domestically-manufactured passenger cars,\55\ which will also be
discussed further in Section VIII below.
---------------------------------------------------------------------------
\52\ 49 U.S.C. 32902(b)(3)(A).
\53\ 49 U.S.C. 32902(a), (g)(2).
\54\ 49 U.S.C. 39202(b)(3)(B).
\55\ 49 U.S.C. 32902(b)(4).
---------------------------------------------------------------------------
For MYs 2011-2020, EISA further required that the separate
standards for passenger cars and for light trucks be set at levels high
enough to ensure that the achieved average fuel economy for the entire
industry-wide combined fleet of new passenger cars and light trucks
reach at least 35 mpg not later than MY 2020, and standards for those
years were also required to ``increase ratably.'' \56\ For model years
after 2020, standards must be set at the maximum feasible level.\57\
---------------------------------------------------------------------------
\56\ 49 U.S.C. 32902(b)(2)(A) and (C). NHTSA has CAFE standards
in place that are projected to result in industry-achieved fuel
economy levels over 35 mpg in MY 2020. EPA typically provides
verified final CAFE data from manufacturers to NHTSA several months
or longer after the close of the MY in question, so the actual MY
2020 fuel economy will not be known until well after MY 2020 has
ended. The standards for all MYs up to and including 2020 are known
and not at issue in this regulatory action, so these provisions are
noted for completeness rather than immediate relevance to this final
rule. Because neither of these requirements apply after MY 2020,
they are not relevant to this rulemaking and will not be discussed
further.
\57\ 49 U.S.C. 32902(b)(2)(B).
---------------------------------------------------------------------------
1. Factors That Must Be Considered in Deciding What Levels of CAFE
Standards are ``Maximum Feasible''
(a) Technological Feasibility
``Technological feasibility'' refers to whether a particular method
of improving fuel economy can be available for commercial application
in the model year for which a standard is being established. Thus, in
determining the level of new standards, the agency is not limited to
technology that is already being commercially applied at the time of
the rulemaking. For this rulemaking, NHTSA has evaluated and considered
all types of technologies that improve real-world fuel economy,
although not every possible technology was expressly included in the
analysis, as discussed in Section VI and also in Section VIII.
(b) Economic Practicability
``Economic practicability'' refers to whether a standard is one
``within the
[[Page 24214]]
financial capability of the industry, but not so stringent as to'' lead
to ``adverse economic consequences, such as a significant loss of jobs
or the unreasonable elimination of consumer choice.'' \58\ The agency
has explained in the past that this factor can be especially important
during rulemakings in which the automobile industry is facing
significantly adverse economic conditions (with corresponding risks to
jobs). Economic practicability is a broad factor that includes
considerations of the uncertainty surrounding future market conditions
and consumer demand for fuel economy in addition to other vehicle
attributes.\59\ In an attempt to evaluate the economic practicability
of different future levels of CAFE standards (i.e., the regulatory
alternatives considered in this rulemaking), NHTSA considers a variety
of factors, including the annual rate at which manufacturers can
increase the percentage of their fleet(s) that employ a particular type
of fuel-saving technology, the specific fleet mixes of different
manufacturers, assumptions about the cost of the standards to
consumers, and consumers' valuation of fuel economy, among other
things, including, in part, safety.
---------------------------------------------------------------------------
\58\ 67 FR 77015, 77021 (Dec. 16, 2002).
\59\ See, e.g., Center for Auto Safety v. NHTSA (``CAS''), 793
F.2d 1322 (D.C. Cir. 1986) (Administrator's consideration of market
demand as component of economic practicability found to be
reasonable); Public Citizen v. NHTSA, 848 F.2d 256 (D.C. Cir. 1988)
(Congress established broad guidelines in the fuel economy statute;
agency's decision to set lower standard was a reasonable
accommodation of conflicting policies).
---------------------------------------------------------------------------
It is important to note, however, that the law does not preclude a
CAFE standard that poses considerable challenges to any individual
manufacturer. The Conference Report for EPCA, as enacted in 1975, makes
clear, and the case law affirms, ``a determination of maximum feasible
average fuel economy should not be keyed to the single manufacturer
which might have the most difficulty achieving a given level of average
fuel economy.'' \60\ Instead, NHTSA is compelled ``to weigh the
benefits to the nation of a higher fuel economy standard against the
difficulties of individual automobile manufacturers.'' \61\
Accordingly, while the law permits NHTSA to set CAFE standards that
exceed the projected capability of a particular manufacturer as long as
the standard is economically practicable for the industry as a whole,
the agency cannot simply disregard that impact on individual
manufacturers.\62\ That said, in setting fuel economy standards, NHTSA
does not seek to maintain competitive positions among the industry
players, and notes that while a particular CAFE standard may pose
difficulties for one manufacturer as being too high or too low, it may
also present opportunities for another. NHTSA has long held that the
CAFE program is not necessarily intended to maintain the competitive
positioning of each particular company. Rather, it is intended to
enhance the fuel economy of the vehicle fleet on American roads, while
protecting motor vehicle safety and paying close attention to the
economic risks.
---------------------------------------------------------------------------
\60\ Center for Auto Safety v. NHTSA (``CAS''), 793 F.2d 1322,
1352 (D.C. Cir. 1986).
\61\ Id.
\62\ Id. (``. . . the Secretary must weigh the benefits to the
nation of a higher average fuel economy standard against the
difficulties of individual automobile manufacturers.'')
---------------------------------------------------------------------------
(c) The Effect of Other Motor Vehicle Standards of the Government on
Fuel Economy
``The effect of other motor vehicle standards of the Government on
fuel economy'' involves an analysis of the effects of compliance with
emission, safety, noise, or damageability standards on fuel economy
capability and thus on average fuel economy. In many past CAFE
rulemakings, NHTSA has said that it considers the adverse effects of
other motor vehicle standards on fuel economy. It said so because, from
the CAFE program's earliest years,\63\ the effects of such compliance
on fuel economy capability over the history of the program have been
negative ones. For example, safety standards that have the effect of
increasing vehicle weight lower vehicle fuel economy capability and
thus decrease the level of average fuel economy that the agency can
determine to be feasible. NHTSA has considered the additional weight
that it estimates would be added in response to new safety standards
during the rulemaking timeframe. NHTSA has also accounted for EPA's
``Tier 3'' standards for criteria pollutants in its estimates of
technology effectiveness.\64\
---------------------------------------------------------------------------
\63\ 42 FR 63184, 63188 (Dec. 15, 1977). See also 42 FR 33534,
33537 (Jun. 30, 1977).
\64\ See Section VI, below.
---------------------------------------------------------------------------
The NPRM also discussed how EPA's CO2 standards for
light-duty vehicles and California's Advanced Clean Cars program fit
into NHTSA's consideration of ``the effect of other motor vehicle
standards of the Government on fuel economy.'' The agencies note that
on September 19, 2019, to ensure One National Program for automobile
fuel economy and carbon dioxide emissions standards, the agencies
finalized regulatory text related to preemption of State tailpipe
CO2 standards and Zero Emission Vehicle (ZEV) mandates under
EPCA and partial withdrawal of a waiver previously provided to
California under the Clean Air Act.\65\ This final rule's impact on
State programs--including California's--will therefore be somewhat
different from the NPRM's consideration. In the interest of brevity,
this preamble will hold further discussion of that point, along with
responses to comments received, until Section VIII.
---------------------------------------------------------------------------
\65\ 84 FR 51310 (Sept. 27, 2019).
---------------------------------------------------------------------------
(d) The Need of the United States To Conserve Energy
``The need of the United States to conserve energy'' means ``the
consumer cost, national balance of payments, environmental, and foreign
policy implications of our need for large quantities of petroleum,
especially imported petroleum.'' \66\ Environmental implications
principally include changes in emissions of carbon dioxide and criteria
pollutants and air toxics. Prime examples of foreign policy
implications are energy independence and security concerns.
---------------------------------------------------------------------------
\66\ 42 FR 63184, 63188 (1977).
---------------------------------------------------------------------------
(1) Consumer Costs and Fuel Prices
Fuel for vehicles costs money for vehicle owners and operators. All
else equal (and this is an important qualification), consumers benefit
from vehicles that need less fuel to perform the same amount of work.
Future fuel prices are a critical input into the economic analysis of
potential CAFE standards because they determine the value of fuel
savings both to new vehicle buyers and to society, the amount of fuel
economy that the new vehicle market is likely to demand in the absence
of new standards, and they inform NHTSA about the consumer cost of the
nation's need for large quantities of petroleum. In this final rule,
NHTSA's analysis relies on fuel price projections estimated using the
version of NEMS used for the U.S. Energy Information Administration's
(EIA) Annual Energy Outlook for 2019.\67\ Federal government agencies
generally use EIA's price projections in their assessment of future
energy-related policies.
---------------------------------------------------------------------------
\67\ The analysis for the proposal relied on fuel price
projections from AEO 2017; the difference in the projections is
discussed in Section VI.
---------------------------------------------------------------------------
(2) National Balance of Payments
Historically, the need of the United States to conserve energy has
included consideration of the ``national balance of payments'' because
of concerns that importing large amounts of oil created a
[[Page 24215]]
significant wealth transfer to oil-exporting countries and left the
U.S. economically vulnerable.\68\ As recently as 2009, nearly half of
the U.S. trade deficit was driven by petroleum,\69\ yet this concern
has largely lain fallow in more recent CAFE actions, in part because
other factors besides petroleum consumption have since played a bigger
role in the U.S. trade deficit.\70\ Given significant recent increases
in U.S. oil production and corresponding decreases in oil imports, this
concern seems likely to remain fallow for the foreseeable future.\71\
Increasingly, changes in the price of fuel have come to represent
transfers between domestic consumers of fuel and domestic producers of
petroleum rather than gains or losses to foreign entities.
---------------------------------------------------------------------------
\68\ See, e.g., 42 FR 63184, 63192 (Dec. 15, 1977) (``A major
reason for this need [to reduce petroleum consumption] is that the
importation of large quantities of petroleum creates serious balance
of payments and foreign policy problems. The United States currently
spends approximately $45 billion annually for imported petroleum.
But for this large expenditure, the current large U.S. trade deficit
would be a surplus.'')
\69\ See ``Today in Energy: Recent improvements in petroleum
trade balance mitigate U.S. trade deficit,'' U.S. Energy Information
Administration (Jul. 21, 2014), available at https://www.eia.gov/todayinenergy/detail.php?id=17191.
\70\ See, e.g., Nida [Ccedil]akir Melek and Jun Nie, ``What
Could Resurging U.S. Energy Production Mean for the U.S. Trade
Deficit,'' Mar. 7, 2018, Federal Reserve Bank of Kansas City.
Available at https://www.kansascityfed.org/publications/research/mb/articles/2018/what-could-resurging-energy-production-mean. The
authors state that ``The decline in U.S. net energy imports has
prevented the total U.S. trade deficit from widening further. . . .
In 2006, petroleum accounted for about 16 percent of U.S. goods
imports and about 3 percent of U.S. goods exports. By the end of
2017, the share of petroleum in total goods imports declined to 8
percent, while the share in total goods exports almost tripled,
shrinking the U.S. petroleum trade deficit. Had the petroleum trade
deficit not improved, all else unchanged, the total U.S. trade
deficit would likely have been more than 35 percent wider by the end
of 2017.''
\71\ For an illustration of recent increases in U.S. production,
see, e.g., `U.S. crude oil and liquid fuels production,'' Short-Term
Energy Outlook, U.S. Energy Information Administration (Aug. 2019),
available at http://www.eia.gov/outlooks/steo/images/Fig16.png. EIA
noted in April 2019 that ``Annual U.S. crude oil production reached
a record level of 10.96 million barrels per day (b/d) in 2018, 1.6
million b/d (17%) higher than 2017 levels. In December 2018, monthly
U.S. crude oil production reached 11.96 million b/d, the highest
monthly level of crude oil production in U.S. history. U.S crude oil
production has increased significantly over the past 10 years,
driven mainly by production from tight rock formations using
horizontal drilling and hydraulic fracturing. EIA projects that U.S.
crude oil production will continue to grow in 2019 and 2020,
averaging 12.3 million b/d and 13.0 million b/d, respectively.''
``Today in Energy: U.S. crude oil production grew 17% in 2018,
surpassing the previous record in 1970,'' EIA, Apr. 9, 2019.
Available at http://www.eia.gov/todayinenergy/detail.php?id=38992.
---------------------------------------------------------------------------
As flagged in the NPRM, some commenters raised concerns about
potential economic consequences for automaker and supplier operations
in the U.S. due to disparities between CAFE standards at home and their
counterpart fuel economy/efficiency and CO2 standards
abroad. NHTSA finds these concerns more relevant to technological
feasibility and economic practicability considerations than to the
national balance of payments. The discussion in Section VIII below
addresses this topic in more detail.
(3) Environmental Implications
Higher fleet fuel economy can reduce U.S. emissions of various
pollutants by reducing the amount of oil that is produced and refined
for the U.S. vehicle fleet, but can also increase emissions by reducing
the cost of driving, which can result in more vehicle miles traveled
(i.e., the rebound effect). Thus, the net effect of more stringent CAFE
standards on emissions of each pollutant depends on the relative
magnitude of both its reduced emissions in fuel refining and
distribution and increases in its emissions from vehicle use. Fuel
savings from CAFE standards also necessarily results in lower emissions
of CO2, the main greenhouse gas emitted as a result of
refining, distributing, and using transportation fuels. Reducing fuel
consumption directly reduces CO2 emissions because the
primary source of transportation-related CO2 emissions is
fuel combustion in internal combustion engines.
NHTSA has considered environmental issues, both within the context
of EPCA and the context of the National Environmental Policy Act
(NEPA), in making decisions about the setting of standards since the
earliest days of the CAFE program. As courts of appeal have noted in
three decisions stretching over the last 20 years,\72\ NHTSA defined
``the need of the United States to conserve energy'' in the late 1970s
as including, among other things, environmental implications. In 1988,
NHTSA included climate change concepts in its CAFE notices and prepared
its first environmental assessment addressing that subject.\73\ It
cited concerns about climate change as one of its reasons for limiting
the extent of its reduction of the CAFE standard for MY 1989 passenger
cars.\74\ Since then, NHTSA has considered the effects of reducing
tailpipe emissions of CO2 in its fuel economy rulemakings
pursuant to the need of the United States to conserve energy by
reducing petroleum consumption.
---------------------------------------------------------------------------
\72\ CAS, 793 F.2d 1322, 1325 n. 12 (D.C. Cir. 1986); Public
Citizen, 848 F.2d 256, 262-63 n. 27 (D.C. Cir 1988) (noting that
``NHTSA itself has interpreted the factors it must consider in
setting CAFE standards as including environmental effects''); CBD,
538 F.3d 1172 (9th Cir. 2007).
\73\ 53 FR 33080, 33096 (Aug. 29, 1988).
\74\ 53 FR 39275, 39302 (Oct. 6, 1988).
---------------------------------------------------------------------------
(4) Foreign Policy Implications
U.S. consumption and imports of petroleum products can impose
additional costs (i.e., externalities) on the domestic economy that are
not reflected in the market price for crude petroleum or in the prices
paid by consumers for petroleum products such as gasoline. NHTSA has
said previously that these costs can include (1) higher prices for
petroleum products resulting from the effect of U.S. oil demand on
world oil prices, (2) the risk of disruptions to the U.S. economy
caused by sudden increases in the global price of oil and its resulting
impact on fuel prices faced by U.S. consumers, and (3) expenses for
maintaining the strategic petroleum reserve (SPR) to provide a response
option should a disruption in commercial oil supplies threaten the U.S.
economy, to allow the U.S. to meet part of its International Energy
Agency obligation to maintain emergency oil stocks, and to provide a
national defense fuel reserve.\75\ Higher U.S. consumption of crude oil
or refined petroleum products increases the magnitude of these external
economic costs, thus increasing the true economic cost of supplying
transportation fuels above the resource costs of producing them.
Conversely, reducing U.S. consumption of crude oil or refined petroleum
products (by reducing motor fuel use) can reduce these external costs.
---------------------------------------------------------------------------
\75\ While the U.S. maintains a military presence in certain
parts of the world to help secure global access to petroleum
supplies, that is neither the primary nor the sole mission of U.S.
forces overseas. Additionally, the scale of oil consumption
reductions associated with CAFE standards would be insufficient to
alter any existing military missions focused on ensuring the safe
and expedient production and transportation of oil around the globe.
See the FRIA's discussion on energy security for more information on
this topic.
---------------------------------------------------------------------------
While these costs are considerations, the United States has
significantly increased oil production capabilities in recent years, to
the extent that the U.S. is currently producing enough oil to satisfy
nearly all of its energy needs and is projected to continue to do so
(or even become a net energy exporter in the near future).\76\ This has
added stable new supply to the global oil market, which ameliorates the
U.S.' need to
[[Page 24216]]
conserve energy from a security perspective even given that oil is a
global commodity. The agencies discuss this issue in more detail in
Section VIII below.
---------------------------------------------------------------------------
\76\ See AEO 2019, at 14 (``In the Reference case, the United
States becomes a net exporter of petroleum liquids after 2020 as
U.S. crude oil production increases and domestic consumption of
petroleum products decreases.''). Available at https://www.eia.gov/outlooks/aeo/pdf/aeo2019.pdf.
---------------------------------------------------------------------------
(2) Factors That NHTSA Is Prohibited From Considering
EPCA states that in determining the level at which it should set
CAFE standards for a particular model year, NHTSA may not consider the
ability of manufacturers to take advantage of several EPCA provisions
that facilitate compliance with CAFE standards and thereby can reduce
their costs of compliance.\77\ As discussed further below, NHTSA cannot
consider compliance credits that manufacturers earn by exceeding the
CAFE standards and then use to achieve compliance in years in which
their measured average fuel economy falls below the standards. NHTSA
also cannot consider the use of alternative fuels by dual-fueled
vehicles (such as plug-in hybrid electric vehicles) nor the
availability of dedicated alternative fuel vehicles (such as battery
electric or hydrogen fuel cell vehicles) in any model year. EPCA
encourages the production of alternative fuel vehicles by specifying
that their fuel economy is to be determined using a special calculation
procedure that results in those vehicles being assigned a higher fuel
economy level than they actually achieve. For non-statutory incentives
that NHTSA developed by regulation, NHTSA does not consider these
incentives subject to the EPCA prohibition on considering
flexibilities. These topics will be addressed further in Section VIII
below.
---------------------------------------------------------------------------
\77\ 49 U.S.C. 32902(h).
---------------------------------------------------------------------------
(3) Other Considerations in Determining Maximum Feasible CAFE Standards
NHTSA historically has interpreted EPCA's statutory factors as
including consideration for potential adverse safety consequences in
setting CAFE standards. Courts have consistently recognized that this
interpretation is reasonable. As courts have recognized, ``NHTSA has
always examined the safety consequences of the CAFE standards in its
overall consideration of relevant factors since its earliest rulemaking
under the CAFE program.'' \78\ The courts have consistently upheld
NHTSA's implementation of EPCA in this manner.\79\ Thus, in evaluating
what levels of stringency would result in maximum feasible standards,
NHTSA assesses the potential safety impacts and considers them in
balancing the statutory considerations and to determine the maximum
feasible level of the standards.\80\ Many commenters addressed the
NPRM's analysis of safety impacts; those comments will be summarized
and responded to in Section VI.D.2 and also in each agency's discussion
in Section VIII.
---------------------------------------------------------------------------
\78\ Competitive Enterprise Institute v. NHTSA, 901 F.2d 107,
120 n. 11 (D.C. Cir. 1990) (``CEI-I'') (citing 42 FR 33534, 33551
(Jun. 30, 1977).
\79\ See, e.g., Competitive Enterprise Institute v. NHTSA, 956
F.2d 321, 322 (D.C. Cir. 1992) (``CEI-II'') (in determining the
maximum feasible fuel economy standard, ``NHTSA has always taken
passenger safety into account,'' citing CEI-I, 901 F.2d at 120 n.
11); Competitive Enterprise Institute v. NHTSA, 49 F.3d 481, 483-83
(D.C. Cir. 1995) (same); Center for Biological Diversity v. NHTSA,
538 F.3d 1172, 1203-04 (9th Cir. 2008) (upholding NHTSA's analysis
of vehicle safety issues with weight in connection with the MYs
2008-2011 light truck CAFE rulemaking).
\80\ NHTSA stated in the NPRM that ``While we discuss safety as
a separate consideration, NHTSA also considers safety as closely
related to, and in some circumstances a subcomponent of, economic
practicability. On a broad level, manufacturers have finite
resources to invest in research and development. Investment into the
development and implementation of fuel saving technology necessarily
comes at the expense of investing in other areas such as safety
technology. On a more direct level, when making decisions on how to
equip vehicles, manufacturers must balance cost considerations to
avoid pricing further consumers out of the market. As manufacturers
add technology to increase fuel efficiency, they may decide against
installing new safety equipment to reduce cost increases. And as the
price of vehicles increase beyond the reach of more consumers, such
consumers continue to drive or purchase older, less safe vehicles.
In assessing practicability, NHTSA also considers the harm to the
nation's economy caused by highway fatalities and injuries.'' 83 FR
at 43209 (Aug. 24, 2018). Many comments were received on this issue,
which will be discussed further in Section VIII below.
---------------------------------------------------------------------------
The above sections explain what Congress thought was important
enough to codify when it directed each agency to regulate, and begin to
explain how the agencies have interpreted those directions over time
and in this final rule. The next section looks more closely at the
interplay between Congress's direction to the agencies and the aspects
of the market that these regulations affect, as follows.
IV. Purpose of Analytical Approach Considered as Part of Decision-
Making
A. Relationship of Analytical Approach to Governing Law
Like the NPRM, today's final rule is supported by extensive
analysis of potential impacts of the regulatory alternatives under
consideration. Below, Section VI reviews the analytical approach,
Section VII summarizes the results of the analysis, and Section VIII
explains how the final standards--informed by this analysis--fulfill
the agencies' statutory obligations. Accompanying today's notice, a
final Regulatory Impact Analysis (FRIA) and, for NHTSA's consideration,
a final Environmental Impact Analysis (FEIS), together provide a more
extensive and detailed enumeration of related methods, estimates,
assumptions, and results. The agencies' analysis has been constructed
specifically to reflect various aspects of governing law applicable to
CAFE and CO2 standards, and has been expanded and improved
in response to comments received to the NPRM and based on additional
work by the agencies. The analysis aided the agencies in implementing
their statutory obligations, including the weighing of competing
considerations, by reasonably informing the agencies about the
estimated effects of choosing different regulatory alternatives.
The agencies' analysis makes use of a range of data (i.e.,
observations of things that have occurred), estimates (i.e., things
that may occur in the future), and models (i.e., methods for making
estimates). Two examples of data include (1) records of actual odometer
readings used to estimate annual mileage accumulation at different
vehicle ages and (2) CAFE compliance data used as the foundation for
the ``analysis fleet'' containing, among other things, production
volumes and fuel economy levels of specific configurations of specific
vehicle models produced for sale in the U.S. Two examples of estimates
include (1) forecasts of future GDP growth used, with other estimates,
to forecast future vehicle sales volumes and (2) the ``retail price
equivalent'' (RPE) factor used to estimate the ultimate cost to
consumers of a given fuel-saving technology, given accompanying
estimates of the technology's ``direct cost,'' as adjusted to account
for estimated ``cost learning effects'' (i.e., the tendency that it
will cost a manufacturer less to apply a technology as the manufacturer
gains more experience doing so).
The agencies' analysis makes use of several models, some of which
are actually integrated systems of multiple models. As discussed in the
NPRM, the agencies' analysis of CAFE and CO2 standards
involves two basic elements: First, estimating ways each manufacturer
could potentially respond to a given set of standards in a manner that
considers potential consumer response; and second, estimating various
impacts of those responses. Estimating manufacturers' potential
responses involves simulating manufacturers' decision-making processes
regarding the year-by-year application of fuel-saving technologies to
specific vehicles. Estimating impacts involves calculating resultant
changes in new vehicle costs, estimating a
[[Page 24217]]
variety of costs (e.g., for fuel) and effects (e.g., CO2
emissions from fuel combustion) occurring as vehicles are driven over
their lifetimes before eventually being scrapped, and estimating the
monetary value of these effects. Estimating impacts also involves
consideration of the response of consumers--e.g., whether consumers
will purchase the vehicles and in what quantities. Both of these basic
analytical elements involve the application of many analytical inputs.
The agencies' analysis uses the CAFE Model to estimate
manufacturers' potential responses to new CAFE and CO2
standards and to estimate various impacts of those responses. The model
may be characterized as an integrated system of models. For example,
one model estimates manufacturers' responses, another estimates
resultant changes in total vehicle sales, and still another estimates
resultant changes in fleet turnover (i.e., scrappage). The CAFE model
makes use of many inputs, values of which are developed outside of the
model and not by the model. For example, the model applies fuel prices;
it does not estimate fuel prices. The model does not determine the form
or stringency of the standards; instead, the model applies inputs
specifying the form and stringency of standards to be analyzed and
produces outputs showing effects of manufacturers working to meet those
standards, which become the basis for comparing between different
potential stringencies.
The agencies also use EPA's MOVES model to estimate ``tailpipe''
(a.k.a. ``vehicle'' or ``downstream'') emission factors for criteria
pollutants,\81\ and use four DOE and DOE-sponsored models to develop
inputs to the CAFE model, including three developed and maintained by
DOE's Argonne National Laboratory. The agencies use the DOE Energy
Information Administration's (EIA's) National Energy Modeling System
(NEMS) to estimate fuel prices,\82\ and use Argonne's Greenhouse gases,
Regulated Emissions, and Energy use in Transportation (GREET) model to
estimate emissions rates from fuel production and distribution
processes.\83\ DOT also sponsored DOE/Argonne to use Argonne's
Autonomie full-vehicle modeling and simulation system to estimate the
fuel economy impacts for roughly a million combinations of technologies
and vehicle types.84 85 Section VI.B.3, below, and the
accompanying final RIA document details of the agencies' use of these
models. In addition, as discussed in the final EIS accompanying today's
notice, DOT relied on a range of climate and photochemical models to
estimate impacts on climate, air quality, and public health. The EIS
discusses and documents the use of these models.
---------------------------------------------------------------------------
\81\ See https://www.epa.gov/moves. Today's final rule used
version MOVES2014b, available at https://www.epa.gov/moves/latest-version-motor-vehicle-emission-simulator-moves.
\82\ See https://www.eia.gov/outlooks/aeo/info_nems_archive.php.
Today's final rule uses fuel prices estimated using the Annual
Energy Outlook (AEO) 2019 version of NEMS (see https://www.eia.gov/outlooks/aeo/data/browser/#/?id=3-AEO2019&cases=ref2019&sourcekey=0).
\83\ Information regarding GREET is available at https://greet.es.anl.gov/index.php. Today's notice uses the 2018 version of
GREET.
\84\ As part of the Argonne simulation effort, individual
technology combinations simulated in Autonomie were paired with
Argonne's BatPAC model to estimate the battery cost associated with
each technology combination based on characteristics of the
simulated vehicle and its level of electrification. Information
regarding Argonne's BatPAC model is available at http://www.cse.anl.gov/batpac/.
\85\ In addition, the impact of engine technologies on fuel
consumption, torque, and other metrics was characterized using GT
POWER simulation modeling in combination with other engine modeling
that was conducted by IAV Automotive Engineering, Inc. (IAV). The
engine characterization ``maps'' resulting from this analysis were
used as inputs for the Autonomie full-vehicle simulation modeling.
Information regarding GT Power is available at https://www.gtisoft.com/gt-suite-applications/propulsion-systems/gt-power-engine-simulation-software.
---------------------------------------------------------------------------
As further explained in the NPRM,\86\ to prepare for analysis
supporting the proposal, DOT expanded the CAFE model to address EPA
statutory and regulatory requirements through a year-by-year simulation
of how manufacturers could comply with EPA's CO2 standards,
including:
---------------------------------------------------------------------------
\86\ 83 FR 42986, 43003 (Aug. 24, 2018).
---------------------------------------------------------------------------
Calculation of vehicle models' CO2 emission
rates before and after application of fuel-saving (and, therefore,
CO2-reducing) technologies;
Calculation of manufacturers' fleet average CO2
emission rates;
Calculation of manufacturers' fleet average CO2
emission rates under attribute-based CO2 standards;
Accounting for adjustments to average CO2
emission rates reflecting reduction of air conditioner refrigerant
leakage;
Accounting for the treatment of alternative fuel vehicles
for CO2 compliance;
Accounting for production ``multipliers'' for PHEVs, BEVs,
compressed natural gas (CNG) vehicles, and fuel cell vehicles (FCVs);
Accounting for transfer of CO2 credits between
regulated fleets; and
Accounting for carried-forward (a.k.a. ``banked'')
CO2 credits, including credits from model years earlier than
modeled explicitly.
As further discussed in the NPRM, although EPA had previously
developed a vehicle simulation tool (``ALPHA'') and a fleet compliance
model (``OMEGA''), and had applied these in prior actions, having
considered the facts before the Agency in 2018, EPA determined that,
``it is reasonable and appropriate to use DOE/Argonne's model for full-
vehicle simulation, and to use DOT's CAFE model for analysis of
regulatory alternatives.'' \87\
---------------------------------------------------------------------------
\87\ 83 FR 42986, 43000 (Aug. 24, 2018).
---------------------------------------------------------------------------
As discussed below and in Section VI.B.3, some commenters--some
citing deliberative EPA staff communications during NPRM development,
and one submitting comments by a former EPA staff member closely
involved in the origination of the above-mentioned OMEGA model--took
strong exception to EPA's decision to rely on DOE/Argonne and DOT-
originated models as the basis for analysis informing EPA's decisions
regarding CO2 standards. Some commenters argued that the EPA
Administrator must consider exclusively models and analysis originating
with EPA staff, and that to do otherwise would be arbitrary and
capricious. As explained below (and as explained in the NPRM), it is
reasonable for the Administrator to consider analysis and information
produced from many sources, including, in this instance, the DOE/
Argonne and DOT models. The Administrator has the discretion to
determine what information reasonably and appropriately informs
decisions regarding emissions standards. Some commenters conflated
models with decisions, suggesting that the former mechanically
determine the latter. The CAA authorizes the EPA Administrator, not a
model, to make decisions about emissions standards, just as EPCA
provides similar authority to the Secretary. Models produce analysis,
the results of which help to inform decisions. However, in making such
decisions, the Administrator may and should consider other relevant
information beyond the outputs of any models--including public
comment--and, in all cases, must exercise judgment in establishing
appropriate standards.
Some commenters conflated models with inputs and/or with results of
the modeling. All of the models mentioned above rely on inputs,
including not only data (i.e., facts), but also estimates (inputs about
the future are estimates, not data). Given these inputs, the models
produce estimates--ultimately, the agencies' reported estimates of the
potential impacts of standards under
[[Page 24218]]
consideration. In other words, inputs do not define models; models use
inputs. Therefore, disagreements about inputs do not logically extend
to disagreements about models. Similarly, while models determine
resulting outputs, they do so based on inputs. Therefore, disagreements
about results do not necessarily imply disagreements about models; they
may merely reflect disagreements about inputs. With respect to the
Administrator's decisions regarding models underlying today's analysis,
comments regarding inputs, therefore, are more appropriately addressed
separately, which is done so below in Section VI.
The EPA Administrator's decision to continue relying on the DOE/
Argonne Autonomie tool and DOT CAFE model rather than on the
corresponding tools developed by EPA staff is informed by consideration
of comments on results and on technical aspects of the models
themselves. As discussed below, some commenters questioned specific
aspects of the CAFE model's simulation of manufacturer's potential
responses to CO2 standards. Considering these comments, the
CAFE model applied in the final rule's analysis includes some revisions
and updates. For example, the ``effective cost'' metric used to select
among available opportunities to apply fuel-saving technologies now
uses a ``cost per credit'' metric rather than the metric used for the
NPRM. Also, the model's representation of sales ``multipliers'' EPA has
included for CNG vehicles, PHEVs, BEVs, and FCVs reflects current EPA
regulations or, as an input-selectable option, an alternative approach
under consideration. On the other hand, some commenters questioning the
CAFE model's approach to some CO2 program features appear to
ignore the fact that prior analysis by EPA (using EPA's OMEGA) model
likewise did not account for the same program features. For example,
some stakeholders took issue with the CAFE model's approach to
accounting for banked CO2 credits and, in particular,
credits banked prior to the model years accounted for explicitly in the
analysis. In the course of updating the basis for analysis fleet from
model year 2016 to model year 2017, the agencies have since updated
corresponding inputs. However, even though the ability to carry forward
credits impacts outcomes, EPA's OMEGA model used in previous
rulemakings never attempted to account for credit banking and, indeed,
lacking a year-by-year structure, cannot account for credit banking.
Therefore, at least with respect to this important CO2
program flexibility, the CAFE model provides a more complete and
realistic basis for estimating actual impacts of new CO2
standards.
For its part, NHTSA remains confident that the combination of the
Autonomie and CAFE models remains the best available for CAFE
rulemaking analysis, and notes, as discussed below, that even the
environmental group coalition stated that the CAFE model is aligned
with EPCA requirements.\88\ In late 2001, after Congress discontinued
an extended series of budget ``riders'' prohibiting work on CAFE
standards, NHTSA and the DOT Volpe Center began development of a
modeling system appropriate for CAFE rulemaking analysis, because other
available models were not designed with this purpose in mind, and
lacked capabilities important for CAFE rulemakings. For example,
although NEMS had procedures to account for CAFE standards, those
procedures did not provide the ability to account for specific
manufacturers, as is especially relevant to the statutory requirement
that NHTSA consider the economic practicability of any new CAFE
standards. Also, as early as the first rulemaking making use of this
early CAFE model, commenters stressed the importance of product
redesign schedules, leading developers to introduce procedures to
account for product cadence. In the 2003 notice regarding light truck
standards for MYs 2005-2007, NHTSA stated that ``we also changed the
methodology to recognize that capital costs require employment of
technologies for several years, rather than a single year. . . . In our
view, this makes the Volpe analysis more consistent with the [manually
implemented] Stage analysis and better reflects actual conditions in
the automotive industry.'' \89\ Since that time, NHTSA and the Volpe
Center have significantly refined the CAFE model with each of
rulemaking. For example, for the 2006 rulemaking regarding standards
for MYs 2008-2011 light trucks, NHTSA introduced the ability to account
for attribute-based standards, account for the social cost of
CO2 emissions, estimate stringencies at which net benefits
would be maximized, and perform probabilistic uncertainty analysis
(i.e., Monte Carlo simulation).\90\ For the 2009 rulemaking regarding
standards for MY 2011 passenger cars and light trucks, we introduced
the ability to account for attribute-based passenger car standards, and
the ability to apply ``synergy factors'' to estimate how some
technology pairings impact fuel consumption,\91\ For the 2010
rulemaking regarding standards for MYs 2012-2016, we introduced
procedures to account for FFV credits, and to account for product
planning as a multiyear consideration.\92\ For the 2012 rulemaking
regarding standards for MYs 2017-2025, we introduced several new
procedures, such as (1) accounting for electricity used to charge
electric vehicles (EVs) and plug-in hybrid electric vehicles (PHEVs),
(2) accounting for use of ethanol blends in flexible-fuel vehicles
(FFVs), (3) accounting for costs (i.e., ``stranded capital'') related
to early replacement of technologies, (4) accounting for previously-
applied technology when determining the extent to which a manufacturer
could expand use of the technology, (5) applying technology-specific
estimates of changes in consumer value, (6) simulating the extent to
which manufacturers might utilize EPCA's provisions regarding
generation and use of CAFE credits, (7) applying estimates of fuel
economy adjustments (and accompanying costs) reflecting increases in
air conditioner efficiency, (8) reporting privately-valued benefits,
(9) simulating the extent to which manufacturers might voluntarily
apply technology beyond levels needed for compliance with CAFE
standards, and (10) estimating changes in highway fatalities
attributable to any applied reductions in vehicle mass.\93\ Also for
the 2012 rulemaking, we began making use of Autonomie to estimate fuel
consumption impacts of different combinations of technologies, using
these estimates to specify inputs to the CAFE model.\94\ In 2016,
providing analyses for both the draft TAR regarding light-duty CAFE
standards and the final rule regarding fuel consumption standards for
heavy-duty pickup trucks and vans, we greatly expanded the agency's use
of Autonomie-based full vehicle simulations and introduced the ability
to simulate compliance with attribute-based standards for heavy-duty
pickups and vans.\95\ And, as discussed at length in the NPRM and
below, for this rulemaking, we have, among other things, refined
procedures to account for impacts on highway travel and safety,
[[Page 24219]]
added procedures to simulate compliance with CO2 standards,
refined procedures to account for compliance credits, and added
procedures to account for impacts on sales, scrappage, and employment.
We have also significantly revised the model's graphical user interface
(GUI) in order to make the model easier to operate and understand. Like
any model, both Autonomie and the CAFE model benefit from ongoing
refinement. However, NHTSA is confident that this combination of models
produces a more realistic characterization of the potential impacts of
new standards than would another combination of available models. Some
stakeholders, while commenting on specific aspects of the inputs,
models, and/or results, commended the agencies' exclusive reliance on
the DOE/Argonne Autonomie tool and DOT CAFE model. With respect to
CO2 standards, these stakeholders noted not only technical
reasons to use these models rather than the EPA models, but also other
reasons such as efficiency, transparency, and ease with which outside
parties can exercise models and replicate the agencies' analysis. These
comments are discussed below and in Section VI.
---------------------------------------------------------------------------
\88\ Environmental group coalition, NHTSA-2018-0067-12000,
Appendix A, at 24-25.
\89\ 68 FR at 16885 (Apr. 7, 2003).
\90\ 71 FR at 17566 et seq. (Apr. 6, 2006).
\91\ 74 FR at 14196 et seq. (Mar. 30, 3009).
\92\ 75 FR at 25599 et seq. (May 7, 2010).
\93\ 77 FR 63009 et seq. (Oct. 15, 2012).
\94\ 77 FR at 62712 et seq. (Oct. 15, 2012).
\95\ 81 FR at 73743 et seq. (Oct. 25, 2016); Draft TAR,
available at Docket No. NHTSA-2016-0068-0001, Chapter 13.
---------------------------------------------------------------------------
Nevertheless, some comments regarding the model's handling of CAFE
and/or CO2 standards, and some comments regarding the
model's estimation of resultant impacts, led the agencies to make
changes to specific aspects of the model. Comments on and changes to
the inputs and model are discussed below and in Section VI; results are
discussed in Section VII and in the accompanying RIA; and the meaning
of results in the context of the applicable statutory requirements is
discussed in Section VIII.
As explained, the analysis is designed to reflect a number of
statutory and regulatory requirements applicable to CAFE and tailpipe
CO2 standard setting. EPCA contains a number of requirements
governing the scope and nature of CAFE standard setting. Among these,
some have been in place since EPCA was first signed into law in 1975,
and some were added in 2007, when Congress passed EISA and amended
EPCA. The CAA, as discussed elsewhere, provides EPA with very broad
authority under Section 202(a), and does not contain EPCA/EISA's
prescriptions. In the interest of harmonization, however, EPA has
adopted some of the EPCA/EISA requirements into its tailpipe
CO2 regulations, and NHTSA, in turn, has created some
additional flexibilities by regulation not expressly envisioned by
EPCA/EISA in order to harmonize better with some of EPA's programmatic
decisions. EPCA/EISA requirements regarding the technical
characteristics of CAFE standards and the analysis thereof include, but
are not limited to, the following, and the analysis reflects these
requirements as summarized:
Corporate Average Standards: 49 U.S.C. 32902 requires standards
that apply to the average fuel economy levels achieved by each
corporation's fleets of vehicles produced for sale in the U.S.\96\ CAA
Section 202(a) does not preclude the EPA Administrator from expressing
CO2 standards as de facto fleet average requirements, and
EPA has adopted a similar approach in the interest of harmonization.
The CAFE Model, used by the agencies to conduct the bulk of today's
analysis, calculates the CAFE and CO2 levels of each
manufacturer's fleets based on estimated production volumes and
characteristics, including fuel economy levels, of distinct vehicle
models that could be produced for sale in the U.S.
---------------------------------------------------------------------------
\96\ This differs from safety standards and traditional
emissions standards, which apply separately to each vehicle. For
example, every vehicle produced for sale in the U.S. must, on its
own, meet all applicable federal motor vehicle safety standards
(FMVSS), but no vehicle produced for sale must, on its own, federal
fuel economy standards. Rather, each manufacturer is required to
produce a mix of vehicles that, taken together, achieve an average
fuel economy level no less than the applicable minimum level.
---------------------------------------------------------------------------
Separate Standards for Passenger Cars and Light Trucks: 49 U.S.C.
32902 requires the Secretary of Transportation to set CAFE standards
separately for passenger cars and light trucks. CAA Section 202(a) does
not preclude the EPA Administrator from specifying CO2
standards separately for passenger cars and light trucks, and EPA has
adopted a similar approach. The CAFE Model accounts separately for
passenger cars and light trucks, including differentiated standards and
compliance.
Attribute-Based Standards: 49 U.S.C. 32902 requires the Secretary
of Transportation to define CAFE standards as mathematical functions
expressed in terms of one or more vehicle attributes related to fuel
economy. This means that for a given manufacturer's fleet of vehicles
produced for sale in the U.S. in a given regulatory class and model
year, the applicable minimum CAFE requirement (i.e., the numerical
value of the requirement) is computed based on the applicable
mathematical function, and the mix and attributes of vehicles in the
manufacturer's fleet. In the 2012 final rule that first established
CO2 standards, EPA also adopted an attribute-based standard
under its broad CAA Section 202(a) authority. The CAFE Model accounts
for such functions and vehicle attributes explicitly.
Separately Defined Standards for Each Model Year: 49 U.S.C. 32902
requires the Secretary to set CAFE standards (separately for passenger
cars and light trucks) at the maximum feasible levels in each model
year. CAA Section 202(a) allows EPA to establish CO2
standards separately for each model year, and EPA has chosen to do so
for this final rule, similar to the approach taken in the previous
light-duty vehicle CO2 standard-setting rules. The CAFE
Model represents each model year explicitly, and accounts for the
production relationships between model years.\97\
---------------------------------------------------------------------------
\97\ For example, a new engine first applied to given vehicle
model/configuration in model year 2020 will most likely be ``carried
forward'' to model year 2021 of that same vehicle model/
configuration, in order to reflect the fact that manufacturers do
not apply brand-new engines to a given vehicle model every single
year.
---------------------------------------------------------------------------
Separate Compliance for Domestic and Imported Passenger Car Fleets:
49 U.S.C. 32904 requires the EPA Administrator to determine CAFE
compliance separately for each manufacturers' fleets of domestic
passenger cars and imported passenger cars, which manufacturers must
consider as they decide how to improve the fuel economy of their
passenger car fleets. CAA 202(a) does not preclude the EPA
Administrator from determining compliance with CO2 standards
separately for a manufacturer's domestic and imported car fleets, but
EPA did not include such a distinction in either the 2010 or 2012 final
rules, and EPA did not propose or ask for comment on taking such an
approach in the proposal. The CAFE Model is able to account explicitly
for this requirement when simulating manufacturers' potential responses
to CAFE standards, but combines any given manufacturer's domestic and
imported cars into a single fleet when simulating that manufacturer's
potential response to CO2 standards.
Minimum CAFE Standards for Domestic Passenger Car Fleets: 49 U.S.C.
32902 requires that domestic passenger car fleets achieve CAFE levels
no less than 92 percent of the industry-wide average level required
under the applicable attribute-based CAFE standard, as projected by the
Secretary at the time the standard is promulgated. CAA 202(a) does not
preclude the EPA Administrator from correspondingly requiring that
domestic passenger car fleets achieve CO2 levels no greater
than 108.7 percent (1/0.92 = 1.087) of the projected industry-wide
average CO2
[[Page 24220]]
requirement under the attribute-based standard, but the GHG program
that EPA designed in the 2010 and 2012 final rules did not include such
a distinction, and EPA did not propose or seek comment on such an
approach in the proposal. The CAFE Model is able to account explicitly
for this requirement for CAFE standards, and sets this requirement
aside for CO2 standards.
Civil Penalties for Noncompliance: 49 U.S.C. 32912 prescribes a
rate (in dollars per tenth of a mpg) at which the Secretary is to levy
civil penalties if a manufacturer fails to comply with a CAFE standard
for a given fleet in a given model year, after considering available
credits. Some manufacturers have historically demonstrated a
willingness to treat CAFE noncompliance as an ``economic'' choice,
electing to pay civil penalties rather than achieving full numerical
compliance across all fleets. The CAFE Model calculates civil penalties
for CAFE shortfalls and provides means to estimate that a manufacturer
might stop adding fuel-saving technologies once continuing to do so
would be effectively more ``expensive'' (after accounting for fuel
prices and buyers' willingness to pay for fuel economy) than paying
civil penalties. In contrast, the CAA does not authorize the EPA
Administrator to allow manufacturers to sell noncompliant fleets and
instead only pay civil penalties; manufacturers who choose to pay civil
penalties for CAFE compliance tend to employ EPA's more-extensive
programmatic flexibilities to meet tailpipe CO2 emissions
standards. Thus, the CAFE Model does not allow civil penalty payment as
an option for CO2 standards.
Dual-Fueled and Dedicated Alternative Fuel Vehicles: For purposes
of calculating CAFE levels used to determine compliance, 49 U.S.C.
32905 and 32906 specify methods for calculating the fuel economy levels
of vehicles operating on alternative fuels to gasoline or diesel
through MY 2020. After MY 2020, methods for calculating alternative
fuel vehicle (AFV) fuel economy are governed by regulation. The CAFE
Model is able to account for these requirements explicitly for each
vehicle model. However, 49 U.S.C. 32902 requires that maximum feasible
CAFE standards be set in a manner that does not presume manufacturers
can respond by producing new dedicated alternative fuel vehicle (AFV)
models. The CAFE model can be run in a manner that excludes the
additional application of dedicated AFV technologies in model years for
which maximum feasible standards are under consideration. As allowed
under NEPA for analysis appearing in EISs informing decisions regarding
CAFE standards, the CAFE Model can also be run without this analytical
constraint. CAA 202(a) does not preclude the EPA Administrator adopting
analogous provisions, but EPA has instead opted through regulation to
``count'' dual- and alternative fuel vehicles on a CO2 basis
(and through MY 2026, to set aside emissions from electricity
generation). The CAFE model accounts for this treatment of dual- and
alternative fuel vehicles when simulating manufacturers' potential
responses to CO2 standards. For natural gas vehicles, both
dedicated and dual-fueled, EPA is establishing a multiplier of 2.0 for
model years 2022-2026.
Creation and Use of Compliance Credits: 49 U.S.C. 32903 provides
that manufacturers may earn CAFE ``credits'' by achieving a CAFE level
beyond that required of a given fleet in a given model year, and
specifies how these credits may be used to offset the amount by which a
different fleet falls short of its corresponding requirement. These
provisions allow credits to be ``carried forward'' and ``carried back''
between model years, transferred between regulated classes (domestic
passenger cars, imported passenger cars, and light trucks), and traded
between manufacturers. However, these provisions also impose some
specific statutory limits. For example, CAFE compliance credits can be
carried forward a maximum of five model years and carried back a
maximum of three model years. Also, EPCA/EISA caps the amount of credit
that can be transferred between passenger car and light truck fleets,
and prohibits manufacturers from applying traded or transferred credits
to offset a failure to achieve the applicable minimum standard for
domestic passenger cars. The CAFE Model explicitly simulates
manufacturers' potential use of credits carried forward from prior
model years or transferred from other fleets.\98\ 49 U.S.C. 32902
prohibits consideration of manufacturers' potential application of CAFE
compliance credits when setting maximum feasible CAFE standards. The
CAFE Model can be operated in a manner that excludes the application of
CAFE credits after a given model year. CAA 202(a) does not preclude the
EPA Administrator adopting analogous provisions. EPA has opted to limit
the ``life'' of compliance credits from most model years to 5 years,
and to limit borrowing to 3 years, but has not adopted any limits on
transfers (between fleets) or trades (between manufacturers) of
compliance credits. The CAFE Model is able to account for the absence
of limits on transfers of CO2 standards. Insofar as the CAFE
model can be exercised in a manner that simulates trading of
CO2 compliance credits, such simulations treat trading as
unlimited.\99\ EPA has considered manufacturers' ability to use credits
as part of its decisions on these final standards, and the CAFE model
is now able to account for that.
---------------------------------------------------------------------------
\98\ As explained in Section VI, the CAFE Model does not
explicitly simulate the potential that manufacturers would carry
CAFE or CO2 credits back (i.e., borrow) from future model
years, or acquire and use CAFE compliance credits from other
manufacturers. At the same time, because EPA has elected to not
limit credit trading, the CAFE Model can be exercised in a manner
that simulates unlimited (a.k.a. ``perfect'') CO2
compliance credit trading throughout the industry (or, potentially,
within discrete trading ``blocs''). The agencies believe there is
significant uncertainty in how manufacturers may choose to employ
these particular flexibilities in the future: for example, while it
is reasonably foreseeable that a manufacturer who over-complies in
one year may ``coast'' through several subsequent years relying on
those credits rather than continuing to make technology
improvements, it is harder to assume with confidence that
manufacturers will rely on future technology investments (that may
not pan out as expected, as if market demand for ``target-beater''
vehicles is lower than expected) to offset prior-year shortfalls, or
whether/how manufacturers will trade credits with market competitors
rather than making their own technology investments. Historically,
carry-back and trading have been much less utilized than carry-
forward, for a variety of reasons including higher risk and
preference not to ``pay competitors to make fuel economy
improvements we should be making'' (to paraphrase one manufacturer),
although the agencies recognize that carry-back and trading are used
more frequently when standards require more technology application
than manufacturers believe their markets will bear. Given the
uncertainty just discussed, and given also the fact that the
agencies have yet to resolve some of analytical challenges
associated with simulating use of these flexibilities, the agencies
consider borrowing and trading to involve sufficient risk that it is
prudent to support today's decisions with analysis that sets aside
the potential that manufacturers could come to depend widely on
borrowing and trading. While compliance costs in real life may be
somewhat different from what is modeled today as a result of this
analytical decision, that is broadly true no matter what, and the
agencies do not believe that the difference would be so great that
it would change the policy outcome.
\99\ To avoid making judgments (that would invariably turn out
to be at least somewhat incorrect) about possible future trading
activity, the model simulates trading by combining all manufacturers
into a single entity, so that the most cost-effective choices are
made for the fleet as a whole.
---------------------------------------------------------------------------
Statutory Basis for Stringency: 49 U.S.C. 32902 requires the
Secretary to set CAFE standards at the maximum feasible levels,
considering technological feasibility, economic practicability, the
need of the Nation to conserve energy, and the impact of other
government standards. EPCA/EISA authorizes the Secretary to interpret
[[Page 24221]]
these factors, and as the Department's interpretation has evolved,
NHTSA has continued to expand and refine its qualitative and
quantitative analysis. For example, as discussed below in Section
VI.B.3, the Autonomie simulations reflect the agencies' judgment that
it would not be economically practicable for a manufacturer to
``split'' an engine shared among many vehicle model/configurations into
a myriad of versions each optimized to a single vehicle model/
configuration. Also responding to evolving interpretation of these
EPCA/EISA factors, the CAFE Model has been expanded to address
additional impacts in an integrated manner. For example, the CAFE Model
version used for the NPRM analysis included the ability to estimate
impacts on labor utilization internally, rather than as an external
``off model'' or ``post processing'' analysis. In addition, NEPA
requires the Secretary to issue an EIS that documents the estimated
impacts of regulatory alternatives under consideration. The EIS
accompanying today's notice documents changes in emission inventories
as estimated using the CAFE model, but also documents corresponding
estimates--based on the application of other models documented in the
EIS, of impacts on the global climate, on tropospheric air quality, and
on human health. Regarding CO2 standards, CAA 202(a)
provides general authority for the establishment of motor vehicle
emissions standards, and the final rule's analysis, like that
accompanying the agencies' proposal, addresses impacts relevant to the
EPA Administrator's decision making, such as technological feasibility,
air quality impacts, costs to industry and consumers, and lead time
necessary for compliance.
Other Factors: Beyond these statutory requirements applicable to
DOT and/or EPA are a number of specific technical characteristics of
CAFE and/or CO2 regulations that are also relevant to the
construction of today's analysis. These are discussed at greater length
in Section II.F. For example, EPA has defined procedures for
calculating average CO2 levels, and has revised procedures
for calculating CAFE levels, to reflect manufacturers' application of
``off-cycle'' technologies that increase fuel economy (and reduce
CO2 emissions) in ways not reflected by the long-standing
test procedures used to measure fuel economy. Although too little
information is available to account for these provisions explicitly in
the same way that the agencies have accounted for other technologies,
the CAFE Model does include and makes use of inputs reflecting the
agencies' expectations regarding the extent to which manufacturers may
earn such credits, along with estimates of corresponding costs.
Similarly, the CAFE Model includes and makes use of inputs regarding
credits EPA has elected to allow manufacturers to earn toward
CO2 levels (not CAFE) based on the use of air conditioner
refrigerants with lower global warming potential (GWP), or on the
application of technologies to reduce refrigerant leakage. In addition,
EPA has elected to provide that through model year 2021, manufacturers
may apply ``multipliers'' to plug-in hybrid electric vehicles,
dedicated electric vehicles, fuel cell vehicles, and hydrogen vehicles,
such that when calculating a fleet's average CO2 levels (not
CAFE), the manufacturer may, for example, ``count'' each electric
vehicle twice. The CAFE Model accounts for these multipliers, based on
either current regulatory provisions or on alternative approaches.
Although these are examples of regulatory provisions that arise from
the exercise of discretion rather than specific statutory mandate, they
can materially impact outcomes. Section VI.B explains in greater detail
how today's analysis addresses them.
Benefits of Analytical Approach
The agencies' analysis of CAFE and CO2 standards
involves two basic elements: First, estimating ways each manufacturer
could potentially respond to a given set of standards in a manner that
considers potential consumer response; and second, estimating various
impacts of those responses. Estimating manufacturers' potential
responses involves simulating manufacturers' decision-making processes
regarding the year-by-year application of fuel-saving technologies to
specific vehicles. Estimating impacts involves calculating resultant
changes in new vehicle costs, estimating a variety of costs (e.g., for
fuel) and effects (e.g., CO2 emissions from fuel combustion)
occurring as vehicles are driven over their lifetimes before eventually
being scrapped, and estimating the monetary value of these effects.
Estimating impacts also involves consideration of the response of
consumers--e.g., whether consumers will purchase the vehicles and in
what quantities. Both of these basic analytical elements involve the
application of many analytical inputs.
As mentioned above, the agencies' analysis uses the CAFE model to
estimate manufacturers' potential responses to new CAFE and
CO2 standards and to estimate various impacts of those
responses. DOT's Volpe National Transportation Systems Center (often
simply referred to as the ``Volpe Center'') develops, maintains, and
applies the model for NHTSA. NHTSA has used the CAFE model to perform
analyses supporting every CAFE rulemaking since 2001, and the 2016
rulemaking regarding heavy-duty pickup and van fuel consumption and
CO2 emissions also used the CAFE model for analysis.\100\
---------------------------------------------------------------------------
\100\ While both agencies used the CAFE Model to simulate
manufacturers' potential responses to standards, some model inputs
differed EPA's and DOT's analyses, and EPA also used the EPA MOVES
model to calculate resultant changes in emissions inventories. See
81 FR 73478, 73743 (Oct. 25, 2016).
---------------------------------------------------------------------------
NHTSA recently arranged for a formal peer review of the model. In
general, reviewers' comments strongly supported the model's conceptual
basis and implementation, and commenters provided several specific
recommendations. The agency agreed with many of these recommendations
and has worked to implement them wherever practicable. Implementing
some of the recommendations would require considerable further
research, development, and testing, and will be considered going
forward. For a handful of other recommendations, the agency disagreed,
often finding the recommendations involved considerations (e.g., other
policies, such as those involving fuel taxation) beyond the model
itself or were based on concerns with inputs rather than how the model
itself functioned. A report available in the docket for this rulemaking
presents peer reviewers' detailed comments and recommendations, and
provides DOT's detailed responses.\101\
---------------------------------------------------------------------------
\101\ Docket No. NHTSA-2018-0067-0055.
---------------------------------------------------------------------------
As also mentioned above, the agencies use EPA's MOVES model to
estimate tailpipe emission factors, use DOE/EIA's NEMS to estimate fuel
prices,\102\ and use Argonne's GREET model to estimate downstream
emissions rates.\103\ DOT also sponsored DOE/Argonne to use the
Autonomie full-vehicle modeling and simulation tool to estimate the
fuel economy impacts for roughly a million
[[Page 24222]]
combinations of technologies and vehicle types.104 105
---------------------------------------------------------------------------
\102\ See https://www.eia.gov/outlooks/aeo/info_nems_archive.php. Today's notice uses fuel prices estimated
using the Annual Energy Outlook (AEO) 2019 version of NEMS (see
https://www.eia.gov/outlooks/archive/aeo19/ and https://www.eia.gov/outlooks/aeo/data/browser/#/?id=3-AEO2019&cases=ref2019&sourcekey=0).
\103\ Information regarding GREET is available at https://greet.es.anl.gov/index.php. Availability of NEMS is discussed at
https://www.eia.gov/outlooks/aeo/info_nems_archive.php. Today's
notice uses fuel prices estimated using the AEO 2019 version of
NEMS.
\104\ As part of the Argonne simulation effort, individual
technology combinations simulated in Autonomie were paired with
Argonne's BatPAC model to estimate the battery cost associated with
each technology combination based on characteristics of the
simulated vehicle and its level of electrification. Information
regarding Argonne's BatPAC model is available at http://www.cse.anl.gov/batpac/.
\105\ Furthermore, the impact of engine technologies on fuel
consumption, torque, and other metrics was characterized using GT
POWER simulation modeling in combination with other engine modeling
that was conducted by IAV Automotive Engineering, Inc. (IAV). The
engine characterization ``maps'' resulting from this analysis were
used as inputs for the Autonomie full-vehicle simulation modeling.
Information regarding GT Power is available at https://www.gtisoft.com/gt-suite-applications/propulsion-systems/gt-power-engine-simulation-software.
---------------------------------------------------------------------------
EPA developed two models after 2009, referred to as the ``ALPHA''
and ``OMEGA'' models, which provide some of the same capabilities as
the Autonomie and CAFE models. EPA applied the OMEGA model to conduct
analysis of tailpipe CO2 emissions standards promulgated in
2010 and 2012, and the ALPHA and OMEGA models to conduct analysis
discussed in the above-mentioned 2016 Draft TAR and Proposed and 2017
Initial Final Determinations regarding standards beyond 2021. In an
August 2017 notice, the agencies requested comments on, among other
things, whether EPA should use alternative methodologies and modeling,
including DOE/Argonne's Autonomie full-vehicle modeling and simulation
tool and DOT's CAFE model.\106\
---------------------------------------------------------------------------
\106\ 82 FR 39551, 39553 (Aug. 21, 2017).
---------------------------------------------------------------------------
Having reviewed comments on the subject and having considered the
matter fully, the agencies have determined it is reasonable and
appropriate to use DOE/Argonne's model for full-vehicle simulation, and
to use DOT's CAFE model for analysis of regulatory alternatives. EPA
interprets Section 202(a) of the CAA as giving the agency broad
discretion in how it develops and sets CO2 emissions
standards for light-duty vehicles. Nothing in Section 202(a) mandates
that EPA use any specific model or set of models for analysis of
potential CO2 standards for light-duty vehicles. EPA weighs
many factors when determining appropriate levels for CO2
standards, including the cost of compliance (see Section 202(a)(2)),
lead time necessary for compliance (id.), safety (see NRDC v. EPA, 655
F.2d 318, 336 n. 31 (D.C. Cir. 1981)) and other impacts on
consumers,\107\ and energy impacts associated with use of the
technology.\108\ Using the CAFE model allows consideration of a number
of factors. The CAFE model explicitly evaluates the cost of compliance
for each manufacturer, each fleet, and each model year; it accounts for
lead time necessary for compliance by directly incorporating estimated
manufacturer production cycles for every vehicle in the fleet, ensuring
that the analysis does not assume vehicles can be redesigned to
incorporate more technology without regard to lead time considerations;
it provides information on safety effects associated with different
levels of standards and information about many other impacts on
consumers, and it calculates energy impacts (i.e., fuel saved or
consumed) as a primary function, besides being capable of providing
information about many other factors within EPA's broad CAA discretion
to consider.
---------------------------------------------------------------------------
\107\ Since its earliest Title II regulations, EPA has
considered the safety of pollution control technologies. See 45 FR
14496, 14503 (1980).
\108\ See George E. Warren Corp. v. EPA, 159 F.3d 616, 623-624
(D.C. Cir. 1998) (ordinarily permissible for EPA to consider factors
not specifically enumerated in the Act).
---------------------------------------------------------------------------
Because the CAFE model simulates a wide range of actual constraints
and practices related to automotive engineering, planning, and
production, such as common vehicle platforms, sharing of engines among
different vehicle models, and timing of major vehicle redesigns, the
analysis produced by the CAFE model provides a transparent and
realistic basis to show pathways manufacturers could follow over time
in applying new technologies, which helps better assess impacts of
potential future standards. Furthermore, because the CAFE model also
accounts fully for regulatory compliance provisions (now including
CO2 compliance provisions), such as adjustments for reduced
refrigerant leakage, production ``multipliers'' for some specific types
of vehicles (e.g., PHEVs), and carried-forward (i.e., banked) credits,
the CAFE model provides a transparent and realistic basis to estimate
how such technologies might be applied over time in response to CAFE or
CO2 standards.
There are sound reasons for the agencies to use the CAFE model
going forward in this rulemaking. First, the CAFE and CO2
fact analyses are inextricably linked. Furthermore, the analysis
produced by the CAFE model and DOE/Argonne's Autonomie addresses the
agencies' analytical needs. The CAFE model provides an explicit year-
by-year simulation of manufacturers' application of technology to their
products in response to a year-by-year progression of CAFE standards
and accounts for sharing of technologies and the implications for
timing, scope, and limits on the potential to optimize powertrains for
fuel economy. In the real world, standards actually are specified on a
year-by-year basis, not simply some single year well into the future,
and manufacturers' year-by-year plans involve some vehicles ``carrying
forward'' technology from prior model years and some other vehicles
possibly applying ``extra'' technology in anticipation of standards in
ensuing model years, and manufacturers' planning also involves applying
credits carried forward between model years. Furthermore, manufacturers
cannot optimize the powertrain for fuel economy on every vehicle model
configuration--for example, a given engine shared among multiple
vehicle models cannot practicably be split into different versions for
each configuration of each model, each with a slightly different
displacement. The CAFE model is designed to account for these real-
world factors.
Considering the technological heterogeneity of manufacturers'
current product offerings, and the wide range of ways in which the many
fuel economy-improving/CO2 emissions-reducing technologies
included in the analysis can be combined, the CAFE model has been
designed to use inputs that provide an estimate of the fuel economy
achieved for many tens of thousands of different potential combinations
of fuel-saving technologies. Across the range of technology classes
encompassed by the analysis fleet, today's analysis involves more than
a million such estimates. While the CAFE model requires no specific
approach to developing these inputs, the National Academy of Sciences
(NAS) has recommended, and stakeholders have commented, that full-
vehicle simulation provides the best balance between realism and
practicality. DOE/Argonne has spent several years developing, applying,
and expanding means to use distributed computing to exercise its
Autonomie full-vehicle modeling and simulation tool over the scale
necessary for realistic analysis of CAFE or average tailpipe
CO2 emissions standards. This scalability and related
flexibility (in terms of expanding the set of technologies to be
simulated) makes Autonomie well-suited for developing inputs to the
CAFE model.
In addition, DOE/Argonne's Autonomie also has a long history of
development and widespread application by a much wider range of users
in government, academia, and industry. Many of these users apply
[[Page 24223]]
Autonomie to inform funding and design decisions. These real-world
exercises have contributed significantly to aspects of Autonomie
important to producing realistic estimates of fuel economy levels and
CO2 emission rates, such as estimation and consideration of
performance, utility, and driveability metrics (e.g., towing
capability, shift business, frequency of engine on/off transitions).
This steadily increasing realism has, in turn, steadily increased
confidence in the appropriateness of using Autonomie to make
significant investment decisions. Notably, DOE uses Autonomie for
analysis supporting budget priorities and plans for programs managed by
its Vehicle Technologies Office (VTO). Considering the advantages of
DOE/Argonne's Autonomie model, it is reasonable and appropriate to use
Autonomie to estimate fuel economy levels and CO2 emission
rates for different combinations of technologies as applied to
different types of vehicles.
Commenters have also suggested that the CAFE model's graphical user
interface (GUI) facilitates others' ability to use the model quickly--
and without specialized knowledge or training--and to comment
accordingly.\109\ For the NPRM, NHTSA significantly expanded and
refined this GUI, providing the ability to observe the model's real-
time progress much more closely as it simulates year-by-year compliance
with either CAFE or CO2 standards.\110\ Although the model's
ability to produce realistic results is independent of the model's GUI,
the CAFE model's GUI appears to have facilitated stakeholders'
meaningful review and comment during the comment period.
---------------------------------------------------------------------------
\109\ From Docket Number EPA-HQ-OAR-2015-0827, see Comment by
Global Automakers, Docket ID EPA-HQ-OAR-2015-0827-9728, at 34.
\110\ The updated GUI provides a range of graphs updated in real
time as the model operates. These graphs can be used to monitor fuel
economy or CO2 ratings of vehicles in manufacturers'
fleets and to monitor year-by-year CAFE (or average CO2
ratings), costs, avoided fuel outlays, and avoided CO2-
related damages for specific manufacturers and/or specific fleets
(e.g., domestic passenger car, light truck). Because these graphs
update as the model progresses, they should greatly increase users'
understanding of the model's approach to considerations such as
multiyear planning, payment of civil penalties, and credit use.
---------------------------------------------------------------------------
The question of whether EPA's actions should consider and be
informed by analysis using non-EPA-staff-developed modeling tools has
generated considerable debate over time. Even prior to the NPRM,
certain commenters had argued that EPA could not consider, in setting
tailpipe CO2 emissions standards, any information derived
from non-EPA-staff-developed modeling. Many of the pre-NPRM concerns
focused on inputs used by the CAFE model for prior rulemaking
analyses.111 112 113 Because inputs are exogenous to any
model, they do not determine whether it would be reasonable and
appropriate for EPA to use NHTSA's model for analysis. Other concerns
focused on certain characteristics of the CAFE model that were
developed to align the model better with EPCA and EISA. The model has
been revised to accommodate both EPCA/EISA and CAA analysis, as
explained further below. Some commenters also argued that use of any
models other than ALPHA and OMEGA for CAA analysis would constitute an
arbitrary and capricious delegation of EPA's decision-making authority
to NHTSA, if NHTSA models are used for analysis instead.\114\ As
discussed above, the CAFE Model--as with any model--is used to provide
analysis, and does not result in decisions. Decisions are made by EPA
in a manner that is informed by modeling outputs, sensitivity cases,
public comments, any many other pieces of information.
---------------------------------------------------------------------------
\111\ For example, EDF previously stated that ``the data that
NHTSA needs to input into its model is sensitive confidential
business information that is not transparent and cannot be
independently verified, . . .'' and it claimed ``the OMEGA model's
focus on direct technological inputs and costs--as opposed to
industry self-reported data--ensures the model more accurately
characterizes the true feasibility and cost effectiveness of
deploying greenhouse gas reducing technologies.'' EDF, EPA-HQ-OAR-
2015-0827-9203, at 12. These statements are not correct, as nothing
about either the CAFE or OMEGA model either obviates or necessitates
the use of CBI to develop inputs.
\112\ As another example, CARB previously stated that ``another
promising technology entering the market was not even included in
the NHTSA compliance modeling'' and that EPA assumes a five-year
redesign cycle, whereas NHTSA assumes a six to seven-year cycle.''
CARB, EPA-HQ-OAR-2015-0827-9197, at 28. Though presented as
criticisms of the models, these comments--at least with respect to
the CAFE model--actually concern model inputs. NHTSA did not agree
with CARB about the commercialization potential of the engine
technology in question (``Atkinson 2'') and applied model inputs
accordingly. Also, rather than applying a one-size-fits-all
assumption regarding redesign cadence, NHTSA developed estimates
specific to each vehicle model and applied these as model inputs.
\113\ As another example, NRDC has argued that EPA should not
use the CAFE model because it ``allows manufacturers to pay civil
penalties in lieu of meeting the standards, an alternative
compliance pathway currently allowed under EISA and EPCA.'' NRDC,
EPA-HQ-OAR-2015-0827-9826, at 37. While the CAFE model can simulate
civil penalty payment, NRDC's comment appears to overlook the fact
that this result depends on model inputs; the inputs can easily be
specified such that the CAFE model will set aside civil penalty
payment as an alternative to compliance.
\114\ See, e.g., CBD et al., NHTSA-2018-0067-12057, at 9.
---------------------------------------------------------------------------
Comments responding to the NPRM's use of the CAFE model and
Autonomie rather than also (for CO2 standards) ALPHA and
OMEGA were mixed. For example, the environmental group coalition stated
that the CAFE model is aligned with EPCA requirements,\115\ but also
argued (1) that EPA is legally prohibited from ``delegat[ing] technical
decision-making to NHTSA;'' \116\ (2) that ``EPA must exercise its
technical and scientific expertise'' to develop CO2
standards and ``Anything less is an unlawful abdication of EPA's
statutory responsibilities;'' \117\ (3) that EPA staff is much more
qualified than DOT staff to conduct analysis relating to standards and
has done a great deal of work to inform development of standards; \118\
(4) that ``The Draft TAR and 2017 Final Determination relied
extensively on use of sophisticated EPA analytic tools and
methodologies,'' i.e., the ``peer reviewed simulation model ALPHA,''
``the agency's vehicle teardown studies,'' and the ``peer-reviewed
OMEGA model to make reasonable estimates of how manufacturers could add
technologies to vehicles in order to meet a fleet-wide [CO2
emissions] standard;'' \119\ (5) that NHTSA had said in the MYs 2012-
2016 rulemaking that the Volpe [CAFE] model was developed to support
CAFE rulemaking and incorporates features ``that are not appropriate
for use by EPA in setting [tailpipe CO2] standards;'' \120\
(6) allegations that some EPA staff had disagreed with aspects of the
NPRM analysis and had requested that ``EPA's name and logo should be
removed from the DOT-NHTSA Preliminary Regulatory Impact Analysis
document'' and stated that ``EPA is relying upon the technical analysis
performed by DOT-NHTSA for the [NPRM];'' \121\ (7) that EPA had
developed ``a range of relevant new analysis'' that the proposal
``failed to consider,'' including ``over a dozen 2017 and 2018 EPA peer
reviewed SAE articles;'' \122\ (8) that EPA's OMEGA modeling undertaken
during NPRM development ``found costs half that of NHTSA's findings,''
``Yet NHTSA did not correct the errors in its modeling and analysis,
and the published proposal drastically overestimates the cost of
complying . . . .;'' \123\ (9) that some EPA staff had requested that
the technology ``HCR2'' be included in the NPRM analysis, ``Yet NHTSA
overruled
[[Page 24224]]
EPA and omitted the technology;'' \124\ (10) that certain EPA staff had
initially ``rejected use of the CAFE model for development of the
proposed [tailpipe CO2] standards;'' \125\ (11) that there
are ``many specific weaknesses of the modeling results derived in this
proposal through use of the Volpe and Autonomie models'' and that the
CAFE model is ``not designed in accordance with'' Section 202(a) of the
CAA because (A) EPA ``is not required to demonstrate that standards are
set at the maximum feasible level year-by-year,'' (B) because EPCA
``preclude[s NHTSA] from considering vehicles powered by fuels other
than gas or diesel'' and EPA is not similarly bound, and (C) because
the CAFE model assumed that the value of an overcompliance credit
equaled $5.50, the value of a CAFE penalty.\126\ Because of all of
these things, the environmental group coalition stated that the
proposal was ``unlawful'' and that ``Before proceeding with this
rulemaking, EPA must consider all relevant materials including these
excluded insights, perform its own analysis, and issue a reproposal to
allow for public comment.'' \127\
---------------------------------------------------------------------------
\115\ Environmental group coalition, NHTSA-2018-0067-12000,
Appendix A, at 24-25.
\116\ Id. at 12.
\117\ Id. at 14.
\118\ Id. at 15-17.
\119\ Id. at 17.
\120\ Id. at 18.
\121\ Id. at 19.
\122\ Id. at 20.
\123\ Id. at 21.
\124\ Id. at 21-22.
\125\ Id. at 23.
\126\ Id. at 24-25.
\127\ Id. at 27.
---------------------------------------------------------------------------
Some environmental organizations and States contracted for external
technical analyses augmenting general comments such as those summarized
above. EDF engaged a consultant, Richard Rykowski, for a detailed
review of the agencies' analysis.\128\ Among Mr. Rykowski's comments, a
few specifically involve differences between these two models. Mr.
Rykowski recommended NHTSA's CAFE model replace its existing
``effective cost'' metric (used to compare available options to add
specific technologies to specific vehicles) with a ``ranking factor''
used for the same purpose. As discussed below in Section VI.A, the
model for today's final rule adopts this recommendation. He also states
that (1) ``EPA has developed a better way to isolate and reject cost
ineffective combinations of technologies . . . [and] includes only
these 50 or so technology combinations in their OMEGA model runs;'' (2)
``NHTSA's arbitrary and rigid designation of leader-follower vehicles
for engine, transmission and platform level technologies
unrealistically slows the rollout of technology into the new vehicle
fleet;'' (3) ``the Volpe Model is not capable of reasonably simulating
manufacturers' ability to utilize CO2 credits to smooth the
introduction of technology throughout their vehicle line-up;'' and (4)
``the Volpe Model is not designed to reflect the use of these [A/C
leakage] technologies and refrigerants.'' \129\
---------------------------------------------------------------------------
\128\ EDF, NHTSA-2018-0067-12108, Appendix B. See also EPA, Peer
Review of the Optimization Model for Reducing Emissions of
Greenhouse Gases from Automobiles (OMEGA) and EPA's Response to
Comments, EPA-420-R-09-016, September 2009.
\129\ EDF, op. cit., at 73-75.
---------------------------------------------------------------------------
Mr. Rogers's analysis focuses primarily on the agencies' published
analysis, but mentions that some engine ``maps'' (estimates--used as
inputs to full vehicle simulation--of engine fuel consumption under a
wide range of engine operating conditions) applied in Autonomie show
greater fuel consumption benefits of turbocharging than those applied
previously by EPA to EPA's ALPHA model, and these benefits could have
caused NHTSA's CAFE model to estimate an unrealistically great tendency
toward turbocharged engines (rather than high compression ratio
engines).\130\ Mr. Rogers also presents alternative examples of year-
by-year technology application to specific vehicle models, contrasting
these with year-by-year results from the agencies' NPRM analysis,
concluding that ``that the use of logical, unrestricted technology
pathways, with incremental benefits supported by industry-accepted
vehicle simulation and dynamic system optimization and calibration,
together with publicly-defensible costs, allows cost-effective
solutions to achieve target fuel economy levels which meet MY 2025
existing standards.'' \131\
---------------------------------------------------------------------------
\130\ Roush Industries, NHTSA-2018-0067-11984, at 17-21.
\131\ Roush Industries, NHTSA-2018-0067-11984, at 17-30.
---------------------------------------------------------------------------
Mr. Duleep's analysis also focuses primarily on the agencies'
published analysis, but does mention that (1) ``the Autonomie modeling
assumes no engine change when drag and rolling resistance reductions
are implemented, as well as no changes to the transmission gear ratios
and axle ratios, . . . [but] the EPA ALPHA model adjusts for this
effect;'' (2) ``baseline differences in fuel economy [between two
manufacturers' different products using similar technologies] are
carried for all future years and this exaggerates the differences in
technology adoption requirements and costs between manufacturers; (3)
``assumptions [that most technology changes are best applied as part of
a vehicle redesign or freshening] result in unnecessary distortion in
technology paths and may bias results of costs for different
manufacturers;'' and (4) that for the sample results shown for the
Chevrolet Equinox ``the publicly available EPA lumped parameter model
(which was used to support the 2016 rulemaking) and 2016 TAR cost data
. . . results in an estimate of attaining 52.2 mpg for a cost of $2110,
which is less than half the cost estimated in the PRIA.'' \132\
---------------------------------------------------------------------------
\132\ H-D Systems, op. cit., at 48, et seq.
---------------------------------------------------------------------------
Beyond these comments regarding differences between EPA's models
and the Argonne and DOT models applied for the NPRM, these and other
technical reviewers had many specific comments about the agencies'
analysis for the NPRM, and these comments are discussed in detail below
in Section VI.B.
Manufacturers, on the other hand, supported the agencies' use of
Autonomie and the CAFE model rather than, in EPA's case, the ALPHA and
OMEGA models. Expressly identifying the distinction between models and
model inputs, Global Automakers stated that:
The agencies provided a new, updated analysis based on the most
up-to-date data, using a proven and long-developed modeling tool,
known as the Volpe model, and offering numerous options to best
determine the right regulatory and policy path for ongoing fuel
efficiency improvements in our nation. Now, all stakeholders have an
opportunity to come to the table as part of the public process to
provide input, data, and information to help shape the final
rule.\133\
---------------------------------------------------------------------------
\133\ Global Automakers, NHTSA-2018-0067-12032, at 2.
---------------------------------------------------------------------------
This NPRM's use of a single model to evaluate alternative
scenarios for both programs provides consistency in the technical
analysis, and Global Automakers supports the Volpe model's use as it
has proven to be a transparent and user-friendly option in this
current analysis. The use of the Volpe model has allowed for a broad
range of stakeholders, with varying degrees of technical expertise,
to review the data inputs to provide feedback on this proposed rule.
The Volpe model's accompanying documentation has historically
provided a clear explanation of all sources of input and constraints
critical to a transparent modeling process. Other inputs have come
from modeling that is used widely by other sources, specifically the
Autonomie model, allowing for a robust validation, review and
reassessment.\134\
---------------------------------------------------------------------------
\134\ Global Automakers, NHTSA-2018-0067-12032, Attachment A, at
A-12.
The Alliance commented, similarly, that ``at least at this time,
NHTSA's modeling systems are superior to EPA's'' and ``as such, we
support the Agencies' decision to use NHTSA's modeling tools for this
rulemaking and recommend that both Agencies continue on this path. We
encourage Agencies to work together to provide input to the single
common set of tools.'' \135\
---------------------------------------------------------------------------
\135\ Alliance, NHTSA-2018-0067-12073, at 134.
---------------------------------------------------------------------------
[[Page 24225]]
Regarding the agencies' use of Argonne's Autonomie model rather
than EPA's ALPHA model, the Alliance commented that (1) ``the benefits
of virtually all technologies and their synergistic effects are now
determined with full vehicle simulations;'' (2) ``vehicle categories
have been increased to 10 to better recognize the range of 0-60
performance characteristics within each of the 5 previous categories,
in recognition of the fact that many vehicles in the baseline fleet
significantly exceeded the previously assumed 0-60 performance metrics.
This provides better resolution of the baseline fleet and more accurate
estimates of the benefits of technology. . . .;'' (3) ``new
technologies (like advanced cylinder deactivation) are included, while
unproven combinations (like Atkinson engines with 14:1 compression,
cooled EGR, and cylinder deactivation in combination) have been
removed;'' (4) ``Consistent with the recommendation of the National
Academy of Sciences and manufacturers, gradeability has been included
as a performance metric used in engine sizing. This helps prevent the
inclusion of small displacement engines that are not commercially
viable and that would artificially inflate fuel savings;'' (5) ``the
Alliance believes NHTSA's tools (Autonomie/Volpe) are superior to EPA's
(APLHA[sic]/LPM/OMEGA). This is not surprising since NHTSA's tools have
had a significant head start in development. . . .'' (6) ``the
Autonomie model was developed at Argonne National Lab with funding from
the Department of Energy going back to the PNGV (Partnership for Next
Generation Vehicles) program in the 1990s. Autonomie was developed from
the start to address the complex task of combining 2 power sources in a
hybrid powertrain. It is a physics-based, forward looking, vehicle
simulator, fully documented with available training,'' and (7) ``EPA's
ALPHA model is also a physics-based, forward looking, vehicle
simulator. However, it has not been validated or used to simulate
hybrid powertrains. The model has not been documented with any
instructions making it difficult for users outside of EPA to run and
interpret the model.'' \136\
---------------------------------------------------------------------------
\136\ Id. at 135.
---------------------------------------------------------------------------
Regarding the use of NHTSA's CAFE model rather than EPA's OMEGA
model, the Alliance stated that (1) NHTSA's model appropriately
differentiate between domestic and imported automobiles; (2) in NHTSA's
model, ``dynamic estimates of vehicle sales and scrappage in response
to price changes replace unrealistic static sales and scrappage
numbers;'' (3) NHTSA's model ``has new capability to perform
[CO2 emissions] analysis with [tailpipe CO2]
program flexibilities;'' (4) ``the baseline fleet [used in NHTSA's
model] has been appropriately updated based on both public and
manufacturer data to reflect the technologies already applied,
particularly tire rolling resistance;'' and (5) ``some technologies
have been appropriately restricted. For example, low rolling resistance
tires are no longer allowed on performance vehicles, and aero
improvements are limited to maximum levels of 15% for trucks and 10%
for minivans.'' \137\ The Alliance continued, noting that ``NHTSA's
Volpe model also predates EPA's OMEGA model. More importantly, the new
Volpe model considers several factors that make its results more
realistic.'' \138\ As factors leading the Volpe model to produce
results that are more realistic than those produced by OMEGA, the
Alliance commented that (1) ``The Volpe model includes estimates of the
redesign and refresh schedules of vehicles based on historical trends,
whereas the OMEGA model uses a fixed, and too short, time interval
during which all vehicles are assumed to be fully redesigned. . . .;''
(2) ``The Volpe model allows users to phase-in technology based on year
of availability, platform technology sharing, phase-in caps, and to
follow logical technology paths per vehicle. . . .;'' (3) ``The Volpe
model produces a year-by year analysis from the baseline model year
through many years in the future, whereas the OMEGA model only analyzes
a fixed time interval. . . .;'' (4) ``The Volpe model recognizes that
vehicles share platforms, engines, and transmissions, and that
improvements to any one of them will likely extend to other vehicles
that use them'' whereas ``The OMEGA model treats each vehicle as an
independent entity. . . .;'' (5) ``The Volpe model now includes sales
and scrappage effects;'' and (6) ``The Volpe model is now capable of
analyzing for CAFE and [tailpipe CO2] compliance, each with
unique program restrictions and flexibilities.'' \139\ The Alliance
also incorporated by reference concerns it raised regarding EPA's
OMEGA-based analysis supporting EPA's proposed and prior final
determinations.\140\
---------------------------------------------------------------------------
\137\ Id. at 134.
\138\ Id. at 135.
\139\ Id. at 135-136.
\140\ Id. at 136.
---------------------------------------------------------------------------
The Alliance further stated that ``For all of the above reasons and
to avoid duplicate efforts, the Alliance recommends that the Agencies
continue to use DOT's Volpe and Autonomie modeling system, rather than
continuing to develop two separate systems. EPA has demonstrated
through supporting Volpe model code revisions and by supplying engine
maps for use in the Autonomie model that their expertise can be
properly represented in the rulemaking process without having to
develop separate or new tools.'' \141\
---------------------------------------------------------------------------
\141\ Id. at 136.
---------------------------------------------------------------------------
Some individual manufacturers provided comments supporting and
elaborating on the above comments by Global Automakers and the
Alliance. For example, FCA commented that ``the modeling performed by
the agencies should illuminate the differences between the CAFE and
[tailpipe CO2 emissions] programs. This cannot be
accomplished when each agency is using different tools and assumptions.
Since we believe NHTSA possesses the better set of tools, we support
both agencies using Autonomie for vehicle modeling and Volpe (CAFE) for
fleet modeling.'' \142\
---------------------------------------------------------------------------
\142\ FCA, NHTSA-2018-0067-11943, at 82.
---------------------------------------------------------------------------
Honda stated that ``The current version of the CAFE model is
reasonably accurate in terms of technology efficiency, cost, and
overall compliance considerations, and reflects a notable improvement
over previous agency modeling efforts conducted over the past few
years. We found the CAFE model's characterization of Honda's
``baseline'' fleet--critical modeling minutiae that provide a technical
foundation of the agencies' analysis--to be highly accurate. We commend
NHTSA and Volpe Center staff on these updates, as well as on the
overall transparency of the model. The model's graphical user interface
(GUI) makes it easier to run, model functionality is thoroughly
documented, and the use of logical, traceable input and output files
accommodates easy tracking of results.'' \143\ Similarly, in an earlier
presentation to the agencies, Honda included the following slide
comparing EPA's OMEGA model to DOT's CAFE (Volpe) model, and making
recommendations regarding future improvements to the latter: \144\
---------------------------------------------------------------------------
\143\ Honda, EPA-HQ-OAR-2018-0283, at 21-22.
\144\ Honda, NHTSA-2018-0067-12019, at 12.
---------------------------------------------------------------------------
[[Page 24226]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.041
Toyota, in addition to arguing that the agencies' application of
model inputs (e.g., an analysis fleet based on MY 2016 compliance data)
produced more realistic results than in the draft TAR and in EPA's
former proposed and final determinations, also stressed the importance
of the CAFE model's year-by-year accounting for product redesigns,
stating that this produces more realistic results than the OMEGA-based
results shown previously by EPA:
The modeling now better accounts for factors that limit the rate
at which new technologies enter and then diffuse through a
manufacturer's fleet. Bringing new or improved vehicles and
technologies to market is a several-year, capital-intensive
undertaking. Once new designs are introduced, a period of stability
is required so investments can be amortized. Vehicle and technology
introductions are staggered over time to manage limited resources.
Agency modeling now better recognizes the inherent constraints
imposed by realities that dictate product cadence. We agree with the
agencies' understanding that ``the simulation of compliance actions
that manufacturers might take is constrained by the pace at which
new technologies can be applied in the new vehicle market,'' and we
are encouraged to learn that ``agency modeling can now account for
the fact that individual vehicle models undergo significant
redesigns relatively infrequently.'' The preamble correctly notes
that manufacturers try to keep costs down by applying most major
changes mainly during vehicle redesigns and more modest changes
during product refresh, and that redesign cycles for vehicle models
can range from six to ten years, and eight to ten-years for
powertrains. This appreciation for standard business practice
enables the modeling to more accurately capture the way vehicles
share engines, transmissions, and platforms. There are now more
realistic limits placed on the number of engines and transmissions
in a powertrain portfolio which better recognizes manufacturers must
manage limited engineering resources and control supplier,
production, and service costs. Technology sharing and inheritance
between vehicle models tends to limit the rate of improvement in a
manufacturer's fleet.\145\
---------------------------------------------------------------------------
\145\ Toyota, NHTSA-2018-0067-12098, Attachment 1, at 3 et seq.
These comments urging EPA to use NHTSA's CAFE model echo comments
provided in response to a 2018 peer review of the model. While
identifying various opportunities for improvement, peer reviewers
expressed strong overall support for the CAFE model's technical
approach and execution. For example, one reviewer, after offering many
---------------------------------------------------------------------------
specific technical recommendations, concluded as follows:
The model is impressive in its detail, and in the completeness
of the input data that it uses. Although the model is complex, the
reader is given a clear account of how variables are variously
divided and combined to yield appropriate granularity and efficiency
within the model. The model tracks well a simplified version of the
real-world and manufacturing/design decisions. The progression of
technology choices and cost benefit choices is clear and logical. In
a few cases, the model simply explains a constraint, or a value
assigned to a variable, without defending the choice of the value or
commenting on real-world variability, but these are not substantive
omissions. The model will lend itself well to future adaptation or
addition of variables, technologies and pathways.\146\
---------------------------------------------------------------------------
\146\ NHTSA, CAFE Model Peer Review, DOT HS 812 590, Available
at https://www.nhtsa.gov/document/cafe-model-peer-review, at 250.
Although the peer review charge focused solely on the CAFE model,
another peer reviewer separately recommended that EPA ``consider
opportunities for EPA to use the output from the Volpe Model in place
of their OMEGA Model output'' \147\
---------------------------------------------------------------------------
\147\ Id. at 287-288 and 304.
---------------------------------------------------------------------------
More recently, in response to the NPRM, Dr. Julian Morris, an
economist at George Washington University, commented extensively on the
superiority of the agencies' NPRM analysis to previous analyses,
offering the following overall assessment:
I have assessed the plausibility of the analyses undertaken by
NHTSA and EPA in relation to the proposed SAFE rule. I found that
the agencies have undertaken a thorough--one might even say
exemplary--analysis, improving considerably on earlier analyses
undertaken by the agencies of previous rules relating to CAFE
standards and [tailpipe CO2] emission standards. Of
particular note, the agencies included more realistic estimates of
the rebound effect, developed a sophisticated model of the
[[Page 24227]]
scrappage effect, and better accounted for various factors affecting
vehicle fatality rates.\148\
---------------------------------------------------------------------------
\148\ Morris, J., OAR-2018-0283-4028, at 6-11.
The agencies carefully considered these and other comments
regarding which models to apply when estimating potential impacts of
each of the contemplated regulatory alternatives. For purposes of
estimating the impacts of CAFE standards, even the coalition of
environmental advocates observed that the CAFE model reflects EPCA's
requirements. As discussed below in Section VI.A, EPCA imposes specific
requirements not only on how CAFE standards are to be structured (e.g.,
including a minimum standard for domestic passenger cars), but also on
how CAFE standards are to be evaluated (e.g., requiring that the
potential to produce additional AFVs be set aside for the model years
under consideration), and the CAFE model reflects these requirements,
and the agencies consider the CAFE model to be the best available tool
for CAFE rulemaking analysis. Regarding the use of Autonomie to
construct fuel consumption (i.e., efficiency) inputs to the CAFE model,
the agencies recognize that other vehicle simulation tools are
available, including EPA's recently-developed ALPHA model. However, as
also discussed in Section VI.B.3, Autonomie has a much longer history
of development and refinement, and has been much more widely applied
and validated. Moreover, Argonne experts have worked carefully for
several years to develop methods for running large arrays of
simulations expressly structured and calibrated for use in DOT's CAFE
model. Therefore, the agencies consider Autonomie to be the best
available tool for constructing such inputs to the CAFE model. While
the agencies have also carefully considered potential specific model
refinements, as well as the merits of potential changes to model inputs
and assumptions, none of these potential refinements and input have led
either agency to reconsider using the CAFE model and Autonomie for CAFE
rulemaking analysis.
With respect to estimating the impacts of CO2 standards,
even though Argonne and the agencies have adapted Autonomie and the
CAFE model to support the analysis of CO2 standards,
environmental groups, California, and other States would prefer that
EPA use the models it developed during 2009-2018 for that purpose.\149\
Arguments that EPA revert to its ALPHA and OMEGA models fall within
three general categories: (1) Arguments that EPA's models would have
selected what commenters consider better (i.e., generally more
stringent) standards, (2) arguments that EPA's models are technically
superior, and (3) arguments that the law requires EPA use its own
models.
---------------------------------------------------------------------------
\149\ The last-finalized versions of EPA's OMEGA model and ALPHA
tools were published in 2016 and 2017, respectively.
---------------------------------------------------------------------------
The first of these arguments--that EPA's models would have selected
better standards--conflates the analytical tool used to inform
decision-making with the action of making the decision. As explained
elsewhere in this document and as made repeatedly clear over the past
several rulemakings, the CAFE model (or, for that matter, any model)
neither sets standards nor dictates where and how to set standards; it
simply informs as to the potential effects of setting different levels
of standards. In this rulemaking, EPA has made its own decisions
regarding what CO2 standards would be appropriate under the
CAA.
The third of these arguments--that EPA is legally required to use
only models developed by its own staff--is also without merit. The CAA
does not require the agency to create or use a specific model of its
own creation in setting tailpipe CO2 standards. The fact
that EPA's decision may be informed by non-EPA-created models does not,
in any way, constitute a delegation of its statutory power to set
standards or decision-making authority.\150\ Arguing to the contrary
would suggest, for example, that EPA's decision would be invalid
because it relied on EIA's Annual Energy Outlook for fuel prices for
all of its regulatory actions rather than developing its own model for
estimating future trends in fuel prices. Yet, all Federal agencies that
have occasion to use forecasts of future fuel prices regularly (and
appropriately) defer to EIA's expertise in this area and rely on EIA's
NEMS-based analysis in the AEO, even when those same agencies are using
EIA's forecasts to inform their own decision-making. Similarly, this
argument would mean that the agencies could not rely on work done by
contractors or other outside consultants, which is contrary to regular
agency practice across the entirety of the Federal Government.
---------------------------------------------------------------------------
\150\ ``[A] federal agency may turn to an outside entity for
advice and policy recommendations, provided the agency makes the
final decisions itself.'' U.S. Telecom. Ass'n v. FCC, 359 F.3d 554,
565-66 (D.C. Cir. 2004). To the extent commenters meant to suggest
outside parties have a reliance interest in EPA using ALPHA and
OMEGA to set standards, EPA and NHTSA do not agree a reliance
interest is properly placed on an analytical methodology, which
consistently evolves from rule to rule. Even if it were, all parties
that closely examined ALPHA and OMEGA-based analyses in the past
either also simultaneously closely examined CAFE and Autonomie-based
analyses in the past, or were fully capable of doing so, and thus,
should face no additional difficulty now they have only one set of
models and inputs/outputs to examine.
---------------------------------------------------------------------------
The specific claim here that use of the CAFE model instead of ALPHA
and OMEGA is somehow illegitimate is similarly unpersuasive. The CAFE
and CO2 rules have, since Massachusetts v. EPA, all been
issued as joint rulemakings, and, thus are the result of a
collaboration between the two agencies. This was true when the
rulemakings used separate models for the different programs and
continues to be true in today's final rule, where the agencies take the
next step in their collaborative approach by now using simply one model
to simulate both programs. In 2007, immediately following this Supreme
Court decision, the agencies worked together toward standards for model
years 2011-2015, and EPA made use of the CAFE model for its work toward
possible future CO2 standards. That the agencies would need
to continue the unnecessary and inefficient process of using two
separate combinations of models as the joint National Program continues
to mature, therefore, runs against the idea that the agencies, over
time, would best combine resources to create an efficient and robust
regulatory program. For the reasons discussed throughout today's final
rule, the agencies have jointly determined that Autonomie and the CAFE
model have significant technical advantages, including important
additional features, and are therefore the more appropriate models to
use to support both analyses.
Further, the fact that Autonomie and CAFE models were initially
developed by DOE/Argonne and NHTSA does not mean that EPA has no role
in either these models or their inputs. That is, the development
process for CAFE and CO2 standards inherently requires
technical and policy examinations and deliberations between staff
experts and decision-makers in both agencies. Such engagements are a
healthy and important part of any rulemaking activity--and particularly
so with joint rulemakings. The Supreme Court stated in Massachusetts v.
EPA that, ``The two obligations [to set CAFE standards under EPCA and
to set tailpipe CO2 emissions standards under the CAA] may
overlap, but there is no reason to think the two agencies cannot both
administer their obligations and yet
[[Page 24228]]
avoid inconsistency.'' \151\ When agency experts consider analytical
issues and agency decision-makers decide on policy, which is informed
(albeit not dictated) by the outcome of that work, they are working
together as the Court appears to have intended in 2007, even if
legislators' intentions have varied in the decades since EPCA and the
CAA have been in place.\152\ Regulatory overlap necessarily involves
deliberation, which can lead to a more balanced, reasonable, and
improved analyses, and better regulatory outcomes. It did here. The
existence of deliberation is not per se evidence of unreasonableness,
even if some commenters believe a different or preferred policy outcome
would or should have resulted.\153\
---------------------------------------------------------------------------
\151\ Massachusetts v. EPA, 549 U.S. 497, 532 (2007).
\152\ For example, when wide-ranging amendments to the CAA were
being debated, S. 1630 contained provisions that, if enacted, would
have authorized automotive CO2 emissions standards and
prescribed specific average levels to be achieved by 1996 and 2000.
In a letter to Senators, then-Administrator William K. Reilly noted
that the Bill ``requires for the first time control of emissions of
carbon dioxide; this is essentially a requirement to improve fuel
efficiency'' and outlined four reasons the H.W. Bush Administration
opposed the requirement, including that ``it is inappropriate to add
this very complex issue to the Clean Air Act which is already full
of complicated and controversial issues.'' Reilly, W., Letter to
U.S. Senators (January 26, 1990). The CAA amendments ultimately
signed into law did not contain these or any other provisions
regarding regulation of CO2 emissions.
\153\ See, e.g., U.S. House of Representatives, Committee on
Oversight and Government Reform, Staff Report, 112th Congress, ``A
Dismissal of Safety, Choice, and Cost: The Obama Administration's
New Auto Regulations,'' August 10, 2012, at 19-21 and 33-34.
---------------------------------------------------------------------------
Over the 44 years since EPCA established the requirement for CAFE
standards, NHTSA, EPA and DOE career staff have discussed, collaborated
on, and debated engineering, economic, and other aspects of CAFE
regulation, through focused meetings and projects, informal exchanges,
publications, conferences and workshops, and rulemakings.
Part of this expanded exchange has involved full vehicle
simulation. While tools such as PSAT (the DOE-sponsored simulation tool
that predated Autonomie) were in use prior to 2007, including for
discrete engineering studies supporting inputs to CAFE rulemaking
analyses, these tools' information and computing requirements were such
that NHTSA had determined (and DOE and EPA had concurred) that it was
impractical to more fully integrate full vehicle simulation into
rulemaking analyses. Since that time, computing capabilities have
advanced dramatically, and the agencies now agree that such integration
of full vehicle simulation--such as the large-scale exercise of
Autonomie to produce inputs to the CAFE Model--can make for more robust
CAFE and CO2 rulemaking analysis. This is not to say,
though, that experts always agree on all methods and inputs involved
with full vehicle simulation. Differences in approach and inputs lead
to differences in results. For example, compared to other publicly
available tools that can be practicably exercised at the scale relevant
to fleetwide analysis needed for CAFE and CO2 rulemaking
analysis, DOE/Argonne's Autonomie model is more advanced, spans a wider
range of fuel-saving technologies, and represents them in more specific
detail, leaving fewer ``gaps'' to be filled with other models (risking
inconsistencies and accompanying errors). These differences discussed
in greater detail below in Section VI.B.3. Perhaps most importantly,
the CAFE model considers fuel prices in determining both which
technologies are applied and the total amount of technology applied, in
the case where market forces demand fuel economy levels in excess of
the standards. While OMEGA can apply technology in consideration of
fuel prices, OMEGA will apply technology to reach the same level of
fuel economy (or CO2 emissions) if fuel prices are 3, 5, or
20 dollars, which violates the SAB's requirement that the analysis
``account for [. . .] future fuel prices .'' \154\ Furthermore, it
produces a counterintuitive result. If fuel prices become exorbitantly
high, we would expect consumers to place an emphasis on additional fuel
efficiency as the potential for extra fuel savings is tremendous.
---------------------------------------------------------------------------
\154\ See SAB Report 10 (``Constructing each of the scenarios is
challenging and involve extensive scientific, engineering, and
economic uncertainties. Projecting the baseline requires the
agencies to account for a wide range of variables including: The
number of new vehicles sold, future fuel prices,. . . .'').
---------------------------------------------------------------------------
Moreover, DOE has for many years used Autonomie (and its precursor
model, PSAT) to produce analysis supporting fuel economy-related
research and development programs involving billions of dollars of
public investment, and NHTSA's CAFE model with inputs from DOE/
Argonne's Autonomie model has produced analysis supporting rulemaking
under the CAA. In 2015, EPA proposed new tailpipe CO2
standards for MY 2021-2027 heavy-duty pickups and vans, finalizing
those standards in 2016. Supporting the NPRM and final rule, EPA relied
on analysis implemented by NHTSA using NHTSA's CAFE model, and NHTSA
used inputs developed by DOE/Argonne using DOE/Argonne's Autonomie
model. CBD questioned this history, asserting that, ``EPA conducted a
separate analysis using a different iteration of the CAFE model rather
than rely on the version which NHTSA used, again resulting and parallel
but corroborative modeling results.'' \155\ CBD's comment
mischaracterizes EPA's actual use of the CAFE Model. As explained in
the final rule, EPA's ``Method B'' analysis was developed as follows:
---------------------------------------------------------------------------
\155\ CBD, et al., 2018-0067-12000, Appendix A, at 27.
In Method B, the CAFE model from the NPRM was used to project a
pathway the industry could use to comply with each regulatory
alternative, along with resultant impacts on per-vehicle costs.
However, the MOVES model was used to calculate corresponding changes
in total fuel consumption and annual emissions for pickups and vans
in Method B. Additional calculations were performed to determine
corresponding monetized program costs and benefits.\156\
---------------------------------------------------------------------------
\156\ 81 FR 73478, 73506-07 (October 25, 2016).
In other words, a version of NHTSA's CAFE Model was used to perform
the challenging part of the analysis--that is, the part that involves
accounting for manufacturers' fleets, accounting for available fuel-
saving technologies, accounting for standards under consideration, and
estimating manufacturers' potential responses to new standards--EPA's
MOVES model was used to perform ``downstream'' calculations of fuel
consumption and tailpipe emissions, and used spreadsheets to calculate
even more straightforward calculations of program costs and benefits.
While some stakeholders perceive these differences as evidencing a
meaningfully independent approach, in fact, the EPA staff's analysis
was, at its core, wholly dependent on NHTSA's CAFE Model, and on that
model's use of Autonomie simulations.
Given the above, the only remaining argument for EPA to revert to
its previously-developed models rather than relying on Autonomie and
the CAFE model would be that the former are so technically superior to
the latter that even model refinements and input changes cannot lead
Autonomie and the CAFE model to produce appropriate and reasonable
results for CO2 rulemaking analysis. As discussed below,
having considered a wide range of technical differences, the agencies
find that the Autonomie and CAFE models currently provide the best
analytical combination for CAFE and tailpipe CO2 emissions
rulemaking analysis. As discussed
[[Page 24229]]
below in Section VI.B.3, Autonomie not only has a longer and wider
history of development and application, but also DOE/Argonne's
interaction with automakers, supplier and academies on continuous bases
had made individual sub-models and assumptions more robust. Argonne has
also been using research from DOE's Vehicle Technology Office (VTO) at
the same time to make continuous improvements in Autonomie.\157\ Also,
while Autonomie uses engine maps as inputs, and EPA developed engine
maps that could have been used for today's analysis, EPA declined to do
so, and those engine maps were only used in a limited capacity for
reasons discussed below in Section VI.C.1.
---------------------------------------------------------------------------
\157\ U.S. DOE Benefits & Scenario Analysis publications is
available at https://www.autonomie.net/publications/fuel_economy_report.html. Last accessed November 14, 2019.
---------------------------------------------------------------------------
As also discussed below in Section VI.A.4, the CAFE model accounts
for some important CO2 provisions that EPA's OMEGA model
cannot account for. For example, the CAFE model estimates the potential
that any given manufacturer might apply CO2 compliance
credits it has carried forward from some prior model year. While one
commenter, Mr. Rykowski, takes issue with how the CAFE model handles
credit banking, he does not acknowledge that EPA's OMEGA model, lacking
a year-by-year representation of compliance, is altogether incapable of
accounting for the earning and use of banked compliance credits. Also,
although Mr. Rykowski's comments regarding A/C leakage and refrigerants
are partially correct insofar as the CAFE model does not account for
leakage-reducing technologies explicitly, the comment is as applicable
to OMEGA as it is to the CAFE model and, in any event, data regarding
which vehicles have which leakage-reducing technologies was not
available for the MY 2016 fleet. Nevertheless, as discussed in Section
VI.A.4, NHTSA has refined the CAFE model's accounting for the cost of
leakage reduction technologies.
The agencies have also considered Mr. Rykowski's comments
suggesting that using OMEGA would be preferable because, rather than
selecting from hundreds of thousands of potential combinations of
technologies, OMEGA includes only the ``50 or so'' combinations that
EPA has already determined to be cost-effective. The ``better way'' of
making this determination is also effectively a model, but the
separation of this model from OMEGA is, as evidenced by manufacturers'
comments, obfuscatory, especially in terms of revealing how specific
vehicle model/configurations initial engineering properties are aligned
with specific initial technology combinations. By using a full set of
technology combinations, the CAFE model makes very clear how each
vehicle model/configuration is assigned to a specific initial
combination and, hence, how subsequently fuel consumption and cost
changes are accounted for. Moreover, EPA's separation of ``thinning''
process from OMEGA's main compliance simulation makes sensitivity
analysis difficult to implement, much less follow. The agencies find,
therefore, that the CAFE model's approach of retaining a full set of
vehicle simulation results throughout the compliance simulation to be
more realistic (e.g., more capable of reflecting manufacturer- and
vehicle-specific factors), more responsive to changes in model inputs
(e.g., changes to fuel prices, which could impact the relative
attractiveness of different technologies), more transparent, and more
amenable to independent corroboration the agencies' analysis.
Regarding comments by Messrs. Duleep, Rogers, and Rykowski
suggesting that the CAFE model, by tying most technology application to
planned vehicle redesigns and freshening, is too restrictive, the
agencies disagree. As illustrated by manufacturers' comments cited
above, as reinforced by both extensive product planning information
provided to the agencies, and as further reinforced by extensive
publicly available information, manufacturers tend to not make major
changes to a specific vehicle model/configuration in one model year,
and then make further major changes to the same vehicle model/
configuration the next model year. There is ample evidence that
manufacturers strive to avoid such discontinuity, complexity, and
waste, and in the agencies' view, while it is impossible to represent
every manufacturer's decision-making process precisely and with
certainty, the CAFE model's approach of using estimated product design
schedules provides a realistic basis for estimating what manufacturers
could practicably do. Also, the relevant inputs are simply inputs to
the CAFE model, and if it is actually more realistic to assume that a
manufacturer can change major technology on all of its products every
year, the CAFE model can easily be operated with every model year
designated as a redesign year for every product, but as discussed
throughout this document, the agencies consider this to be extremely
unrealistic. While this means the CAFE model can be run without a year-
by-year representation that carries forward technologies between model
years, doing so would be patently unrealistic (as reflected in some
stakeholders' comments in 2002 on the first version of the CAFE model).
Conversely, the OMEGA model cannot be operated in a way that accounts
for what the agencies consider to be very real product planning
considerations.
However, having also considered Mr. Rykowski's comments about the
CAFE model's ``effective cost'' metric, and having conducted side-by-
side testing documented in the accompanying FRIA, the agencies are
satisfied that an alternative ``cost per credit'' metric is also a
reasonable metric to use for estimating how manufacturers might
selected among available options to add specific fuel-saving
technologies to specific vehicles.\158\ Therefore, NHTSA has revised
the CAFE model accordingly, as discussed below in Section VI.A.4.
---------------------------------------------------------------------------
\158\ As discussed in the FRIA, results vary with model inputs,
among manufacturers, and across model years, but compared to the
NPRM's ``effective cost'' metric, the ``cost per credit'' metric
appears to more frequently produce less expensive solutions than
more expensive solutions, at least when simulating compliance with
CO2 standards. Differences are more mixed when simulating
compliance with CAFE standards, and even when simulating compliance
with CO2 standards, results simulating ``perfect'' trading of
CO2 compliance credits are less intuitive when the ``cost
per credit metric.'' Nevertheless, and while less expensive
solutions are not necessarily ``optimal'' solutions (e.g., if
gasoline costs $7 per gallon and electricity is free, expensive
electrification could be optimal), the agencies consider it
reasonable to apply the ``cost per credit'' metric for the analysis
supporting today's rulemaking.
---------------------------------------------------------------------------
Section VI.C.1 also addresses Mr. Rogers's comments on engine maps
used as estimates to full vehicle simulation. In any event, because
engine maps are inputs to full vehicle modeling and simulation, the
relative merits of specific maps provide no basis to prefer one vehicle
simulation modeling system over another. Similarly, Section VI.B.3 also
addresses Mr. Duleep's comments preferring EPA's prior approach, using
ALPHA, of effectively assuming that a manufacturer would incur no
additional cost by reoptimizing every powertrain to extract the full
fuel economy potential of even the smallest incremental changes to
aerodynamic drag and tire rolling resistance. Mr. Duleep implies that
Autonomie is flawed because the NPRM analysis did not apply Autonomie
in a way that makes such assumptions. The agencies discuss powertrain
sizing and calibration in Section VI.B.3, and note here that such
assumptions are not inherent to
[[Page 24230]]
Autonomie; like engine maps, these are inputs to full vehicle
simulation. Therefore, neither of these comments by Mr. Rogers and Mr.
Duleep lead the agencies to find reason not to use Autonomie.
None of this is to say that Autonomie and the CAFE model as
developed and applied for the NPRM left no room for improvement. In the
NPRM and RIA, the agencies discussed plans to continue work in a range
of specific technical areas, and invited comment on all aspects of the
analysis. As discussed below in Chapter VI, the agencies received
extensive comment on the published model, inputs, and analysis, both in
response to the NPRM and, for newly-introduced modeling capabilities
(estimation of sales, scrappage, and employment effects), in response
to additional peer review conducted in 2019. The agencies have
carefully considered these comments, refined various specific technical
aspects of the CAFE model (like the ``effective cost'' metric mentioned
above), and have also updated inputs to both Autonomie and the CAFE
model. Especially given these refinements and updates, as discussed
throughout this rule, EPA maintains that for CO2 rulemaking
analysis, Autonomie and the CAFE model have advantages that warrant
relying on them rather than on EPA's ALPHA and OMEGA models. Some
examples of such advantages include: A longer history of ongong
development and application for rulemaking, including by EPA;
documentation and model operation stakeholders have found to be
comparatively clear and enabling of independent replication of agency
analyses; a mechanism to explicitly reflect the fact that
manufacturers' product decisions are likely to be informed by fuel
prices; better integration of various model functions, enabling
efficient sensitivity analysis; and an annual time step that makes it
possible to conduct report results on both a calendar year and model
year basis, to estimate accruing impacts on vehicle sales and
scrappage, and to account for the fact that not every vehicle can be
designed in every model year; and other advantages discussed throughout
today's notice. Therefore, recognizing that models inform but do not
make regulatory decisions, EPA has elected to rely solely on the
Autonomie and CAFE models to produce today's analysis of regulatory
alternatives for CO2 standards.
The following sections provide a brief technical overview of the
CAFE model, including changes NHTSA made to the model since 2012, and
differences between the current analysis, the analysis for the 2016
Draft TAR and for the 2017 Proposed Determination/2018 Final
Determination, and the 2018 NPRM, before discussing inputs to the model
and then diving more deeply into how the model works. For more
information on the latter topic, see the CAFE model documentation,
available in the docket for this rulemaking and on NHTSA's website.
1. What assumptions have changed since the 2012 final rule?
Any analysis of regulatory actions that will be implemented several
years in the future, and whose benefits and costs accrue over decades,
requires a large number of assumptions. Over such time horizons, many,
if not most, of the relevant assumptions in such an analysis are
inevitably uncertain.\159\ The 2012 CAFE/CO2 rule considered
regulatory alternatives for model years through MY 2025 (17 model years
after the 2008 market information that formed the basis of the
analysis) that accrued costs and benefits into the 2060s. Not only was
the new vehicle market in 2025 unlikely to resemble the market in 2008,
but so, too, were fuel prices. It is natural, then, that each
successive CAFE/CO2 analysis should update assumptions to
reflect better the current state of the world and the best current
estimates of future conditions.\160\ However, beyond the issue of
unreliable projections about the future, a number of agency assertions
have proven similarly problematic. In fact, Securing America's Future
Energy (SAFE) stated in their comments on the NPRM:
---------------------------------------------------------------------------
\159\ As often stated, ``It's difficult to make predictions,
especially about the future.'' See, e.g., https://quoteinvestigator.com/2013/10/20/no-predict/.
\160\ See, e.g., 77 FR 62785 (Oct. 15, 2012) (``If EPA initiates
a rulemaking [to revise standards for MYs 2022-2025], it will be a
joint rulemaking with NHTSA. . . . NHTSA's development of its
proposal in that later rulemaking will include the making of
economic and technology analyses and estimates that are appropriate
for those model years and based on then-current information.'').
Although the agencies argue ``circumstances have changed'' and
``analytical methods and inputs have been updated,'' a thorough
analysis should provide a side-by-side comparison of the changing
circumstances, methods, and inputs used to arrive at this
determination . . . They represent a rapid, dramatic departure from
the agencies' previous analyses, without time for careful review and
consideration.\161\
---------------------------------------------------------------------------
\161\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 39.
We describe in detail (below) the changes to critical assumptions,
perspectives, and modeling techniques that have created substantive
differences between the current analysis and the analysis conducted in
2012 to support the final rule. To the greatest extent possible, we
have calculated the impacts of these changes on the 2012 analysis.
a) The Value of Fuel Savings
The value of fuel savings associated with the preferred alternative
in the 2012 final rule is primarily a consequence of two assumptions:
\162\ The fuel price forecast and the assumed growth in fuel economy in
the baseline alternative against which savings are measured. Therefore,
as the value of fuel savings accounted for nearly 80 percent of the
total benefits of the 2012 rule, each of these assumptions is
consequential. With a lower fuel price projection and an expectation
that new vehicle buyers respond to fuel prices, the 2012 rule would
have shown much smaller fuel savings attributable to the more stringent
standards. Projected fuel prices are considerably lower today than in
2012, the agencies now understand new vehicle buyers to be at least
somewhat responsive to fuel prices, and the agencies have therefore
updated corresponding model inputs to produce an analysis the agencies
consider to be more realistic.
---------------------------------------------------------------------------
\162\ The value of fuel savings is also affected by the rebound
effect assumption, assumed lifetime VMT accumulation, and the
simulated penetration of alternative fuel technologies. However,
each of these ancillary factors is small compared to the impact of
the two factors discussed in this subsection.
---------------------------------------------------------------------------
The first of these assumptions, fuel prices, was simply an artifact
of the timing of the rule. Following recent periodic spikes in the
national average gasoline price and continued volatility after the
great recession, the fuel price forecast then produced by EIA (as part
of AEO 2011) showed a steady march toward historically high, sustained
gasoline prices in the United States. However, the actual series of
fuel prices has skewed much lower. As it has turned out, the observed
fuel price in the years between the 2012 final rule and this rule has
frequently been lower than the ``Low Oil Price'' sensitivity case in
the 2011 AEO, even when adjusted for inflation. The following graph
compares fuel prices underlying the 2012 final rule to fuel prices
applied in the analysis reported in today's notice, expressing both
projections in 2010 dollars. The differences are clear and significant:
[[Page 24231]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.042
The discrepancy in fuel prices is important to the discussion of
differences between the current rule and the 2012 final rule, because
that discrepancy leads in turn to differences in analytical outputs and
thus to differences in what the agencies consider in assessing what
levels of standards are reasonable, appropriate, and/or maximum
feasible. As an example, the agencies discuss in Sections VI.D.3
Simulating Environmental Impacts of Regulatory Alternatives and
VIII.A.3 EPA's Conclusion that the Final CO2 Standards are
Appropriate and Reasonable that fuel price projections from the 2012
rule were one assumption, among others, that could have led to
overestimates of the health benefits that resulted from reducing
criteria pollutant emissions. Yet the agencies caution readers not to
interpret this discrepancy as a reflection of negligence on the part of
the agencies, or on the part of EIA. Long-term predictions are
challenging and the fuel price projections in the 2012 rule were within
the range of conventional wisdom at the time. However, it does suggest
that fuel economy and tailpipe CO2 regulations set almost
two decades into the future are vulnerable to surprises, in some ways,
and reinforces the value of being able to adjust course when critical
assumptions are proven inaccurate. This value was codified in
regulation when EPA bound itself to the mid-term evaluation process as
part of the 2012 final rule.\163\
---------------------------------------------------------------------------
\163\ See 40 CFR 86-1818-12(h).
---------------------------------------------------------------------------
To illustrate this point clearly, substituting the current (and
observed) fuel price forecast for the forecast used in the 2012 final
rule creates a significant difference in the value of fuel savings.
Even under identical discounting methods (see Section 2, below), and
otherwise identical inputs in the 2012 version of the CAFE Model, the
current (and historical) fuel price forecast reduces the value of fuel
savings by $150 billion--from $525 billion to $375 billion (in 2009
dollars).
The second assumption employed in the 2012 (as well as the 2010)
final rule, that new vehicle fuel economy never improves unless
manufacturers are required to increase fuel economy in the new vehicle
market by increasingly stringent regulations, is more problematic.
Despite the extensive set of recent academic studies showing, as
discussed in Section VI.D.1.a)(2), that consumers value at least some
portion, and in some studies nearly all, of the potential fuel savings
from higher levels of fuel economy at the time they purchase vehicles,
the agencies assumed in past rulemakings that buyers of new vehicles
would never purchase, and manufacturers would never supply, vehicles
with higher fuel economy than those in the baseline (MY 2016 in the
2012 analysis), regardless of technology cost or prevailing fuel prices
in future model years. In calendar year 2025, the 2012 final rule
assumed gasoline would cost nearly $4.50/gallon in today's dollars, and
continue to rise in subsequent years. Even recognizing that higher
levels of fuel economy would be achieved under the augural/existing
standards than without them, the assertion that fuel economy and
CO2 emissions would not improve beyond 2016 levels in the
presence of nearly $5/gallon gasoline is not supportable. This is
highlighted by the observed increased consumer demand for higher-fuel-
economy vehicles during the gas price spike of 2008, when average U.S.
prices briefly broke $4/gallon. In the 2012 final rule, this
assumption--that fuel economy and emissions would never improve absent
regulation--created a persistent gap in fuel economy between
[[Page 24232]]
the baseline and augural standards that grew to 13 mpg (at the industry
average, across all vehicles) by MY 2025. In the 2016 Draft TAR,
NHTSA's analysis included the assumption that manufacturers would
deploy, and consumers would demand, any technology that recovered its
own cost in the first year of ownership through avoided fuel costs.
However, in both the Draft TAR and the Proposed and Final Determination
documents, EPA's analysis assumed that the fuel economy levels achieved
to reach compliance with MY 2021 standards would persist indefinitely,
regardless of fuel prices or technology costs.
By substituting the conservative assumption that consumers are
willing to purchase fuel economy improvements that pay for themselves
with avoided fuel expenditures over the first 2.5 years \164\
(identical to the assumption in this final rule's central analysis) the
gap in industry average fuel economy between the baseline and augural
scenarios narrows from 13 mpg in 2025 to 6 mpg in 2025. As a corollary,
acknowledging that fuel economy would continue to improve in the
baseline under the fuel price forecast used in the final rule erodes
the value of fuel savings attributable to the preferred alternative.
While each gallon is still worth as much as was assumed in 2012, fewer
gallons are consumed in the baseline due to higher fuel economy levels
in new vehicles. In particular, the number of gallons saved by the
preferred alternative selected in 2012 drops from about 180 billion to
50 billion once we acknowledge the existence of even a moderate market
for fuel economy.\165\ The value of fuel savings is similarly eroded,
as higher fuel prices lead to correspondingly higher demand for fuel
economy even in the baseline--reducing the value of fuel savings from
$525 billion to $190 billion.
---------------------------------------------------------------------------
\164\ Greene, D.L. and Welch, J.G., ``Impacts of fuel economy
improvements on the distribution of income in the U.S.,'' Energy
Policy, Volume 122, November 2018, pp. 528-41 (``Four nationwide
random sample surveys conducted between May 2004 and January 2013
produced payback period estimates of approximately three years,
consistent with the manufacturers' perceptions.'') (The 2018 article
succeeds Greene and Welch's 2017 publication titled ``The Impact of
Increased Fuel Economy for Light-Duty Vehicles on the Distribution
of Income in the U.S.: A Retrospective and Prospective Analysis,''
Howard H. Baker Jr. Center for Public Policy, March 2017, which
Consumers Union, CFA, and ACEEE comments include as Attachment 4,
Docket NHTSA-2018-0067-11731).
\165\ Readers should note that this is not an estimate of the
total amount of fuel that will be consumed or not consumed by the
fleet as a whole, but simply the amount of fuel that will be
consumed or not consumed as a direct result of the regulation. As
illustrated in Section VII, light-duty vehicles in the U.S. would
continue to consume considerable quantities of fuel and emit
considerable quantities of CO2 even under the baseline/
augural standards, and agencies' analysis shows that the standards
finalized today will likely increase fuel consumption and
CO2 emissions by a small amount.
---------------------------------------------------------------------------
The magnitude of the fuel economy improvement in the baseline is a
consequence of both the fuel prices assumed in the 2012 rule (already
discussed as being higher than both subsequent observed prices and
current projections) and the assumed technology costs. In 2012, a
number of technologies were assumed to have negative incremental
costs--meaning that applying those technologies to existing vehicles
would both improve their fuel economy and reduce the cost to produce
them. Asserting that the baseline would experience no improvement in
fuel economy without regulation is equivalent to asserting that
manufacturers, despite their status as profit maximizing entities,
would not apply these cost-saving technologies unless forced to do so
by regulation. While this issue is discussed in greater detail in
Section VI.B the combination of inexpensive (or free) technology and
high fuel prices created a logically inconsistent perspective in the
2012 rule--where consumers never demanded additional fuel economy,
despite high fuel costs, and manufacturers never supplied additional
fuel economy, despite the availability of inexpensive (or cost saving)
technology to do so.
Many commenters on earlier rules supported the assumption that fuel
economy would not improve at all in the absence of standards. In fact,
some commenters still support this position. For example, EDF commented
to the NPRM that, ``NHTSA set the Volpe model to project that, with
CAFE standards remaining flat at MY 2020 levels through MY 2026,
automakers would over-comply with the MY 2020 standards by 9 grams/mile
of CO2 for cars and 15 g/mi of CO2 for light
trucks during the 2029-2032 timeframe, plus 1%/year improvements beyond
MY 2032. This assumption unreasonably obscures the impacts of the
rollback and is not reflective of historical compliance performance.''
\166\
---------------------------------------------------------------------------
\166\ EDF, NHTSA-2018-0067-11996, Comments to DEIS, at 4.
---------------------------------------------------------------------------
EDF is mistaken in two different ways: (1) By acknowledging the
existence of a well-documented market for fuel economy, rather than
erroneously inflating the benefits associated with increasing
standards, this assumption serves to isolate the benefits actually
attributable to each regulatory alternative, and (2) it is, indeed,
reflective of historical compliance performance. While the agencies
rely on the academic literature (and comments from companies that build
and sell automobiles) to defend the assertion that a market for fuel
economy exists, the industry's historical CAFE compliance performance
is a matter of public record.\167\ As shown in Figure IV-3, Figure IV-
4, and Figure IV-5 for more than a decade, the industry average CAFE
has exceeded the standard for each regulatory class--by several mpg
during periods of high fuel prices.
---------------------------------------------------------------------------
\167\ Data from CAFE Public Information Center (PIC), https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm, last accessed 10/08/2019.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24233]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.043
[[Page 24234]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.044
BILLING CODE 4910-59-C
While this rulemaking has shown the impact of deviations from the
2012 rule assumptions individually, these two assumptions affect the
value of fuel savings jointly. Replacing the fuel price forecast with
the observed historical and current projected prices, and including any
technology that pays for itself in the first 2.5 years of ownership
through avoided fuel expenditures, reduces the value of fuel savings
from $525 billion in the 2012 rule to $140 billion, all else equal.
Interestingly, this reduction in the value of fuel savings is smaller
than the result when assuming only that the desired payback period is
nonzero. While it may seem counterintuitive, it is entirely consistent.
The number of gallons saved under the preferred alternative is
actually higher when modifying both assumptions, compared to only
modifying the payback period. Updating both assumptions leads to about
100 billion gallons saved by the preferred alternative in 2012,
compared to only 50 billion from changing only the payback period, and
180 billion in the 2012 analysis. This occurs because the fuel economy
in the baseline is lower when updating both the fuel price and the
payback period--the gap between the augural standards and the baseline
grows to 9 mpg, rather than only 6 mpg when updating only the payback
period. Despite the existence of inexpensive
[[Page 24235]]
technology in both cases, with lower fuel prices there are fewer
opportunities to apply technology that will pay back quickly. As a
consequence, the number of gallons saved by the preferred alternative
in 2012 increases--but each gallon saved is worth less because the
price of fuel is lower.
b) Technology Cost
While the methods used to identify cost-effective technologies to
improve fuel economy in new vehicles have continuously evolved since
2012 (as discussed further in Section IV.B.1), as have the estimated
cost of individual technologies, the inclusion of a market response in
all scenarios (including the baseline) has changed the total technology
cost associated with a given alternative. As also discussed in Section
VI.B, acknowledging the existence of a market for fuel economy leads to
continued application of the most cost-effective technologies in the
baseline--and in other less stringent alternatives--up to the point at
which there are no remaining technologies whose cost is fully offset by
the value of fuel saved in the first 30 months of ownership. The
application of this market-driven technology has implications for fuel
economy levels under lower stringencies (as discussed earlier), but
also for the incremental technology cost associated with more stringent
alternatives. As lower stringency alternatives (including the 2012
baseline) accrue more technology, the incremental cost of more
stringent alternatives decreases.
By including a modest market for fuel economy, and preserving all
other assumptions from the 2012 final rule, the incremental cost of
technology attributable to the preferred alternative decreases from
about $140 billion to about $72 billion. This significant reduction in
technology cost is somewhat diminished by the associated reduction in
the value of fuel savings (a decrease of $385 billion) when
acknowledging the existence of a market for fuel economy. Another
consequence of these changes is that the incremental cost of fuel
economy technology is responsive to fuel price, as it should be. Under
higher prices (as were assumed in 2012), consumers demand higher fuel
economy in the new vehicle market. Under lower prices (as have occurred
since the 2012 rule) consumers demand less fuel economy than would have
been consistent with the fuel price assumptions in 2012.\168\ Including
a market response in the analysis ensures that, in each case, the cost
of fuel economy technology within an alternative is consistent with
those assumptions. Using the same fuel price forecast that supports
this rule, and the same estimate of market demand for fuel economy, the
incremental cost of technology in the preferred alternative would rise
back up to about $110 billion.
---------------------------------------------------------------------------
\168\ This is why dozens of studies examining the ability of
fuel taxes (and carbon taxes, which produce the same result for
transportation fuels) to reduce CO2 emissions have found
cost-effective opportunities available for those pricing mechanisms.
---------------------------------------------------------------------------
c) The Social Cost of Carbon (SCC) Emissions
As discussed extensively in the NPRM, the agencies' perspective
regarding the social cost of carbon has narrowed in focus. While the
2012 final rule considered the net present value of global damages
resulting from carbon emitted by vehicles sold in the U.S. between MY
2009 and MY 2025, the NPRM (and this final rule) consider only those
damages that occur to the United States and U.S. territories. As a
result of this change in perspective, the value of estimated damages
per-ton of carbon is correspondingly smaller. Had the 2012 final rule
utilized the same perspective on the social cost of carbon, the
benefits associated with the preferred alternative would have been
about $11 billion, rather than $53 billion. However, the savings
associated with carbon damages are a consequence of both the assumed
cost per-ton of damages and the number of gallons of fuel saved. As
discussed above, the gallons saved in the 2012 final rule were likely
inflated as a result of both fuel price forecasts and the assumption
that no market exists for fuel economy improvements. Correcting the
estimate of gallons saved from the preferred alternative in the 2012
rule and considering only the domestic social cost of carbon further
reduces the savings in carbon damages to $6 billion.
d) Safety Neutrality
In the 2012 final rule, the agencies showed a ``safety neutral''
compliance solution; that is, a compliance solution that produced no
net increase in on-road fatalities for MYs 2017-2025 vehicles as a
result of technology changes associated with the preferred alternative.
In practice, safety neutrality was achieved by expressly limiting the
availability of mass reduction technology to only those vehicles whose
usage causes fewer fatalities with decreased mass. This result was
discussed as one possible solution, where manufacturers chose
technology solutions that limited the amount of mass reduction applied,
and concentrated the application on vehicles that improve the safety of
other vehicles on the roads (primarily by reducing the mass
differential in collisions). However, it implicitly assumed that each
and every manufacturer would leave cost-effective technologies unused
on entire market segments of vehicles in order to preserve a safety
neutral outcome at the fleet level for a given model year (or set of
model years) whose useful lives stretched out as far as the 2060s.
Removing these restrictions tells a different story.
When mass reduction technology, determined in the model to be a
cost-effective solution (particularly in later model years, when more
advanced levels of mass reduction were expected to be possible), is
unrestricted in its application, the 2012 version of the CAFE Model
chooses to apply it to vehicles in all segments. This has a small
effect on technology costs, increasing compliance costs in the earliest
years of the program by a couple billion dollars, and reducing
compliance costs for MYs 2022--2025 by a couple billion dollars.
However, the impact on safety outcomes is more pronounced.
Also starting with the model and inputs used for the 2012 final
rule (and, as an example, focusing on that rule's 2008-based market
forecast), removing the restrictions on the application of mass
reduction technology results in an additional 3,400 fatalities over the
full lives of MYs 2009-2025 vehicles in the baseline,\169\ and another
6,900 fatalities over those same vehicle lives under the preferred
alternative. The result, a net increase of 3,500 fatalities under the
preferred alternative relative to the baseline, also produces a net
social cost of $18 billion. The agencies' current treatment of both
mass reduction technology, which can greatly improve the effectiveness
of certain technology packages by reducing road load, and estimated
fatalities and now account for both general exposure (omitted in the
2012 final rule modeling) and fatality risk by age of the vehicle,
further changes the story around mass reduction technology application
for compliance and its relationship to on-road safety.
---------------------------------------------------------------------------
\169\ Relative to the continuation of vehicle mass from the 2008
model year carried forward into the future.
---------------------------------------------------------------------------
2. What methods have changed since the 2012 final rule?
Simulating how manufacturers might respond to CAFE/CO2
standards
[[Page 24236]]
requires information about existing products being offered for sale, as
well as information about the costs and effectiveness of technologies
that could be applied to those vehicles to bring the fleets in which
they reside into compliance with a given set of standards. Following
extensive additional work and consideration since the 2012 analysis,
both agencies now use the CAFE Model to simulate these compliance
decisions. This has several practical implications which are discussed
in greater detail in Section VI.A. Briefly, this change represents a
shift toward including a number of real-world production constraints--
such as component sharing across a manufacturer's portfolio--and
product cadence, where only a subset of vehicles in a given model year
are redesigned (and thus eligible to receive fuel economy technology).
Furthermore, the year-by-year accounting ensures a continuous evolution
of a manufacturer's product portfolio that begins with the market data
of an initial model year (model year 2017, in this analysis) and
continues through the last year for which compliance is simulated.
Finally, the modeling approach has migrated from one that relied on the
simple product of single values to estimate technology effectiveness to
a model that relies on full vehicle simulation to determine the
effectiveness of any combination of fuel economy technologies. The
combination of these changes has greatly improved the realism of
simulated vehicle fuel economy for combinations of technologies across
vehicle systems and classes.
In addition to these changes to the portions of the analysis that
represent the supply of fuel economy (by manufacturer, fleet, and model
year) in the new vehicle market, this analysis contains changes to the
representation of consumer demand for fuel economy. One such measure
was discussed above--the notion that consumers will demand some amount
of fuel economy improvement over time, consistent with technology costs
and fuel prices. However, another deviation from the 2012 final rule
analysis reflects overall demand for new vehicles. Across ten
alternatives, ranging from the baseline (freezing future standards at
2016 levels) to scenarios that increased stringency by seven percent
per year, from 2017 through 2025, the 2012 analysis showed no response
in new vehicle sales, down to the individual model level. This implied
that, regardless of changes to vehicle cost or attributes driven by
stringency increases, no fewer (or possibly more) units of any single
model would be sold in any year, in any alternative. Essentially, that
analysis asserted that the new vehicle market does not respond, in any
way, to average new vehicle prices across the alternatives--regardless
of whether the incremental cost is $1,600/vehicle (as it was estimated
to be under the preferred alternative) or nearly $4,000/vehicle (as it
was in under the 7 percent alternative). Both the NPRM and this final
rule, while not employing pricing models or full consumer choice models
to address differentiated demand within brands or manufacturer
portfolios, have incorporated a modeled sales response that seeks to
quantify what was not quantified in previous rulemakings.
An important accounting method has also changed since the 2012
final rule was published. At the time of that rule, the agencies used
an approach to discounting that combined attributes of a private
perspective and a social perspective in their respective benefit cost
analyses. This approach was logically inconsistent, and further
reinforced some of the exaggerated estimates of fuel savings, social
benefits (from reduced externalities), and technology costs described
above. The old method discounted the value of all incremental
quantities, whether categorized as benefits or costs, to the model year
of the vehicle to which they accrued. This approach is largely
acceptable for use in a private benefit cost analysis, where the costs
and benefits accrue to the buyer of a new vehicle (in the case of this
policy) who weighs their discounted present values at the time of
purchase. However, the private perspective would not include any costs
or benefits that are external to the buyer (e.g., congestion or the
social cost of carbon emissions). For an analysis that compares
benefits and costs from the social perspective, attempting to estimate
the relative value of a policy to all of society rather than just
buyers of new vehicles, this approach is more problematic.
The discounting approach in the 2012 final rule was particularly
distortionary for a few reasons. The fact that benefits and costs
occurred over long time periods in the 2012 rule, and the standards
isolated the most aggressive stringency increases in the latter years
of the program, served to allow benefits that occurred in 2025 (for
example) to enter the accounting without being discounted, provided
that they accrued to the affected vehicles during their first year of
ownership. In a setting where numerous inputs (e.g., fuel price and
social cost of carbon) increase over time, benefits were able to grow
faster than the discount rate in some cases--essentially making them
infinite. The interpretation of discounting for externalities was
equally problematic. For example, the discounting approach in the 2012
final rule would have counted a ton of CO2 not emitted in CY
2025 in multiple ways, despite the fact that the social cost of carbon
emissions was inherently tied to the calendar year in which the
emissions occurred. Were the ton avoided by a MY 2020 vehicle, which
would have been five years old in CY 2025, the value of that ton would
have been the social cost of carbon times 0.86, but would have been
undiscounted if that same ton had been saved by a MY 2025 vehicle in
its initial year of usage.
This approach was initially updated in the 2016 Draft TAR to be
consistent with common economic practice for benefit-cost analysis, and
this analysis continues that approach. In the social perspective, all
benefits and costs are discounted back to the decision year based on
the calendar year in which they occur. Had the agencies utilized such
an approach in the 2012 final rule, net benefits would have been
reduced by about 20 percent, from $465 billion to $374 billion--not
accounting for any of the other adjustments discussed above.
3. How have conditions changed since the 2012 final rule was published?
The 2012 final rule relied on market and compliance information
from model year 2008 to establish standards for model years 2017-2025.
However, in the intervening years, both the market and the industry's
compliance positions have evolved. The industry has undergone a
significant degree of change since the MY 2008 fleet on which the
2012FR was based. Entire brands (Pontiac, Oldsmobile, Saturn, Hummer,
Mercury, etc.) and companies (Saab, Suzuki, Lotus) have exited the U.S.
market, while others (most notably Tesla) have emerged. Several dozen
nameplates have been retired and dozens of other created in that time.
Overall, the industry has offered a diverse set of vehicle models that
have generally higher fuel economy than the prior generation, and an
ever-increasing set of alternative fuel powertrains.
As Table IV-1 shows, alternative powertrains have steadily
increased under CAFE/CO2 regulations. Under the standards
between 2011 and 2018, the number of electric vehicle offerings in the
market has increased from 1 model to 57 models (inclusive of all plug-
in vehicles that are rated for use on the highway), and hybrids (like
the Toyota Prius) have increased from 20 models to
[[Page 24237]]
43 models based on data from DOE's Alternative Fuels Data Center. Fuel
efficient diesel vehicles have similarly been on the rise in that
period, more than doubling the number of offerings. Flexible fuel
vehicles (FFVs), capable of operating on both gasoline and E85 remain
readily available in the market, but have been excluded from the table
due to both their lower fuel economy and demonstrated consumer
reluctance to operate FFVs on E85. They have historically been used to
improve a manufacturer's compliance position, rather than other
alternative fuel systems that reduce fuel consumption and save buyers
money.
[GRAPHIC] [TIFF OMITTED] TR30AP20.045
Not only have alternative powertrain options proliferated since the
2012 FR, the average fuel economy of new vehicles within each body
style has increased. However, the more dramatic effect may lie in the
range of fuel economies available within each body style. Figure IV-6
shows the distribution of new vehicle fuel economy (in miles per gallon
equivalent) by body style for MYs 2008, 2016, and 2020 (simulated).
Each box represents the 25th and 75th percentiles, where 25 and 75
percent of new models offered are less fuel efficient than that level.
Not only has the median fuel economy improved (the median shows the
point at which 50 percent of new models are less efficient) under the
CAFE/CO2 programs, but the range of available fuel economies
(determined by the length of the boxes and their whiskers) has
increased as well. For example, the 25th percentile of pickup truck
fuel economy in 2020 is expected to be significantly more efficient
than 75 percent of the pickups offered in 2008. In MY 2008, there were
only a few SUVs offered with rated fuel economies above 34MPG. By MY
2020 almost half of the SUVs offered will have higher fuel economy
ratings--with almost 20 percent of offerings exceeding 40MPG.
The improvement in passenger car styles has been no less dramatic.
As with the other styles, the range of available fuel economies has
increased under the CAFE/CO2 programs and the distribution
of available fuel economies skewed higher--with 40 percent of MY 2020
models exceeding 40MPG. The attribute-based standards are designed to
encourage manufacturers to improve vehicle fuel economy across their
portfolios, and they have clearly done so. Not only have the higher
ends of the distributions increased, the lower ends in all body styles
have improved as well, where the least fuel efficient 25 percent of
vehicles offered in MY 2016 (and simulated in 2020) are more fuel
efficient than the most efficient 25 percent of vehicles offered in MY
2008.
BILLING CODE 4910-59-P
[[Page 24238]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.046
BILLING CODE 4910-59-C
Some commenters have argued that consumers will be harmed by any
set of standards lower than the baseline (augural) standards because
buyers of new vehicles will be forced to spend more on fuel than they
would have under the augural standards. However, as Figure IV-6 shows,
the range of fuel economies available in the new market is already
sufficient to suit the needs of buyers who desire greater fuel economy
rather than interior volume or some other attributes. Full size pickup
trucks are now available with smaller turbocharged engines paired with
8 and 10-speed transmissions and some mild electrification. Buyers
looking to transport a large family can choose to purchase a plug-in
hybrid minivan. There were 57 electric models available in 2018, and
hybrid powertrains are no longer limited to compact cars (as they once
were). Buyers can choose hybrid SUVs with all-wheel and four-wheel
drive. While these kinds of highly efficient options were largely
absent from some body styles in MY 2008, this is no longer the case.
Given that high-MPG vehicles are widely available, consumers must also
value other vehicle attributes (e.g., acceleration and load-carrying
capacity) that can can also be improved with the same technologies that
can be used to improve fuel economy.
---------------------------------------------------------------------------
\170\ Circles represent specific outlying vehicle models.
---------------------------------------------------------------------------
Manufacturers have accomplished a portfolio-wide improvement by
improving the combustion efficiency of engines (through direct
injection and
[[Page 24239]]
turbocharging), migrating from four and five speed transmissions to 8
and 10 speed transmissions, and electrifying to varying degrees. All of
this has increased both production costs and fuel efficiency during a
period of economic expansion and low energy prices. While the vehicles
offered for sale have increased significantly in efficiency between MY
2008 and MY 2020, the sales-weighted average fuel economy has achieved
less improvement. Despite stringency increases of about five percent
(year-over-year) between 2012 and 2016, the sales-weighted average fuel
economy increased marginally. Figure IV-7 shows an initial increase in
average new vehicle fuel economy (the heavy solid line, shown in mpg as
indicated on the left y axis), followed by relative stagnation as fuel
prices (the light dashed lines, shown in dollars per gallon as
indicated on the right y axis) fell and remained low.\171\ It is worth
noting that average new vehicle fuel economy observed a brief spike
during the year that the Tesla Model 3 was introduced (as a consequence
of strong initial sales volumes, as pre-orders were satisfied, and fuel
economy ratings that are significantly higher than the industry
average), and settled around 27.5 MPG in Fall 2019. Average fuel
economy receded further over the next several months to 26.6 MPG in
February 2020.\172\
---------------------------------------------------------------------------
\171\ Ward's Automotive, https://www.wardsauto.com/industry/fuel-economy-index-shows-slow-improvement-april. Last accessed Dec.
13, 2019.
\172\ Ward's Automotive, https://wardsintelligence.informa.com/WI964622/Fuel-Economy-Slightly-Down-in-February. Last accessed Mar.
9, 2020.
[GRAPHIC] [TIFF OMITTED] TR30AP20.047
In their NPRM comments, manufacturers expressed concern that CAFE
standards had already increased to the point where the price increases
necessary to recoup manufacturers' increased costs for providing
further increases in fuel economy outweigh the value of fuel
savings.173 174 The agencies do not agree that this point
has already been reached by previous stringency increases, but
acknowledge the reality of diminishing marginal returns to improvements
in fuel economy. A driver with a 40MPG vehicle uses about 300 gallons
of fuel per year. Increasing the fuel economy of that vehicle to 50MPG,
a 25 percent increase, would likely be over $1000 in additional
technology cost. However, that driver would only save 25 percent of
their annual fuel consumption, or 75 gallons out of 300 gallons. Even
at $3/gallon, higher than the current national average, that represents
$225 per year in fuel savings. That means that the buyer's $1000
investment in additional fuel economy pays back in just under 4.5 years
(undiscounted). The agencies' respective programs have created greater
access to high MPG vehicles in all classes and encouraged the
proliferation of alternative fuels and powertrains. But if the value of
the fuel savings is insufficient to motivate buyers to invest in ever
greater levels of fuel economy, manufacturers will face challenges in
the market.
---------------------------------------------------------------------------
\173\ NHTSA-2018-0067-12064-25.
\174\ NHTSA-2018-0067-12073-2.
---------------------------------------------------------------------------
While Figure IV-3 through Figure IV-5 illustrate the trends in
historical CAFE compliance for the entire industry, the figures contain
another relevant fact. After several consecutive years of increasing
standards, the achieved and required levels converge. When the
standards began increasing again for passenger cars in 2011, the prior
year had industry CAFE levels 5.6 mpg and 7.7 mpg in excess of their
standards for domestic cars and imported cars, respectively. Yet, by
2016, the consecutive year-over-year increases had eroded the levels of
over-compliance. Light trucks similarly exceeded their standard prior
to increasing standards, which began in 2005. Yet, by 2011, after
several consecutive years of stringency increases, the industry light-
truck average CAFE was merely compliant with the rising standard.
This is largely due to the fact that stringency requirements have
increased at a faster rate than achieved fuel
[[Page 24240]]
economy levels for several years. The attribute-based standards took
effect in 2011 for all regulatory classes, although light truck CAFE
standards had been increasing since 2005. Since 2004, light truck
stringency has increased an average of 2.7 percent per year, while
light truck's compliance fuel economy has increased by an average of
1.7 percent over the same period.\175\ For the passenger classes, a
similar story unfolds over a shorter period of time. Year over year
stringency increases have averaged 4.7 percent per year for domestic
cars (though increases in the first two years were about 8 percent--
with lower subsequent increases), but achieved fuel economy increases
averaged only 2.2 percent per year over the same period. Imported
passenger cars were similar to domestic cars, with average annual
stringency increases of 4.4 percent but achieved fuel economy levels
increasing an average of only 1.4 percent per year from 2011 through
2017. Given that each successive percent increase in stringency is
harder to achieve than the previous one, long-term discrepancies
between required and achieved year-over-year increases cannot be offset
indefinitely with existing credit banks, as they have been so far.
---------------------------------------------------------------------------
\175\ Both the standards and these calculations are defined in
consumption space--gallons per mile--which also translates directly
into CO2 based on the carbon content of the fuel
consumed.
---------------------------------------------------------------------------
With the fuel price increases fresh in the minds of consumers, and
the great recession only recently passed, the CAFE stringency increases
that began in 2011 (and subsequent CAFE/CO2 stringency
increases after EPA's program was first enforced in MY 2012) had
something of a head start. As Figure IV-3 through Figure IV-5
illustrate, the standards were not binding in MY 2011--even
manufacturers that had historically paid civil penalties were earning
credits for overcompliance. It took two years of stringency increase to
catch up to the CAFE levels already present in MY 2011. However, seven
consecutive years of increases for passenger cars and a decade of
increases for light trucks has changed the credit situation. Figure IV-
8 shows CAFE credit performance for regulated fleets--the solid line
represents the number of fleets generating shortfalls and the dashed
line represents the number of fleets earning credits in each model
year.
[GRAPHIC] [TIFF OMITTED] TR30AP20.048
Fewer than half as many fleets earned surplus credits for over-
compliance in MY 2017 compared to MY 2011--and this trend is
persistent. The story varies from one manufacturer to another, but it
seems sufficient to state the obvious--when the agencies conducted the
analysis to establish standards through MY 2025 back in 2012, most (if
not all) manufacturers had healthy credit positions. That is no longer
the case, and each successive increase requires many fleets to not only
achieve the new level from the resulting increase, but to resolve
deficits from the prior year as well. The large sums of credits, which
last five years under both programs, have allowed most manufacturers to
resolve shortfalls. But the light truck fleet, in particular, has a
dwindling supply of credits available for purchase or trade. The
CO2 program has a provision that allows credits earned
during the early years of over-compliance to be applied through MY
2021. This has reduced the compliance burden in the last several years,
as intended, but will not mitigate the compliance challenges some OEMs
would face if the baseline standards remained in place and energy
prices persisted at current levels.
BILLING CODE 4910-59-P
[[Page 24241]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.049
BILLING CODE 4910-59-C
Table IV-2 shows the credits earned by each manufacturer over
time.\176\ As the table shows, when the agencies considered future
standards in 2012, most manufacturers were earning credits in at least
one fleet. However, the bold values show years with deficits and even
some manufacturers who started out in strong positions, such as Ford's
passenger car fleet, have seen growing deficits in recent years. While
[[Page 24242]]
the initial banks for early-action years eases the burden of
CO2 compliance for many OEMs, the year-to-year compliance
story is similar to CAFE, see Table IV-3.
---------------------------------------------------------------------------
\176\ MY 2017 values represent estimated earned credits based on
MY 2017 final compliance data.
[GRAPHIC] [TIFF OMITTED] TR30AP20.050
Credit position and shortfall rates clearly illustrate
manufacturers' fleet performance relative to the standards. Recognizing
that manufacturers plan compliance over several model years at any
given time, sporadic shortfalls may not be evidence of undue
difficulty, but sustained, widespread, growing shortfalls should
probably be viewed as evidence that standards previously believed to be
manageable might no longer be so. While NHTSA is prohibited by statute
from considering availability of credits (and thus, size of credit
banks) in determining maximum feasible standards, it does consider
shortfalls as part of its determination. EPA has no such prohibition
under the CAA and is free to consider both credits and shortfalls.
These increasing credit shortfalls are occurring at a time that the
industry is deploying more technology than the agencies anticipated
when establishing future standards in 2012. The agencies' projections
of transmission technologies were mixed. While the agencies expected
the deployment of 8-speed transmissions to about 25 percent of the
market by MY 2018, transmissions with eight or more gears account for
almost 30 percent of the market. However, the agencies projected no CVT
transmissions in future model years, instead projecting high
penetration of DCTs. However, CVTs currently make up more than 20
percent of new transmissions. The tradeoff between advanced engines and
electrification was also underestimated. While the agencies projected
penetration rates of turbocharged engines that are higher than we've
observed in the market (45 percent compared to 30 percent), the
estimated penetration of electric technologies was significantly lower.
The agencies projected a couple percent of strong hybrids--which we've
seen--but virtually no PHEVs or EVs. While the volumes of those
vehicles are still only a couple percent of the new vehicle market,
they are heavily credited under both programs and can significantly
improve compliance positions even at smaller volumes. Even lower-level
electrification technologies, like stop-start systems, are
significantly more prevalent than we anticipated (stop-start systems
were projected to be in about 2 percent of the market, compared to over
20 percent in the 2018 fleet). Despite technology deployment that is
comparable to 2012 projections, and occasionally more aggressive,
passenger car and light truck fleets have slightly lower fuel economy
than projected. As fleet volumes have shifted along the footprint
curve, the standards have decreased as well (relative to the
expectation in 2012), but less so. While compliance deficits have been
modest, they have been accompanied by record sales for several years.
This has not only depleted existing credit banks, but created
significant shortfalls that may be more difficult to overcome if sales
recede from record levels.
V. Regulatory Alternatives Considered
Agencies typically consider regulatory alternatives in proposals as
a way of evaluating the comparative effects of different potential ways
of accomplishing their desired goal. NEPA
[[Page 24243]]
requires agencies (in this case, NHTSA, but not EPA) to compare the
potential environmental impacts of their proposed actions to those of a
reasonable range of alternatives. Executive Orders 12866 and 13563 and
OMB Circular A-4 also encourage agencies to evaluate regulatory
alternatives in their rulemaking analyses. Alternatives analysis begins
with a ``no-action'' alternative, typically described as what would
occur in the absence of any regulatory action. This final rule, like
the proposal, includes a no-action alternative, described below, as
well as seven ``action alternatives.'' The final standards may, in
places, be referred to as the ``preferred alternative,'' which is NEPA
parlance, but NHTSA and EPA intend ``final standards'' and ``preferred
alternative'' to be used interchangeably for purposes of this
rulemaking.
In the proposal, NHTSA and EPA defined the different regulatory
alternatives (other than the no-action alternative) in terms of
percent-increases in CAFE and CO2 stringency from year to
year. Percent increases in stringency referred to changes in the
standards year over year--as in, standards that become 1 percent more
stringent each year. Readers should recognize that those year-over-year
changes in stringency are not measured in terms of mile per gallon or
CO2 gram per mile differences (as in, 1 percent more
stringent than 30 miles per gallon in one year equals 30.3 miles per
gallon in the following year), but in terms of shifts in the footprint
functions that form the basis for the actual CAFE and CO2
standards (as in, on a gallon or gram per mile basis, the CAFE and
CO2 standards change by a given percentage from one model
year to the next). Under some alternatives, the rate of change was the
same for both passenger cars and light trucks; under others, the rate
of change differed. Like the no-action alternative, all of the
alternatives considered in the proposal were more stringent than the
preferred alternative.
Alternatives considered in the proposal also varied in other
significant ways. Alternatives 3 and 7 in the proposal involved a
gradual discontinuation of CAFE and average CO2 adjustments
reflecting the use of technologies that improve air conditioner
efficiency or otherwise improve fuel economy under conditions not
represented by long-standing fuel economy test procedures (off-cycle
adjustments, described in further detail in Section IX, although the
proposal itself would have retained these flexibilities. Commenters
responding to the request for comment on phasing out these
flexibilities generally supported maintaining the existing program, as
proposed. Some commenters suggested changes to the existing program
that were not discussed in the NPRM. Such changes would be beyond the
scope of this rulemaking and were not considered. Section IX contains a
more thorough summary of these comments and the issues they raise, as
well as the agencies' responses. Consistent with the decision to retain
these flexibilities in the final rule, alternatives reflecting their
phase-out have not been considered in this final rule.
Additionally, in the NPRM for this rule, EPA proposed to exclude
the option for manufacturers partially to comply with tailpipe
CO2 standards by generating CO2-equivalent
emission adjustments associated with air conditioning refrigerants and
leakage after MY 2020. This approach was proposed in the interest of
improved harmonization between the CAFE and tailpipe CO2
emissions programs because this optional flexibility cannot be
available in the CAFE program.\177\ Alternatives 1 through 8 excluded
this option. EPA requested comment ``on whether to proceed with [the]
proposal to discontinue accounting for A/C leakage, methane emissions,
and nitrous oxide emissions as part of the CO2 emissions
standards to provide for better harmony with the CAFE program, or
whether to continue to consider these factors toward compliance and
retain that as a feature that differs between the programs.'' \178\ EPA
stated that if EPA were to proceed with excluding A/C refrigerant
credits as proposed, ``EPA would consider whether it is appropriate to
initiate a new rulemaking to regulate these programs independently . .
. .'' \179\ EPA also stated that ``[i]f the agency decides to retain
the A/C leakage . . . provisions for CO2 compliance, it
would likely re-insert the current A/C leakage offset and increase the
stringency levels for CO2 compliance by the offset amounts
described above (i.e., 13.8 g/mi equivalent for passenger cars and 17.2
g/mi equivalent for light trucks). EPA received comments from a wide
range of stakeholders, most of whom opposed the elimination of these
flexibility provisions.
---------------------------------------------------------------------------
\177\ For the CAFE program, carbon-based tailpipe emissions
(including CO2, HC, and CO) are measured, and fuel
economy is calculated using a carbon balance equation. EPA also uses
carbon-based emissions (CO2, HC, and CO, the same as for
CAFE) to calculate tailpipe CO2 for use in determining
compliance with its standards. In addition, under the no-action
alternative for the proposal and under all alternatives in the final
rule, in determining compliance, EPA includes on a CO2
equivalent basis (using Global Warming Potential (GWP) adjustment)
A/C refrigerant leakage credits, at the manufacturer's option, and
nitrous oxide and methane emissions. EPA also has separate emissions
standards for methane and nitrous oxides. The CAFE program does not
include or account for A/C refrigerant leakage, nitrous oxide and
methane emissions because they do not impact fuel economy. Under
Alternatives 1-8 in the proposal, the standards were closely aligned
for gasoline powered vehicles because compliance with the fleet
average standard for such vehicles is based on tailpipe
CO2, HC, and CO for both programs and not emissions
unrelated to fuel economy, although diesel and alternative fuel
vehicles would have continued to be treated differently between the
CAFE and CO2 programs. While such an approach would have
significantly improved harmonization between the programs, standards
would not have been fully aligned because of the small fraction of
the fleet that uses diesel and alternative fuels (as described in
the proposal, such vehicles made up approximately four percent of
the MY 2016 fleet), as well as differences involving EPCA/EISA
provisions EPA has not adopted, such as minimum standards for
domestic passenger cars and limits on credit transfers between
regulated fleets. The proposal to eliminate flexibilities associated
with A/C refrigerants and leakage was not adopted for this final
rule, and the reasons for and implications of that decision are
discussed further below.
\178\ 83 FR at 43193 (Aug. 24, 2018).
\179\ Id. at 43194.
---------------------------------------------------------------------------
Specifically, the two major trade organizations of auto
manufacturers, as well as some individual automakers, supported
retaining these provisions. Global Automakers commented that ``[a]ir
conditioning refrigerant leakage . . . should be included for
compliance with the EPA standards for all MYs, even if it means a
divergence from the NHTSA standards.'' \180\ Global provides several
detailed reasons for their comments, including that the existing
provisions are ``. . . important to maintaining regulatory flexibility
through real [CO2] emission reductions and would prevent the
potential for additional bifurcated, separate programs at the state
level.'' \181\ The Alliance similarly commented that it ``supports
continuation of the full air conditioning refrigerant leakage credits
under the [CO2] standards.'' \182\ Some individual
[[Page 24244]]
manufacturers, including General Motors,\183\ Fiat Chrysler,\184\ and
BMW,\185\ also commented in support of maintaining the current A/C
refrigerant and leakage credits.
---------------------------------------------------------------------------
\180\ Global, NHTSA-2018-0067-12032, Appendix A at A-5.
\181\ Id. Global also stated that excluding A/C leakage credits
would ``. . . greatly limit the ability [of manufacturers] to select
the most cost-effective approach for technology improvements and
result in a costlier, separate set of regulations that actually
relate to the overall GHG standards.'' Global also expressed concern
that issuing separate regulations for A/C leakage could take too
long and create a gap in which States might act to separately
regulate or even ban refrigerants, and supported continued inclusion
of A/C leakage and refrigerant regulation in EPA's GHG program to
avoid risk of an ensuing patchwork. Global argued that manufacturers
had already invested to meet the existing program, and that ``the
proposed phase-out also creates another risk that manufacturers will
have stranded capital in technologies that are not fully
amortized.'' Global Automakers, EPA-HQ-OAR-2018-0283-5704,
Attachment A, at A.43-44.
\182\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 12.
Alliance also expressed concern about stranded capital and risk of
patchwork of state regulation if MAC direct credits were not
retained in the Federal GHG program. Id. at 80-81.
\183\ General Motors, NHTSA-2018-0067-11858, Appendix 4, at 1
(``General Motors supports the extensive comments from the Alliance
of Automobile Manufacturers regarding flexibility mechanisms, and
incorporates them by reference. In particular, the Alliance cites
the widening gap between the regulatory standards and actual
industry-wide new vehicle average fuel economy that has become
evident since 2016, despite the growing use of improvement `credits'
from various flexibility mechanisms, such as off-cycle technology
credits, mobile air conditioner efficiency credits, mobile air
conditioner refrigerant leak reduction credits and credits from
electrified vehicles.'')
\184\ FCA, NHTSA-2018-0067-11943, at 8. FCA also expressed
concern about patchwork in the absence of a federal rule. Id.
\185\ BMW, EPA-HQ-OAR-2018-4204, at 3. BMW stated that ``Today's
rules allow flexibilities to be used by the motor vehicle
manufacturers for fuel saving technologies and efficiency gains
which are not covered in the applicable test procedures. To enhance
the future use of these technologies and to reward motor vehicle
manufacturer's investments taken for future innovations, the
agencies should consider the continuation of current flexibilities
for the model years 2021 to 2026.''
---------------------------------------------------------------------------
Auto manufacturing suppliers who addressed A/C refrigerant and
leakage credits also generally supported retaining the existing
provisions. MEMA commented that ``It is essential for supplier
investment and jobs, and continuous innovation and improvements in the
technologies that the credit programs continue and expand to broaden
the compliance pathways. MEMA urges the agencies to continue the
current credit and incentives programs . . . . '' \186\ DENSO also
supported maintaining the current provisions.\187\ However, BorgWarner
supported the proposed removal of A/C refrigerant credits ``for
harmonization reasons,'' while encouraging EPA to regulate A/C
refrigerants and leakage separately from the CO2
standards.\188\
---------------------------------------------------------------------------
\186\ MEMA, available at https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf, comment at p. 2. MEMA also expressed
concern about stranded capital investments by suppliers and supplier
jobs if the direct MAC credits were not available; stated that the
credits were an important compliance flexibility and ``one of the
highest values of any credit offered in the EPA program;'' and
stated that ``Harmonizing the programs does not require making them
identical or equivalent. Rather, harmonization can be achieved by
better coordinating the two programs to the extent feasible while
allowing each agency to implement its separate and distinct
mandate.'' Id. at 15-16.
\187\ DENSO, NHTSA-2018-0067-11880, at 8.
\188\ BorgWarner, NHTSA-2018-0067-11895, at 10.
---------------------------------------------------------------------------
The two producers of a lower GWP refrigerant, Chemours and
Honeywell, commented extensively in support of continuing to allow A/C
refrigerant and leakage credits for CO2 compliance, making
both economic and legal arguments. Both Chemours and Honeywell stated
that A/C refrigerant and leakage credits were a highly cost-effective
way for OEMs to comply with the CO2 standards,\189\ with
Chemours suggesting that OEM compliance strategies are based on the
assumption that these credits will be available for CO2
compliance \190\ and that any increase in stringency above the proposal
effectively necessitates that the credits remain part of the
program.\191\ Honeywell stated that all OEMs (and a variety of other
parties) supported retaining the credits for CO2
compliance,\192\ and Chemours, Honeywell, and CBD et al. all noted that
OEMs are already using the credits for low GWP refrigerants in more
than 50 percent of the MY 2018 vehicles produced for sale in the
U.S.\193\ The American Chemistry Council also stated that the ``auto
industry widely supports the credits, and U.S. chemical manufacturers
are at a loss as to why EPA would propose to eliminate such a
successful flexible compliance program.'' \194\ In response to NPRM
statements expressing concern that the A/C refrigerant and leakage
credits could be market distorting, both Chemours and Honeywell
disagreed,\195\ arguing that the credits were simply a highly cost-
effective means of complying with the CO2 standards,\196\
and that removal of the credits at this point would, itself, distort
the market for refrigerants. Honeywell argued that eliminating the A/C
credit program from CO2 compliance would put the U.S. at a
competitive disadvantage with other countries, and would risk U.S.
jobs.\197\
---------------------------------------------------------------------------
\189\ Chemours at 1 (``MVAC credits many times offer the `least
cost' approach to compliance . . .'') and 9; Honeywell at 6.
\190\ Chemours at 6-7; both Chemours and Honeywell expressed
concern about OEM reliance on the expectation that HFC credits would
continue to be part of the CO2 program (Chemours at 31;
Honeywell at 16-20) and that investments in alternative refrigerants
would be stranded (Chemours at 1, 3, 4-6; Honeywell at 2, 7-8).
\191\ Chemours at 7.
\192\ Honeywell at 8-11.
\193\ Chemours at 4; Honeywell at 6-7; CBD et al. at 46-47.
\194\ American Chemistry Council, EPA-HQ-OAR-2018-0283-1415, at
9-10 (comments similar to Chemours and Honeywell).
\195\ Chemours at 1; Honeywell at 13.
\196\ Chemours at 29-30; Honeywell at 13-14.
\197\ Honeywell at 20-21.
---------------------------------------------------------------------------
Regarding the NPRM's statements that removing the A/C refrigerant
and leakage credits from CO2 compliance would promote
harmonization with the CAFE program, these commenters argued that
harmonization was not a valid basis for that aspect of the proposal.
Chemours, Honeywell, and CBD et al. all argued that Section 202(a)
creates no obligation to harmonize the [CO2] program with
the CAFE program.\198\ Chemours further argued that to the extent
disharmonization between the programs existed, it should be addressed
via stringency changes (i.e., reducing CAFE stringency relative to
CO2 stringency) rather than ``dropping low-cost compliance
options.'' \199\
---------------------------------------------------------------------------
\198\ Chemours at 23-24; Honeywell at 11-12; CBD et al. at 47.
\199\ Chemours at 9-11.
---------------------------------------------------------------------------
These commenters also expressed concern that the proposal
constituted an EPA decision not to regulate HFC emissions from motor
vehicles at all. Commenters argued that the NPRM provided no legal
analysis or reasoned explanation for stopping regulation of HFCs,\200\
and that Massachusetts v. EPA requires any final rule to regulate all
greenhouse gases from motor vehicles and not CO2 alone,\201\
suggesting that there was a high likelihood of a lapse in regulation
because EPA had not yet proposed a new way of regulating HFC
emissions.\202\ Because the NPRM provided no specific information about
how EPA might regulate non-CO2 emissions separately,
commenters argued that the lack of clarity was inherently disruptive to
OEMs.\203\ CBD et al. argued that any lapse in regulation is ``illegal
on its face'' and that even creating a separate standard for HFC
emissions would be ``illegal'' because it ``would increase costs to
manufacturers and result in environmental detriment by removing any
incentive to use the most aggressive approaches to curtail emissions of
these highly potent GHGs.'' \204\
---------------------------------------------------------------------------
\200\ Chemours at 1-2; Honeywell at 11.
\201\ Chemours at 18-19; Honeywell at 14-16.
\202\ Chemours at 6; Honeywell at 16.
\203\ Chemours at 21; Honeywell at 16; ICCT at I-39.
\204\ CBD et al. at 46.
---------------------------------------------------------------------------
Environmental organizations,\205\ other NGOs, academic
institutions, consumer organizations, and state governments \206\ also
commented in support of continuing the existing provisions.
---------------------------------------------------------------------------
\205\ ICCT, NHTSA-2018-0067-11741, Full Comments, at 4
(describing ``air conditioning GHG-reduction technologies [as]
available, cost-effective, and experiencing increased deployment by
many companies due to the standards.''); CBD et al., Appendix A, at
45-47.
\206\ CARB, NHTSA-2018-0067-11873, Detailed Comments, at 120-
121; Washington State Department of Ecology, NHTSA-2018-0067-11926,
at 6 (HFCs are an important GHG; compliance flexibility is
important).
---------------------------------------------------------------------------
EPA has considered its proposed approach to A/C refrigerant and
leakage
[[Page 24245]]
credits in light of these comments. EPA believes that maintaining this
element of its program is consistent with EPA's authority under Section
202(a) to establish standards for reducing emissions from LDVs. Thus,
maintaining the optional HFC credit program is appropriate. In
addition, EPA recognizes the value of regulatory flexibility and
compliance options, and has concluded that the advantages from
retaining the existing A/C refrigerant/leakage credit program and
associated offset between the CO2 and CAFE standards--in
terms of providing for a more-comprehensive regulation of emissions
from light-duty vehicles--outweigh the disadvantages resulting from the
lack of harmonization.
Regarding the comment from BorgWarner about how having a separate
A/C refrigerant and leakage regulation would allow for better
harmonization between the programs, the agencies accept this to be an
accurate statement, but believe the benefits of continued refrigerant
regulation as an option for CO2 compliance outweigh the
problems associated with lack of harmonization with the CAFE program.
For these reasons, EPA is not finalizing the proposed provisions,
and is making no changes in the A/C refrigerant and leakage-related
provisions of the current program. In light of this conclusion, EPA
does not need to address the legal arguments made by CBD et al. and
CARB about regulating refrigerant-related emissions separately, or
potential lapses in regulation of refrigerant emissions while such a
program could be developed.
As with A/C refrigerant and leakage credits, EPA proposed to
exclude nitrous oxide and methane from average performance calculations
after model year 2020, thereby removing these optional program
flexibilities. Alternatives 1 through 8 excluded this option. EPA
sought comment on whether to remove those aspects of the program that
allow a manufacturer to use nitrous oxide and methane emissions
reductions for compliance with its CO2 average fleet
standards because such a flexibility is not allowed in the NHTSA CAFE
program, or whether to retain the flexibilities as a feature that
differs between the programs. Further, EPA sought comment on whether to
change the existing methane and nitrous oxide standards. Specifically,
EPA requested information from the public on whether the existing
standards are appropriate, or whether they should be revised to be less
stringent or more stringent based on any updated data.
The Alliance in its comments may have misunderstood EPA's proposal
to mean that EPA was proposing to eliminate regulation of methane and
nitrous oxide emissions altogether. The Alliance commented in support
of such a proposal as they understood it, to eliminate the standards to
provide better harmony between the two compliance programs.\207\ The
Alliance commented that ``[n]ot only is emission of these two
substances from vehicles a relatively minor contribution to GHG
emissions, the Alliance has continuing concern regarding measurement
and testing technologies for nitrous oxide.'' \208\ The Alliance
commented further that if ``EPA decides instead to continue to regulate
methane and nitrous oxide, the Alliance recommends that EPA re-assess
whether the levels of the standards remain appropriate and to retain
the current compliance flexibilities. Furthermore, in this scenario,
the Alliance also recommends that methane and nitrous oxide standards
be assessed as a fleet average and as the average of FTP and HFET test
cycles.'' \209\ Several individual manufacturers submitted similar
comments, including Ford,\210\ FCA,\211\ Volvo,\212\ and Mazda.\213\
Ford also commented that it does not support the proposal to maintain
the existing N2O/CH4 standards while removing the
program flexibilities.\214\
---------------------------------------------------------------------------
\207\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 13.
\208\ Id.
\209\ Id.
\210\ Ford, EPA-HQ-OAR-2018-0283-5691, at 4.
\211\ FCA, NHTSA-2018-0067-11943, at 9.
\212\ Volvo, NHTSA-2018-0067-12036, at 5.
\213\ Mazda, NHTSA-2018-0067-11727, at 3 (``In reality, these
emissions are at deminimis levels and have very little, if any,
impact on global warming. So, the need to regulate these emissions
as part of the GHG program, or separately, is unclear. Although most
current engines can comply with the existing requirements, there are
some existing and upcoming new technologies that may not be able to
fully comply. These technologies can provide substantial
CO2 reductions.'').
\214\ Ford, at 4 (``Finally, without the ability to incorporate
exceedances into CREE, each vehicle will need to employ hardware
solutions if they do not comply. We do not believe it was EPA's
intent in the original rulemaking to require additional after-
treatment, with associated cost increases, explicitly for the
control and reduction of an insignificant contributor to GHG
emissions. Therefore, we do not support the proposal to maintain the
existing N2O/CH4 standards while removing the
CREE exceedance pathway.'').
---------------------------------------------------------------------------
The Alliance further commented that ``data from the 2016 EPA report
on light-duty vehicle emissions supports the position that
CH4 and N2O have minimal impact on total GHG
emissions, reporting only 0.045 percent in exceedance of the standard.
This new information makes it apparent that CH4 and
N2O contribute a de minimis amount to GHG emissions.
Additionally, gasoline CH4 and N2O performance is
within the current standards. Finally, the main producers of
CH4 and N2O emissions are flex fuel (E85) and
diesel vehicles, and these vehicles have been declining in sales as
compared to gasoline-fueled vehicles.'' \215\ The Alliance also
commented that CH4 and N2O have minimal
opportunities to be catalytically treated, as N2O is
generated in the catalyst and CH4 has a low conversion
efficiency compared to other emissions. EPA did not intend that
additional hardware should be required to comply with the
CH4 or N2O standards on any vehicle.'' \216\
---------------------------------------------------------------------------
\215\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 43.
\216\ Id. at 44.
---------------------------------------------------------------------------
Global Automakers commented in support of continuing inclusion of
nitrous oxide and methane emissions standards for all MYs, even if it
means a divergence from the NHTSA standards for these program elements
in the regulations, ``because they are complementary to EPA's program,
and are better managed through a coordinated federal policy. They are
also important to maintaining regulatory flexibility through real
[CO2] emission reductions and would prevent the potential
for additional bifurcated, separate programs at the state level.''
\217\ Global Automakers recommended that they remain in place per the
existing program but continued to support that the N2O
testing is not necessary. Global Automakers commented that it
``strongly recommends reducing the need for N2O testing or
eliminating these test requirements in their entirety. It should be
sufficient to allow manufacturers to attest to compliance with the
N2O capped standards based upon good engineering judgment,
development testing, and correlation to NOX emissions. EPA
could, however, maintain the option to request testing to be performed
for new technologies only, which could have unknown impacts on
N2O emissions.'' \218\ Hyundai \219\ and Kia \220\ submitted
similar comments.
---------------------------------------------------------------------------
\217\ Global, NHTSA-2018-0067-12032, at 4, 5.
\218\ Global, Appendix A, NHTSA-2018-0067-12032, at A-44, fn.
89.
\219\ Hyundai, EPA-HQ-OAR-2018-0283-4411, at 7.
\220\ Kia, EPA-HQ-OAR-2018-0283-4195, at 8-9.
---------------------------------------------------------------------------
Others commented in support of retaining the existing program. MECA
commented that it supports the existing standards for methane and
nitrous oxide because catalyst technologies provided by MECA members
that reduce these climate forcing gases are readily
[[Page 24246]]
available and cost-effective.\221\ MECA also commented that the ability
to trade reductions in these pollutants in exchange for CO2
gives vehicle manufacturers the flexibilities they need to comply with
the emission limits by the most cost-effective means.\222\ CBD et al.
commented that the alternative compliance mechanisms currently
available in the program exist to provide cost-effective options for
compliance, and were considered by manufacturers to be a necessary
element of the program for certain types of vehicles.\223\ CBD et al.
further argued that ``[e]liminating these flexibilities consequently
imposes costs on manufacturers without discernible environmental
benefits,'' and suggested that harmonization with the CAFE program was
not a relevant decision factor for EPA.\224\ Several other parties
commented generally in support of retaining the existing program for A/
C leakage credits, discussed above, and N2O and
CH4 standards.\225\
---------------------------------------------------------------------------
\221\ MECA, NHTSA-2018-0067-11994, at 12.
\222\ Id.
\223\ CBD et al. at 48.
\224\ Id.
\225\ Washington State Department of Ecology, NHTSA-2018-0067-
11926, at 6.
---------------------------------------------------------------------------
After considering these comments, EPA is retaining the regulatory
provisions related to the N2O and CH4 standards
with no changes, specifically including the existing flexibilities that
accompany those standards. EPA is not adopting its proposal to exclude
nitrous oxide and methane emissions from average performance
calculations after model year 2020 or any other changes to the program.
The standards continue to serve their intended purpose of capping
emissions of those pollutants and providing for more-comprehensive
regulation of emissions from light-duty vehicles. The standards were
intended to prevent future emissions increases, and these standards
were generally not expected to result in the application of new
technologies or significant costs for manufacturers using current
vehicle designs.\226\ The program flexibilities are working as intended
and all manufacturers are successfully complying with the standards.
Most vehicle models are well below the standards and for those that are
above the standards, manufacturers have used the flexibilities to
offset exceedances with CO2 improvements to demonstrate
compliance. EPA did not receive any data in response to its request for
comments supporting potential alternative levels of stringency.
---------------------------------------------------------------------------
\226\ 77 FR 62624, at 62799 (Oct 15, 2012).
---------------------------------------------------------------------------
While the Alliance and several individual manufacturers recommended
eliminating the standards altogether, EPA did not propose to eliminate
the standards, but to eliminate the optional flexibilities, and
solicited comment on adjusting the standards to be more or less
stringent. Thus, EPA does not believe it would be appropriate to
eliminate completely the standards in this final rule without providing
an opportunity for comment on that idea. Furthermore, as noted above,
EPA believes the standards are continuing to serve their intended
purpose of capping emissions and remain appropriate. Manufacturers have
been subject to the standards for several years, and the Alliance
acknowledges in their comments that the exceedance of the standards,
which is offset by manufacturers using compliance flexibilities, is
very small and that most vehicles meet the standards. Regarding the
Alliance comments that the standards should be based on a fleet average
approach, EPA notes that the purpose of the standards is to cap
emissions, not to achieve fleet-wide reductions.\227\ The fleet average
emissions for N2O and CH4 are well below the
numerical level of the cap standards and therefore the existing cap
standards would not be an appropriate fleet average standard. Adopting
a fleet average approach using the same numerical level as the
established cap standards would not achieve the intended goal of
capping emissions at current levels. If technologies lead to
exceedances of the caps, automakers have the opportunity to apply
appropriate flexibilities under the current program to achieve GHG
emission neutrality. EPA is not aware of any manufacturer that has been
prevented from bringing a technology to the marketplace because of the
current cap levels or approach. EPA believes it would need to consider
all options further, with an opportunity for public comment, before
adopting such a significant change to the program.
---------------------------------------------------------------------------
\227\ Relatedly, the Alliance and Global Automakers raised
concerns in their comments regarding N2O measurement and
testing burden. EPA did not propose any changes in testing
requirements and at this time EPA is not adopting any changes.
Manufacturers have been measuring N2O emissions and have
successfully certified vehicles to the N2O standards for
several years and EPA does not believe N2O measurement is
an issue needing regulatory change. EPA continues to believe direct
measurement is the best way for manufacturers to demonstrate
compliance with the N2O standards and is more appropriate
than an engineering statement without direct measurement.
---------------------------------------------------------------------------
As explained above, the agencies have changed the alternatives
considered for the final rule, partly in response to comments. The
basic form of the standards represented by the alternatives--footprint-
based, defined by particular mathematical functions--remains the same
and as described in the NPRM. For the EPA program, EPA has chosen in
this final rule to retain the existing program for regulation of A/C
refrigerant leakage, nitrous oxide, and methane emissions as part of
the CO2 standard. This allows manufacturers to continue to
rely on this flexibility which they describe as extremely important for
compliance, although it results in continued differences between EPA's
and NHTSA's programs. This approach also avoids the possibility of gaps
in the regulation of HFCs, CH4, and N2O while EPA
developed a different way of regulating the non-CO2
emissions as part of or concurrent with the NPRM, and thereby allows
EPA to continue to regulate GHE emissions from light-duty vehicles on a
more-comprehensive basis. Thus, all alternatives considered in this
final rule reflect inclusion of CH4, N2O, and HFC
in EPA's overall ``CO2'' (more accurately, CO2-
equivalent, or CO2e) requirements. Besides this change, the
alternatives considered for the final rule differ from the NPRM in two
additional ways: First, alternatives reflecting the phase-out of the A/
C efficiency and off-cycle programs have been dropped in response to
certain comments and in recognition of the potential real-world
benefits of those programs. And second, the preferred alternative for
this final rule reflects a 1.5 percent year-over-year increase for both
passenger cars and light trucks. These changes will be discussed
further below, following a brief discussion of the form of the
standards.
A. Form of the Standards
As in the CAFE and CO2 rulemakings in 2010 and 2012,
NHTSA and EPA proposed in the NPRM to set attribute-based CAFE and
CO2 standards defined by a mathematical function of vehicle
footprint, which has observable correlation with fuel economy and
vehicle emissions. EPCA, as amended by EISA, expressly requires that
CAFE standards for passenger cars and light trucks be based on one or
more vehicle attributes related to fuel economy and be expressed in the
form of a mathematical function.\228\ While the CAA includes no
specific requirements regarding CO2 regulation, EPA has
chosen to adopt attribute-based CO2 standards consistent
with NHTSA's EPCA/EISA requirements in the interest of harmonization
and simplifying compliance. Such an approach is permissible under
section 202(a) of the
[[Page 24247]]
CAA, and EPA has used the attribute-based approach in issuing standards
under analogous provisions of the CAA. Thus, both the proposed and
final standards take the form of fuel economy and CO2
targets expressed as functions of vehicle footprint (the product of
vehicle wheelbase and average track width). Section V.A.2 below
discusses the agencies' continued reliance on footprint as the relevant
attribute.
---------------------------------------------------------------------------
\228\ 49 U.S.C. 32902(a)(3)(A).
---------------------------------------------------------------------------
Under the footprint-based standards, the function defines a
CO2 or fuel economy performance target for each unique
footprint combination within a car or truck model type. Using the
functions, each manufacturer thus will have a CAFE and CO2
average standard for each year that is almost certainly unique to each
of its fleets,\229\ based upon the footprints and production volumes of
the vehicle models produced by that manufacturer. A manufacturer will
have separate footprint-based standards for cars and for trucks. The
functions are mostly sloped, so that generally, larger vehicles (i.e.,
vehicles with larger footprints) will be subject to lower CAFE mpg
targets and higher CO2 grams/mile targets than smaller
vehicles. This is because, generally speaking, smaller vehicles are
more capable of achieving higher levels of fuel economy/lower levels of
CO2 emissions, mostly because they tend not to have to work
as hard (and therefore require as much energy) to perform their driving
task. Although a manufacturer's fleet average standards could be
estimated throughout the model year based on the projected production
volume of its vehicle fleet (and are estimated as part of EPA's
certification process), the standards to which the manufacturer must
comply are determined by its final model year production figures. A
manufacturer's calculation of its fleet average standards as well as
its fleets' average performance at the end of the model year will thus
be based on the production-weighted average target and performance of
each model in its fleet.\230\
---------------------------------------------------------------------------
\229\ EPCA/EISA requires NHTSA to separate passenger cars into
domestic and import passenger car fleets whereas EPA combines all
passenger cars into one fleet.
\230\ As discussed in prior rulemakings, a manufacturer may have
some vehicle models that exceed their target and some that are below
their target. Compliance with a fleet average standard is determined
by comparing the fleet average standard (based on the production-
weighted average of the target levels for each model) with fleet
average performance (based on the production-weighted average of the
performance of each model).
---------------------------------------------------------------------------
For passenger cars, consistent with prior rulemakings, NHTSA is
defining fuel economy targets as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.051
where:
TARGETFE is the fuel economy target (in mpg) applicable to a
specific vehicle model type with a unique footprint combination,
a is a minimum fuel economy target (in mpg),
b is a maximum fuel economy target (in mpg),
c is the slope (in gallons per mile per square foot, or gpm, per
square foot) of a line relating fuel consumption (the inverse of
fuel economy) to footprint, and
d is an intercept (in gpm) of the same line.
Here, MIN and MAX are functions that take the minimum and maximum
values, respectively, of the set of included values. For example,
MIN[40,35] = 35 and MAX(40, 25) = 40, such that MIN[MAX(40, 25), 35] =
35.
For light trucks, also consistent with prior rulemakings, NHTSA is
defining fuel economy targets as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.052
where:
TARGETFE is the fuel economy target (in mpg) applicable to a
specific vehicle model type with a unique footprint combination,
a, b, c, and d are as for passenger cars, but taking values specific
to light trucks,
e is a second minimum fuel economy target (in mpg),
f is a second maximum fuel economy target (in mpg),
g is the slope (in gpm per square foot) of a second line relating
fuel consumption (the inverse of fuel economy) to footprint, and
h is an intercept (in gpm) of the same second line.
Although the general model of the target function equation is the
same for each vehicle category (passenger cars and light trucks) and
each model year, the parameters of the function equation differ for
cars and trucks. For MYs 2020-2026, the parameters are unchanged,
resulting in the same stringency in each of those model years.
Mathematical functions defining the CO2 targets are
expressed as functions that are similar, with coefficients a-h
corresponding to those listed above.\231\ For passenger cars, EPA is
defining CO2 targets mathematically equivalent to the
following:
---------------------------------------------------------------------------
\231\ EPA regulations use a different but mathematically
equivalent approach to specify targets. Rather than using a function
with nested minima and maxima functions, EPA regulations specify
requirements separately for different ranges of vehicle footprint.
Because these ranges reflect the combined application of the listed
minima, maxima, and linear functions, it is mathematically
equivalent and more efficient to present the targets as in this
Section.
---------------------------------------------------------------------------
TARGETCO2 = MIN[b, MAX[a, c x FOOTPRINT + d]]
where:
TARGETCO2 is the is the CO2 target (in grams per mile, or
g/mi) applicable to a specific vehicle model configuration,
a is a minimum CO2 target (in g/mi),
b is a maximum CO2 target (in g/mi),
c is the slope (in g/mi, per square foot) of a line relating
CO2 emissions to footprint, and
d is an intercept (in g/mi) of the same line.
For light trucks, CO2 targets are defined as follows:
TARGETCO2 = MIN[MIN[b, MAX[a, c x FOOTPRINT + d]], MIN[f, MAX[e, g x
FOOTPRINT + h]]
[[Page 24248]]
where:
TARGETCO2 is the is the CO2 target (in g/mi) applicable
to a specific vehicle model configuration,
a, b, c, and d are as for passenger cars, but taking values specific
to light trucks,
e is a second minimum CO2 target (in g/mi),
f is a second maximum CO2 target (in g/mi),
g is the slope (in g/mi per square foot) of a second line relating
CO2 emissions to footprint, and
h is an intercept (in g/mi) of the same second line.
To be clear, as has been the case since the agencies began
establishing attribute-based standards, no vehicle need meet the
specific applicable fuel economy or CO2 targets, because
compliance with either CAFE or CO2 standards is determined
based on corporate average fuel economy or fleet average CO2
emission rates. In this respect, CAFE and CO2 standards are
unlike, for example, safety standards and traditional vehicle emissions
standards. CAFE and CO2 standards apply to the average fuel
economy levels and CO2 emission rates achieved by
manufacturers' entire fleets of vehicles produced for sale in the U.S.
Safety standards apply on a vehicle-by-vehicle basis, such that every
single vehicle produced for sale in the U.S. must, on its own, comply
with minimum FMVSS. Similarly, criteria pollutant emissions standards
are applied on a per-vehicle basis, such that every vehicle produced
for sale in the U.S. must, on its own, comply with all applicable
emissions standards. When first mandating CAFE standards in the 1970s,
Congress specified a more flexible averaging-based approach that allows
some vehicles to ``under comply'' (i.e., fall short of the overall flat
standard, or fall short of their target under attribute-based
standards) as long as a manufacturer's overall fleet is in compliance.
The required CAFE level applicable to a given fleet in a given
model year is determined by calculating the production-weighted
harmonic average of fuel economy targets applicable to specific vehicle
model configurations in the fleet, as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.053
where:
CAFErequired is the CAFE level the fleet is required to achieve,
i refers to specific vehicle model/configurations in the fleet,
PRODUCTIONi is the number of model configuration i produced for sale
in the U.S., and
TARGETFE,i the fuel economy target (as defined above) for model
configuration i.
Similarly, the required average CO2 level applicable to
a given fleet in a given model year is determined by calculating the
production-weighted average (not harmonic) of CO2 targets
applicable to specific vehicle model configurations in the fleet, as
follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.054
where:
CO2required is the average CO2 level the fleet is
required to achieve,
i refers to specific vehicle model/configurations in the fleet,
PRODUCTIONi is the number of model configuration i produced for sale
in the U.S., and
TARGETCO2,i is the CO2 target (as defined above) for
model configuration i.
Section VI.A.1 describes the advantages of attribute standards,
generally. Section VI.A.2 explains the agencies' specific decision to
use vehicle footprint as the attribute over which to vary stringency
for past and current rules. Section VI.A.3 discusses the policy
considerations in selecting the specific mathematical function. Section
VI.A.4 discusses the methodologies used to develop current attribute-
based standards, and the agencies' current proposal to continue to do
so for MYs 2021-2026. Section VI.A.5 discusses the methodologies used
to reconsider the mathematical function for the proposed standards.
1. Why attribute-based standards, and what are the benefits?
Under attribute-based standards, every vehicle model has fuel
economy and CO2 targets, the levels of which depend on the
level of that vehicle's determining attribute (for the MYs 2021-2026
standards, footprint is the determining attribute, as discussed below).
The manufacturer's fleet average CAFE performance is calculated by the
harmonic production-weighted average of those targets, as defined
below:
[GRAPHIC] [TIFF OMITTED] TR30AP20.055
Here, i represents a given model \232\ in a manufacturer's
fleet, Productioni represents the U.S. production of that model, and
Targeti represents the target as defined by the attribute-based
standards. This means no vehicle is required to meet its target;
instead, manufacturers are free to balance improvements however they
deem best within (and, given credit transfers, at least partially
across) their fleets.
\232\ If a model has more than one footprint variant, here each
of those variants is treated as a unique model, i, since each
footprint variant will have a unique target.
---------------------------------------------------------------------------
Because CO2 is on a gram per mile basis rather a mile
per gallon basis,
[[Page 24249]]
harmonic averaging is not necessary when calculating required
CO2 levels:
[GRAPHIC] [TIFF OMITTED] TR30AP20.056
The idea is to select the shape of the mathematical function
relating the standard to the fuel economy-related attribute to reflect
the trade-offs manufacturers face in producing more of that attribute
over fuel efficiency (due to technological limits of production and
relative demand of each attribute). If the shape captures these trade-
offs, every manufacturer is more likely to continue adding fuel-
efficient technology across the distribution of the attribute within
their fleet, instead of potentially changing the attribute--and other
correlated attributes, including fuel economy--as a part of their
compliance strategy. Attribute-based standards that achieve this have
several advantages.
First, assuming the attribute is a measurement of vehicle size,
attribute-based standards help to at least partially reduce the
incentive for manufacturers to respond to CAFE and CO2
standards by reducing vehicle size in ways harmful to safety, as
compared to ``flat,'' non-attribute based standards.\233\ Larger
vehicles, in terms of mass and/or crush space, generally consume more
fuel and produce more carbon dioxide emissions, but are also generally
better able to protect occupants in a crash.\234\ Because each vehicle
model has its own target (determined by a size-related attribute),
properly fitted attribute-based standards reduce the incentive to build
smaller vehicles simply to meet a fleet-wide average, because smaller
vehicles are subject to more stringent compliance targets.
---------------------------------------------------------------------------
\233\ The 2002 NAS Report described at length and quantified the
potential safety problem with average fuel economy standards that
specify a single numerical requirement for the entire industry. See
Transportation Research Board and National Research Council. 2002.
Effectiveness and Impact of Corporate Average Fuel Economy (CAFE)
Standards, Washington, DC: The National Academies Press (``2002 NAS
Report'') at 5, finding 12, available at https://www.nap.edu/catalog/10172/effectiveness-and-impact-of-corporate-average-fuel-economy-cafe-standards (last accessed June 15, 2018). Ensuing
analyses, including by NHTSA, support the fundamental conclusion
that standards structured to minimize incentives to downsize all but
the largest vehicles will tend to produce better safety outcomes
than flat standards.
\234\ Bento, A., Gillingham, K., & Roth, K. (2017). The Effect
of Fuel Economy Standards on Vehicle Weight Dispersion and Accident
Fatalities. NBER Working Paper No. 23340. Available at http://www.nber.org/papers/w23340 (last accessed June 15, 2018).
---------------------------------------------------------------------------
Second, attribute-based standards, if properly fitted, provide
automakers with more flexibility to respond to consumer preferences
than do single-valued standards. As discussed above, a single-valued
standard encourages a fleet mix with a larger share of smaller vehicles
by creating incentives for manufacturers to use downsizing the average
vehicle in their fleet (possibly through fleet mixing) as a compliance
strategy, which may result in manufacturers building vehicles for
compliance reasons that consumers do not want. Under a size-related,
attribute-based standard, reducing the size of the vehicle for
compliance's sake is a less-viable strategy because smaller vehicles
have more stringent regulatory targets. As a result, the fleet mix
under such standards is more likely to reflect aggregate consumer
demand for the size-related attribute used to determine vehicle
targets.
Third, attribute-based standards provide a more equitable
regulatory framework across heterogeneous manufacturers who may each
produce different shares of vehicles along attributes correlated with
fuel economy.\235\ An industry-wide single-value CAFE standard imposes
disproportionate cost burden and compliance challenges on manufacturers
who produce more vehicles with attributes inherently correlated with
lower fuel economy--i.e. manufacturers who produce, on average, larger
vehicles. As discussed above, retaining flexibility for manufacturers
to produce vehicles which respect heterogeneous market preferences is
an important consideration. Since manufacturers may target different
markets as a part of their business strategy, ensuring that these
manufacturers do not incur a disproportionate share of the regulatory
cost burden is an important part of conserving consumer choices within
the market.
---------------------------------------------------------------------------
\235\ 2002 NAS Report at 4-5, finding 10.
---------------------------------------------------------------------------
Industry commenters generally supported attribute-based standards,
while other commenters questioned their benefits. IPI argued that
preserving the current vehicle mix was not necessarily desirable or
necessary for consumer welfare, and suggested that some vehicle
downsizing in the fleet might be beneficial both for safety and for
compliance.\236\ IPI also argued that compliance credit trading would
``help smooth out any disproportionate impacts on certain
manufacturers'' and ``ensure that manufacturers with relatively
efficient fleets still have an incentive to continue improving fuel
economy (in order to generate credits)'' \237\ Similarly, citing Ito
and Sallee, Kathryn Doolittle commented that ``. . . Ito and Sallee
(2018) have found ABR [``attribute-based regulations''] inefficient in
cost when juxtaposed with flat standard with compliance trading.''
\238\
---------------------------------------------------------------------------
\236\ IPI, NHTSA-2018-0067-12362, at 14-15.
\237\ IPI, NHTSA-2018-0067-12362, at 14.
\238\ Doolittle, K, NHTSA-2018-0067-7411. See also Ito, K and
Sallee, J. ``The Economics of Attribute-Based Regulation: Theory and
Evidence from Fuel Economy Standards.'' The Review of Economics and
Statistics (2018), 100(2), pp. 319-36.
---------------------------------------------------------------------------
The agencies have considered these comments. IPI incorrectly
characterizes the agencies' prior statements as claims that it is
important to preserve the current vehicle mix. EPA and NHTSA have never
claimed, and are not today claiming that it is important to preserve
the current fleet mix. The agencies have said, and are today
reiterating, that it is reasonable to expect that reducing the tendency
of standards to distort the market should reduce at least part of the
tendency of standards to reduce consumer welfare. Or, more concisely,
it is better to work with the market than against it. Single-value (aka
flat) CAFE standards in place from the 1970s through 2010 were clearly
distortionary. Recognizing this, the National Academy of Sciences
recommended in 2002 that NHTSA adopt attribute-based CAFE standards.
NHTSA did so in 2006, for light trucks produced starting MY 2008. As
mentioned above, in 2007, Congress codified the requirement for
attribute-based passenger car and light truck CAFE standards. Agreeing
with this history, premise, and motivation, EPA has also adopted
attribute-based CO2 standards. None of this is to say the
agencies consider it important to hold fleet mix constant. Rather, the
agencies expect that, compared to flat standards, attribute-based
standards can allow the market--including fleet mix--to better
[[Page 24250]]
follow its natural course, and all else equal, consumer acceptance is
likely to be greater if the market does so.
The agencies also disagree with comments implying that compliance
credit trading can address all of the market distortion that flat
standards would entail. Evidence thus far suggests that trading is
fragmented, with some manufacturers apparently willing to trade only
with some other specific manufacturers. The Ito and Sallee article
cited by one commenter is a highly idealized theoretical construction,
with the authors noting, inter alia, that their model ``assumes perfect
competition.'' \239\ Its findings regarding comparative economic
efficiency of flat- and attribute-based standards are, therefore,
merely hypothetical, and the agencies find little basis in recent
transactions to suggest the compliance credit trading market reflects
the authors' idealized assumptions. Even if the agencies did expect
credit trading markets to operate as in an idealized textbook example,
basing the structure of standards on the presumption of perfect trading
would not be appropriate. FCA commented that ``. . . when flexibilities
are considered while setting targets, they cease to be flexibilities
and become simply additional technology mandates,'' and the Alliance
commented, similarly, that ``the Agencies should keep `flexibilities'
as optional ways to comply and not unduly assume that each flexibility
allows additional stringency of footprint-based standards.'' \240\
Perhaps recognizing this reality, Congress has barred NHTSA from
considering manufacturers' ability to use compliance credits (even
credits earned and used by the same OEM, much less credits traded
between OEMs). As discussed further in Section VIII.A.2, EPA believes
that while credit trading may be a useful flexibility to reduce the
overall costs of the program, it is important to set standards in a way
that does not rely on credit purchasing availability as a compliance
mechanism.
---------------------------------------------------------------------------
\239\ Ito and Sallee, op. cit., Supplemental Appendix, at A-15,
available at https://www.mitpressjournals.org/doi/suppl/10.1162/REST_a_00704/suppl_file/REST_a_00704-esupp.pdf (accessed October 29,
2019).
\240\ FCA, NHTSA-2018-0067-11943, at 6; Alliance, NHTSA-2018-
0067-12073, Full Comment Set, at 40, fn. 82.
---------------------------------------------------------------------------
Considering these comments and realities, considering EPCA's
requirement for attribute-based CAFE standards, and considering the
benefits of regulatory harmonization, the agencies are, again,
finalizing attribute-based CAFE and CO2 standards rather
than, for either program, finalizing flat standards.
Why footprint as the attribute?
It is important that the CAFE and CO2 standards be set
in a way that does not unnecessarily incentivize manufacturers to
respond by selling vehicles that are less safe. Vehicle size is highly
correlated with vehicle safety--for this reason, it is important to
choose an attribute correlated with vehicle size (mass or some
dimensional measure). Given this consideration, there are several
policy and technical reasons why footprint is considered to be the most
appropriate attribute upon which to base the standards, even though
other vehicle size attributes (notably, curb weight) are more strongly
correlated with fuel economy and tailpipe CO2 emissions.
First, mass is strongly correlated with fuel economy; it takes a
certain amount of energy to move a certain amount of mass. Footprint
has some positive correlation with frontal surface area, likely a
negative correlation with aerodynamics, and therefore fuel economy, but
the relationship is less deterministic. Mass and crush space
(correlated with footprint) are both important safety considerations.
As discussed below and in the accompanying PRIA, NHTSA's research of
historical crash data indicates that holding footprint constant, and
decreasing the mass of the largest vehicles, will result in a net
positive safety impact to drivers overall, while holding footprint
constant and decreasing the mass of the smallest vehicles will result
in a net decrease in fleetwide safety. Properly fitted footprint-based
standards provide little, if any, incentive to build smaller footprint
vehicles to meet CAFE and CO2 standards, and therefore help
minimize the impact of standards on overall fleet safety.
Second, it is important that the attribute not be easily
manipulated in a manner that does not achieve the goals of EPCA or
other goals, such as safety. Although weight is more strongly
correlated with fuel economy than footprint, there is less risk of
artificial manipulation (i.e., changing the attribute(s) to achieve a
more favorable target) by increasing footprint under footprint-based
standards than there would be by increasing vehicle mass under weight-
based standards. It is relatively easy for a manufacturer to add enough
weight to a vehicle to decrease its applicable fuel economy target a
significant amount, as compared to increasing vehicle footprint, which
is a much more complicated change that typically takes place only with
a vehicle redesign.
Further, some commenters on the MY 2011 CAFE rulemaking were
concerned that there would be greater potential for such manipulation
under multi-attribute standards, such as those that also depend on
weight, torque, power, towing capability, and/or off-road capability.
As discussed in NHTSA's MY 2011 CAFE final rule,\241\ it is anticipated
that the possibility of manipulation is lowest with footprint-based
standards, as opposed to weight-based or multi-attribute-based
standards. Specifically, standards that incorporate weight, torque,
power, towing capability, and/or off-road capability in addition to
footprint would not only be more complex, but by providing degrees of
freedom with respect to more easily adjusted attributes, they could
make it less certain that the future fleet would actually achieve the
projected average fuel economy and CO2 levels. This is not
to say that a footprint-based system eliminates manipulation, or that a
footprint-based system eliminates the possibility that manufacturers
will change vehicles in ways that compromise occupant protection, but
footprint-based standards achieve the best balance among affected
considerations.
---------------------------------------------------------------------------
\241\ See 74 FR at 14359 (Mar. 30, 2009).
---------------------------------------------------------------------------
Several stakeholders commented on whether vehicular footprint is
the most suitable attribute upon which to base standards. IPI commented
that ``. . . footprint-based standards may be unnecessary to respect
consumer preferences, may negatively impact safety, and may be overall
inefficient. Several arguments call into question the footprint-based
approach, but a particularly important one is that large vehicles can
impose a negative safety externality on other drivers.'' \242\ IPI
commented, further, that the agencies should consider the relative
merits of other vehicle attributes, including vehicle fuel type,
suggesting that it would be more difficult for manufacturers to
manipulate a flatter standard or one ``differentiated by fuel type.''
\243\ Similarly, Michalek and Whitefoot recommended ``that the agencies
reexamine automaker response to the footprint-based standards to
determine if adjustments should be made to avoid inducing increases to
vehicle size.'' \244\
---------------------------------------------------------------------------
\242\ IPI, NHTSA-2018-0067-12362, at 12.
\243\ IPI, NHTSA-2018-0067-12362, at 13 et seq.
\244\ Michalek, J. and Whitefoot, K., NHTSA-2018-0067-11903, at
13.
---------------------------------------------------------------------------
[[Page 24251]]
Conversely, ICCT commented that ``the switch to footprint-based
CAFE and [CO2] standards has been widely credited with
diminishing safety concerns with efficiency standards. Footprint
standards encourage larger vehicles with wider track width, which
reduces rollovers, and longer wheelbase, which increases the crush
space and reduces deceleration forces for both vehicles in a two-
vehicle collision.'' \245\ Similarly, BorgWarner commented that ``the
use of a footprint standard not only provides greater incentive for
mass reduction, but also encourages a larger footprint for a given
vehicle mass, thus providing increased safety for a given mass
vehicle,'' \246\ and the Aluminum Association commented footprint based
standards drive ``fuel-efficiency improvement across all vehicle
classes,'' ``eliminate the incentive to shift fleet volume to smaller
cars which has been shown to slightly decrease safety in vehicle-to-
vehicle collisions,'' and provide ``an incentive for reducing weight in
the larger vehicles, where weight reduction is of the most benefit for
societal safety,'' citing Ford's aluminum-intensive F150 pickup truck
as an example.\247\ NADA urged the agencies to continue basing
standards on vehicle footprint, as doing so ``serves both to require
and allow OEMs to build more fuel-efficient vehicles across the
broadest possible light-duty passenger car and truck spectrum,'' \248\
and UCS commented that footprint-based standards ``increase consumer
choice, ensuring that the vehicles available for purchase in every
vehicle class continue to get more efficient.'' \249\ Furthermore,
regarding concerns that footprint-based standards may be susceptible to
manipulation, the Alliance commented that ``the data above [from
Novation Analytics] shows there are no systemic footprint increases (or
any type of target manipulation) occurring.'' \250\ While FCA's
comments supported this Alliance comment, FCA commented further that,
lacking some utility-related vehicle attributes such as towing
capability, 4-wheel-drive, and ride height, ``it is clear the footprint
standard does not fully account for pickup truck capability and the
components needed such as larger powertrains, greater mass and frontal
area,'' and requested the agencies ``correct LDT standards to reflect
the current market preference for capability over efficiency, and
introduce mechanisms into the regulation that can adjust for efficiency
and capability tradeoffs that footprint standards currently ignore.''
\251\
---------------------------------------------------------------------------
\245\ ICCT, NHTSA-2018-0067-11741, at B-4.
\246\ BorgWarner, NHTSA-2018-0067-11893, at 10.
\247\ Aluminum Association, NHTSA-2018-0067-11952, at 3.
\248\ NADA, NHTSA-2018-0067-12064, at 13.
\249\ UCS, UCS, NHTSA-2018-0067-12039, at 46.
\250\ Alliance, NHTSA-2018-0067-12073, at 123.
\251\ FCA, NHTSA-2018-0067-11943, at 49.
---------------------------------------------------------------------------
When first electing to adopt footprint-based standards, NHTSA
carefully considered other alternatives, including vehicle mass and
``shadow'' (overall width multiplied by overall length). Compared to
both of these other alternatives, footprint is much less susceptible to
gaming, because while there is some potential to adjust track width,
wheelbase is more expensive to change, at least outside a planned
vehicle redesign. EPA agreed with NHTSA's assessment, nothing has
changed the relative merits of at least these three potential
attributes, and nothing in the evolution of the fleet demonstrates that
footprint-based standards are leading manufacturers to increase the
footprint of specific vehicle models by more than they would in
response to customer demand. Also, even if footprint-based standards
are encouraging some increases in vehicle size, NHTSA continues to
maintain, and EPA to agree, that such increases should tend to improve
overall highway safety rather than degrading it. Regarding FCA's
request that the agencies adopt an approach that accounts for a wider
range of vehicle attributes related to both vehicle fuel economy and
customer-facing vehicle utility, the agencies are concerned that doing
so could further complicate already-complex standards and also lead to
unintended consequences. For example, it is not currently clear how a
multi-attribute approach would appropriately balance emphasis between
vehicle attributes (e.g., how much relative fuel consumption should be
attributed to, respectively, vehicle footprint, towing capacity, drive
type, and ground clearance). Also, basing standards on, in part, ground
clearance would encourage manufacturers to increase ride height,
potentially increasing the frequency of vehicle rollover crashes.
Regarding IPI's recommendation that fuel type be included as a vehicle
attribute for attribute-based standards, the agencies note that both
CAFE and CO2 standards already account for fuel type in the
procedures for measuring fuel economy levels and CO2
emission rates, and for calculating fleet average CAFE and
CO2 levels.
Therefore, having considered public comments on the choice of
vehicle attributes for CAFE and CO2 standards, the agencies
are finalizing standards that, as proposed, are defined in terms of
vehicle footprint.
3. What mathematical function should be used to specify footprint-based
standards?
In requiring NHTSA to ``prescribe by regulation separate average
fuel economy standards for passenger and non-passenger automobiles
based on 1 or more vehicle attributes related to fuel economy and
express each standard in the form of a mathematical function,'' EPCA/
EISA provides ample discretion regarding not only the selection of the
attribute(s), but also regarding the nature of the function. The CAA
provides no specific direction regarding CO2 regulation, and
EPA has continued to harmonize this aspect of its CO2
regulations with NHTSA's CAFE regulations. The relationship between
fuel economy (and CO2 emissions) and footprint, though
directionally clear (i.e., fuel economy tends to decrease and
CO2 emissions tend to increase with increasing footprint),
is theoretically vague, and quantitatively uncertain; in other words,
not so precise as to a priori yield only a single possible curve.
The decision of how to specify this mathematical function therefore
reflects some amount of judgment. The function can be specified with a
view toward achieving different environmental and petroleum reduction
goals, encouraging different levels of application of fuel-saving
technologies, avoiding any adverse effects on overall highway safety,
reducing disparities of manufacturers' compliance burdens, and
preserving consumer choice, among other aims. The following are among
the specific technical concerns and resultant policy tradeoffs the
agencies have considered in selecting the details of specific past and
future curve shapes:
Flatter standards (i.e., curves) increase the risk that
both the size of vehicles will be reduced, potentially compromising
highway safety, and reducing any utility consumers would have gained
from a larger vehicle.
Steeper footprint-based standards may create incentives to
upsize vehicles, potentially oversupplying vehicles of certain
footprints beyond what consumers would naturally demand, and thus
increasing the possibility that fuel savings and CO2
reduction benefits will be forfeited artificially.
Given the same industry-wide average required fuel economy
or CO2 standard, flatter standards tend to place greater
compliance burdens on full-line manufacturers.
Given the same industry-wide average required fuel economy
or CO2
[[Page 24252]]
standard, dramatically steeper standards tend to place greater
compliance burdens on limited-line manufacturers (depending of course,
on which vehicles are being produced).
If cutpoints are adopted, given the same industry-wide
average required fuel economy, moving small-vehicle cutpoints to the
left (i.e., up in terms of fuel economy, down in terms of
CO2 emissions) discourages the introduction of small
vehicles, and reduces the incentive to downsize small vehicles in ways
that could compromise overall highway safety.
If cutpoints are adopted, given the same industry-wide
average required fuel economy, moving large-vehicle cutpoints to the
right (i.e., down in terms of fuel economy, up in terms of
CO2 emissions) better accommodates the design requirements
of larger vehicles--especially large pickups--and extends the size
range over which downsizing is discouraged.
4. What mathematical functions have been used previously, and why?
Notwithstanding the aforementioned discretion under EPCA/EISA, data
should inform consideration of potential mathematical functions, but
how relevant data is defined and interpreted, and the choice of
methodology for fitting a curve to that data, can and should include
some consideration of specific policy goals. This section summarizes
the methodologies and policy concerns that were considered in
developing previous target curves (for a complete discussion see the
2012 FRIA).
As discussed below, the MY 2011 final curves followed a constrained
logistic function defined specifically in the final rule.\252\ The MYs
2012-2021 final standards and the MYs 2022-2025 augural standards are
defined by constrained linear target functions of footprint, as shown
below: \253\
---------------------------------------------------------------------------
\252\ See 74 FR 14196, 14363-14370 (Mar. 30, 2009) for NHTSA
discussion of curve fitting in the MY 2011 CAFE final rule.
\253\ The right cutpoint for the light truck curve was moved
further to the right for MYs 2017-2021, so that more possible
footprints would fall on the sloped part of the curve. In order to
ensure that, for all possible footprints, future standards would be
at least as high as MY 2016 levels, the final standards for light
trucks for MYs 2017-2021 is the maximum of the MY 2016 target curves
and the target curves for the give MY standard. This is defined
further in the 2012 final rule. See 77 FR 62624, at 62699-700 (Oct.
15, 2012).
[GRAPHIC] [TIFF OMITTED] TR30AP20.057
Here, Target is the fuel economy target applicable to vehicles
of a given footprint in square feet (Footprint). The upper
asymptote, a, and the lower asymptote, b, are specified in mpg; the
reciprocal of these values represent the lower and upper asymptotes,
respectively, when the curve is instead specified in gallons per
mile (gpm). The slope, c, and the intercept, d, of the linear
portion of the curve are specified as gpm per change in square feet,
---------------------------------------------------------------------------
and gpm, respectively.
The min and max functions will take the minimum and maximum values
within their associated parentheses. Thus, the max function will first
find the maximum of the fitted line at a given footprint value and the
lower asymptote from the perspective of gpm. If the fitted line is
below the lower asymptote it is replaced with the floor, which is also
the minimum of the floor and the ceiling by definition, so that the
target in mpg space will be the reciprocal of the floor in mpg space,
or simply, a. If, however, the fitted line is not below the lower
asymptote, the fitted value is returned from the max function and the
min function takes the minimum value of the upper asymptote (in gpm
space) and the fitted line. If the fitted value is below the upper
asymptote, it is between the two asymptotes and the fitted value is
appropriately returned from the min function, making the overall target
in mpg the reciprocal of the fitted line in gpm. If the fitted value is
above the upper asymptote, the upper asymptote is returned is returned
from the min function, and the overall target in mpg is the reciprocal
of the upper asymptote in gpm space, or b.
In this way curves specified as constrained linear functions are
specified by the following parameters:
a = upper limit (mpg)
b = lower limit (mpg)
c = slope (gpm per sq.ft.)
d = intercept (gpm)
The slope and intercept are specified as gpm per sq. ft. and gpm
instead of mpg per sq. ft. and mpg because fuel consumption and
emissions appear roughly linearly related to gallons per mile (the
reciprocal of the miles per gallon).
a) NHTSA in MY 2008 and MY 2011 CAFE (Constrained Logistic)
For the MY 2011 CAFE rule, NHTSA estimated fuel economy levels by
footprint from the MY 2008 fleet after normalization for differences in
technology,\254\ but did not make adjustments to reflect other vehicle
attributes (e.g., power-to-weight ratios). Starting with the
technology-adjusted passenger car and light truck fleets, NHTSA used
minimum absolute deviation (MAD) regression without sales weighting to
fit a logistic form as a starting point to develop mathematical
functions defining the standards. NHTSA then identified footprints at
which to apply minimum and maximum values (rather than letting the
standards extend without limit) and transposed these functions
vertically (i.e., on a gallons-per-mile basis, uniformly downward) to
produce the promulgated standards. In the preceding rule, for MYs 2008-
2011 light truck standards, NHTSA examined a range of potential
functional forms, and concluded that, compared to other considered
forms, the constrained logistic form provided the expected and
appropriate trend (decreasing fuel economy as footprint increases), but
avoided creating ``kinks'' the agency was concerned would provide
distortionary incentives for vehicles with neighboring footprints.\255\
---------------------------------------------------------------------------
\254\ See 74 FR 14196, 14363-14370 (Mar. 30, 2009) for NHTSA
discussion of curve fitting in the MY 2011 CAFE final rule.
\255\ See 71 FR 17556, 17609-17613 (Apr. 6, 2006) for NHTSA
discussion of ``kinks'' in the MYs 2008-2011 light truck CAFE final
rule (there described as ``edge effects''). A ``kink,'' as used
here, is a portion of the curve where a small change in footprint
results in a disproportionally large change in stringency.
---------------------------------------------------------------------------
b) MYs 2012-2016 Standards (Constrained Linear)
For the MYs 2012-2016 rule, potential methods for specifying
mathematical functions to define fuel economy and CO2
standards were reevaluated. These methods were fit to the same MY 2008
data as the MY 2011 standard. Considering these further specifications,
the constrained logistic form, if applied to post-MY 2011 standards,
would likely contain a steep mid-section that would provide undue
incentive to increase the footprint of midsize passenger cars.\256\ A
range of
[[Page 24253]]
methods to fit the curves would have been reasonable, and a minimum
absolute deviation (MAD) regression without sales weighting on a
technology-adjusted car and light truck fleet was used to fit a linear
equation. This equation was used as a starting point to develop
mathematical functions defining the standards. Footprints were then
identified at which to apply minimum and maximum values (rather than
letting the standards extend without limit). Finally, these
constrained/piecewise linear functions were transposed vertically
(i.e., on a gpm or CO2 basis, uniformly downward) by
multiplying the initial curve by a single factor for each MY standard
to produce the final attribute-based targets for passenger cars and
light trucks described in the final rule.\257\ These transformations
are typically presented as percentage improvements over a previous MY
target curve.
---------------------------------------------------------------------------
\256\ 75 FR at 25362.
\257\ See generally 74 FR at 49491-96; 75 FR at 25357-62.
---------------------------------------------------------------------------
c) MYs 2017 and Beyond Standards (Constrained Linear)
The mathematical functions finalized in 2012 for MYs 2017 and
beyond changed somewhat from the functions for the MYs 2012-2016
standards. These changes were made both to address comments from
stakeholders, and to consider further some of the technical concerns
and policy goals judged more preeminent under the increased uncertainty
of the impacts of finalizing and proposing standards for model years
further into the future.\258\ Recognizing the concerns raised by full-
line OEMs, it was concluded that continuing increases in the stringency
of the light truck standards would be more feasible if the light truck
curve for MYs 2017 and beyond was made steeper than the MY 2016 truck
curve and the right (large footprint) cut-point was extended only
gradually to larger footprints. To accommodate these considerations,
the 2012 final rule finalized the slope fit to the MY 2008 fleet using
a sales-weighted, ordinary least-squares regression, using a fleet that
had technology applied to make the technology application across the
fleet more uniform, and after adjusting the data for the effects of
weight-to-footprint. Information from an updated MY 2010 fleet was also
considered to support this decision. As the curve was vertically
shifted (with fuel economy specified as mpg instead of gpm or
CO2 emissions) upwards, the right cutpoint was progressively
moved for the light truck curves with successive model years, reaching
the final endpoint for MY 2021.
---------------------------------------------------------------------------
\258\ The MYs 2012-2016 final standards were signed April 1st,
2010--putting 6.5 years between its signing and the last affected
model year, while the MYs 2017-2021 final standards were signed
August 28th, 2012--giving just more than nine years between signing
and the last affected final standards.
---------------------------------------------------------------------------
5. Reconsidering the Mathematical Functions for Today's Rulemaking
a) Why is it important to reconsider the mathematical functions?
By shifting the developed curves by a single factor, it is assumed
that the underlying relationship of fuel consumption (in gallons per
mile) to vehicle footprint does not change significantly from the model
year data used to fit the curves to the range of model years for which
the shifted curve shape is applied to develop the standards. However,
it must be recognized that the relationship between vehicle footprint
and fuel economy is not necessarily constant over time; newly developed
technologies, changes in consumer demand, and even the curves
themselves could influence the observed relationships between the two
vehicle characteristics. For example, if certain technologies are more
effective or more marketable for certain types of vehicles, their
application may not be uniform over the range of vehicle footprints.
Further, if market demand has shifted between vehicle types, so that
certain vehicles make up a larger share of the fleet, any underlying
technological or market restrictions which inform the average shape of
the curves could change. That is, changes in the technology or market
restrictions themselves, or a mere re-weighting of different vehicles
types, could reshape the fit curves.
For the above reasons, the curve shapes were reconsidered in the
proposal using the newest available data from MY 2016. With a view
toward corroboration through different techniques, a range of
descriptive statistical analyses were conducted that do not require
underlying engineering models of how fuel economy and footprint might
be expected to be related, and a separate analysis that uses vehicle
simulation results as the basis to estimate the relationship from a
perspective more explicitly informed by engineering theory was
conducted as well. Despite changes in the new vehicle fleet both in
terms of technologies applied and in market demand, the underlying
statistical relationship between footprint and fuel economy has not
changed significantly since the MY 2008 fleet used for the 2012 final
rule; therefore, EPA and NHTSA proposed to continue to use the curve
shapes fit in 2012. The analysis and reasoning supporting this decision
follows.
b) What statistical analyses did EPA and NHTSA consider?
In considering how to address the various policy concerns discussed
above, data from the MY 2016 fleet was considered, and a number of
descriptive statistical analyses (i.e., involving observed fuel economy
levels and footprints) using various statistical methods, weighting
schemes, and adjustments to the data to make the fleets less
technologically heterogeneous were performed. There were several
adjustments to the data that were common to all of the statistical
analyses considered.
With a view toward isolating the relationship between fuel economy
and footprint, the few diesels in the fleet were excluded, as well as
the limited number of vehicles with partial or full electric
propulsion; when the fleet is normalized so that technology is more
homogenous, application of these technologies is not allowed. This is
consistent with the methodology used in the 2012 final rule.
The above adjustments were applied to all statistical analyses
considered, regardless of the specifics of each of the methods,
weights, and technology level of the data, used to view the
relationship of vehicle footprint and fuel economy. Table V-1, below,
summarizes the different assumptions considered and the key attributes
of each. The analysis was performed considering all possible
combinations of these assumptions, producing a total of eight footprint
curves.
[[Page 24254]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.058
(1) Current Technology Level Curves
The ``current technology'' level curves exclude diesels and
vehicles with electric propulsion, as discussed above, but make no
other changes to each model year fleet. Comparing the MY 2016 curves to
ones built under the same methodology from previous model year fleets
shows whether the observed curve shape has changed significantly over
time as standards have become more stringent. Importantly, these curves
will include any market forces which make technology application
variable over the distribution of footprint. These market forces will
not be present in the ``maximum technology'' level curves: By making
technology levels homogenous, this variation is removed. The current
technology level curves built using both regression types and both
regression weight methodologies from the MY 2008, MY 2010, and MY 2016
fleets, shown in more detail in Chapter 4.4.2.1 of the PRIA, support
the curve slopes finalized in the 2012 final rule. The curves built
from most methodologies using each fleet generally shift, but remain
very similar in slope. This suggests that the relationship of footprint
to fuel economy, including both technology and market limits, has not
significantly changed.
(2) Maximum Technology Level Curves
As in prior rulemakings, technology differences between vehicle
models were considered to be a significant factor producing uncertainty
regarding the relationship between fuel consumption and footprint.
Noting that attribute-based standards are intended to encourage the
application of additional technology to improve fuel efficiency and
reduce CO2 emissions across the distribution of footprint in
the fleet, approaches were considered in which technology application
is simulated for purposes of the curve fitting analysis in order to
produce fleets that are less varied in technology content. This
approach helps reduce ``noise'' (i.e., dispersion) in the plot of
vehicle footprints and fuel consumption levels and identify a more
technology-neutral relationship between footprint and fuel consumption.
The results of updated analysis for maximum technology level curves are
also shown in Chapter 4.4.2.2 of the PRIA. Especially if vehicles
progress over time toward more similar size-specific efficiency,
further removing variation in technology application both better
isolates the relationship between fuel consumption and footprint and
further supports the curve slopes finalized in the 2012 final rule.
c) What other methodologies were considered?
The methods discussed above are descriptive in nature, using
statistical analysis to relate observed fuel economy levels to observed
footprints for known vehicles. As such, these methods are clearly based
on actual data, answering the question ``how does fuel economy appear
to be related to footprint?'' However, being independent of explicit
engineering theory, they do not answer the question ``how might one
expect fuel economy to be related to footprint?'' Therefore, as an
alternative to the above methods, an alternative methodology was also
developed and applied that, using full-vehicle simulation, comes closer
to answering the second question, providing a basis either to
corroborate answers to the first, or suggest that further investigation
could be important.
As discussed in the 2012 final rule, several manufacturers have
confidentially shared with the agencies what they described as
``physics-based'' curves, with each OEM showing significantly different
shapes for the footprint-fuel economy relationships. This variation
suggests that manufacturers face different curves given the other
attributes of the vehicles in their fleets (i.e., performance
[[Page 24255]]
characteristics) and/or that their curves reflected different levels of
technology application. In reconsidering the shapes of the proposed MYs
2021-2026 standards, a similar estimation of physics-based curves
leveraging third-party simulation work form Argonne National
Laboratories (Argonne) was developed. Estimating physics-based curves
better ensures that technology and performance are held constant for
all footprints; augmenting a largely statistical analysis with an
analysis that more explicitly incorporates engineering theory helps to
corroborate that the relationship between fuel economy and footprint is
in fact being characterized.
Tractive energy is the amount of energy it will take to move a
vehicle.\259\ Here, tractive energy effectiveness is defined as the
share of the energy content of fuel consumed which is converted into
mechanical energy and used to move a vehicle--for internal combustion
engine (ICE) vehicles, this will vary with the relative efficiency of
specific engines. Data from Argonne simulations suggest that the limits
of tractive energy effectiveness are approximately 25 percent for
vehicles with internal combustion engines which do not possess
integrated starter generator, other hybrid, plug-in, pure electric, or
fuel cell technology.
---------------------------------------------------------------------------
\259\ Thomas, J. ``Drive Cycle Powertrain Efficiencies and
Trends Derived from EPA Vehicle Dynamometer Results,'' SAE Int. J.
Passeng. Cars--Mech. Syst. 7(4):2014, doi:10.4271/2014-01-2562.
Available at https://www.sae.org/publications/technical-papers/content/2014-01-2562/ (last accessed June 15, 2018).
---------------------------------------------------------------------------
A tractive energy prediction model was also developed to support
today's proposal. Given a vehicle's mass, frontal area, aerodynamic
drag coefficient, and rolling resistance as inputs, the model will
predict the amount of tractive energy required for the vehicle to
complete the Federal test cycle. This model was used to predict the
tractive energy required for the average vehicle of a given footprint
\260\ and ``body technology package'' to complete the cycle. The body
technology packages considered are defined in Table V-2, below. Using
the absolute tractive energy predicted and tractive energy
effectiveness values spanning possible ICE engines, fuel economy values
were then estimated for different body technology packages and engine
tractive energy effectiveness values.
---------------------------------------------------------------------------
\260\ The mass reduction curves used elsewhere in this analysis
were used to predict the mass of a vehicle with a given footprint,
body style box, and mass reduction level. The `Body style Box' is 1
for hatchbacks and minivans, 2 for pickups, and 3 for sedans, and is
an important predictor of aerodynamic drag. Mass is an essential
input in the tractive energy calculation.
[GRAPHIC] [TIFF OMITTED] TR30AP20.059
Chapter 6 of the PRIA show the resultant CAFE levels estimated for
the vehicle classes Argonne simulated for this analysis, at different
footprint values and by vehicle ``box.'' Pickups are considered 1-box,
hatchbacks and minivans are 2-box, and sedans are 3-box. These
estimates are compared with the MY 2021 standards finalized in 2012.
The general trend of the simulated data points follows the pattern of
the previous MY 2021 standards for all technology packages and tractive
energy effectiveness values presented in the PRIA. The tractive energy
curves are intended to validate the curve shapes against a physics-
based alternative, and the analysis suggests that the curve shapes
track the physical relationship between fuel economy and tractive
energy for different footprint values.
Physical limitations are not the only forces manufacturers face;
their success is dependent upon producing vehicles that consumers
desire and will purchase. For this reason, in setting future standards,
the analysis will continue to consider information from statistical
analyses that do not homogenize technology applications in addition to
statistical analyses which do, as well as a tractive energy analysis
similar to the one presented above.
The relationship between fuel economy and footprint remains
directionally discernable but quantitatively uncertain. Nevertheless,
each standard must commit to only one function. Approaching the
question ``how is fuel economy related to footprint'' from different
directions and applying different approaches has given EPA and NHTSA
confidence that the function applied here appropriately and reasonably
reflects the relationship between fuel economy and footprint.
The agencies invited comments on this conclusion and the supporting
analysis. IPI raised concerns that ``. . . several dozen models (mostly
subcompacts and sports cars) fall in the 30-40 square feet range, which
are all subject to the same standards'' and that ``manufacturers of
these models may have an incentive to decrease footprints as a
compliance strategy, since doing so would not trigger more stringent
standards.'' \261\ NHTSA and EPA agree that, all else equal, downsizing
the smallest cars (e.g., Chevrolet Spark, Ford Fiesta, Mini Cooper,
Mazda MX-5, Porsche 911, Toyota Yaris) would most likely tend to
degrade overall highway safety. At the same time, as discussed above,
the agencies recognize that small vehicles do appear attractive to some
market segments (although obviously the Ford Fiesta and Porsche 911
compete in different segments).
[[Page 24256]]
Therefore, there is a tension between on one hand, avoiding standards
that unduly encourage safety-eroding downsizing and, on the other,
avoiding standards that unduly penalize the market for small vehicles.
The agencies examined this issue, and note that the market for the
smallest vehicles has not evolved at all as estimated in the analysis
supporting the 2012 final rule, and attribute this more to fuel prices
and consumer demand for larger vehicles than to attribute-based CAFE
and CO2 standards. For example, the market for vehicles with
footprints less than 40 square foot was about 45 percent smaller in MY
2017 than in MY 2010. The agencies also found that among the smallest
vehicle models produced throughout MYs 2010-2017, most have become
larger, not smaller. For example, while the Mazda MX-5's footprint
decreased by 0.1 square foot (0.3 percent) during that time, the MY
2017 versions of the Mini Cooper, Smart fortwo, Porsche 911, and Toyota
Yaris had larger footprints than in MY 2010. With the market for very
small vehicles shrinking, and with manufacturers not evidencing a
tendency to make the smallest vehicles even smaller, the agencies are
satisfied that it would be unwise to change the target functions such
that targets never stop becoming more stringent as vehicle footprint
becomes ever smaller, because doing so could further impede an already-
shrinking market.
---------------------------------------------------------------------------
\261\ IPI, NHTSA-2018-0067-12362, p. 14.
---------------------------------------------------------------------------
B. No-Action Alternative
As in the proposal, the No-Action Alternative applies the augural
CAFE and final CO2 targets announced in 2012 for MYs 2021-
2025.\262\ For MY 2026, this alternative applies the same targets as
for MY 2025. The carbon dioxide equivalent of air conditioning
refrigerant leakage credits, nitrous oxide, and methane emissions are
included for compliance with the EPA standards for all model years
under the no-action alternative.\263\
---------------------------------------------------------------------------
\262\ https://www.govinfo.gov/content/pkg/CFR-2014-title40-vol19/pdf/CFR-2014-title40-vol19-sec86-1818-12.pdf
\263\ EPA regulations use a different but mathematically
equivalent approach to specify targets. Rather than using a function
with nested minima and maxima functions, EPA regulations specify
requirements separately for different ranges of vehicle footprint.
Because these ranges reflect the combined application of the listed
minima, maxima, and linear functions, it is mathematically
equivalent and more efficient to present the targets as in this
Section.
[GRAPHIC] [TIFF OMITTED] TR30AP20.060
[[Page 24257]]
In comments on the DEIS, CBD et al. indicated that it was
appropriate for NHTSA to use the augural CAFE standards as the baseline
No Action regulatory alternative.\264\ However, CARB commented that the
baseline regulatory alternative should include CARB's ZEV mandate, in
part because EPA must consider ``other regulations promulgated by EPA
or other government entities,'' and, according to CARB, there will be
much more vehicle electrification in the future as manufacturers
respond to market demand and also work to comply with the ZEV
mandate.\265\ Similarly, EPA's Science Advisory Board recommended--
despite the action taken in the One National Program Action--that the
baseline include state ZEV mandates ``to be consistent with policies
that would prevail in the absence of the rule change.'' \266\ EPA's
Science Advisory Board further recommended including sensitivity
analyses with different penetration rates of ZEVs.
---------------------------------------------------------------------------
\264\ CBD et al., NHTSA-2018-0067-12123, Attachment 1, at 13.
\265\ CARB, NHTSA-2018-0067-11873, at 124-125.
\266\ SAB at 12 and 29-30.
---------------------------------------------------------------------------
On the other hand, arguing for consideration of standards less
stringent than those proposed in the NPRM, Walter Kreucher commented
that rather than using the augural standards as the baseline, ``a
better approach would be to assume a clean sheet of paper and start
from the existing 2016MY fleet and its associated standards as the
baseline using 0%/year increases for both passenger cars and light
trucks for MYs 2017-2026.'' \267\ Similarly, AVE argued that because
previously-promulgated standards for MYs 2018-2021 already present a
significant challenge that ``will likely require almost every automaker
to continue using credits for compliance, . . . AVE believes this
rulemaking should reset . . . the current compliance baseline for cars
and light trucks at MY 2018 . . .'' \268\ BorgWarner commented
similarly that ``Beginning in MY 2018, standards should be reset to the
levels the industry actually achieved. For MY 2018 and beyond,
succeeding model year targets should be set with an annual rate of
improvement defined by the slope of improvement the industry has
achieved over the last six years. . . . Based on these data, our
analysis suggests the most reasonable and logical rate of improvement
falls between 2.0% to 2.6% for cars and trucks. Additionally, a single
rate of improvement for the combined fleet should be considered.''
\269\
---------------------------------------------------------------------------
\267\ Kreucher, W., NHTSA-2018-0067-0444, at 8.
\268\ AVE, NHTSA-2018-0067-11696, at 8-9.
\269\ BorgWarner, NHTSA-2018-0067-11895, at 3, 6.
---------------------------------------------------------------------------
The No-Action Alternative represents expectations regarding the
world in the absence of a proposal, accounting for applicable laws
already in place. Although manufacturers are already making significant
use of compliance credits toward compliance with even MY 2017
standards, the agencies are obligated to evaluate regulatory
alternatives against the standards already in place through MY 2025.
Similarly, even though manufacturers are already producing electric
vehicles, EPA and NHTSA appropriately excluded California's ZEV mandate
from the No-Action alternative for the NPRM, for several reasons.
First, the ZEV mandate is not Federal law; second, as described in the
proposal and subsequently finalized in regulatory text, the ZEV mandate
is expressly and impliedly preempted by EPCA; third, EPA proposed to
withdraw the waiver of CAA preemption in the NPRM and subsequently
finalized this withdrawal. Accordingly, the agencies have, therefore,
appropriately excluded the ZEV mandate from the No-Action alternative.
However, as discussed below, the agencies' analysis does account for
the potential that under every regulatory alternative, including the
No-Action Alternative, vehicle electrification could increase in the
future, especially if batteries become less expensive as gasoline
becomes more expensive.
C. Action Alternatives
1. Alternatives in Final Rule
Table V-5 below shows the different alternatives evaluated in
today's notice.
[GRAPHIC] [TIFF OMITTED] TR30AP20.061
[[Page 24258]]
With one exception, the alternatives considered in the NPRM
included the changes in stringency for the above alternatives.
Alternative 3, the preferred alternative, is newly included for today's
notice.\270\
---------------------------------------------------------------------------
\270\ As the agencies indicated in the NPRM, they were
considering and taking comment ``on a wide range of alternatives and
have specifically modeled eight alternatives.'' 83 FR at 42990 (Aug.
24, 2018). The preferred alternative in this final rule was within
the range of alternatives considered in the proposal, although it
was not specifically modeled at that time. This issue is discussed
in further detail below.
---------------------------------------------------------------------------
Regulations regarding implementation of NEPA requires agencies to
``rigorously explore and objectively evaluate all reasonable
alternatives, and for alternatives which were eliminated from detailed
study, briefly discuss the reasons for their having been eliminated.''
\271\ This does not amount to a requirement that agencies evaluate the
widest conceivable spectrum of alternatives. For example, a State
considering adding a single travel lane to a preexisting section of
highway would not be required to consider adding three lanes, or to
consider dismantling the highway altogether.
---------------------------------------------------------------------------
\271\ 40 CFR 1502.14.
---------------------------------------------------------------------------
Among thousands of individual comments that mentioned the proposed
standards very generally, some comments addressed the range and
definition of these regulatory alternatives in specific terms, and
these specific comments include comments on the stringency, structure,
and particular provisions defining the set of regulatory alternatives
under consideration.
As discussed throughout today's notice, the agencies have updated
and otherwise revised many aspects of the analysis. The agencies have
also reconsidered whether the set of alternatives studied in detail
should be expanded to include standards less stringent than the
proposal's preferred alternative, or to include standards more
stringent than the proposal's no-action alternative. On one hand,
comments from Walter Kreucher and AVE cited above indicate the agencies
should consider relaxing standards below MY 2020 levels, and CEI
challenged the agencies' failure to include less-stringent alternatives
in the following comments on this question:
DOT failed to consider the possibility of freezing CAFE at an
even more lenient standard than currently exists, nor did it
consider making its proposed freeze take effect sooner than MY 2020.
However, as DOT's own analysis strongly indicates, doing so would
lead to even greater benefits and an even greater reduction in CAFE-
related deaths and injuries. In short, DOT's failure to consider
this possibility is arbitrary and capricious. It has an opportunity
to remedy this in its final rule, and it should do so by selecting a
standard that is even more lenient than the one it proposed. . . .
It should have gone beyond its original set of alternatives and
examined less stringent ones as well--until it found one that, for
some reason or another, failed to produce greater safety benefits or
failed to meet the statutory factors.\272\
---------------------------------------------------------------------------
\272\ CEI, NHTSA-2018-0067-12015, at 1.
On the other hand, a coalition of ten environmental advocacy
organizations stated that the agencies should consider alternatives
more stringent than those defining the baseline no action alternative,
arguing that in light of CEQ guidance and the 2018 IPCC report on
climate change, ``the increasing danger, increasing urgency, and
increasing importance of vehicle emissions all rationally counsel for
strengthening emission standards.'' \273\ CBD et al. observe that
``none of these alternatives [considered in the NPRM] increases fuel
economy in comparison with the No Action Alternative, none conserves
energy . . .'' and go on to assert that ``none represents maximum
feasible CAFE standards.'' \274\ Similarly, EDF commented that ``. . .
given its clear statutory directive to maximize fuel savings, NHTSA
should have considered a range of alternatives that would be more
protective than the existing standards,'' \275\ and three State
agencies in Minnesota commented that ``more stringent standards are
consistent with EPCA's purpose of energy conservation and the CAA's
purpose of reducing harmful air pollutants.'' \276\ The North Carolina
Department of Environmental Quality acknowledged the agencies'
determination in the proposal that alternatives beyond the augural
standards might be economically impracticable, but nevertheless argued
that ``alternatives that exceed the stringency of the current standards
are consistent with EPCA's purpose'' \277\ In oral testimony before the
agencies, the New York State Attorney General also indicated that the
agencies should consider alternatives more stringent than the augural
standards.\278\ A coalition of States and cities commented that ``at a
minimum, the existing standards should be left in place, but EPA should
also consider whether to make the standards more stringent, not less,
just as it has done in prior proposals.'' \279\ More specifically,
through International Mosaic, some individuals commented that the
agencies must ``fully and publicly consider a few options that require
at least a seven annual percent [sic] improvement in vehicle fleet
mileage.'' \280\ In comments on the DEIS, CBD, et al. went further,
commenting that ``NHTSA's most stringent alternative must be set at no
lower than a 9 percent improvement per year.'' \281\ Most manufacturers
who commented on stringency did not identify specific regulatory
alternatives that the agencies should consider, although Honda
suggested that standards be set to increase in stringency at 5 percent
annually for both passenger cars and light trucks throughout model
years 2021-2026.282 283
---------------------------------------------------------------------------
\273\ CBD, et al., NHTSA-2018-0067-12057 p. 10. Also, see
comments from Senator Tom Carper, NHTSA-2018-0067-11910, at 8-9, and
from UCS, NHTSA-2018-0067-12039, at 3.
\274\ CBD, et al., NHTSA-2018-0067-12123, at 12-13.
\275\ EDF, NHTSA-2018-0067-11996, at 20.
\276\ Minnesota Pollution Control Agency, Department of
Transportation, and Department of Health, NHTSA-2018-0067-11706, at
5.
\277\ North Carolina Department of Environmental Quality, NHTSA-
2018-0067-12025, at 37-38.
\278\ New York State Attorney General, Testimony of Austin
Thompson, NHTSA-2018-0067-12305, at 13.
\279\ NHTSA-2018-0067-11735, at 49.
\280\ International Mosaic NHTSA-2018-0067-11154, at 1
\281\ CBD, et al., NHTSA-2018-0067-12123, at 17.
\282\ Honda, NHTSA-2018-0067-12019, EPA-HQ-OAR-2018-0283, at 54.
\283\ In model year 2021, the baseline standards for passenger
cars and light trucks increase by about 4% and 6.5%, respectively,
relative to standards for model year 2020. Depending on the
composition of the future new vehicle fleet (i.e., the footprints
and relative market shares of passenger cars and light trucks), this
amounts to an overall average stringency increase of about 5.5%
relative to model year 2020.
---------------------------------------------------------------------------
The agencies carefully considered these comments to expand the
range of stringencies to be evaluated as possible candidates for
promulgation. To inform this consideration, the agencies used the CAFE
model to examine a progression of stringencies extending outside the
range presented in the proposal and draft EIS, and as a point of
reference, using a case that reverts to MY 2018 standards starting in
MY 2021. Scenarios included in this initial screening exercise ranged
as high as increasing annually at 9.5 percent during MYs 2021-2026,
reaching average CAFE and CO2 requirements of 66 mpg and 120
g/mi, respectively. Results of this analysis are presented in the
following tables and charts. Focusing on MY 2029, the tables show
average required and achieved CAFE (as mpg) and CO2 (as g/
mi) levels for each scenario, along with average per-vehicle costs (in
2018 dollars, relative to retaining MY 2017 technologies). The proposed
(0%/0%), final (1.5%/1.5%), and baseline augural standards are shown in
bold type. The charts present
[[Page 24259]]
the same results on a percentage basis, relative to values shown below
for the scenario that reverts to MY 2018 standards starting in MY 2021.
For example, reverting to the MY 2018 CAFE standards starting in MY
2021 yields an average CAFE requirement of 35 mpg by MY 2029, with the
industry exceeding that standard by 5 mpg at an average cost of $1,255
relative to MY 2017 technology. Under the augural standards, the MY
2029 requirement increases to 47 mpg, the average compliance margin
falls to 1 mpg, and the average cost increases to $2,770. In other
words, compared to the scenario that reverts to MY 2018 stringency
starting in MY 2021, the augural standards increase stringency by 34
percent (from 35 to 47 mpg), increase average fuel economy by 20
percent (from 40 to 48 mpg), and increase costs by 121 percent (from
$1,255 to $2,770).
As indicated in the following two charts, the reality of
diminishing returns clearly applies in both directions. On one hand,
relaxing stringency below the proposed standards by reverting to MY
2018 or MY 2019 standards reduces average MY 2029 costs by only modest
amounts ($54-$121). As discussed in Section VIII, the agencies' updated
analysis indicates that the proposed standards would not be maximum
feasible considering the EPCA/EISA statutory factors, and would not be
appropriate under the CAA after considering the appropriate factors. If
further relaxation of standards appeared likely to yield more
significant cost reductions, it is conceivable that such savings could
outweigh further foregoing of energy and climate benefits. However,
this screening analysis does not show dramatic cost reductions.
Therefore, the agencies did not include these two less stringent
alternatives in the detailed analysis presented in Section VII.
On the other hand, increases in stringency beyond the baseline
augural standards show relative costs continuing to accrue much more
rapidly than relative CAFE and CO2 improvements. As
discussed below in Section VIII, even the no action alternative is
already well beyond levels that can be supported under the CAA and
EPCA. If further stringency increases appeared likely to yield more
significant additional energy and environmental benefits, it is
conceivable that these could outweigh these significant additional cost
increases. However, this screening analysis shows no dramatic relative
acceleration of energy and environmental benefits. Therefore, the
agencies did not include stringencies beyond the augural standards in
the detailed analysis presented in Section VII.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.062
[[Page 24260]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.063
[GRAPHIC] [TIFF OMITTED] TR30AP20.064
[[Page 24261]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.065
BILLING CODE 4910-59-C
Specific to model year 2021, some commenters argued that EPCA's
lead time requirement prohibits NHTSA from revising CAFE standards for
model year 2021.\284\ Regarding the revision of standards for model
year 2021, NHTSA did consider EPCA's lead time requirement, and
determined that while the agency would need to finalize a stringency
increase at least 18 months before the beginning of the first affected
model year, the agency can finalize a stringency decrease closer (or
even after) the beginning of the first affected model year. The
agency's reasoning is explained further in Section VIII. Therefore,
NHTSA did not change regulatory alternatives to avoid any relaxation of
stringency in model year 2021.
---------------------------------------------------------------------------
\284\ State of California, et al., NHTSA-2018-0067-11735, at
78.; CBD, et al., NHTSA-2018-0067-12000, Appendix A, at 66.;
National Coalition for Advanced Transportation, NHTSA-2018-0067-
11969, at 46.
---------------------------------------------------------------------------
The Auto Alliance stated that ``the truck increase rate should be
no greater than the car rate of increase and should be the `equivalent
task' per fleet.'' \285\ Supporting these Alliance comments, FCA
elaborated by commenting that ``(1) in MY2017, the latest data we have
available, most trucks have a larger gap to standards than cars, and
(2) all of the truck segments are challenged because consumers are
placing a greater emphasis on capability than fuel economy.'' \286\
Similarly, Ford commented that ``. . . the rates of increase in the
stringency of the standards should remain equivalent between passenger
cars and light duty trucks.'' \287\ Other commenters expressed general
support for equalizing the rates at which the stringencies of passenger
car and light truck standards increase.\288\
---------------------------------------------------------------------------
\285\ Alliance, NHTSA-2018-0067-12073, at 7-8
\286\ FCA, NHTSA-2018-0067-11943, at 46-47.
\287\ Ford, NHTSA-2018-0067-11928, at 3.
\288\ See, e.g., Global, NHTSA-2018-0067-12032, at 4; NADA,
NHTSA-2018-0067-12064, at 13; BorgWarner, NHTSA-2018-0067-11895, at
6.
---------------------------------------------------------------------------
For the final rule, the agencies have added an alternative in which
stringency for both cars and trucks increases at 1.5 percent. This is
consistent with comments received requesting that both fleets'
standards increase in stringency by the same amount, and 1.5 percent
represents a rate of increase within the range of rates of increase
considered in the NPRM.
Throughout the NPRM, the agencies described their consideration as
covering a range of alternatives.\289\ The preferred alternative for
this final rule, an increase in stringency of 1.5 percent for both cars
and trucks, falls squarely
[[Page 24262]]
within the range of alternatives proposed by the agencies.
---------------------------------------------------------------------------
\289\ 83 FR at 42986 (Aug. 24, 2018) (explaining, in ``Summary''
section of NPRM, that ``comment is sought on a range of alternatives
discussed throughout this document''); id. at 42988 (stating that
the agencies are ``taking comment on a wide range of alternatives,
including different stringencies and retaining existing
CO2 standards and the augural CAFE standards''); 42990
(``As explained above, the agencies are taking comment on a wide
range of alternatives and have specifically modeled eight
alternatives (including the proposed alternative) and the current
requirements (i.e., baseline/no action).''); 43197 (``[T]oday's
notice also presents the results of analysis estimating impacts
under a range of other regulatory alternatives the agencies are
considering.''); 43229 (explaining that ``technology availability,
development and application, if it were considered in isolation, is
not necessarily a limiting factor in the Administrator's selection
of which standards are appropriate within the range of the
Alternatives presented in this proposal.''); 43369 (``As discussed
above, a range of regulatory alternatives are being considered.'').
---------------------------------------------------------------------------
The NPRM alternatives were bounded on the upper end by the
baseline/no action alternative, and the proposed alternative on the
lower end (0 percent per year increase in stringency for both cars and
trucks). For passenger cars, the agencies considered a range of
stringency increases between 0 percent and 2 percent per year for
passenger cars, in addition to the baseline/no action alternative. For
light trucks, the agencies considered a range of stringency increases
between 0 percent and 3 percent per year, in addition to the baseline/
no action alternative.
The agencies considered the same range of alternatives for this
final rule. As with the proposal, the alternatives for stringency are
bounded on the upper end by the baseline/no action alternative and on
the lower end by 0 percent per year increases for both passenger cars
and light trucks. Consistent with the proposal, for this final rule,
the agencies considered stringency increases of between 0 and 2 percent
per year for passenger cars and between 0 and 3 percent per year for
light trucks, in addition to the baseline/no action alternative.
While it was not specifically modeled in the NPRM, the new
preferred alternative of an increase in stringency of 1.5 percent for
both cars and trucks was well within the range of alternatives
considered. The proposal described the alternatives specifically
modeled as options for the agencies, but also gave notice that they did
not limit the agencies in selecting from among the range of
alternatives under consideration.\290\
---------------------------------------------------------------------------
\290\ See, e.g., 83 FR at 43003 (Aug. 24, 2018) (``These
alternatives were examined because they will be considered as
options for the final rule. The agencies seek comment on these
alternatives, seek any relevant data and information, and will
review responses. That review could lead to the selection of one of
the other regulatory alternatives for the final rule or some
combination of the other regulatory alternatives (e.g., combining
passenger cars standards from one alternative with light truck
standards from a different alternative).''); id. at 43229
(describing a factor relevant to ``the Administrator's selection of
which standards are appropriate within the range of the Alternatives
presented in this proposal'').
---------------------------------------------------------------------------
The agencies explained in the proposal that they were ``taking
comment on a wide range of alternatives and have specifically modeled
eight alternatives.'' \291\ As with the proposal, for the final rule,
the agencies specifically modeled the upper and lower bounds of the
baseline/no action alternative and 0 percent per year stringency
increases for both passenger cars and light trucks. In both the
proposal and the final rule, the agencies also modeled a stringency
increase of 2 percent per year for passenger cars and 3 percent per
year for light trucks, as well as a variety of other specific increases
between 0 and 2 percent for passenger cars and 0 and 3 percent for
light trucks.
---------------------------------------------------------------------------
\291\ 83 FR at 42990 (Aug. 24, 2018).
---------------------------------------------------------------------------
The specific alternatives the agencies modeled for the final rule
reflect their consideration of public comments. As discussed above,
multiple commenters expressed support for equalizing the rates at which
the stringencies of passenger car and light truck standards increase.
To help the agencies evaluate alternatives that include the same
stringency increase for passenger cars and light trucks, three of the
seven alternatives (in addition to the baseline/no action alternative)
that the agencies specifically modeled for the final rule included the
same stringency increase for passenger cars and light trucks. This
includes the new preferred alternative of an increase in stringency of
1.5 percent for both cars and trucks. This alternative, and all others
specifically modeled for the final rule, falls within the range of
alternatives for stringency considered by the agencies in the proposal.
Beyond these stringency provisions discussed in the NPRM, the
agencies also sought comment on a number of additional compliance
flexibilities for the programs, as discussed in Section IX.
2. Additional Alternatives Suggested by Commenters
Beyond the comments discussed above regarding the shapes of the
functions defining fuel economy and CO2 targets, regarding
the inclusion of non-CO2 emissions, and regarding the
stringencies to be considered, the agencies also received a range of
other comments regarding regulatory alternatives.
Some of these additional comments involved how CAFE and
CO2 standards compare to one another for any given
regulatory alternative. With a view toward maximizing harmonization of
the standards, the Alliance, supported by some of its members'
individual comments, indicated that ``to the degree flexibilities and
incentives are not completely aligned between the CAFE and
[CO2] programs, there must be an offset in the associated
footprint-based targets to account for those differences. Some areas of
particular concerns are air conditioning refrigerant credits, and
incentives for advanced technology vehicles. The Alliance urges the
Agencies to seek harmonization of the standards and flexibilities to
the greatest extent possible. . . .'' \292\
---------------------------------------------------------------------------
\292\ Alliance, NHTSA-2018-0067-12073, at 40. See also FCA,
NHTSA-2018-0067-11943, at 6-7.
---------------------------------------------------------------------------
On the other hand, discussing consideration of compliance credits
but making a more general argument, the NYU Institute for Policy
Integrity commented that ``. . . EPA is not allowed to set lower
standards just for the sake of harmonization; to the contrary, full
harmonization may be inconsistent with EPA's statutory
responsibilities.'' \293\ Similarly, ACEEE argued that ``any
consideration of an extension or expansion of credit provisions under
the [carbon dioxide] or CAFE standards program should take as a
starting point the assumption that the additional credits will allow
the stringency of the standards to be increased.'' \294\
---------------------------------------------------------------------------
\293\ IPI, NHTSA-2018-0067-12213, at 21.
\294\ ACEEE, NHTSA-2018-0067-12122, at 3.
---------------------------------------------------------------------------
EPCA's requirement that NHTSA set standards at the maximum feasible
levels is separate and ``wholly independent'' from the CAA's
requirement, per Massachusetts v. EPA, that EPA issue regulations
addressing pollutants that EPA has determined endanger public health
and welfare.\295\ Nonetheless, as recognized by the Supreme Court,
``there is no reason to think the two agencies cannot both administer
their obligations and yet avoid inconsistency.'' \296\ This conclusion
was reached despite the fact that EPCA has a range of very specific
requirements about how CAFE standards are to be structured, how
manufacturers are to comply, what happens when manufacturers are unable
to comply, and how NHTSA is to approach setting standards, and despite
the fact that the CAA has virtually no such requirements. This means
that while nothing about either EPCA or the CAA, much less the
combination of the two, guarantees ``harmonization'' defining ``One
National Program,'' the agencies are expected to be able to work out
the differences.
---------------------------------------------------------------------------
\295\ Massachusetts v. EPA, 549 U.S. 497, 532 (2007).
\296\ Id.
---------------------------------------------------------------------------
Since tailpipe CO2 standards are de facto fuel economy
standards, the more differences there are between CO2 and
CAFE standards and compliance provisions, the more challenging it is
for manufacturers to plan year-by-year production that responses to
both, and the more difficult it is for affected stakeholders and the
general public to understand regulation in this space. Therefore, even
if the two statutes, taken together, do not guarantee ``full
harmonization,'' steps toward greater
[[Page 24263]]
harmonization help with compliance planning and transparency--and meet
the expectations set forth by the Supreme Court that the agencies avoid
inconsistencies.
The agencies have taken important steps toward doing so. For
example, EPA has adopted separate footprint-based CO2
standards for passenger cars and light trucks, and has redefined CAFE
calculation procedures to introduce recognition for the application of
real-world fuel-saving technology that is not captured with traditional
EPA two-cycle compliance testing. Detailed aspects of both sets of
standards and corresponding compliance provisions are discussed at
length in Section IX. The agencies never set out with the primary goal
of achieving ``full harmonization,'' such that both sets of standards
would lead each manufacturer to respond in exactly the same way in
every model year.\297\ For example, EPA did not adopt the EPCA
requirement that domestic passenger car fleets each meet a minimum
standard, or the EPCA cap on compliance credit transfers between
passenger car fleets. On the other hand, EPA also did not adopt the
EPCA civil penalty provisions that have allowed some manufacturers to
pay civil penalties as an alternative method of meeting EPCA
obligations. These and other differences provide that even if CAFE and
CO2 standards are ``mathematically'' harmonized, for any
given manufacturer, the two sets of standards will not be identically
burdensome in each model year. Inevitably, one standard will be more
challenging than the other, varying over time, between manufacturers,
and between fleets. This means manufacturers need to have compliance
plans for both sets of standards.
---------------------------------------------------------------------------
\297\ Full harmonization would mean that, for example, if Ford
would do some set of things over time in response to CAFE standards
in isolation, it would do exactly the same things on exactly the
same schedule in response to CO2 standards in isolation.
---------------------------------------------------------------------------
In 2012, recognizing that EPCA provides no clear basis to address
HFC, CH4, or N2O emissions directly, the agencies
``offset'' CO2 targets from fuel economy targets (after
converting the latter to a CO2 basis) by the amounts of
credit EPA anticipated manufacturers would, on average, earn in each
model years by reducing A/C leakage and adopting refrigerants with
reduced GWPs. In 2012, EPA assumed that by 2021, all manufacturers
would be earning the maximum available credit, and EPA's analysis
assumed that all manufacturers would make progress at the same rate.
However, as discussed above, data highlighted in comments by Chemours,
Inc., demonstrate that actual manufacturers' adoption of lower-GWP
refrigerants thus far ranges widely, with some manufacturers (e.g.,
Nissan) having taken no such steps to move toward lower-GWP
refrigerants, while others (e.g., JLR) have already applied lower-GWP
refrigerants to all vehicles produced for sale in the U.S. Therefore,
at least in practice, HFC provisions thus far continue to leave a gap
(in terms of harmonization) between the two sets of standards. The
proposal would have taken the additional step of decoupling provisions
regarding HFC (i.e., A/C leakage credits), CH4, and
N2O emissions from CO2 standards, addressing
these in separate regulations to be issued in a new proposal. As
discussed above, EPA did not finalize this proposal. Accordingly, for
the regulatory alternatives considered today, EPA has reinstated
offsets of CO2 targets from fuel economy targets, reflecting
the assumption that all manufacturers will be earning the maximum
available A/C leakage credit by MY 2021.
In addition to general comments on harmonization, the agencies
received a range of comments on specific provisions--especially
involving ``flexibilities''--that may or may not impact harmonization.
With a view toward encouraging further electrification, NCAT proposed
that EPA extend indefinitely the exclusion of upstream emissions from
electricity generation, and also extend and potentially restructure
production multipliers for PHEVs, EVs, and FCVs.\298\ On the other
hand, connecting its comments back to the stringency of standards, NCAT
also commented that ``. . . expansion of compliance flexibilities in
the absence of any requirement to improve [CO2] reduction or
fuel economy (as under the agencies' preferred option) could result in
an effective deterioration of existing [CO2] and fuel
economy performance, as well as little or no effective support for
advanced vehicle technology development or deployment.'' \299\ Global
Automakers indicated that the final rule ``should include a package of
programmatic elements that provide automakers with flexible compliance
options that promote the full breadth of vehicle technologies,'' such
options to include the extension of ``advanced technology'' production
multipliers through MY 2026, the indefinite exclusion of emissions from
electricity generation, the extension to passenger cars of credits
currently granted for the application of ``game changing'' technologies
(e.g., HEVs) only to full-size pickup trucks, an increase (to 15 g/mi)
of the cap on credits for off-cycle technologies, an updated credit
``menu'' of off-cycle technologies, and easier process for handling
applications for off-cycle credits.\300\ The Alliance also called for
expanded sales multipliers and a permanent exclusion of emissions from
electricity generation.\301\ Walter Kreucher recommended the agencies
consider finalizing the proposed standards but also keeping the augural
standards as ``voluntary targets'' to ``provide compliance with the
statutes and an aspirational goal for manufacturers.'' \302\
---------------------------------------------------------------------------
\298\ NCAT, NHTSA-2018-0067-11969, at 3-5.
\299\ Id.
\300\ Global Automakers, NHTSA-2018-0067-12032, at 4 et seq.
\301\ Alliance, NHTSA-2018-0067-12073, at 8.
\302\ Kreucher, W., NHTSA-2018-0067-0444, at 9.
---------------------------------------------------------------------------
The agencies have carefully considered these comments, and have
determined that the current suite of ``flexibilities'' generally
provide ample incentive more rapidly to develop and apply advanced
technologies and technologies that produce fuel savings and/or
CO2 reductions that would otherwise not count toward
compliance. The agencies also share some stakeholders' concern that
expanding these flexibilities could increase the risk of ``gaming''
that would make compliance less transparent and would unduly compromise
energy and environmental benefits. Nevertheless, as discussed in
Section IX, EPA is adopting new multiplier incentives for natural gas
vehicles. EPA is also finalizing some changes to procedures for
evaluating applications for off-cycle credits, and expects these
changes to make this process more accurate and more efficient. Also,
EPA is revising its regulations to not require manufacturers to account
for upstream emissions associated with electricity use for electric
vehicles and plug-in hybrid electric vehicles through model year 2026;
compliance will instead be based on tailpipe emissions performance only
and not include emissions from electricity generation until model year
2027. As discussed below, even with this change, and even accounting
for continued increases in fuel prices and reductions in battery
prices, BEVs are projected in this final rule analysis to continue to
account for less than 5 percent of new light vehicle sales in the U.S.
through model year 2026. To the extent that this projection turns out
to reflect reality, this means that the impact of upstream emissions
from electricity use on the projected CO2
[[Page 24264]]
reductions associated with these standards would likely remain small.
Regarding comments suggesting that the augural standards should be
finalized as ``voluntary targets,'' the agencies have determined that
having such targets exist alongside actual regulatory requirements
would be, at best, unnecessary and confusing.
Beyond these additional proposals, some commenters' proposals
clearly fell outside authority provided under EPCA or the CAA. Ron
Lindsay recommended the agencies ``consider postponing the rule changes
until the U.S. can establish a legally binding national and
international carbon budget and a binding mechanism to adhere to it.''
\303\ EPCA requires NHTSA to issue standards for MY 2022 by April 1,
2020, and previously-issued EPA regulations commit EPA to revisiting MY
2021-2025 standards on a similar schedule. These statutory and
regulatory provisions do not include a basis to delay decisions pending
an international negotiation for which prospects and schedules are both
unknown.
---------------------------------------------------------------------------
\303\ Ron Lindsay, EPA-HQ-OAR-2018-0283-1414, at 6.
---------------------------------------------------------------------------
SCAQMD, supported by Shyam Shukla, indicated that the agencies
should consider an alternative that keeps the waiver for California's
CO2 standards in place.\304\ NCAT and the North Carolina DEQ
offered similar comments and CBD, et al. commented that ``among the set
of more stringent alternatives that NEPA requires the agency to
consider, NHTSA must include action alternatives that retain the
standards California and other states have lawfully adopted.'' \305\ As
discussed above, the agencies recently issued a final rule addressing
the issue of California's authority. NEPA does not require NHTSA to
include action alternatives that cannot be lawfully realized.
---------------------------------------------------------------------------
\304\ SCAQMD, NHTSA-2018-0067-5666, at 1-2; Shyam Shukla, NHTSA-
2018-0067-5793, at 1-2.
\305\ NCAT, NHTSA-2018-0067-11969, at 64; NCDEQ, NHTSA-2018-
0067-12025, at 38; CBD et al., NHTSA-2018-0067-12123, Attachment 1,
at 18.
---------------------------------------------------------------------------
International Mosiac commented that NHTSA's DEIS ``is fatally
flawed . . . because it does not consider any market-based alternatives
(e.g., a `cap and trade' type option).'' \306\ While EPCA/EISA does
include very specific provisions regarding trading of CAFE compliance
credits, the statute provides no authority for a broad-based cap-and-
trade program involving other sectors. Similarly, Michalek, et al.
wrote that ``a more economically efficient approach of, taxing
emissions and fuel consumption at socially appropriate levels would
allow households to determine whether to reduce fuel consumption and
emissions by driving less, by buying a vehicle with more fuel saving
technologies, or by buying a smaller vehicle--or, alternatively, not to
reduce fuel consumption and emissions at all but rather pay a cost
based on the damages they cause. Forcing improvements only through one
mechanism (fuel-saving technologies) increases the cost of achieving
these outcomes.'' \307\ While some economists would agree with these
comments, Congress has provided no clear authority for NHTSA or EPA to
implement either an emissions tax or a broad-based cap-and-trade
program in which motor vehicles could participate.
---------------------------------------------------------------------------
\306\ International Mosaic, NHTSA-2018-0067-11154, at 1-2.
\307\ Michalek, et al., NHTSA-2018-0067-11903, at 13.
---------------------------------------------------------------------------
3. Details of Alternatives Considered in Final Rule
a) Alternative 1
Alternative 1 holds the stringency of targets constant and MY 2020
levels through MY 2026.
[[Page 24265]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.066
b) Alternative 2
Alternative 2 increases the stringency of targets annually during
MYs 2021-2026 (on a gallon per mile basis, starting from MY 2020) by
0.5 percent for passenger cars and 0.5 percent for light trucks.
[[Page 24266]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.067
c) Alternative 3
Alternative 3; the final standards promulgated today, increases the
stringency of targets annually during MYs 2021-2026 (on a gallon per
mile basis, starting from MY 2020) by 1.5 percent for passenger cars
and 1.5 percent for light trucks.
[[Page 24267]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.068
d) Alternative 4
Alternative 4 increases the stringency of targets annually during
MYs 2021-2026 (on a gallon per mile basis, starting from MY 2020) by
1.0 percent for passenger cars and 2.0 percent for light trucks.
[[Page 24268]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.069
e) Alternative 5
Alternative 5 increases the stringency of targets annually during
MYs 2022-2026 (on a gallon per mile basis, starting from MY 2021) by
1.0 percent for passenger cars and 2.0 percent for light trucks.
[GRAPHIC] [TIFF OMITTED] TR30AP20.070
[[Page 24269]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.071
f) Alternative 6
Alternative 6 increases the stringency of targets annually during
MYs 2021-2026 (on a gallon per mile basis, starting from MY 2020) by
2.0 percent for passenger cars and 3.0 percent for light trucks.
[GRAPHIC] [TIFF OMITTED] TR30AP20.072
[GRAPHIC] [TIFF OMITTED] TR30AP20.073
[[Page 24270]]
g) Alternative 7
Alternative 7 increases the stringency of targets annually during
MYs 2022-2026 (on a gallon per mile basis, starting from MY 2021) by
2.0 percent for passenger cars and 3.0 percent for light trucks.
[GRAPHIC] [TIFF OMITTED] TR30AP20.074
[GRAPHIC] [TIFF OMITTED] TR30AP20.075
EPCA, as amended by EISA, requires that any manufacturer's
domestically-manufactured passenger car fleet must meet the greater of
either 27.5 mpg on average, or 92 percent of the average fuel economy
projected by the Secretary for the combined domestic and non-domestic
passenger automobile fleets manufactured for sale in the U.S. by all
manufacturers in the model year, which projection shall be published in
the Federal Register when the standard for that model year is
promulgated in accordance with 49 U.S.C. 32902(b).\308\ Any time NHTSA
establishes or changes a passenger car standard for a model year, the
MDPCS for that model year must also be evaluated or re-evaluated and
established accordingly. Thus, this final rule establishes the
applicable MDPCS for MYs 2021-2026. Table V-22 lists the minimum
domestic passenger car standards.
---------------------------------------------------------------------------
\308\ 49 U.S.C. 32902(b)(4).
[GRAPHIC] [TIFF OMITTED] TR30AP20.076
[[Page 24271]]
VI. Analytical Approach as Applied to Regulatory Alternatives
A. Overview of Methods
Like analyses accompanying the NPRM and past CAFE and CAFE/
CO2 rulemakings, the analysis supporting today's notice
spans a range of technical topics, uses a range of different types of
data and estimates, and applies several different types of computer
models. The purpose of the analysis is not to determine the standards,
but rather to provide information for consideration in doing so. The
analysis aims to answer the question ``what impacts might each of these
regulatory alternatives have?''
Over time, NHTSA's and, more recently, NHTSA's and EPA's analyses
have expanded to address an increasingly wide range of types of
impacts. Today's analysis involves, among other things, estimating how
the application of various combinations of technologies could impact
vehicles' costs and fuel economy levels (and CO2 emission
rates), estimating how vehicle manufacturers might respond to standards
by adding fuel-saving technologies to new vehicles, estimating how
changes in new vehicles might impact vehicle sales and operation, and
estimating how the combination of these changes might impact national-
scale energy consumption, emissions, highway safety, and public health.
In addition, the EIS accompanying today's notice addresses impacts on
air quality and climate. The analysis of these factors informs and
supports both NHTSA's application of the statutory requirements
governing the setting of ``maximum feasible'' fuel-economy standards
under EPCA, including, among others, technological feasibility and
economic practicability, and EPA's application of the CAA requirements
for tailpipe emissions.
Supporting today's analysis, the agencies have brought to bear a
variety of different types of data, a few examples of which include
fuel economy compliance reports, historical sales and average
characteristics of light-duty vehicles, historical economic and
demographic measures, historical travel demand and energy prices and
consumption, and historical measures of highway safety. Also supporting
today's analysis, the agencies have applied several different types of
estimates, a few examples of which include projections of the future
cost of different fuel-saving technologies, projections of future GDP
and the number of households, estimates of the ``gap'' between
``laboratory'' and on-road fuel economy, and estimates of the social
cost of CO2 emissions and petroleum ``price shocks.''
With a view toward transparency, repeatability, and efficiency, the
agencies have used a variety of computer models to conduct the majority
of today's analysis. For example, the agencies have applied DOE/EIA's
National Energy Modeling System (NEMS) to estimate future energy
prices, EPA's MOVES model to estimate tailpipe emission rates for ozone
precursors and other criteria pollutants, DOE/Argonne's GREET model to
estimate emission rates for ``upstream'' processes (e.g., petroleum
refining), and DOE/Argonne's Autonomie simulation tool to estimate the
fuel consumption impacts of different potential combinations of fuel-
saving technology. In addition, the EIS accompanying today's notice
applies photochemical models to estimate air quality impacts, and
applies climate models to estimate climate impacts of overall emissions
changes.
Use of these different types of data, estimates, and models is
discussed further below in the most closely relevant sections. For
example, the agencies' use of NEMS is discussed below in the portion of
Section VI that addresses the macroeconomic context, which includes
fuel prices, and the agencies use of Autonomie is discussed in the
portion of Section VI.B.3 that addresses the agencies' approach to
estimating the effectiveness of various technologies (in reducing fuel
consumption and CO2 emissions).
Providing an integrated means to estimate both vehicle
manufacturers' potential responses to CAFE or CO2 standards
and, in turn, many of the different potential direct results (e.g.,
changes in new vehicle costs) and indirect impacts (e.g., changes in
rates of fleet turnover) of those responses, the CAFE Model plays a
central role in the agencies' analysis supporting today's notice. The
agencies used the specific models mentioned above to develop inputs to
the CAFE model, such as fuel prices and emission factors. Outputs from
the CAFE Model are discussed in Sections VII and VIII of today's
notice, and in the accompanying RIA. The EIS accompanying today's
notice makes use of the CAFE Model's estimates of changes in total
emissions from light-duty vehicles, as well as corresponding changes in
upstream emissions. These changes in emissions are included in the set
of inputs to the models used to estimate air quality and climate
impacts.
The remainder of this overview focuses on the CAFE Model. The
purpose of this overview is not to provide a comprehensive technical
description of the model,\309\ but rather to give an overview of the
model's functions, to explain some specific aspects not addressed
elsewhere in today's notice, and to discuss some model aspects that
were the subject of significant public comment. Some model functions
and related comments are addressed in other parts of today's notice.
For example, the model's handling of Autonomie-based fuel consumption
estimates is addressed in the portion of Section VI.B.3 that discusses
the agencies' application of Autonomie. The model documentation
accompanying today's notice provides a comprehensive and detailed
description of the model's functions, design, inputs, and outputs.
---------------------------------------------------------------------------
\309\ The CAFE Model is available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system with documentation and all inputs and outputs supporting
today's notice.
---------------------------------------------------------------------------
1. Overview of CAFE Model
The basic design of the CAFE Model is as follows: The system first
estimates how vehicle manufacturers might respond to a given regulatory
scenario, and from that potential compliance solution, the system
estimates what impact that response will have on fuel consumption,
emissions, and economic externalities. A regulatory scenario involves
specification of the form, or shape, of the standards (e.g., flat
standards, or linear or logistic attribute-based standards), scope of
passenger car and truck regulatory classes, and stringency of the CAFE
and CO2 standards for each model year to be analyzed.
Manufacturer compliance simulation and the ensuing effects
estimation, collectively referred to as compliance modeling, encompass
numerous subsidiary elements. Compliance simulation begins with a
detailed user-provided initial forecast of the vehicle models offered
for sale during the simulation period. The compliance simulation then
attempts to bring each manufacturer into compliance with the standards
defined by the regulatory scenario contained within an input file
developed by the user. For example, a regulatory scenario may define
CAFE or CO2 standards that increase in stringency by 4
percent per year for 5 consecutive years.
The model applies various technologies to different vehicle models
in each manufacturer's product line to simulate how each manufacturer
might make progress toward compliance with the specified standard.
Subject to a variety of user-controlled constraints, the model applies
technologies based on
[[Page 24272]]
their relative cost-effectiveness, as determined by several input
assumptions regarding the cost and effectiveness of each technology,
the cost of compliance (determined by the change in CAFE or
CO2 credits, CAFE-related civil penalties, or value of
CO2 credits, depending on the compliance program being
evaluated and the effective-cost mode in use), and the value of avoided
fuel expenses. For a given manufacturer, the compliance simulation
algorithm applies technologies either until the manufacturer runs out
of cost-effective technologies, until the manufacturer exhausts all
available technologies, or, if the manufacturer is assumed to be
willing to pay civil penalties, until paying civil penalties becomes
more cost-effective than increasing vehicle fuel economy. At this
stage, the system assigns an incurred technology cost and updated fuel
economy to each vehicle model, as well as any civil penalties incurred
by each manufacturer. This compliance simulation process is repeated
for each model year available during the study period.
This point marks the system's transition between compliance
simulation and effects calculations. At the conclusion of the
compliance simulation for a given regulatory scenario, the system
contains multiple copies of the updated fleet of vehicles corresponding
to each model year analyzed. For each model year, the vehicles'
attributes, such as fuel types (e.g., diesel, electricity), fuel
economy values, and curb weights have all been updated to reflect the
application of technologies in response to standards throughout the
study period. For each vehicle model in each of the model year specific
fleets, the system then estimates the following: Lifetime travel, fuel
consumption, carbon dioxide and criteria pollutant emissions, the
magnitude of various economic externalities related to vehicular travel
(e.g., noise), and energy consumption (e.g., the economic costs of
short-term increases in petroleum prices). The system then aggregates
model-specific results to produce an overall representation of modeling
effects for the entire industry.
Different categorization schemes are relevant to different types of
effects. For example, while a fully disaggregated fleet is retained for
purposes of compliance simulation, vehicles are grouped by type of fuel
and regulatory class for the energy, carbon dioxide, criteria
pollutant, and safety calculations. Therefore, the system uses model-
by-model categorization and accounting when calculating most effects,
and aggregates results only as required for efficient reporting.
2. Representation of the Market
As a starting point, the model needs enough information to
represent each manufacturer covered by the program. As discussed below
in Section VI.B.1, the MY 2017 analysis fleet contains information
about each manufacturer's:
Vehicle models offered for sale--their current (i.e., MY
2017) production volumes, manufacturer suggested retail prices (MSRPs),
fuel saving technology content and other attributes (curb weight, drive
type, assignment to technology class and regulatory class);
Production considerations--product cadence of vehicle
models (i.e., schedule of model redesigns and ``freshenings''), vehicle
platform membership, degree of engine and/or transmission sharing (for
each model variant) with other vehicles in the fleet; and
Compliance constraints and flexibilities--preference for
full compliance or penalty payment/credit application, willingness to
apply additional cost-effective fuel saving technology in excess of
regulatory requirements, projected applicable flexible fuel credits,
and current credit balance (by model year and regulatory class) in
first model year of simulation.
Representation of Fuel-Saving Technologies
The modeling system defines technology pathways for grouping and
establishing a logical progression of technologies that can be applied
to a vehicle. Technologies that share similar characteristics form
cohorts that can be represented and interpreted within the CAFE Model
as discrete entities. The following Table VI-1 shows the technologies
available within the modeling system used for this final rule. Each
technology is discussed in detail below. However, an understanding of
the technologies considered and how they are defined in the model
(e.g., a 6-speed manual transmission is defined as ``MT6'') is helpful
for the following explanation of the compliance simulation and the
inputs required for that simulation.
BILLING CODE 4910-59-P
[[Page 24273]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.077
[[Page 24274]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.078
BILLING CODE 4910-59-C
These entities are then laid out into pathways (or paths), which
the system uses to define relations of mutual exclusivity between
conflicting sets of technologies. For example, as presented in the next
section, technologies on the Turbo Engine path are incompatible with
those on the HCR Engine or the Diesel Engine paths. As such, whenever a
vehicle uses a technology from one pathway (e.g., turbo), the modeling
system immediately disables the incompatible technologies from one or
more of the other pathways (e.g., HCR and diesel).
In addition, each path designates the direction in which vehicles
are allowed to advance as the modeling system evaluates specific
technologies for application. Enforcing this directionality within the
model ensures that a vehicle that uses a more advanced or more
efficient technology (e.g., AT8) is not allowed to ``downgrade'' to a
less efficient option (e.g., AT5). Visually, as portrayed in the charts
in the sections that follow, this is represented by an arrow leading
from a preceding technology to a succeeding one, where vehicles begin
at the root of each path, and traverse to each successor technology in
the direction of the arrows.
The modeling system incorporates twenty technology pathways for
evaluation as shown below. Similar to individual technologies, each
path carries an intrinsic application level that denotes the scope of
applicability of all technologies present within that path, and whether
the pathway is evaluated on one vehicle at a time, or on a collection
of vehicles that share a common platform, engine, or transmission.
[[Page 24275]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.079
Even though technology pathways outline a logical progression
between related technologies, all technologies available to the system
are evaluated concurrently and independently of each other. Once all
technologies have been examined, the model selects a solution deemed to
be most cost-effective for application on a vehicle. If the modeling
system applies a technology that resides later in the pathway, it will
subsequently disable all preceding technologies from further
consideration to prevent a vehicle from potentially downgrading to a
less advanced option. Consequently, the system skips any technology
that is already present on a vehicle (either those that were available
on a vehicle from the input fleet or those that were previously applied
by the model). This ``parallel technology'' approach, unlike the
``parallel path'' methodology utilized in the preceding versions of the
model, allows the system always to consider the entire set of available
technologies instead of foregoing the application of potentially more
cost-effective options that happen to reside further down the
pathway.\310\ This revised approach addresses comments summarized
below, and allows the system to analyze all available technology
options concurrently and independently of one other without having to
first apply one or more ``predecessor'' technologies. For example, if
model inputs are such that a 7-speed transmission is cost-effective,
but not as cost-effective as an 8-speed transmission, the revised
approach enables the model to skip over the 7-speed transmission
entirely, whereas the NPRM version of the model might first apply the
7-speed transmission and then consider whether to proceed immediately
to the 8-speed transmission. As such, the model's choices for
evaluation of new technology solutions becomes slightly less
restrictive, allowing it immediately to consider and apply more
advanced options, and increasing the likelihood that the a globally
optimum solution is selected.
---------------------------------------------------------------------------
\310\ Previous versions of the CAFE Model followed a ``low-
cost'' first approach where the system would stop evaluating
technologies residing within a given pathway as soon as the first
cost-effective option within that path was reached.
---------------------------------------------------------------------------
Some commenters supported the agencies' use of such pathways in the
simulation of manufacturers' potential application of technologies. As
one of a dozen examples of CAFE model design elements that lead to the
transparent representation of real-world factors, the Alliance
highlighted ``recognition of the need for manufacturers to follow
`technology' pathways that retain capital and implementation expertise,
such as specializing in one type of engine or transmission instead of
following an unconstrained optimization that would cause manufacturers
to leap to unrelated technologies and show overly optimistic costs and
benefits.'' \311\ Similarly, Toyota commented that ``the inertia of
capital investments and engineering expertise dedicated to one
compliance technology or set of technologies makes it unreasonable for
manufacturers to immediately switch to another technology path.'' \312\
---------------------------------------------------------------------------
\311\ Alliance, NHTSA-2018-0067-12073, at 9.
\312\ Toyota, NHTSA-2018-0067-12098, at 7.
---------------------------------------------------------------------------
Other commenters cited the use of technology pathways as inherently
overly restrictive. For example, as an example of ``arbitrary model
constraints,'' a coalition of commenters cited the fact the model
``prohibit[s] manufacturers from switching vehicle technology
pathways.'' \313\ Also, EDF, UCS, and CARB cited the combination of
technology pathways, decision making criteria, and model inputs as
producing unrealistic results.\314\ Regarding the technology pathways,
specifically, EDF's consultant argued that the technology paths are not
[[Page 24276]]
transparent, and cited the potential that specific paths may not
necessarily be arranged in progression from least to most cost-
effective--that ``NHTSA ignores the cost of the technology when
developing this list.'' \315\ Relatedly, as EDF's consultant commented:
---------------------------------------------------------------------------
\313\ CBD, et al., NHTSA-2018-0067-12057, at 3.
\314\ EDF, NHTSA-2018-0067-12108, Appendix A, at 57 et seq.;
UCS, NHTSA-2018-0067-12039, Appendix, at 25 et seq.; Roush
Industries, NHTSA-2018-0067-11984, at 5.
\315\ EDF, NHTSA-2018-0067-12108, Appendix B, at 69.
[T]he Volpe Model is not designed to look backwards along its
technology paths. Thus, the opportunity to recover the expenditure
of inefficient technology is missed. NHTSA might argue that a
manufacturer will not invest in 10-speed transmissions, for example,
and then return to an older design. Whether or not this is true in
real life, such a view would put too much stake in the Volpe Model
projections. The model simply projects what could be done, not what
will be. Anyone examining the progression of technology and noting
the reversion of transmission technology could easily modify the
model inputs to avoid this. Also, if NHTSA evaluated combinations of
technologies prior to entering them in the model piecemeal, it would
automatically avoid such apparent problems.\316\
---------------------------------------------------------------------------
\316\ Ibid., at 70.
The agencies also received additional public comments on specific
paths and specific interactions between paths (e.g., involving engines
and hybridization). These comments are addressed below.
The agencies have carefully considered these comments and the
approach summarized below reflects some corresponding revision. As
mentioned above, the CAFE model now approaches the technology paths in
a such way that, faced with two cost-effective technologies on the same
path, the model can proceed directly to the more advanced technology if
that technology is the more cost effective of the two.
However, the agencies reject assertions that the model's use of
technology paths is not transparent. The agencies provided extensive
explanatory text, figures, model documentation, and model source code
specifically addressing these paths (and other model features). This
transparency appears evident in that commenters (sometimes while
claiming that a specific feature of the model is not transparent)
presented analytical results involving changes to corresponding inputs
that required a detailed understanding of that feature's operation.
Regarding comments that the technology paths should be arranged in
order of cost-effectiveness, the agencies note that such comments
presume, without merit, that costs, fuel consumption impacts, and other
inputs (e.g., fuel prices) that logically impact manufacturers'
decision-making are not subject to uncertainty. These inputs are all
subject to uncertainty, and the CAFE Model's arrangement of
technologies into several paths is responsive to these uncertainties.
Nevertheless, the agencies maintain that some technologies do reflect a
higher level of advancement than others (e.g., 10-speed transmissions
vs. 5-speed transmissions), and while manufacturers may, in practice,
occasionally revert to less advanced technologies, it is appropriate
and reasonable to conduct the agencies' analysis in a manner that
assumes manufacturers will continue to make forward progress. As
observed by EDF's consultant's remarks, the CAFE Model ``simply
projects what could be done, not what will be.'' While no model, much
less any model relying on information that can be made publicly
available, can hope to represent precisely each manufacturers' actual
detailed constrains related to product development and planning, such
constraints are real and important. The agencies agree that the CAFE
Model's representation of such constraints--including the Model's use
of technology paths--provides a reasonable means of accounting for
them.
4. Compliance Simulation
The CAFE model provides a way of estimating how vehicle
manufacturers could attempt to comply with a given CAFE standard by
adding technology to fleets that the agencies anticipate they will
produce in future model years. This exercise constitutes a simulation
of manufacturers' decisions regarding compliance with CAFE or
CO2 standards.
This compliance simulation begins with the following inputs: (a)
The analysis fleet of vehicles from model year 2017 discussed below in
Section VI.B.1, (b) fuel economy improving technology estimates
discussed below in Section VI.C, (c) economic inputs discussed below in
Section VI.D, and (d) inputs defining baseline and potential new CAFE
or CO2 standards discussed above in Section V. For each
manufacturer, the model applies technologies in both a logical sequence
and a cost-optimizing strategy in order to identify a set of
technologies the manufacturer could apply in response to new CAFE or
CO2 standards. The model applies technologies to each of the
projected individual vehicles in a manufacturer's fleet, considering
the combined effect of regulatory and market incentives while
attempting to account for manufacturers' production constraints.
Depending on how the model is exercised, it will apply technology until
one of the following occurs:
(1) The manufacturer's fleet achieves compliance \317\ with the
applicable standard and adding additional technology in the current
model year would be attractive neither in terms of stand-alone (i.e.,
absent regulatory need) cost-effectiveness nor in terms of facilitating
compliance in future model years;
---------------------------------------------------------------------------
\317\ When determining whether compliance has been achieved in
the CAFE program, existing CAFE credits that may be carried over
from prior model years or transferred between fleets are also used
to determine compliance status. For purposes of determining the
effect of maximum feasible CAFE standards, however, EPCA prohibits
NHTSA from considering these mechanisms for years being considered
(though it does so for model years that are already final) and the
agency runs the CAFE model without enabling these options. 49 U.S.C.
32902(h)(3).
---------------------------------------------------------------------------
(2) The manufacturer ``exhausts'' available technologies; \318\ or
---------------------------------------------------------------------------
\318\ In a given model year, it is possible that production
constraints cause a manufacturer to ``run out'' of available
technology before achieving compliance with standards. This can
occur when: (a) An insufficient volume of vehicles are expected to
be redesigned, (b) vehicles have moved to the ends of each
(relevant) technology pathway, after which no additional options
exist, or (c) engineering aspects of available vehicles make
available technology inapplicable (e.g., secondary axle disconnect
cannot be applied to two-wheel drive vehicles).
---------------------------------------------------------------------------
(3) For manufacturers assumed to be willing to pay civil penalties
(in the CAFE program), the manufacturer reaches the point at which
doing so would be more cost-effective (from the manufacturer's
perspective) than adding further technology.
The model accounts explicitly for each model year, applying
technologies when vehicles are scheduled to be redesigned or freshened
and carrying forward technologies between model years once they are
applied (until, if applicable, they are superseded by other
technologies). The model then uses these simulated manufacturer fleets
to generate both a representation of the U.S. auto industry and to
modify a representation of the entire light-duty registered vehicle
population. From these fleets, the model estimates changes in physical
quantities (gallons of fuel, pollutant emissions, traffic fatalities,
etc.) and calculates the relative costs and benefits of regulatory
alternatives under consideration.
The CAFE model accounts explicitly for each model year, in turn,
because manufacturers actually ``carry forward'' most technologies
between model years, tending to concentrate the application of new
technology to vehicle redesigns or mid-cycle ``freshenings,'' and
design cycles vary widely among manufacturers and specific products.
[[Page 24277]]
Comments by manufacturers and model peer reviewers strongly support
explicit year-by-year simulation. Year-by-year accounting also enables
accounting for credit banking (i.e., carry-forward), as discussed
above, and at least four environmental organizations recently submitted
comments urging the agencies to consider such credits, citing NHTSA's
2016 results showing impacts of carried-forward credits.\319\ Moreover,
EPCA/EISA requires that NHTSA make a year-by-year determination of the
appropriate level of stringency and then set the standard at that
level, while ensuring ratable increases in average fuel economy through
MY 2020. The multi-year planning capability, simulation of ``market-
driven overcompliance,'' and EPCA credit mechanisms (again, for
purposes of modeling the CAFE program) increase the model's ability to
simulate manufacturers' real-world behavior, accounting for the fact
that manufacturers will seek out compliance paths for several model
years at a time, while accommodating the year-by-year requirement. This
same multi-year planning structure is used to simulate responses to
standards defined in grams CO2/mile, and utilizing the set
of specific credit provisions defined under EPA's program.
---------------------------------------------------------------------------
\319\ Comment by Environmental Law & Policy Center, Natural
Resources Defense Council (NRDC), Public Citizen, and Sierra Club,
Docket ID EPA-HQ-OAR-2015-0827-9826, at 28-29.
---------------------------------------------------------------------------
After the light-duty rulemaking analysis accompanying the 2012
final rule that finalized NHTSA's standards through MY 2021, NHTSA
began work on changes to the CAFE model with the intention of better
reflecting constraints of product planning and cadence for which
previous analyses did not account. This involves accounting for
expected future schedules for redesigning and ``freshening'' vehicle
models, and accounting for the fact that a given engine or transmission
is often shared among more than one vehicle model, and a given vehicle
production platform often includes more than one vehicle model. These
real product planning considerations are explained below.
Like earlier versions, the current CAFE model provides the
capability for integrated analysis spanning different regulatory
classes, accounting both for standards that apply separately to
different classes and for interactions between regulatory classes.
Light vehicle CAFE and CO2 standards are specified
separately for passenger cars and light trucks. However, there is
considerable sharing between these two regulatory classes, where a
single engine, transmission, or platform can appear in both the
passenger car and light truck regulatory class. For example, some
sport-utility vehicles are offered in 2WD versions (classified as
passenger cars for compliance purposes) and 4WD versions (classified as
light trucks for compliance purposes). Integrated analysis of
manufacturers' passenger car and light truck fleets provides the
ability to account for such sharing and reduces the likelihood of
finding solutions that could involve introducing impractical and
unrealistic levels of complexity in manufacturers' product lines. In
addition, integrated fleet analysis provides the ability to simulate
the potential that manufacturers could earn CAFE and CO2
credits by over complying with the standard in one fleet and use those
credits toward compliance with the standard in another fleet (i.e., to
simulate credit transfers between regulatory classes).\320\
---------------------------------------------------------------------------
\320\ Note, however, that EPCA prohibits NHTSA from considering
the availability of such credit trading when setting maximum
feasible fuel economy standards. 49 U.S.C. 32902(h)(3).
---------------------------------------------------------------------------
The CAFE model also accounts for EPCA's requirement that compliance
be determined separately for fleets of domestic passenger cars and
fleets of imported passenger cars. The model accounts for all three
CAFE regulatory classes simultaneously (i.e., in an integrated way) yet
separately: Domestic passenger cars, imported passenger cars, and light
trucks. The model further accounts for two related specific statutory
requirements specifically involving this distinction between domestic
and imported passenger cars. First, EPCA/EISA requires that any given
fleet of domestic passenger cars meet a minimum standard, irrespective
of any available compliance credits. Second, EPCA/EISA requires
compliance with the standards applicable to the domestic passenger car
fleet without regard to traded or transferred credits.\321\
---------------------------------------------------------------------------
\321\ 49 U.S.C. 32903(f)(2) and (g)(4).
---------------------------------------------------------------------------
However, the CAA has no such limitation regarding compliance by
domestic and imported vehicles; EPA did not adopt provisions similar to
the aforementioned EPCA/EISA requirements and is not doing so today.
Therefore, the CAFE model determines compliance for manufacturers'
overall passenger car and light truck fleets for EPA's program.
Each manufacturer's regulatory requirement represents the
production-weighted harmonic mean of their vehicle's targets in each
regulated fleet. This means that no individual vehicle has a
``standard,'' merely a target, and each manufacturer is free to
identify a compliance strategy that makes the most sense given its
unique combination of vehicle models, consumers, and competitive
position in the various market segments. As the CAFE model provides
flexibility when defining a set of regulatory standards, each
manufacturer's requirement is dynamically defined based on the
specification of the standards for any simulation and the distribution
of footprints within each fleet.
Given this information, the model attempts to apply technology to
each manufacturer's fleet in a manner that, given product planning and
engineering-related considerations, optimizes the selected cost-related
metric. The metric supported by the NPRM version of the model is termed
``effective cost.'' The effective cost captures more than the
incremental cost of a given technology; it represents the difference
between their incremental cost and the value of fuel savings to a
potential buyer over the first 30 months of ownership.\322\ In addition
to the technology cost and fuel savings, the effective cost also
includes the change in CAFE civil penalties from applying a given
technology and any estimated welfare losses associated with the
technology (e.g., earlier versions of the CAFE model simulated low-
range electric vehicles that produced a welfare loss to buyers who
valued standard operating ranges between re-fueling events). Comments
on this metric are discussed below, as are model changes responding to
these comments.
---------------------------------------------------------------------------
\322\ The length of time over which to value fuel savings in the
effective cost calculation is a model input that can be modified by
the user. This analysis uses 30 months' worth of fuel savings in the
effective cost calculation, using the price of fuel at the time of
vehicle purchase.
---------------------------------------------------------------------------
This construction allows the model to choose technologies that both
improve a manufacturer's regulatory compliance position and are most
likely to be attractive to its consumers. This also means that
different assumptions about future fuel prices will produce different
rankings of technologies when the model evaluates available
technologies for application. For example, in a high fuel price regime,
an expensive but very efficient technology may look attractive to
manufacturers because the value of the fuel savings is sufficiently
high both to counteract the higher cost of the technology and,
implicitly, to satisfy consumer demand to balance price increases with
reductions in operating cost.
[[Page 24278]]
In general, the model adds technology for several reasons but
checks these sequentially. The model then applies any ``forced''
technologies. Currently, only variable valve timing (VVT) is forced to
be applied to vehicles at redesign since it is the root of the engine
path and the reference point for all future engine technology
applications.\323\ The model next applies any inherited technologies
that were applied to a leader vehicle on the same vehicle platform and
carried forward into future model years where follower vehicles (on the
shared system) are freshened or redesigned (and thus eligible to
receive the updated version of the shared component). In practice, very
few vehicle models enter without VVT, so inheritance is typically the
first step in the compliance loop. Next, the model evaluates the
manufacturer's compliance status, applying all cost-effective
technologies regardless of compliance status.\324\ Then the model
applies expiring overcompliance credits (if allowed to do so under the
perspective of either the ``unconstrained'' or ``standard setting''
analysis, for CAFE purposes).\325\ At this point, the model checks the
manufacturer's compliance status again. If the manufacturer is still
not compliant (and is unwilling to pay civil penalties, again for CAFE
modeling), the model will add technologies that are not cost-effective
until the manufacturer reaches compliance. If the manufacturer exhausts
opportunities to comply with the standard by improving fuel economy/
reducing emissions (typically due to a limited percentage of its fleet
being redesigned in that year), the model will apply banked CAFE or
CO2 credits to offset the remaining deficit. If no credits
exist to offset the remaining deficit, the model will reach back in
time to alter technology solutions in earlier model years.
---------------------------------------------------------------------------
\323\ As a practical matter, this affects very few vehicles.
More than 95 percent of vehicles in the market file either already
have VVT present or have surpassed the basic engine path through the
application of hybrids or electric vehicles.
\324\ For further explanation of how the CAFE model considers
the effective cost of applying different technologies see the CAFE
Model Documentation for the final rule, at S5.3 Compliance
Simulation Algorithm.
\325\ As mentioned above, EPCA prohibits consideration of
available credits when setting maximum feasible fuel economy
standards. 49 U.S.C. 32902(h)(3).
---------------------------------------------------------------------------
The CAFE model implements multi-year planning by looking back,
rather than forward. When a manufacturer is unable to comply through
cost-effective (i.e., producing effective cost values less than zero)
technology improvements or credit application in a given year, the
model will ``reach back'' to earlier years and apply the most cost-
effective technologies that were not applied at that time and then
carry those technologies forward into the future and re-evaluate the
manufacturer's compliance position. The model repeats this process
until compliance in the current year is achieved, dynamically
rebuilding previous model year fleets and carrying them forward into
the future, and accumulating CAFE or CO2 credits from over-
compliance with the standard wherever appropriate.
In a given model year, the model determines applicability of each
technology to each vehicle platform, model, engine, and transmission.
The compliance simulation algorithm begins the process of applying
technologies based on the CAFE or CO2 standards specified
during the current model year. This involves repeatedly evaluating the
degree of noncompliance, identifying the next ``best'' technology
(ranked by the effective cost discussed earlier) available on each of
the parallel technology paths described above and applying the best of
these. The algorithm combines some of the pathways, evaluating them
sequentially instead of in parallel, to ensure appropriate incremental
progression of technologies.
The algorithm first finds the best next applicable technology in
each of the technology pathways and then selects the best among these.
For CAFE purposes, the model applies the technology to the affected
vehicles if a manufacturer is either unwilling to pay penalties or if
applying the technology is more cost-effective than paying penalties.
Afterwards, the algorithm reevaluates the manufacturer's degree of
noncompliance and continues application of technology. Once a
manufacturer reaches compliance (i.e., the manufacturer would no longer
need to pay penalties), the algorithm proceeds to apply any additional
technology determined to be cost-effective (as discussed above).
Conversely, if a manufacturer is assumed to prefer to pay penalties,
the algorithm only applies technology up to the point where doing so is
less costly than paying penalties. The algorithm stops applying
additional technology to this manufacturer's products once no more
cost-effective solutions are encountered. This process is repeated for
each manufacturer present in the input fleet. It is then repeated for
each model year. Once all model years have been processed, the
compliance simulation algorithm concludes. The process for
CO2 standard compliance simulation is similar, but without
the option of penalty payment, such that technologies are applied until
compliance (accounting for any modeled application of credits) is
achieved. For both CAFE and CO2 standards, the model also
applies any additional (i.e., beyond required for compliance)
technology that ``pays back'' within a specified period (for the NPRM
and today's analysis, 30 months).
Some commenters argued that the CAFE model applies constraints that
excessively limit options manufacturers have to add technology, causing
the model to overestimate costs to achieve a given level of
improvement.\326\ Some of these commenters further argued that the
agencies should assume greater potential to apply technologies that
contribute to compliance by improving air conditioner efficiency or
otherwise reducing ``off cycle'' fuel consumption and CO2
emissions.\327\ Other commenters argued that such constraints, while
warranting some refinements, help the model to simulate manufacturers'
decision making realistically and to estimate technology effectiveness
and costs reasonably.328 329
---------------------------------------------------------------------------
\326\ NHTSA-2018-0067-12057, CBD, et. al, p. 3.
\327\ NHTSA-2018-0067-11741, ICCT, Attachment 2, p. 4.
\328\ NHTSA-2018-0067-12073, Alliance of Automobile
Manufacturers, pp. 134-36.
\329\ American Honda Motor Co., ``Honda Comments on the NPRM and
various proposals contained therein--Prepared for NHTSA, EPA and
ARB,'' October 17, 2018, pp. 12-16.
---------------------------------------------------------------------------
Some commenters questioned the ``effective cost'' metric the model
uses to decide among available options, claiming that the metric also
causes the model to avoid selection of pathways that are not always
economically optimal.\330\ One of these commenters recommended the
agencies modify the effective cost metric for CO2 compliance
by removing the term placing a monetary value on progress toward
compliance, and instead dividing the remaining net cost (i.e., the
increase in technology costs minus a portion of the fuel outlays
expected to be avoided) by the additional CO2 credits
earned.\331\ Another of these commenters claimed on one hand, that the
effective cost metric ``does not include a measurement of the
technology's reduction in fuel consumption or CO2
emissions'' and, on the other, that the metric inappropriately places a
value on avoided fuel consumption.\332\
---------------------------------------------------------------------------
\330\ NHTSA-2018-0067-11741, ICCT, Attachment 3, p. I-62.
\331\ NHTSA-2018-0067-12039, UCS, Technical Appendix, pp. 28-32.
\332\ NHTSA-2018-0067-12108, EDF, Appendix B, p. 67.
---------------------------------------------------------------------------
One commenter claimed that the model inappropriately allows earned
[[Page 24279]]
credits (including CO2 program credits for which EPA has
granted a one-time exemption from carry-forward limits) to expire while
also showing undue degrees overcompliance with standards, and further
proposed that the model be modified to simulate both credit ``carry
back'' (aka ``borrowing'') and credit trading between
manufacturers.\333\
---------------------------------------------------------------------------
\333\ NHTSA-2018-0067-12039, UCS, Technical Appendix, pp. 36-40.
---------------------------------------------------------------------------
In addition, some commenters indicated that the agencies' analysis
(impliedly, its modeling) should account for some States' mandates that
manufacturers sell minimum quantities of ``Zero Emission Vehicles''
(ZEVs).334 335
---------------------------------------------------------------------------
\334\ NHTSA-2018-0067-12036, Volvo, p. 5.
\335\ NHTSA-2018-0067-11813, South Coast AQMD, Attachment 1, p.
4 and EIS comments, p. 9.
---------------------------------------------------------------------------
Regarding the model's representation of engineering and product
planning constraints, the agencies maintain that having such
constraints produces more realistic potential (as mentioned above, not
``predicted'') pathways forward from manufacturers' current fleets than
would be the case were these constraints removed. For example, while
manufacturers' product plans are protected as confidential business
information (CBI), some manufacturers' public comments demonstrate
year-by-year balancing such as the CAFE model emulates.\336\ Also, even
manufacturers that have invested in technologies such as hybrid
electric powertrains and Atkinson cycle engines have commented that a
manufacturers' past investments will constrain the pathways it can
practicably take.\337\ Therefore, the agencies have retained the
model's basic structural constraints, have updated and expanded the
model's technology paths (and, as discussed, the model's logic for
approaching these paths), and have updated inputs defining the range of
manufacturer-, technology-, and product-specific constraints. These
updates are discussed below at greater length.
---------------------------------------------------------------------------
\336\ See, e.g., FCA, pp. 5-6.
\337\ Toyota, Attachment 1, p. 10.
---------------------------------------------------------------------------
The agencies have also reconsidered opportunities manufacturers may
have to expand the application of technologies that contribute to
compliance by improving air conditioner efficiency or otherwise
reducing ``off cycle'' fuel consumption and CO2 emissions,
or to earn credit toward CO2 compliance by using
refrigerants with lower global warming potential (GWP) or reducing the
potential for refrigerant leaks. The version of the model used for the
proposal accommodates inputs that, for each of these adjustments or
credits, applies the same value to every model year. The agencies have
revised the model to accommodate inputs that specify the degree of
adjustment or credit separately for each model year, and have applied
inputs that assume manufacturers will increase application of these
improvements to the highest levels reported within the industry.
Regarding comments on the effective cost metric the model uses to
compare and select among available options to add technology, the
agencies have considered changes such as those mentioned above. Given
the myriad of factors that manufacturers can consider, any weighing to
be conducted using publicly-available information will constitute a
simplified representation. Nevertheless, within the model's context, it
is obvious that any weighing of options should, at a minimum, consider
some measure of each option's costs and benefits. Since this aspect of
the model involves simulating manufacturers' decisions, it is also
clearly appropriate that these costs and benefits be considered from a
manufacturer perspective rather than a social perspective.
The effective cost metric used for the NPRM version of the model
represents the cost of a given option as the cost to apply a given
technology to a given set of vehicles, and represents the benefit of
the same option as the extent to which the manufacturer might expect
buyers would be willing to pay for fuel economy (as represented by a
portion of the projected fuel savings), combined with any reduction in
CAFE civil penalties that the manufacturer might ultimately need to
pass along to buyers. The reduction in CAFE civil penalties places a
value on progress made toward compliance with CAFE standards. The CAA
provides no direction regarding CO2 standards, so the model
accepts inputs specifying an analogous basis for valuing changes in the
quantity of CO2 credits earned from (or required by) a
manufacturer's fleet. Because each of these three components
(technology cost, fuel benefit, and compliance benefit) is expressed in
dollars, subtracting benefits from costs produces a net cost, and after
dividing net costs by the number of affected vehicles, it is logical
to, at each step, select the option that produces the most negative net
unit cost. This approach can be interpreted as maximizing net benefits
(to the manufacturer).
As an alternative, the agencies considered a simpler metric that
considers only the cost of the option and the extent to which the
option increases the quantity of earned credits, and does not require
input assumptions regarding how to value progress toward compliance.
Such a metric is expressed in dollars per ton or dollars per gallon
such that seeking options that produce the smallest (positive) values
can be interpreted as maximizing cost effectiveness (of progress toward
compliance). However, simply comparing technology costs to
corresponding compliance improvements would implicitly assume that
manufacturers do not respond at all to fuel prices. This assumption is
clearly unrealistic. For example, if diesel fuel costs $5 per gallon
and gasoline costs $2 per gallon, manufacturers will be reluctant to
respond to stringent CAFE or CO2 standards by replacing
gasoline engines with diesel engines. Manufacturers' comments credibly
assert that fuel prices matter, and in the agencies' judgment,
simulations of decisions between available options should continue to
account for avoided fuel outlays.
On the other hand, while any metric should incorporate some measure
of progress toward compliance, it is not obvious that this progress
must be expressed in monetary terms. While the CAFE civil penalty
provisions provide a logical basis for doing so with respect to CAFE,
the recently-introduced (through EISA) option to trade credit between
manufacturers adds an alternative basis that is undefined and
uncertain, in part because terms of past trades are not known to the
agencies. Also, as mentioned above, EPCA/EISA's civil penalty
provisions are not applicable to noncompliance with CO2
standards.
Therefore, for the purpose of selecting among available options to
add technology, the agencies consider it reasonable to use the degree
of compliance improvement in ``raw'' (i.e., not monetized) form, and to
divide net costs (i.e., technology costs minus a portion of expected
avoided fuel outlays) by this improvement. Under a range of side-by-
side tests, this change to the effective cost metric most frequently
produced lower overall estimates of compliance costs. However,
differences vary among manufacturers, model years, and regulatory
alternatives, and also depend on other model inputs. For example, at
high fuel prices, the new metric tends to select more expensive
pathways than the NPRM's metric, and with the new metric, a case
simulating ``perfect trading'' of CO2 compliance credits
tends to show such trading increasing compliance costs rather than, as
expected, decreasing such costs.
The version of the model used for the proposal simulates the
potential that, for
[[Page 24280]]
a given fleet in a given model year, a manufacturer might be able to
use credits from an earlier model year or a different fleet. This
version of the model did not explicitly simulate the potential that,
for a given fleet in a given model year, a manufacturer might be to use
credits from a future model year or a different manufacturer. However,
the agencies did apply model inputs that reflected assumptions
regarding possible trading of credits actually earned prior to model
year 2016 (the earliest represented in detail in the agencies'
analysis), and the agencies did examine a case (included in the
sensitivity analysis) involving hypothetical ``perfect'' trading of
CO2 credits among manufacturers by treating the industry as
a single ``manufacturer.'' Although past versions of the CAFE Model had
included code under development with a view toward eventually
simulating one or both of these provisions, this code had never
proceeded beyond preliminary experimentation, and had never been the
focus of peer reviews or application in published analyses.
Nevertheless, the agencies considered expanding the model to
simulate credit ``carry back'' (or ``borrowing'') and trading
(explicitly, rather than in an idealized hypothetical way). The
agencies closely examined the corresponding model revisions proposed by
UCS and determined that such methods would not produce repeatable
results. This is because the approach proposed by UCS ``randomly swaps
items in list to minimize trading bias.'' \338\
---------------------------------------------------------------------------
\338\ UCS, NHTSA-2018-0067-12039, Technical Appendix, at 84-87.
---------------------------------------------------------------------------
Even if such revisions could be modified to produce non-random
results, including credit banking and trading would introduce highly
speculative elements into the agencies' analysis. While manufacturers
have occasionally indicated plans to carry back credits from future
model years, those plans have sometimes backfired when projected
credits have failed to materialize, e.g., by misjudging consumer demand
for more efficient vehicles. In the agencies' judgment, it would be
inappropriate to set standards based on an analysis that relies on the
type of borrowing that has been known to fail. To rely also on credit
trading during the model years included in the analysis would compound
this undue speculation. For example, including credit borrowing and
trading throughout the analysis, as some commenters proposed, would
lead to an analysis that depends on the potential that, in order to
comply with the MY 2022 standard for light trucks, FCA could use
credits it expects to be able to buy from another manufacturer in MY
2025. Even if the agencies' analysis had knowledge of and made use of
manufacturers' actual product plans, expectations about the ability to
borrow others' unearned credits would necessarily be considered risky
and unreliable. Within an analysis that, to provide for public
disclosure, extrapolates forward many years from the most recent
observed fleet, such transactions would add an unreasonable level of
speculation. Therefore, the agencies have declined to introduce credit
borrowing and trading into the model's logic.
The analysis presented in the proposal applied inputs reflecting
potential application of credits earned earlier than the first year
modeled explicitly. However, as observed by some commenters, those
inputs did not fully account for the one-time exemption from the 5-year
limit on the extent to which manufacturers may carry forward
CO2 credits. The agencies have updated the analysis fleet to
MY 2017 and, in doing so, have updated inputs specifying how credits
earned to MY 2017 might be applied. These updates implement a
reasonably full accounting of these ``legacy'' credits, including of
the one-time exemption from the credit life limit.
As mentioned above, some commenters also indicated that the model
is unrealistically ``reluctant'' to apply credits carried forward from
early model years. As explained in the proposal and in the model
documentation, the model's application of carried-forward credits is
partially controlled by model inputs, which, for the proposal, were set
to assume that manufacturers would tend to retain credits as long as
possible. This assumption is entirely consistent with manufacturers'
past practice and logical in a context wherein the stringency of
standards is generally increasing over time. Even though using credits
in some model years might seem initially advantageous, doing so means
foregoing actual improvements likely to be needed in later model years.
Regarding the model's treatment of mandates and credits for the
sale of ZEVs, as indicated in the model documentation accompanying the
proposal, these capabilities were experimental in that version of the
model. The reference case analysis for today's notice, like that for
the proposal, does not simulate compliance with ZEV mandates.\339\
---------------------------------------------------------------------------
\339\ The agencies note their finalization of the One National
Program Final Action, in which EPA partially withdrew a waiver of
CAA preemption previously granted to the State of California
relating to its ZEV mandate, and NHTSA finalized regulations
providing that State ZEV mandates are impliedly and expressly
preempted by EPCA. This joint action is available at 84 FR 51310.
---------------------------------------------------------------------------
For the NPRM, the CAFE model was exercised with inputs extending
this explicit simulation of technology application through MY 2032, as
the agencies anticipated this was sufficiently beyond MY 2026 that
nearly all multiyear planning attributable to MY 2026 standards should
be accounted for, and any compliance credits carried forward from MY
2026 would have expired. The analysis met this expectation, and the
agencies presented analysis of the resultant estimated impacts over the
useful lives of vehicles produced prior to MY 2030. The agencies
invited comment on all aspects of the analysis, and relevant to this
aspect of the analysis--i.e., its perspective and temporal span--EDF
stated that that these led the agencies to overstate the proposal's
positive impacts on safety, in part because by explicitly representing
vehicle model years only through 2032, the agencies had failed to
account for the impact of distant model years prices and fuel economy
levels on the retention and scrappage of vehicles produced through MY
2029.\340\ For example, some vehicles produced in MY 2026 will likely
still be on the road during calendar years (CY) 2033-2050 and the rates
at which these MY 2026 vehicles will be scrapped during CYs 2033-2050
will be impacted by the prices and fuel economy levels of vehicles
produced during MYs 2033-2050.
---------------------------------------------------------------------------
\340\ EDF, NHTSA-2018-0067-12108, Attachment A at 11 and
Attachment B at 11-28.
---------------------------------------------------------------------------
The agencies have addressed this comment by expanding model inputs
to extend the explicit simulation of technology application through MY
2050. Most of these expanded model inputs involve the analysis fleet
and inputs defining the cost and availability of various fuel-saving
technologies. These inputs are discussed below. The agencies also made
minor modifications to the model in order to extend model outputs to
cover this wider span and to carry forward each regulatory
alternative's standards automatically through the last year to be
modeled (e.g., extending standards without change from MY 2032 through
MY 2050). The model documentation discusses these
[[Page 24281]]
minor changes.\341\ In addition, although the agencies published
detailed model output files documenting all estimated annual impacts
through calendar year 2089, the notice and PRIA both emphasized the
above-mentioned ``model year'' perspective, as in past regulatory
analyses supporting CAFE and CO2 standards. Recognizing that
an alternative ``calendar year'' perspective is of interest to EDF and,
perhaps other stakeholders, the agencies have expanded the presentation
of results in today's notice and FRIA by presenting some physical
impacts (e.g., fuel consumption and CO2 emissions) as well
as monetized benefits, costs, and net benefits for each of CYs 2017-
2050. All of these results appear in the model output files published
with today's notice, as do corresponding results for more specific
impacts (e.g., year-by-year components of monetized social costs).\342\
---------------------------------------------------------------------------
\341\ The model and documentation are available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
\342\ Detailed model inputs and outputs are available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
5. Calculation of Physical Impacts
Once it has completed the simulation of manufacturers' potential
application of technology in response to CAFE/CO2 standards
and fuel prices, the CAFE Model calculates impacts of the resultant
changes in new vehicle fuel economy levels and prices. This involves
several steps.
The model calculates changes in the total quantity of new vehicles
sold in each model year as well as the relative shares passenger cars
and light trucks comprise of the overall new vehicle market. The
agencies received many comments on the estimation of sales impacts, and
as discussed below, today's analysis applies methods and corresponding
estimates that reflect careful consideration of these comments. Related
to these calculations, the model now operates in an iterated fashion
with a view toward obtaining sales impacts that are balanced with
changes in vehicle prices and fuel economy levels. This involves
solving for compliance, calculating sales impacts, re-solving for
compliance, and repeating these steps as many times as specified in
model inputs. For today's analysis, the agencies operated the model
with four iterations, as early testing suggested three iterations
should be sufficient for fleetwide results to converge between
iterations. The model documentation describes the procedures for
iteration in detail.
The impacts on outlays for new vehicles occur coincident with the
sale of these vehicles so the model can simply calculate and record
these for each model year included in the analysis. However, virtually
all other impacts result from vehicle operation that extends long after
a vehicle is produced. Like other models (including, e.g., NEMS), the
CAFE Model includes procedures (sometimes referred to as ``stock
models'' or as models of fleet turnover) to estimate annual rates at
which new vehicles are used and subsequently scrapped. The agencies
received many comments on procedures for estimating vehicle scrappage
and on procedures for estimating annual quantities of highway travel,
accounting for the elasticity of travel demand with respect to per-mile
costs for fuel. Below, Section VI.D.1 discusses these comments and
reviews procedures and corresponding estimates that also reflect
careful consideration of these comments.
For each vehicle model in each model year, these procedures result
in estimates of the number of vehicles remaining in service in each
calendar year, as well as the annual mileage accumulation (i.e.,
vehicle miles traveled, or VMT) in each calendar year. As mentioned
above, most of the physical impacts of interest derive from this
vehicle operation. Also discussed above, the simulated application of
technology results in ``initial'' and ``final'' estimates of the cost,
fuel type, fuel economy, and fuel share (for, in particular, PHEVs that
can run on gasoline or electricity) applicable to each vehicle model in
each model year. Together with quantities of travel, and with estimates
of the ``gap'' between ``laboratory'' and ``on-road'' fuel economy,
these enable calculation of quantities of fuel consumed in each year
during the useful life of each vehicle model produced in each model
year.\343\ The model documentation provides specific procedures and
formulas implementing these calculations.
---------------------------------------------------------------------------
\343\ The agencies have applied the same estimates of the ``on
road gap'' as applied for the analysis supporting the NPRM. For
operation on gasoline, diesel, E85, and CNG, this gap is 20 percent;
for electricity and hydrogen, 30 percent.
---------------------------------------------------------------------------
As for the NPRM, the model calculates emissions of CO2
and other air pollutants, reporting emissions both from vehicle
tailpipes and from upstream processes (e.g., petroleum refining)
involved in producing and supplying fuels. Section VI.D.3 below reviews
methods, models, and estimates used in performing these calculations.
The model also calculates impacts on highway safety, accounting for
changes in travel demand, changes in vehicle mass, and continued past
and expected progress in vehicle safety (through, e.g., the application
of new crash avoidance systems). Section VI.D.2 discusses methods, data
sources, and estimates involved in estimating safety impacts, comments
on the same, and changes included in today's analysis. In response to
the NPRM, some comments urged the agencies also to quantify different
types of health impacts from changes in air pollution rather than only
accounting for such impacts in aggregate estimates of the social costs
of air pollution. Considering these comments, the agencies added such
calculations to the model, as discussed in Section VI.D.3.
6. Calculation of Benefits and Costs
Having estimated how technologies might be applied going forward,
and having estimated the range of resultant physical impacts, the CAFE
Model calculates a variety of private and social benefits and costs,
reporting these from the consumer, manufacturer, and social
perspectives, both in undiscounted and discounted present value form
(given inputs specifying the corresponding discount rate and present
year). Estimates of regulatory costs are among the direct outputs of
the simulation of manufacturers' potential responses to new standards.
Other benefits and costs are calculated based on the above-mentioned
estimates of travel demand, fuel consumption, emissions, and safety
impacts. The agencies received many comments on the NPRM's calculation
of benefits and costs, and Section VI.D.1 discusses these comments and
presents the methods, data sources, and estimates used in calculating
benefits and costs reported here.
7. Structure of Model Inputs and Outputs
All CAFE Model inputs and outputs described above are specified in
Microsoft Excel format, and the user can define and edit all inputs to
the system. Table VI-3 describes (non-exhaustively) which inputs are
contained within each input file and Table VI-4 describes which outputs
are contained in each output file. This is important for three reasons:
(1) Each file is discussed throughout the following sections; (2)
several commenters conflated aspects of the model with its inputs; and
(3) several commenters seemed confused about where to find specific
information in the output files. This information was described in
detail in the NPRM CAFE Model Documentation, but is reproduced here for
quick reference. When specifically referencing the input
[[Page 24282]]
or output file used for the NPRM or final rule in the following
discussion, NPRM or FRM, respectively, will precede the file name.
[GRAPHIC] [TIFF OMITTED] TR30AP20.080
[GRAPHIC] [TIFF OMITTED] TR30AP20.081
A catalog of the Argonne National Laboratory Autonomie fuel economy
technology effectiveness value output files are reproduced in the
following Table VI-5 as well. The left column shows the terminology
used in this text to refer to the file, while the right column
describes each file. NPRM or FRM, respectively, may precede the
terminology in the text as appropriate.
[[Page 24283]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.082
Finally, Table VI-6 lists the terminologies used to refer to other
model-related documents which are referred to frequently throughout the
text. NPRM or FRM, respectively, may precede the terminology in the
text as appropriate.
[[Page 24284]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.083
B. What inputs does the compliance analysis require?
1. Analysis Fleet
The starting point for the evaluation of the potential feasibility
of different stringency levels for future CAFE and CO2
standards is the analysis fleet, which is a snapshot of the recent
vehicle market. The analysis fleet provides a baseline from which to
project what and how additional technologies could feasibly be applied
to vehicles in a cost-effective manner to raise those vehicles' fuel
economy and lower their CO2 emission levels.\344\ The fleet
characterization also provides a reference point with data for other
factors considered in the analysis, including environmental effects and
effects estimated by the economic modules (i.e., sales, scrappage, and
labor utilization). When the scope of the analysis widens, another
piece of data must be included for each vehicle in the analysis fleet
to map a given element of the fleet appropriately onto an analysis
module.
---------------------------------------------------------------------------
\344\ The CAFE model does not generate compliance paths a
manufacturer should, must, or will deploy. It is intended as a tool
to demonstrate a compliance pathway a manufacturer could choose. It
is almost certain all manufacturers will make compliance choices
differing from those projected by the CAFE model.
---------------------------------------------------------------------------
For the analysis presented in this final rule, the analysis fleet
includes information about vehicles that is essential for each analysis
module. The first part of projecting how additional technologies could
be applied to vehicles is knowing which vehicles are produced by which
manufacturers, the fuel economies of those vehicles, how many of each
are sold, whether they are passenger cars or light trucks, and their
footprints. This is important because it improves understanding of the
overall impacts of different levels of CAFE and CO2
standards; overall impacts that result from industry's response to
standards, and industry's response, is made up of individual
manufacturer responses to the standards in light of the overall market
and their individual assessment of consumer acceptance. Establishing an
accurate representation of manufacturers' existing fleets (and the
vehicle models in them) that will be subject to future standards helps
in predicting potential individual manufacturer responses to those
future standards in addition to potential changes in those standards.
Another part of projecting how additional fuel economy improving
technologies could be applied to vehicles is knowing which fuel saving
technologies manufacturers have equipped on which vehicles. In many
cases, the agencies also collect and reference additional information
on other vehicle attributes to help with this process.\345\ Accounting
for technologies already applied to vehicles helps avoid ``double-
counting'' the value of those technologies, by assuming they are still
available to be applied to improve fuel economy and reduce
CO2 emissions. It also promotes more realistic
determinations of what additional technologies can feasibly be applied
to those vehicles: If a manufacturer has already started down a
technological path to fuel economy or performance improvements, the
agencies do not assume it will completely abandon that path because
doing so would be unrealistic and fails to represent accurately
manufacturer responses to standards. Each vehicle model (and
configurations of each model) in the analysis fleet, therefore, has a
comprehensive list of its technologies, which is important because
different configurations may have different technologies applied to
them.\346\ In addition, to properly account for technology costs, the
agencies assign each vehicle to a technology class and an engine class.
Technology classes reference each vehicle to a set of full vehicle
simulations, so that the agencies may project fuel efficiency with
combinations of additional fuel saving equipment and hybrid and
electric vehicle battery costs.
---------------------------------------------------------------------------
\345\ For instance, curb weight, horsepower, drive
configuration, pickup bed length, oil type, body style, aerodynamic
drag coefficients, and rolling resistance coefficients, and (if
applicable) battery sizes are all required to assign technology
content properly.
\346\ Considering each vehicle model/configuration also improves
the ability to consider the differential impacts of different levels
of potential standards on different manufacturers, since all vehicle
model/configurations ``start'' at different places, in terms of
technologies already used and how those technologies are used.
---------------------------------------------------------------------------
Yet another part of projecting which vehicles might exist in future
model years is developing reasonable real-world assumptions about when
and how manufacturers might apply certain technologies to vehicles. The
analysis fleet accounts for links between vehicles, recognizing vehicle
platforms will share technologies, and the vehicles that make up that
platform should receive (or not receive) additional technological
improvements together. Shared engines, shared transmissions, and shared
vehicle platforms for mass reduction technology are considered. In
addition, each vehicle model/configuration in the analysis fleet also
has information about its redesign
[[Page 24285]]
schedule, i.e., the last year it was redesigned and when the agencies
expect it to be redesigned again. Redesign schedules are a key part of
manufacturers' business plans, as each new product can cost more than
$1B, and involve a significant portion of a manufacturer's scarce
research, development, and manufacturing and equipment budgets and
resources.\347\ Manufacturers have repeatedly told the agencies that
sustainable business plans require careful management of resources and
capital spending, and that the length of time each product remains in
production is crucial to recouping the upfront product development and
plant/equipment costs, as well as the capital needed to fund the
development and manufacturing equipment needed for future products.
Because the production volume of any given vehicle model varies within
a manufacturer's product line, and varies among different
manufacturers, redesign schedules typically vary for each model and
manufacturer. Some (relatively few) technological improvements are
small enough that they can be applied in any model year; a few other
technological improvements may be applied during a refreshening (when a
few additional changes are made, but well short of a full redesign),
but others are major enough that they can only be cost-effectively
applied at a vehicle redesign, when many other things about the vehicle
are already changing. Ensuring the CAFE model makes technological
improvements to vehicles only when it is feasible to do so also helps
the analysis better represent manufacturer responses to different
levels of standards.
---------------------------------------------------------------------------
\347\ Shea, T., Why Does It Cost So Much For Automakers To
Develop New Models? Autoblog (Jul. 27, 2010), https://www.autoblog.com/2010/07/27/why-does-it-cost-so-much-for-automakers-to-develop-new-models/.
---------------------------------------------------------------------------
Finally, the agencies restrict the applications of some
technologies on some vehicles upon determining the technology is not
compatible with the functional and performance requirements of the
vehicle, or if the manufacturers are unlikely to apply a specific
technology to a specific vehicle for reasons articulated with
confidential business information that the agencies found credible.
Other data important for the analysis that are referenced to the
analysis fleet include baseline economic, environmental, and safety
information. Vehicle fuel tank size is required to estimate range and
refueling benefit while curb weights and safety class assignments help
the agencies consider how changes in vehicle mass may affect safety.
The agencies identify the final assembly location for each vehicle,
engine, and transmission, as well as the percent of U.S. content to
support the labor impact analysis. In addition, the aforementioned
accounting for first-year vehicle production volumes (i.e., the number
of vehicles of each new model sold in MY 2017, for this analysis) is
the foundation for estimating how future vehicle sales might change in
response to different potential standards.
The input file for the CAFE model characterizing the analysis
fleet, referred to as the ``market inputs'' file or ``market data''
file, accordingly includes a large amount of data about vehicles, their
technological characteristics, the manufacturers and fleets to which
they belong, and initial prices and production volumes, which provide
the starting points for projection (by the sales model) to ensuing
model years. In the Draft TAR (which utilized a MY 2015 analysis fleet)
and NPRM (which utilized a MY 2016 analysis fleet), the agencies needed
to populate about 230,000 cells in the market data file to characterize
the fleet. For this final rule (which utilized a MY 2017 analysis
fleet), the agencies populated more than 400,000 cells to characterize
the fleet. While the fleet is not actually much more heterogeneous in
reality,\348\ the agencies have provided and collected more data to
justify the characterization of the analysis fleet, and to support the
functionality of modules in the CAFE model.
---------------------------------------------------------------------------
\348\ The expansion of cells is primarily due to (1) considering
more technologies, and (2) listing trim levels separately, which
often yields more precise curb weights and more accurate
manufacturer suggested retail prices.
---------------------------------------------------------------------------
A solid characterization of a recent model year as an analytical
starting point helps realistically estimate ways manufacturers could
potentially respond to different levels of standards, and the modeling
strives to simulate realistically how manufacturers could progress from
that starting point. While manufacturers can respond in many ways
beyond those represented in the analysis (e.g., applying other
technologies, shifting production volumes, changing vehicle footprint),
such that it is impossible to predict with any certainty exactly how
each manufacturer will respond, it is still important to establish a
solid foundation from which to estimate potential costs and benefits of
potential future standards. The following sections discuss aspects of
how the analysis fleet was built for this analysis, and includes
discussion of the comments on fleet that the agencies received on the
proposed rule.
a) Principles on Data Sources Used To Populate the Analysis Fleet
The source data for vehicles in the analysis fleet and their
technologies is a central input for the analysis. The sections below
discuss pros and cons of different potential sources and what the
agencies used for this analysis, and responds to comments the agencies
received on data sources in the proposal.
(1) Use of Confidential Business Information Versus Publicly-Releasable
Sources
Since 2001, CAFE analysis has used either confidential, forward-
estimating product plans from manufacturers, or publicly available data
on vehicles already sold as a starting point for determining what
technologies can be applied to what vehicles in response to potential
different levels of standards. The use of either data source requires
certain tradeoffs. Confidential product plans comprehensively represent
what vehicles a manufacturer expects to produce in coming years,
accounting for plans to introduce new vehicles and fuel-saving
technologies and, for example, plans to discontinue other vehicles and
even brands. This information can be very thorough and can improve the
accuracy of the analysis, but cannot be publicly released. This makes
it difficult for public commenters to reproduce the analysis for
themselves as they develop their comments. Some non-industry commenters
have also expressed concern about manufacturers having an incentive in
the submitted plans to underestimate (deliberately or not) their future
fuel economy capabilities and overstate their expectations about, for
example, the levels of performance of future vehicle models in order to
affect the analysis. Accordingly, since 2010, EPA and NHTSA have based
analysis fleets almost exclusively on information from commercial and
public sources, starting with CAFE compliance data and adding
information from other sources.
An analysis fleet based primarily on public sources can be released
to the public, solving the issue of commenters being unable to
reproduce the overall analysis. However, industry commenters have
argued such an analysis fleet cannot accurately reflect manufacturers'
actual plans to apply fuel-saving technologies (e.g., manufacturers may
apply turbocharging to improve not just fuel economy, but also to
improve vehicle performance) or manufacturers' plans to change product
offerings by introducing some vehicles and brands and discontinuing
other
[[Page 24286]]
vehicles and brands, precisely because that information is typically
confidential business information (CBI). A fully-publicly-releasable
analysis fleet holds vehicle characteristics unchanged over time and
lacks some level of accuracy when projected into the future. For
example, over time, manufacturers introduce new products and even
entire brands. On the other hand, plans announced in press releases do
not always ultimately bear out, nor do commercially available third-
party forecasts. Assumptions could be made about these issues to
improve the accuracy of a publicly releasable analysis fleet, but
concerns include that this information would either be largely
incorrect, or, if the assumptions were correct, information would be
released that manufacturers would consider CBI.
Furthermore, some technologies considered in the rulemaking are
difficult to observe in the analysis fleet without expensive teardown
study and time-consuming benchmarking. Not giving credit for these
technologies puts the analysis at significant risk of double-counting
the effectiveness of these technologies, as manufacturers cannot equip
technologies twice to the same vehicle for double the fuel economy
benefit. As discussed in the Draft TAR, the agencies assigned little
(if any) technology application in the baseline fleet for some of these
technologies.\349\ For the NPRM MY 2016 fleet development process, the
agencies again offered the manufacturers the opportunity to volunteer
CBI to the agencies to help inform the technology content of the
analysis fleet, and many manufacturers did. The agencies were able to
confirm that many manufacturers had already included many hard-to-
observe technologies in the MY 2016 fleet (which they were not properly
given credit for in the characterization of the MY 2014 and MY 2015
fleets presented in Draft TAR) so the agencies reflected this new
information in the NPRM analysis and in the analysis presented today.
---------------------------------------------------------------------------
\349\ These technologies include low rolling resistance
technology (incorrectly applied to zero baseline vehicles in Draft
TAR), low-drag brakes (incorrectly applied to zero baseline vehicles
in Draft TAR), electric power steering (incorrectly applied to too
few vehicles in Draft TAR), accessory drive improvements
(incorrectly applied to zero baseline vehicles in Draft TAR), engine
friction reduction (previously named LUBEFR1, LUBEFR2, and LUBEFR3),
secondary axle disconnect and transmission improvements.
---------------------------------------------------------------------------
In addition, many manufacturers provided confidential comment on
the potential applicability of fuel-saving technologies to their fleet.
In particular, many manufacturers confidentially identified specific
engine technologies that they will not use in the near term, either on
specific vehicles, or at all. Reasons varied: Some manufacturers cited
intellectual property concerns, and others stated functional
performance concerns for some engine types on some vehicles. Other
manufacturers shared forward-looking product plans, and explained that
it would be cost prohibitive to scrap significant investments in one
technology in favor of another. This topic is discussed in more detail
in Section VI.B.1.b)(6), below.
The agencies sought comment on how to address this issue going
forward, recognizing both the competing interests involved and the
typical timeframes for CAFE and CO2 standards rulemakings.
Many commenters expressed concern with the agencies using any CBI
as part of the rulemaking process. Some commenters expressed concern
that use of CBI would make the CAFE model subject to inaccuracies
because manufacturers would only provide additional information in
situations in which a correction to the agencies' baseline assumptions
would favor the manufacturers.\350\ The agencies recognize this as a
reasonable concern, but the analysis presented in the Draft TAR
consistently assumed very little (if any) technology had been applied
in the baseline. In addition, many manufacturers shared information on
advanced technologies that were not yet in production in MY 2017, but
could be used in the future; manufacturer contributions helped the
agencies better model many advanced engine technologies and to include
them in today's analysis, and inclusion of these technologies (and
costs) in the analysis sometimes lowered the projected cost of
compliance for stringent alternatives. Other commenters expressed
concern that automakers would supply false or incomplete information
that would unduly restrict what technologies can be deployed.\351\ When
possible, the agencies sought independently to verify manufacturer CBI
(or claims made by other stakeholders) through lab testing and
benchmarking.\352\ The agencies found no evidence of misrepresentation
of engineering specifications in the MY 2017 fleet in manufacturer CBI;
instead, the agencies were able to verify independently many CBI
submissions, and confirm the credibility of information provided from
those sources.
---------------------------------------------------------------------------
\350\ NHTSA-2018-0067-12039, Union of Concerned Scientists.
\351\ NHTSA-2018-0067-11741, ICCT.
\352\ For instance, the agencies continue to evaluate tire
rolling resistance on production vehicles via independent lab
testing, and the agencies bench-marked the operating behavior and
calibration of many engines and transmissions.
---------------------------------------------------------------------------
Some commenters requested that more CBI be used in the analysis.
For instance, some commenters suggested that the agencies should return
to the use of product plans and announcements regarding future fleets
because manufacturers had already committed investments to bring
announced products to market.\353\ However, if the agencies were to
assume that these commitments were already in the baseline, the
agencies would underestimate the cost of compliance for stringent
alternatives. Moreover, while upfront investments to bring technologies
to market are significant, the total marginal costs of components are
typically large in comparison over the entire product life-cycle, and
these costs have not yet been realized in vehicles not yet produced.
---------------------------------------------------------------------------
\353\ NHTSA-2018-0067-11956, PA Department of Environmental
Protection.
---------------------------------------------------------------------------
The agencies did make use of some forward-looking CBI in the
analysis. The agencies received many comments from manufacturers on the
technological feasibility, or functional applicability of some fuel
saving technologies to certain vehicles, or certain vehicle
applications, and the agencies took this information into consideration
when projecting compliance pathways. These cases are discussed
generally in Section VI.B.1.b)(6), below, and specifically for each
technology in those technology sections. Some commenters expressed that
the use of CBI for future product plans would be acceptable, but only
if the agencies disclosed the CBI affecting all vehicles through MY
2025 at the time of publication.\354\ Functionally, this is not
possible. Manufacturer's confidential product plans cannot be made
public, as prohibited under NHTSA's regulations at 49 CFR part 512, and
if the information meets the requirements of section 208(c) of the
Clean Air Act. If the agencies disclosed confidential information, it
would not only violate the terms on which the agencies obtained the
CBI, but it is unlikely that manufacturers would continue to offer CBI,
which in turn would likely degrade the quality of the analysis. The
agencies believe that the use of CBI in the NPRM and final rule
analysis--to confirm, reference, or to otherwise modify aspects of the
analysis that can be made public--threads the needle between a more
accurate but less transparent analysis (using more CBI) and a less
accurate but more transparent analysis (using less CBI).
---------------------------------------------------------------------------
\354\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
[[Page 24287]]
(2) Source Data and Vintage Used in the Analysis
Based on the assumption that a publicly-available analysis fleet
continued to be desirable, manufacturer compliance submissions to EPA
and NHTSA were used as a starting point for the NPRM and final rule
analysis fleets. Generally, manufacturer compliance submissions break
down vehicle fuel economy and production volume by regulatory class,
and include some very basic product information (typically including
vehicle nameplate, engine displacement, basic transmission information,
and drive configuration). Many different trim levels of a product are
typically rolled up and reported in an aggregated fashion, and these
groupings can make decomposition of different fuel-saving, road load
reducing technologies extremely difficult. For instance, vehicles in
different test weight classes, with different tires or aerodynamic
profiles may be aggregated and reported together.\355\ A second portion
of the compliance submission summarizes production volume by vehicle
footprints (a key compliance measure for standard setting) by
nameplate, and includes some basic information about engine
displacement, transmission, and drive configuration. Often these
production volumes by footprint do not fit seamlessly together with the
production volumes for fuel economy, so the agencies must reconcile
this information.
---------------------------------------------------------------------------
\355\ Some fuel-economy compliance information for pickup trucks
span multiple cab and box configurations, but manufacturers reported
these disparate vehicles together.
---------------------------------------------------------------------------
Information from the MY 2016 fleet was chosen as the foundation for
the NPRM analysis fleet because, at the time the rulemaking analysis
was initiated, the 2016 fleet represented the most up-to-date
information available in terms of individual vehicle models and
configurations, production technology levels, and production volumes.
If MY 2017 data had been used while this analysis was being developed,
the agencies would have needed to use product planning information that
could not be made available to the public until a later date.
The NPRM analysis fleet was initially developed with 2016 mid-model
year compliance data because final compliance data was not available at
that time, and the timing provided manufacturers the opportunity to
review and comment on the characterization of their vehicles in the
fleet. With a view toward developing an accurate characterization of
the 2016 fleet to serve as an analytical starting point, corrections
and updates to mid-year data (e.g., to production estimates) were
sought, in addition to corroboration or correction of technical
information obtained from commercial and other sources (to the extent
that information was not included in compliance data), although future
product planning information from manufacturers (e.g., future product
offerings, products to be discontinued) was not requested, as most
manufacturers view such information as CBI. Manufacturers offered a
range of corrections to indicate engineering characteristics (e.g.,
footprint, curb weight, transmission type) of specific vehicle model/
configurations, as well as updates to fuel economy and production
volume estimates in mid-year reporting. After following up on a case-
by-case basis to investigate significant differences, the analysis
fleet was updated.
Sales, footprint, and fuel economy values with final compliance
data were also updated if that data was available. In a few cases,
final production and fuel economy values were slightly different for
specific MY 2016 vehicle models and configurations than were indicated
in the NPRM analysis; however, other vehicle characteristics (e.g.,
footprint, curb weight, technology content) important to the analysis
were reasonably accurate. While some commenters have, in the past,
raised concerns that non-final CAFE compliance data is subject to
change, the potential for change is likely not significant enough to
merit using final data from an earlier model year reflecting a more
outdated fleet. Moreover, even ostensibly final CAFE compliance data is
frequently subject to later revision (e.g., if errors in fuel economy
tests are discovered), and the purpose of the analysis was not to
support enforcement actions but rather to provide a realistic
assessment of manufacturers' potential responses to future standards.
Manufacturers integrated a significant amount of new technology in
the MY 2016 fleet, and this was especially true for newly-designed
vehicles launched in MY 2016. While subsequent fleets will involve even
further application of technology, using available data for MY 2016
provided the most realistic detailed foundation for analysis that could
be made available publicly in full detail, allowing stakeholders to
reproduce the analysis presented in the proposal independently. Insofar
as future product offerings are likely to be more similar to vehicles
produced in 2016 than to vehicles produced in earlier model years,
using available data regarding the 2016 model year provided the most
realistic, publicly releasable foundation for constructing a forecast
of the future vehicle market for this proposal. Many comments
responding to the Draft TAR, EPA's Proposed Determination, EPA's 2017
Request for Comment, and the NPRM preceding today's notice stated that
the most up-to-date analysis fleet possible should be used, because a
more up-to-date analysis fleet will better capture how manufacturers
apply technology and will account better for vehicle model/
configuration introductions and deletions.356 357
---------------------------------------------------------------------------
\356\ 82 FR 39551 (Aug. 21, 2017).
\357\ For example, in 2016 comments to dockets EPA-HQ-OAR-2015-
0827 and NHTSA-2016-0068, the Alliance of Automobile Manufacturers
commented that ``the Alliance supports the use of the most recent
data available in establishing the baseline fleet, and therefore
believes that NHTSA's selection [of, at the time, model year 2015]
was more appropriate for the Draft TAR.'' Alliance at 82, Docket ID.
EPA-HQ-OAR-2015-0827-4089. Global Automakers commented that ``a one-
year difference constitutes a technology change-over for up to 20%
of a manufacturer's fleet. It was also generally understood by
industry and the agencies that several new, and potentially
significant, technologies would be implemented in MY 2015. The use
of an older, outdated baseline can have significant impacts on the
modeling of subsequent Reference Case and Control Case
technologies.'' Global Automakers at A-10, Docket ID EPA-HQ-OAR-
2015-0827-4009.
---------------------------------------------------------------------------
On the other hand, some commenters suggested that because
manufacturers continue improving vehicle performance and utility over
time, an older analysis fleet should be used to estimate how the fleet
could have evolved had manufacturers applied all technological
potential to fuel economy rather than continuing to improve vehicle
performance and utility.\358\ Because manufacturers change and improve
product offerings over time, conducting analysis with an older analysis
fleet (or with a fleet using fuel economy levels and CO2
emissions rates that have been adjusted to reflect an assumed return to
levels of performance and utility typical of some past model year)
would miss this real-world trend. While such an analysis could project
what industry could do if, for example, manufacturers devoted all
technological improvements toward raising fuel economy and reducing
CO2 emissions (and if consumers decided to purchase these
vehicles), the agencies do not believe it would be consistent with a
transparent examination of what effects different levels of standards
would have
[[Page 24288]]
on individual manufacturers and the fleet as a whole.
---------------------------------------------------------------------------
\358\ For example, in 2016 comments to dockets EPA-HQ-OAR-2015-
0827 and NHTSA-2016-0068, UCS stated ``in modeling technology
effectiveness and use, the agencies should use 2010 levels of
performance as the baseline.'' UCS at 4, Docket ID. EPA-HQ-OAR-2015-
0827-4016.
---------------------------------------------------------------------------
All else being equal, using a newer analysis fleet will produce
more realistic estimates of impacts of potential new standards than
using an outdated analysis fleet. However, among relatively current
options, a balance must be struck between input freshness, and input
completeness and accuracy.\359\ During assembly of the inputs for the
NPRM analysis, final compliance data was available for the MY 2015
model year but not, in a few cases, for MY 2016. However, between mid-
year compliance information and manufacturers' specific updates
discussed above, a robust and detailed characterization of the MY 2016
fleet was developed. While information continued to develop regarding
the MY 2017 and, to a lesser extent MY 2018 and even MY 2019 fleets,
this information was--even in mid-2017--too incomplete and inconsistent
to be assembled with confidence into an analysis fleet for modeling
supporting deliberations regarding the NPRM analysis.
---------------------------------------------------------------------------
\359\ Comments provided through a recent peer review of the CAFE
model recognize the competing interests behind this balance. For
example, referring to NHTSA's 2016 Draft TAR analysis, one of the
peer reviewers commented as follows: ``The NHTSA decision to use MY
2015 data is wise. In the TAR they point out that a MY 2016
foundation would require the use of confidential data, which is less
desirable. Clearly they would also have a qualitative vision of the
MY 2016 landscape while employing MY 2015 as a foundation. Although
MY 2015 data may still be subject to minor revision, this is
unlikely to impact the predictive ability of the model . . . A more
complex alternative approach might be to employ some 2016 changes in
technology, and attempt a blend of MY 2015 and MY 2016, while
relying of estimation gained from only MY 2015 for sales. This
approach may add some relevancy in terms of technology, but might
introduce substantial error in terms of sales.''
---------------------------------------------------------------------------
Manufacturers requested that the baseline fleet supporting the
final rule incorporate the MY 2018 or most recent information
available.\360\ Other commenters expressed desire for multiple fleets
of various vintages to compare the updated model outputs with those of
previous rule-makings. Specifically, some commenters requested that
older fleet vintages (MY 2010, for instance) be developed in parallel
with the MY 2017 fleet so that those too may be used as inputs for the
model.\361\
---------------------------------------------------------------------------
\360\ NHTSA-2018-0067-12150, Toyota North America.
\361\ NHTSA-2018-0067-11741, ICCT.
---------------------------------------------------------------------------
Between the NPRM and this final rule, manufacturers submitted final
compliance data for the MY 2017 fleet. When the agencies pulled
together information for the fleet for the final rule, the agencies
decided to use the highest-quality, most up-to-date information
available. Given that pulling this information together takes some
time, and given that ``final'' compliance submissions often lag
production by a few years, the agencies decided to use 2017 model year
as the base year for the analysis fleet, as the agencies stated in the
NPRM.\362\ While the agencies could have used preliminary 2018 data or
even very early 2019 data, this information was not available in time
to support the final rulemaking. Likewise, the agencies chose not to
revert to a previous model year (for instance 2016 or 2012) because
many manufacturers have incorporated fuel savings technologies over the
last few years, realized some benefits for fuel economy, and adjusted
the performance or sales mix of vehicles to remain competitive in the
market. Also, using an earlier model year would provide less accurate
projections because the analysis would be based on what manufacturers
could have done in past model years and would have estimated the fuel
economy improvements instead of using known information on the
technologies that were employed and the actual fuel economy that
resulted from applying those technologies.
---------------------------------------------------------------------------
\362\ 83 FR 43006 (``If newer compliance data (i.e., MY 2017)
becomes available and can be analyzed during the pendency of this
rulemaking, and if all other necessary steps can be performed, the
analysis fleet will be updated, as feasible, and made publicly
available.'').
---------------------------------------------------------------------------
Some additional information (about off-cycle technologies, for
instance) was often not reported by manufacturers in MY 2017 formal
compliance submissions in a way that provided clear information on
which technologies were included on which products. As part of the
formal compliance submission, some manufacturers voluntarily submitted
additional information (about engine technologies, for instance). While
this data was generally of very high quality, there were some mistakes
or inconsistencies with publicly available information, causing the
agencies to contact the manufacturers to understand and correct
identified issues. In most cases, however, the formal compliance data
was very limited in nature, and the agencies collected additional
information necessary to characterize fully the fleet from other
sources, and scrutinized additional information submitted by
manufacturers carefully, independently verifying when possible.
Specifically, the agencies downloaded and reviewed numerous
marketing brochures and product launch press releases to confirm
information submitted by manufacturers and to fill in information
necessary for the analysis fleet that was not provided in the
compliance data. Product brochures often served as the basis for the
curb weights used in the analysis. This publicly available manufacturer
information sometimes also included aerodynamic drag coefficients,
information about steering architecture, start-stop systems, pickup bed
lengths, fuel tank capacities, and high-voltage battery capacities. The
agencies recorded vehicle horsepower, compression ratio, fuel-type, and
recommended oil weight rating from a combination of manufacturer
product brochures and owner's manuals. The product brochures, as well
as online references such as Autobytel, informed which combinations of
fuel saving technologies were available on which trim levels, and what
the manufacturer suggested retail price was for many products. Overall
this information proved helpful for assigning technologies to vehicles,
and for getting data (such as fuel tank size \363\) necessary for the
analysis. These reference materials have been included in the
rulemaking documentation.\364\
---------------------------------------------------------------------------
\363\ The quality of data for today's analysis fleet is notably
improved for fuel tank capacity, which factors into the calculation
of refueling time benefits. In many previous analyses, fuel tank
sizes were often stated as estimates or proxies, and not sourced so
carefully.
\364\ Publicly available data used to supplement analysis fleet
information is available in the docket.
---------------------------------------------------------------------------
The agencies elected not to develop fleets of previous model year
vintages that could be used in parallel as an input to the CAFE model.
Developing a detailed characterization of the fleet of any vintage
would be a huge undertaking with few benefits. As the scope has
increased, and as additional modules are added, going back in time to
re-characterize a previous fleet in a format that works with CAFE model
updates can be time- and resource-prohibitive for the agencies, even if
that work is adapting a fleet that was used in previous rule-making
analysis. Doing so also offers little value in determining what
potential fuel saving technology can be added to a more recent fleet
during the rulemaking timeframe.
The MY 2017 manufacturer-submitted data, verified and supplemented
by the agencies with publicly-available information, therefore
presented the fullest, most up-to-date data set that the agencies could
have used to support this analysis.
[[Page 24289]]
b) Characterizing Vehicles and Their Technology Content
The starting point for projecting what additional fuel economy
improving technologies could feasibly be applied to vehicles is knowing
what vehicles are produced by which manufacturers and what technologies
exist on those vehicles. Rows in the market data file are the smallest
portion of the fleet to which technology may be applied as part of a
projected compliance pathway. For the analysis presented in this final
rule, the agencies, when possible, attempted to include vehicle trim
level information in discrete rows. A manufacturer, for example GM, may
produce one or more vehicle makes (or brands), for example Chevrolet,
Buick and others. Each vehicle make may offer one or more vehicle
models, for example Malibu, Traverse and others. And each vehicle model
may be available in one or more trim levels (or standard option
levels), for example ``RS,'' ``Premier'' and others, which have
different levels of standard options, and in some cases, different
engines and transmissions.
Manufacturer compliance submissions, discussed above, were used as
a starting point to define working rows in the market data file;
however, often the rows needed to be further disaggregated to correctly
characterize vehicle information covered in the scope of the analysis,
and analysis fleet. Manufacturers often grouped vehicles with multiple
trim levels together because they often included the same fuel-saving
technologies and may be aggregated to simplify reporting. However, the
manufacturer suggested retail prices of different trim levels are
certainly different, and other features relevant to the analysis are
occasionally different.
As a result of further disaggregating compliance information, the
number of rows in the market data file increased from 1,667 rows used
in the NPRM to 2,952 rows for this final rule analysis. The agencies do
not have data on sales volumes for each nameplate by trim level, and
used an approach that evenly distributed volume across offered trim
levels, within the defined constraints of the compliance data.\365\
Evenly distributing the volume across trim levels is a simplification,
but this action should (1) highlight some difficulties that could be
encountered when acquiring data for a full-vehicle consumer choice
model should the agencies pursue developing one in the future
(discussed further, below), and (2) lower the average sales volume per
row in the market data file, thereby allowing the application of very
advanced electrification technologies in smaller lumps. The latter
effect is responsive to comments (discussed below) that suggested
electrification technologies could be more cost-effectively deployed in
lower volumes, and that the CAFE model artificially constrains cost
effective technologies that may be deployed, resulting in higher costs
and large over-compliance.
---------------------------------------------------------------------------
\365\ The sum of volumes by nameplate configuration, for fuel
economy value, and for footprint value remains the same.
---------------------------------------------------------------------------
(1) Assigning Vehicle Technology Classes
While each vehicle in the analysis fleet has its list of observed
technologies and equipment, the ways in which manufacturers apply
technologies and equipment do not always coincide perfectly with how
the analysis characterizes the various technologies that improve fuel
economy and reduce CO2 emissions. To improve how the
observed vehicle fleet ``fits into'' the analysis, each vehicle model/
configuration is ``mapped'' to the full-vehicle simulation modeling by
Argonne National Laboratory that is used to estimate the effectiveness
of the fuel economy-improving/CO2 emissions-reducing
technologies considered. Argonne produces full-vehicle simulation
modeling for many combinations of technologies, on many types of
vehicles, but it did not simulate literally every single manufacturer's
vehicle model/configuration in the analysis fleet because it would be
impractical to assemble the requisite detailed information--much of
which would likely only be provided on a confidential basis--specific
to each vehicle model/configuration and because the scale of the
simulation effort would correspondingly increase by at least two orders
of magnitude. Instead, Argonne simulated 10 different vehicle types
corresponding to the ``technology classes'' generally used in CAFE
analysis over the past several rulemakings (e.g., small car, small
performance car, pickup truck, etc.). Each of those 10 different
vehicle types was assigned a set of ``baseline characteristics'' to
which Argonne added combinations of fuel-saving technologies and then
ran simulations to determine the fuel economy achieved when applying
each combination of technologies to that vehicle type given its
baseline characteristics.
BILLING CODE 4910-59-P
[[Page 24290]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.089
BILLING CODE 4910-59-C
In the analysis fleet, inputs assign each specific vehicle model/
configuration to a technology class, and once there, map to the
simulation within that technology class most closely matching the
combination of observed technologies and equipment on that vehicle.
This mapping to a specific simulation result most closely representing
a given vehicle model/configuration's initial technology ``state''
enables the CAFE model to estimate the same vehicle model/
configuration's fuel economy after application of some other
combination of technologies, leading to an alternative technology
state.
(2) Assigning Vehicle Technology Content
As explained above, the analysis fleet is defined not only by the
vehicles it contains, but also by the technologies on those vehicles.
Each vehicle in the analysis fleet has an associated list of observed
technologies and equipment that can improve fuel economy and reduce
CO2 emissions.\366\ With a portfolio of descriptive
technologies arranged by manufacturer and model, the analysis fleet can
be summarized and project how vehicles in that fleet may increase fuel
economy over time via the application of additional technology.
---------------------------------------------------------------------------
\366\ These technologies are generally grouped into the
following categories: Vehicle technologies include mass reduction,
aerodynamic drag reduction, low rolling resistance tires, and
others. Engine technologies include engine attributes describing
fuel type, engine aspiration, valvetrain configuration, compression
ratio, number of cylinders, size of displacement, and others.
Transmission technologies include different transmission
arrangements like manual, 6-speed automatic, 10-speed automatic,
continuously variable transmission, and dual-clutch transmissions.
Hybrid and electric powertrains may complement traditional engine
and transmission designs or replace them entirely.
---------------------------------------------------------------------------
In many cases, vehicle technology is clearly observable from the
2017 compliance data (e.g., compliance data indicates clearly which
vehicles have turbochargers and which have continuously variable
transmissions), but in some cases technology levels are less
observable. For the latter, like levels of mass reduction, the analysis
categorized levels of technology already used in a given vehicle.
Similarly, engineering judgment was used to determine if higher mass
reduction levels may be used practicably and safely for a given
vehicle.
Either in mid-year compliance data for MY 2016, final compliance
data for MY 2017, or separately and at the agencies' invitation prior
to the NPRM or in comments in responses to the NPRM, most manufacturers
provided guidance on the technology already present in each of their
vehicle model/configurations. This information was not as complete for
all manufacturers' products as needed for the analysis, so, in some
cases, information was supplemented with publicly available data,
typically from manufacturer media sites. In limited cases,
manufacturers did not supply information, and
[[Page 24291]]
information from commercial and publicly available sources was used.
The agencies continued to evaluate emerging technologies in the
analysis. In response to comments,\367\ and given recent product
launches for MY 2020, and some very recently announced future product
offerings, the agencies elevated some technologies that were discussed
in the NPRM to the compliance simulation. As a result, several
additional engine technologies, expanded levels of mass reduction
technology, and some additional combinations of engines with plug-in
hybrid, or strong hybrid technology are available in the compliance
pathways for the final rule analysis.
---------------------------------------------------------------------------
\367\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
In addition, some redundant technologies, or technologies that were
inadvertently represented on the technology tree as being available to
be applied twice, have been consolidated. For instance, previous basic
versions of engine friction reduction were layered on top of basic
engine maps, but the efficiency in many modern engine maps already
include the benefits of that engine friction reduction technology. The
following Table VI-8 lists the technologies considered in the final
rule analysis, with the data sources used to map those technologies to
vehicles in the analysis fleet.
BILLING CODE 4910-59-P
[[Page 24292]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.090
[[Page 24293]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.091
[[Page 24294]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.092
BILLING CODE 4910-59-C
Industry commenters generally stated the MY 2016 baseline
technology content presented in the NPRM as an improvement over
previous analyses because it more accurately accounted for technology
already used in the fleet.368 369 In contrast, some
commenters expressed preference for EPA's baseline technology
assignment assumptions presented in the Draft TAR for mass reduction,
tire rolling resistance, and aerodynamic drag because those assumptions
projected very few technology improvements were present in the baseline
fleet. In assessing the comments, the agencies found that
[[Page 24295]]
using the EPA Draft TAR approach would lead to projected compliance
pathways with overestimated fuel economy improvements and
underestimated costs.\370\
---------------------------------------------------------------------------
\368\ NHTSA-2018-0067-12073, Alliance of Automobile
Manufacturers.
\369\ NHTSA-2018-0067-12150, Toyota North America.
\370\ NHTSA-2018-0067-11741, ICCT.
---------------------------------------------------------------------------
Many of those assumptions were neither scientifically meritorious,
nor isolated examples. For instance, for the EPA Draft TAR and Proposed
Determination analyses, the BMW i3, a vehicle with full carbon fiber
bodysides and downsized, mass-reduced wheels and tires (some of the
most advanced mass reducing technologies commercialized in the
automotive industry), was assumed to have 1.0 percent mass reduction (a
very minor level of mass reduction). Similarly, previous analyses
assigned the Chevrolet Corvette, a performance vehicle that has long
been a platform for commercializing advanced weight saving
technologies,\371\ with zero mass reduction. For aerodynamic drag,
previous EPA analysis assumed that pickup trucks could achieve the
aerodynamic drag profile typical of a sedan, with little regard for
form drag constraints or frontal area (and headroom, or ground
clearance) considerations. These assumptions commonly led to
projections of a 20 percent improvement in mass, aerodynamic drag, and
tire rolling resistance, even when a large portion of those
improvements had either already been implemented, or were not
technologically feasible. On the other hand, in the Draft TAR, NHTSA
presented methodologies to evaluate content for mass reduction
technology, aerodynamic drag improvements, and rolling resistance
technologies that better accounted for the actual level of technologies
in the analysis fleet. Throughout the rulemaking process, the agencies
reconciled these differences, jointly presented improved approaches in
the NPRM similar to what NHTSA presented in the Draft TAR, and again
used those reconciled approaches in today's analysis.\372\
---------------------------------------------------------------------------
\371\ See, e.g., Fiberglass to Carbon Fiber: Corvette's
Lightweight Legacy, GM (August 2012), https://media.gm.com/media/us/en/gm/news.detail.html/content/Pages/news/us/en/2012/Aug/0816_corvette.html.
\372\ Because these road load technologies are no longer double
counted, the projected compliance pathway in the NPRM, and in
today's analysis for stringent alternatives, often requires more
advanced fuel saving technologies than previously projected,
including higher projected penetration rates of hybrid and electric
vehicle technologies.
---------------------------------------------------------------------------
Many commenters correctly observed that the analysis fleet in the
NPRM recognized more technology content in the baseline than in the
Draft TAR (with higher penetration rates of tire rolling resistance and
aerodynamic drag improvements, for instance), but also that the fuel
economy values of the fleet had not improved all that much from the
previous year. Some commenters concluded that the NPRM baseline
technology assignment process was arbitrary and overstated the
technology content already present in the baseline
fleet.373 374 The agencies agree that there was a large
increase in the amount of road load technology credited in the baseline
fleet between EPA's Draft TAR and the jointly produced NPRM, and
clarify that this change was largely due to a recognition of
technologies that were actually present in the fleet, but not properly
accounted for in previous analyses. The change in penetration rates of
road load technologies (after accounting for glider share updates,
which is discussed in more detail in the mass reduction technology
section) between the NPRM and today's analysis is relatively small.
---------------------------------------------------------------------------
\373\ NHTSA-2018-0067-11741, ICCT.
\374\ NHTSA-2018-0067-12039, Union of Concerned Scientists.
---------------------------------------------------------------------------
Many commenters noted that the different baseline road load
assumptions (and other technology modeling) materially affect
compliance pathways, and projected costs.\375\ ICCT commented that the
agencies should conduct sensitivity analyses assuming every vehicle in
the analysis fleet is set to zero percent road load technology
improvement, to demonstrate how the technology content of the analysis
fleet affected the compliance scenarios.\376\
---------------------------------------------------------------------------
\375\ NHTSA-2018-0067-11928, Ford Motor Company.
\376\ NHTSA-2018-0067-11741, ICCT.
---------------------------------------------------------------------------
While the agencies have clearly described the methods by which
initial road load technologies are assigned in Section VI.C.4 Mass
Reduction, Section VI.C.5 Aerodynamics, and Section VI.C.6 Tire Rolling
Resistance below, the agencies considered a sensitivity case that
assumed no mass reduction, rolling resistance, or aerodynamic
improvements had been made to the MY 2017 fleet (i.e., setting all
vehicle road levels to zero--MRO, AERO and ROLL0). While this is an
unrealistic characterization of the initial fleet, the agencies
conducted a sensitivity analysis to understand any affect it may have
on technology penetration along other paths (e.g. engine and hybrid
technology). Under the CAFE program, the sensitivity analysis shows a
slight decrease in reliance on engine technologies (HCR engines,
turbocharge engines, and engines utilizing cylinder deactivation) and
hybridization (strong hybrids and plug-in hybrids) in the baseline
(relative to the central analysis). The consequence of this shift to
reliance on lower-level road load technologies is a reduction in
compliance cost in the baseline of about $300 per vehicle (in MY 2026).
As a result, cost savings in the preferred alternative are reduced by
about $200 per vehicle. Under the CO2 program, the general
trend in technology shift is less dramatic (though the change in BEVs
is larger) than the CAFE results. The cost change is also comparable,
but slightly smaller ($200 per vehicle in the baseline) than the CAFE
program results. Cost savings under the preferred alternative are
further reduced by about $100. With the lower technology costs in all
cases, the consumer payback periods decreased as well. These results
are consistent with the approach taken by manufacturers who have
already deployed many of the low-level road load reduction
opportunities to improve fuel economy.
Some commenters preferred that the agencies develop a different
methodology based on reported road load coefficients (``A,'' ``B'' and
``C'' coastdown coefficients) to estimate levels of aerodynamic drag
improvement and rolling resistance in the baseline fleet that did not
rely on CBI.\377\ The agencies considered this, but determined that
using CBI to assign baseline aerodynamic drag levels and rolling
resistance values was more accurate and appropriate. Estimating
aerodynamic drag levels and rolling resistance levels from coastdown
coefficients is not straightforward, and to do it well would require
information the agencies do not have (much of which is also CBI). For
instance, rotational inertias of wheel, tire, and brake packages can
affect coastdown, so mass of the vehicle is not sufficient. The frontal
area of the vehicles, a key component for calculating aerodynamic drag,
is rarely known, and often requires manufacturer input to get an
accurate value. Other important vehicle features like all-wheel-drive
should also be accounted for, and the agencies would struggle to
correctly identify improvements in rolling resistance, low-drag brakes,
and secondary axle disconnect, because all of these technologies would
present similar signature on a coast down test. All of these
technologies are represented as technology pathways in today's
analysis. Manufacturers acknowledged the possibility of using road load
coefficients to estimate rolling resistance and aerodynamic features,
but warned that the process ``required
[[Page 24296]]
various assumptions and is not very accurate,'' and stated that the use
of CBI to assess aerodynamic and rolling resistance technologies is an
``accurate and practical solution'' to assign these difficult to
observe technologies.\378\
---------------------------------------------------------------------------
\377\ NHTSA-2018-0067-11741, ICCT.
\378\ NHTSA-2018-0067-12073, Alliance of Automobile
Manufacturers.
---------------------------------------------------------------------------
(3) Assigning Engine Configurations
Engine technology costs can vary significantly by the configuration
of the engine. For instance, adding variable valve lift to each
cylinder on an engine would cost more for an engine with eight
cylinders than an engine with four cylinders. Similarly, the cost of
adding a turbocharger to an engine and downsizing the engine would be
different going from a naturally aspirated V8 to a turbocharged V6 than
going from a naturally aspirated V6 to a turbocharged I4. As discussed
in detail in the engine technology section of this document, the cost
files for the CAFE model account for instances such as these examples.
Information in the analysis fleet enables the CAFE model to
reference the intended engine costs. The ``Engine Technology Class
(Observed)'' lists the architecture of the observed engine. Notably,
the analysis assumes that nearly all turbo charged engines take
advantage of downsizing to optimize fuel efficiency, minimize the cost
of turbo charging, and to maintain performance (to the extent
practicable) with the naturally aspirated counterpart engine.
Therefore, engines observed in the fleet that have already been down-
sized must reference costs for a larger basic engine, which assumes
down-sizing with the application of turbo technology. In these cases,
the ``Engine Technology Class'' which is used to reference costs will
be larger than the ``Engine Technology Class (Observed).''
This is the same process agencies used in the NPRM, and it corrects
a previous error in the Draft TAR analysis, which incorrectly
underestimated turbocharged engine costs.\379\ Some commenters
expressed confusion and disagreement with this correction, with some
even commenting that the analysis baselessly inflated costs of
turbocharging technologies between the Draft TAR and the NPRM.\380\ To
be clear, this was a correction so that the costs used to calculate
turbocharged engine costs accurately reflected the total costs for a
turbocharged engine.
---------------------------------------------------------------------------
\379\ For instance, the Draft TAR engine costs would map an
observed V6 Turbo engine to I4 Turbo engine costs, by referencing a
4C1B engine cost.
\380\ NHTSA-2018-0067-11741, ICCT.
---------------------------------------------------------------------------
(4) Characterizing Shared Vehicle Platforms, Engines, and Transmissions
Another aspect of characterizing vehicle model/configurations in
the analysis fleet is based on whether they share a ``platform'' with
other vehicle model/configurations. A ``platform'' refers to engineered
underpinnings shared on several differentiated products. Manufacturers
share and standardize components, systems, tooling, and assembly
processes within their products (and occasionally with the products of
another manufacturer) to manage complexity and costs for development,
manufacturing, and assembly.
The concept of platform sharing has evolved over time. Years ago,
manufacturers rebadged vehicles and offered luxury options only on
premium nameplates (and manufacturers shared some vehicle platforms in
limited cases). Today, manufacturers share parts across highly
differentiated vehicles with different body styles, sizes, and
capabilities that may share the same platform. For instance, the Honda
Civic and Honda CR-V share many parts and are built on the same
platform. Engineers design chassis platforms with the ability to vary
wheelbase, ride height, and even driveline configuration. Assembly
lines can produce hatchbacks and sedans to cost-effectively utilize
manufacturing capacity and respond to shifts in market demand. Engines
made on the same line may power small cars or mid-size sport utility
vehicles. In addition, although the agencies' analysis, like past CAFE
analyses, considers vehicles produced for sale in the U.S., the agency
notes these platforms are not constrained to vehicle models built for
sale in the U.S.; many manufacturers have developed, and use, global
platforms, and the total number of platforms is decreasing across the
industry. Several automakers (for example, General Motors and Ford)
either plan to, or already have, reduced their number of platforms to
less than 10 and account for the overwhelming majority of their
production volumes on that small number of platforms.
Vehicle model/configurations derived from the same platform are so
identified in the analysis fleet. Many manufacturers' use of vehicle
platforms is well documented in the public record and widely recognized
among the vehicle engineering community. Engineering knowledge,
information from trade publications, and feedback from manufacturers
and suppliers was also used to assign vehicle platforms in the analysis
fleet.
When the CAFE model is deciding where and how to add technology to
vehicles, if one vehicle on the platform receives new technology, other
vehicles on the platform also receive the technology as part of their
next major redesign or refresh.\381\ Similar to vehicle platforms,
manufacturers create engines that share parts. For instance,
manufacturers may use different piston strokes on a common engine
block, or bore out common engine block castings with different
diameters to create engines with an array of displacements. Head
assemblies for different displacement engines may share many components
and manufacturing processes across the engine family. Manufacturers may
finish crankshafts with the same tools to similar tolerances. Engines
on the same architecture may share pistons, connecting rods, and the
same engine architecture may include both six and eight cylinder
engines. One engine family may appear on many vehicles on a platform,
and changes to that engine may or may not carry through to all the
vehicles. Some engines are shared across a range of different vehicle
platforms. Vehicle model/configurations in the analysis fleet that
share engines belonging to the same platform are also identified as
such.
---------------------------------------------------------------------------
\381\ The CAFE model assigns mass reduction technology at a
platform level, but many other technologies may be assigned and
shared at a vehicle nameplate or vehicle model level.
---------------------------------------------------------------------------
It is important to note that manufacturers define common engines
differently. Some manufacturers consider engines as ``common'' if the
engines shared an architecture, components, or manufacturing processes.
Other manufacturers take a narrower definition, and only assume
``common'' engines if the parts in the engine assembly are the same. In
some cases, manufacturers designate each engine in each application as
a unique powertrain. For example, a manufacturer may have listed two
engines separately for a pair that share designs for the engine block,
the crank shaft, and the head because the accessory drive components,
oil pans, and engine calibrations differ between the two. In practice,
many engines share parts, tooling, and assembly resources, and
manufacturers often coordinate design updates between two similar
engines. Engine families, designated in the analysis using ``engine
codes,'' for each manufacturer were tabulated and assigned based on
data-driven criteria. If engines shared a common cylinder count and
configuration, displacement, valvetrain, and fuel type, those engines
[[Page 24297]]
may have been considered together. In addition, if the compression
ratio, horsepower, and displacement of engines were only slightly
different, those engines were considered the same for the purposes of
redesign and sharing.
Vehicles in the analysis fleet with the same engine family will,
therefore, adopt engine technology in a coordinated fashion.
Specifically, if such vehicles have different design schedules (i.e.,
refresh and redesign schedules), and a subset of vehicles using a given
engine add engine technologies during of a redesign or refresh that
occurs in an early model year (e.g., 2018), other vehicles using the
same engine ``inherit'' these technologies at the soonest ensuing
refresh or redesign. This is consistent with a view that, over time,
most manufacturers are likely to find it more practicable to shift
production to a new version of an engine than to continue production of
both the new engine and a ``legacy'' engine indefinitely. By grouping
engines together, the CAFE model controls future engine families to
ensure reasonable powertrain complexity. This means, however, that for
manufacturers that submitted highly atomized engine and transmission
portfolios, there is a practical cap on powertrain complexity and the
ability of the manufacturer to optimize the displacement of (i.e.,
``right size'') engines perfectly for each vehicle configuration. This
concept is discussed further in Section VI.B.4.a), below.
Like with engines, manufacturers often use transmissions that are
the same or similar on multiple vehicles. Manufacturers may produce
transmissions that have nominally different machining to castings, or
manufacturers may produce transmissions that are internally identical,
except for the final gear ratio. In some cases, manufacturers sub-
contract with suppliers that deliver whole transmissions. In other
cases, manufacturers form joint ventures to develop shared
transmissions, and these transmission platforms may be offered in many
vehicles across manufacturers. Manufacturers use supplier and joint-
venture transmissions to a greater extent than they do with engines. To
reflect this reality, shared transmissions were considered for
manufacturers as appropriate. Transmission configurations are referred
to in the analysis as ``transmission codes.'' Like the inheritance
approach outlined for engines, if one vehicle application of a shared
transmission family upgraded the transmission, other vehicle
applications also upgraded the transmission at the next refresh or
redesign year. To define common transmissions, the agencies considered
transmission type (manual, automatic, dual-clutch, continuously
variable), number of gears, and vehicle architecture (front-wheel-
drive, rear-wheel-drive, all-wheel-drive based on a front-wheel drive
platform, or all-wheel-drive based on a rear-wheel-drive platform). If
vehicles shared these attributes, these transmissions were grouped for
the analysis. Vehicles in the analysis fleet with the same transmission
configuration will adopt transmission technology together, as described
above.
Having all vehicles that share a platform (or engines that are part
of a family) adopt fuel economy-improving/CO2 emissions-
reducing technologies together, subject to refresh/redesign
constraints, reflects the real-world considerations described above,
but also overlooks some decisions manufacturers might make in the real
world in response to market pull. Accordingly, even though the analysis
fleet is incredibly complex, it is also over-simplified in some
respects compared to the real world. For example, the CAFE model does
not currently attempt to simulate the potential for a manufacturer to
shift the application of technologies to improve performance rather
than fuel economy. Therefore, the model's representation of the
``inheritance'' of technology can lead to estimates a manufacturer
might eventually exceed fuel economy standards as technology continues
to propagate across shared platforms and engines. While the agencies
have previously seen examples of extended periods during which some
manufacturers exceeded one or both CAFE and/or CO2
standards, in plenty of other examples, manufacturers chose to
introduce (or even reintroduce) technological complexity into their
vehicle lineups in response to buyer preferences. Going forward, and
recognizing the recent trend for consolidating platforms, it seems
likely manufacturers will be more likely to choose efficiency over
complexity in this regard; therefore, the potential should be lower
that today's analysis turns out to be oversimplified compared to the
real world.
Manufacturers described shared engines, transmissions, and vehicle
platforms as ``standard business practice'' and they were encouraged
that the NHTSA analysis in the Draft TAR, and the jointly issued NPRM
placed realistic limits on the number of unique engines and
transmissions in a powertrain portfolio.\382\ In previous rulemakings,
stakeholders pointed out that shared parts and portfolio complexity
should be considered (but were not), and that the proliferation of
unique technology combinations resulting from unconstrained compliance
pathways would jeopardize economies of scale in the real world.\383\
---------------------------------------------------------------------------
\382\ NHTSA-2018-0067-12150, Toyota North America.
\383\ Alliance of Automobile Manufacturers, EPA-HQ-OAR-0827 and
NHTSA-2016-0068.
---------------------------------------------------------------------------
HD Systems acknowledged that previous rulemakings did not
appropriately consider part sharing, but contended that in today's
global marketplace, manufacturers have flexibility to compete in new
ways that break old part sharing rules.\384\ The agencies acknowledge
that some transmissions are now sourced through suppliers, and that
economies of scale could, in the future be achieved at an industry
level instead of a manufacturer level; however, even when manufacturers
outsource a transmission, recent history suggests they apply that
transmission to multiple vehicles to control assembly plant and service
parts complexity, as they would if they were making the transmission
themselves. Similarly, even for global platforms, or global
powertrains, there is little evidence that manufacturers fragment
powertrain line-ups for a vehicle, or a set of vehicles that have
typically used the same engine. The agencies will continue to consider
how to capture more accurately the ways vehicles share engines,
transmissions, and platforms in future rulemakings, but the part-
sharing and modeling approach presented in the NPRM and this final rule
represents a marked improvement over previous analysis.
---------------------------------------------------------------------------
\384\ NHTSA-2018-0067-11985, HD Systems.
---------------------------------------------------------------------------
(5) Characterizing Production Design Cycles
Another aspect of characterizing vehicles in the analysis fleet is
based on when they can next be refreshed or redesigned. Redesign
schedules play an important role in determining when new technologies
may be applied. Many technologies that improve fuel economy and reduce
CO2 emissions may be difficult to incorporate without a
major product redesign. Therefore, each vehicle model in the analysis
fleet has an associated redesign schedule, and the CAFE model uses that
schedule to implement significant advances in some technologies (like
major mass reduction) to redesign years, while allowing manufacturers
to include minor advances (such as improved tire rolling
[[Page 24298]]
resistance) during a vehicle ``refresh,'' or a smaller update made to a
vehicle, which can happen between redesigns. In addition to refresh and
redesign schedules associated with vehicle model/configurations,
vehicles that share a platform subsequently have platform-wide refresh
and redesign schedules for mass reduction technologies.
To develop the refresh/redesign cycles used for the NPRM vehicles
in the analysis fleet, information from commercially available sources
was used to project redesign cycles through MY 2022, as was done for
NHTSA's analysis for the 2016 Draft TAR.\385\ Commercially available
sources' estimates through MY 2022 are generally supported by detailed
consideration of public announcements plus related intelligence from
suppliers and other sources, and recognize that uncertainty increases
considerably as the forecasting horizon is extended. For MYs 2023-2035,
in recognition of that uncertainty, redesign schedules were extended
considering past pacing for each product, estimated schedules through
MY 2022, and schedules for other products in the same technology
classes. As mentioned above, potentially confidential forward-looking
information was not requested from manufacturers; nevertheless, all
manufacturers had an opportunity to review the estimates of product-
specific redesign schedules. A few manufacturers provided related
forecasts and, for the most part, that information corroborated the
estimates.
---------------------------------------------------------------------------
\385\ In some cases, data from commercially available sources
was found to be incomplete or inconsistent with other available
information. For instance, commercially available sources identified
some newly imported vehicles as new platforms, but the international
platform was midway through the product lifecycle. While new to the
U.S. market, treating these vehicles as new entrants would have
resulted in artificially short redesign cycles if carried forward,
in some cases. Similarly, commercially available sources labeled
some product refreshes as redesigns, and vice versa. In these
limited cases, the data was revised to be consistent with other
available information or typical redesign and refresh schedules for
CAFE modeling. In these limited cases, the forecast time between
redesigns and refreshes was updated to match the observed past
product timing.
---------------------------------------------------------------------------
Some commenters suggested supplanting these estimated redesign
schedules with estimates applying faster cycles (e.g., four to five
years), and this approach was considered for the analysis. Some
manufacturers tend to operate with faster redesign cycles and may
continue to do so, and manufacturers tend to redesign some products
more frequently than others. However, especially considering that
information presented by manufacturers largely supports estimates
discussed above, applying a ``one size fits all'' acceleration of
redesign cycles would not improve the analysis; instead, assuming a
fixed, shortened redesign schedule across the industry would likely
reduce consistency with the real world, especially for light trucks,
which are redesigned, on average, no less than every six years (see
Table VI-9, below). Moreover, if some manufacturers accelerate
redesigns in response to new standards, doing so would likely involve
costs (greater levels of stranded capital, reduced opportunity to
benefit from ``learning''-related cost reductions) greater than
reflected in other inputs to the analysis.
As discussed in the NPRM, manufacturers use diverse strategies with
respect to when, and how often they update vehicle designs. While most
vehicles have been redesigned sometime in the last five years, many
vehicles have not. In particular, vehicles with lower annual sales
volumes tend to be redesigned less frequently, perhaps giving
manufacturers more time to recoup the investment needed to bring the
product to market. In some cases, manufacturers continue to produce and
sell vehicles designed more than a decade ago.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.093
[[Page 24299]]
Each manufacturer may use different strategies throughout their
product portfolio, and a component of each strategy may include the
timing of refresh and redesign cycles. Table VI-10 summarizes the
average time between redesigns, by manufacturer, by vehicle technology
class. Dashes mean the manufacturer has no volume in that vehicle
technology class in the MY 2017 analysis fleet. Across the industry,
manufacturers average 6.6 years between product redesigns.
[GRAPHIC] [TIFF OMITTED] TR30AP20.094
Trends on redesign schedules identified in the NPRM remain in place
for today's analysis. Pick-up trucks have much longer redesign
schedules than small cars. Some manufacturers redesign vehicles often,
while other manufacturers redesign vehicles less often. Even if two
manufacturers have similar redesign cadence, the model years in which
the redesigns occur may still be different and dependent on where each
of the manufacturer's products are in their life cycle.
Table VI-11 summarizes the average age of manufacturers' offering
by vehicle technology class. A value of ``0.0'' means that every
vehicle for a manufacturer in the vehicle technology class, represented
by the MY 2017 analysis fleet was new in MY 2017. Across the industry
manufacturers redesigned MY 2017 vehicles an average of 3.5 years
earlier, meaning the average MY 2017 vehicle was last redesigned in
approximately MY 2013, also on average near a midpoint in their product
lifecycle.
[[Page 24300]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.095
BILLING CODE 4910-59-C
Some commenters cited examples of vehicles in the NPRM analysis
fleet where the redesign years were off by a year here or there in the
2017-2022 timeframe relative to the most recent public announcements,
or that the extended forecasts were too rigid.\386\ The CAFE model
structurally requires an input for the redesign years, and the agencies
worked to make these generally representative without disclosing
precise CBI product plans. Many of the redesign schedules were carried
over from the NPRM, with a few minor updates.
---------------------------------------------------------------------------
\386\ NHTSA-2018-0067-11723, Natural Resources Defense Council.
---------------------------------------------------------------------------
Some commenters contended that the agencies should not look at the
historical data to project the timing between redesigns (``business as
usual''), but should instead adopt a ``policy case'' with an
accelerated pace of redesigns and refreshes.\387\ Some commenters
suggested that the agencies use a standard 5 or 6 year redesign
schedule for all manufacturers and all products as a way to lower
projected costs.\388\ Other stakeholders commented that the entire
industry should be modeled with the ability to redesign everything at
one time in the near term because that would not presuppose precisely
how manufacturers may adjust their fleet.\389\
---------------------------------------------------------------------------
\387\ NHTSA-2018-0067-11723, Natural Resources Defense Council.
\388\ NHTSA-2018-0067-11985, HD Systems.
\389\ NHTSA-2018-0067-12039, Union of Concerned Scientists.
---------------------------------------------------------------------------
If the agencies were to implement any such approaches, the agencies
would need to more precisely account for tooling costs, research and
development costs, and product lifecycle marketing costs, or risk
missing ``hidden costs'' of a shortened cadence. To account properly
for these, the CAFE model would require major changes, and would
require specific inputs that are currently covered generically under
the retail price equivalency (RPE) factor.\390\ The agencies considered
these comments, and decided the process for refresh and redesign
outlined in the NPRM was a reasonable and realistic approach to
characterize product changes. The agencies conducted sensitivity
analysis with compressed redesign and refresh schedules, though these
ignore the resulting compressed amortization schedules, missing
important costs that are incorporated in the current RPE assumptions.
---------------------------------------------------------------------------
\390\ Shorter redesign schedules are likely to put upward
pressure on RPE, as the manufacturers would have less time to recoup
investments.
---------------------------------------------------------------------------
Some commenters claimed that the agency had extraordinarily
extended redesign schedule of 17.7 years for FCA between 2021-2025, and
an average redesign time of 25.8 years for Ford between 2022-2025.\391\
The agencies found these claims inaccurate and without basis. Table VI-
10, ``Summary of Sales Weighted Average Time
[[Page 24301]]
between Engineering Redesigns, by Manufacturer, by Vehicle Technology
Class'' summarizes the data used in today's analysis (which is very
similar to the information used in the NPRM, with some minor
adjustments and updates to the fleet), and the detailed information
vehicle-by-vehicle is reported in the ``market data'' file. The
agencies recognize that the natural sequence of redesigns for some
manufacturers and some products is not ideal to meet stringent
alternatives, which is part of the consideration for economic
practicability and technological feasibility. Manufacturers commented
supportively on the idea of vehicle specific redesign schedules, and
the redesign cadence used in the NPRM, as these contribute to realistic
assessments of new technology penetration within the fleet, and
acknowledge the heterogeneity in the product development approaches and
business practices for each manufacturer.\392\ One commenter recognized
that redesign and refresh schedules represented a vast improvement over
phase-in caps to model the adoption of mature technologies.\393\
---------------------------------------------------------------------------
\391\ NHTSA-2018-0067-11723, Natural Resources Defense Council.
\392\ NHTSA-2018-0067-11928, Ford Motor Company.
\393\ NHTSA-2018-0067-0444, Walter Kreucher.
---------------------------------------------------------------------------
Other commenters argued that the structural construct of
technologies only being available at redesign or at refresh (via
inheritance) did not reflect real world actions and was not supported
by any actual data.\394\ Other commenters acknowledged the inheritance
of engine and transmission technologies at refresh as an important,
positive feature of the CAFE model.\395\ HD Systems argued that an
engine or transmission package available in other markets on a global
platform could be imported to the U.S. market during refresh, and did
not require a ``leader'' at redesign in the U.S. market to seed
adoption. HDS cited a few examples where manufacturers have introduced
strong hybrid powertrains on an existing vehicle a year or two after
the product launch, not associated with any particular vehicle redesign
or refresh.
---------------------------------------------------------------------------
\394\ NHTSA-2018-0067-11985, HD Systems.
\395\ NHTSA-2018-0067-11723, Natural Resources Defense Council.
---------------------------------------------------------------------------
The agencies carefully considered these comments, and observed that
some relatively low volume hybrid options may appear after launch, or
that some transmissions were quickly replaced shortly after a major
redesign. In many of these cases, launch delays, warranty claims, or
other external factors contributed to, at least in part, an atypically
timed introduction of fuel saving technology to the fleet.\396\ At this
point, this does not appear to be a mainstream, or preferred industry
practice. However, the agencies will continue to evaluate this. For
future rulemaking, the agencies may consider engine refresh and
redesign cycles for engines and transmissions. These may be separate
from vehicle redesign and refresh schedules because the powertrain
product lifecycles may be longer on average than the typical vehicle
redesign schedules. This approach, if researched and implemented in
future analysis, could provide some opportunity for manufacturers to
introduce new powertrain technologies independent of the vehicle
redesign schedules, in addition to inheriting advanced powertrain
technology as refresh as already modeled in the NPRM and today's
analysis.
---------------------------------------------------------------------------
\396\ Such instances are observable in detailed CAFE and
CO2 compliance data submitted to EPA and NHTSA.
---------------------------------------------------------------------------
For today's analysis, the agencies, with a few exceptions based on
updated publicly available information, carried over redesign cadences
for each vehicle nameplate as presented in the NPRM. The agencies do
not claim that the projected redesign years will perfectly match what
industry does--notably because refresh and redesign information is CBI
and the agencies have applied more generalized schedules to protect the
CBI. Also, what any individual manufacturer may choose to do today
could be completely different than what it chooses to do tomorrow due
to changing business circumstances and plans--but the agencies have
worked to ensure the timing of redesigns will be roughly correct
(especially in the near term), and that the time between redesigns will
continue forward for each manufacturer as it has based on recent
history. The agencies have also increased the frequency of refreshes in
response to comments about the proliferation of some engine and
transmission families through manufacturers' product portfolios.
Also for today's analysis, the agencies now explicitly model CAFE
compliance pathways out through 2050. For the model to work as
intended, the agencies must project refresh and redesign schedules out
through 2050. The agencies recognize that the accuracy of predictions
about the distant future, particularly about refresh and redesign
cycles through the 2030-2050 timeframe, are likely to be poor. If
historical evolution of the industry continues, many of the nameplates
carried forward in the fleet are likely to be out of production, and
new nameplates not considered in the analysis are sure to emerge.
Still, carrying forward the MY 2017 fleet with the current refresh and
redesign cadences is consistent with the current analysis, and imposing
an alternative schedule on the fleet, or making up new nameplates and
retiring older nameplates without a clear basis, would lack proper
foundation.
BILLING CODE 4910-59-P
[[Page 24302]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.096
BILLING CODE 4910-59-C
(6) Defining Technology Adoption Features
In some circumstances, the agencies may reference full vehicle
simulation effectiveness data for technology combinations that are not
able to be, or are not likely to be applied to all vehicles. In some
cases, a specific technology as modeled only exists on paper, and
questions remain about the technological feasibility of the efficiency
characterization.\397\ Or, a technology may perform admirably on the
test cycle, but fail to meet all functional, or performance
requirements for certain vehicles.\398\ In other cases, the
intellectual property landscape may make commercialization of one
technology risky for a manufacturer without the consent of the
intellectual property owner.\399\ In such cases, the agencies may not
allow a technology to be applied to a certain vehicle. The agencies
designate this in the ``market data'' file with a ``SKIP'' for the
technology and vehicle. The logic is explained technology by technology
in this document, as the logic was explained in the PRIA for this rule.
---------------------------------------------------------------------------
\397\ High levels of aerodynamic drag reduction for some body
styles, or EPA's previous, speculative characterization of ``HCR2''
engines, for example.
\398\ Examples of applications that are unsuitable for certain
technologies include low end torque requirements for HCR engines on
high load vehicles, or towing and trailering applications,
continuously variable transmissions in high torque applications, and
low rolling resistance tires on vehicles built for precision
cornering and high lateral forces, or instant acceleration from a
stand still.
\399\ Variable compression ratio engines, for example.
---------------------------------------------------------------------------
Some commenters argued that the restrictions of technologies on a
case-by-case basis required case-by-case explanation (and not objective
specification defined cut-offs), and that the use of CBI for
performance considerations was unacceptable unless fully
disclosed.\400\ As discussed above, the agencies are not able to
disclose CBI. Stakeholders have had plenty of opportunities to comment
on the applicability of technologies, including the few that have used
SKIP logic restrictions for a portion of the fleet.
---------------------------------------------------------------------------
\400\ NHTSA-2018-0067-11741, ICCT.
---------------------------------------------------------------------------
Other commenters suggested an optimistic and wholly unfounded
approach to manufacturer innovation, arguing that costs would continue
to come down (beyond what is currently modeled with cost learning), and
the list of fuel-saving technologies would continually regenerate
itself (even if the technological mechanism for fuel saving
technologies was not yet identified).\401\ Therefore, the argument goes
that people will figure out new ways to improve fuel saving
technologies to increase their applicability, and the current
technology characterization should be enabled for selection with no
restriction--not because the commenter knows how the technology will be
adapted, but that the commenter believes the technology could,
eventually, within the timeline of the rulemaking, be adapted, brought
to market, and be accepted by consumers. While the agencies recognize
the improvements that many manufacturers
[[Page 24303]]
have achieved in fuel saving technologies, some of which were difficult
to foresee, the agencies have an obligation under the law to be
judicious and specific about technological feasibility, and to avoid
speculative conclusions about technologies to justify the rulemaking.
---------------------------------------------------------------------------
\401\ NHTSA-208-0067-12122-33, American Council for an Energy-
Efficient Economy.
---------------------------------------------------------------------------
c) Other Analysis Fleet Data
(1) Safety Classes
The agencies referenced the mass-size-safety analysis to project
the effects changes in weight may have on crash fatalities. That
analysis, discussed in more detail in Section VI.D.2, considers how
weight changes may affect safety for cars, crossover utility vehicles
and sport utility vehicles, and pick-up trucks. To consider these
effects, the agencies mapped each vehicle in the analysis fleet to the
appropriate ``Safety Class.''
(2) Labor Utilization
The analysis fleet summarizes components of direct labor for each
vehicle considered in the analysis. The labor is split into three
components: (1) Dealership hours worked on sales functions per vehicle,
(2) direct assembly labor for final assembly, engine, and transmission,
and (3) percent U.S. content.
In the MY 2016 fleet for the NPRM, the agencies catalogued
production locations and plant employment, reviewed annual reports from
the North American Dealership Association to estimate dealership
employment (27.8 hours per vehicle sold), and estimated the industry
average labor hours for final assembly of vehicles (30 hours per
vehicle produced), engine machining and assembly (4 hours per engine
produced), and transmission production (5 hours per transmission
produced).
Today's analysis fleet carries over the estimated labor
coefficients for sales and production, but references the most recent
Part 583 American Automobile Labeling Act Report for percent U.S.
content and for the location of vehicle assembly, engine assembly, and
transmission assembly.\402\
---------------------------------------------------------------------------
\402\ Part 583 American Automobile Labeling Act Report,
available at https://www.nhtsa.gov/part-583-american-automobile-labeling-act-reports.
---------------------------------------------------------------------------
(3) Production Volumes for Sales Analysis
A final important aspect of projecting what vehicles will exist in
future model years and potential manufacturer responses to standards is
estimating how future sales might change in response to different
potential standards. If potential future standards appear likely to
have major effects in terms of shifting production from cars to trucks
(or vice versa), or in terms of shifting sales between manufacturers or
groups of manufacturers, that is important for the agencies to
consider. For previous analyses, the CAFE model used a static forecast
contained in the analysis fleet input file, which specified changes in
production volumes over time for each vehicle model/configuration. This
approach yielded results that, in terms of production volumes, did not
change between scenarios or with changes in important model inputs. For
example, very stringent standards with very high technology costs would
result in the same estimated production volumes as less stringent
standards with very low technology costs. For this analysis, as in the
proposal, the CAFE model begins with the first-year production volumes
(i.e., MY 2017 for today's analysis) and adjusts ensuing sales mix year
by year (between cars and trucks, and between manufacturers)
endogenously as part of the analysis, rather than using external
forecasts of future car/truck split and future manufacturer sales
volumes. This leads the model to produce different estimates of future
production volumes under different standards and in response to
different inputs, reflecting the expectation that regulatory standards
and other external factors will, in fact, impact the market.
(4) Comments on Other Analysis Fleet Data
Some commenters suggest that the CAFE model should run as a full
consumer choice model (and this idea is discussed in more detail in
Section VI.D.1). While this sounds like a reasonable request on the
surface, such an approach would place enormous new demands on the data
characterized in the fleet (and preceding fleets, which may be needed
to calibrate a model properly). For instance, some model concepts may
depend on a bevy of product features, such as interior cargo room,
artistic appeal of the design, and perceived quality of the vehicle.
But product features alone may not be sufficient. Additional
information about dealership channels, product awareness and
advertising effectiveness, and financing terms also may be required.
Such information could dramatically increase the scope of work needed
to characterize the analysis fleet for future rulemakings. As described
in Section VI.D.1.b)(2)(d) Using Vehicle Choice Models in Rulemaking
Analysis. Accordingly, the agencies decided not to develop such a model
for this rulemaking.
2. Treatment of Compliance Credit Provisions
Today's final rule involves a variety of provisions regarding
``credits'' and other compliance flexibilities. Some recently
introduced regulatory provisions allow a manufacturer to earn
``credits'' that will be counted toward a vehicle's rated
CO2 emissions level, or toward a fleet's rated average
CO2 or CAFE level, without reference to required levels for
these average levels of performance. Such flexibilities effectively
modify emissions and fuel economy test procedures, or methods for
calculating fleets' CAFE and average CO2 levels. Such
provisions are discussed below in Section VI.B.2. Other provisions (for
CAFE, statutory provisions) allow manufacturers to earn credits by
achieving CAFE or average CO2 levels beyond required levels;
these provisions may hence more appropriately be termed ``compliance
credits.''
EPCA has long provided that, by exceeding the CAFE standard
applicable to a given fleet in a given model year, a manufacturer may
earn corresponding ``credits'' that the same manufacturer may, within
the same regulatory class, apply toward compliance in a different model
year. EISA amended these provisions by providing that manufacturers
may, subject to specific statutory limitations, transfer compliance
credits between regulatory classes, and trade compliance credits with
other manufacturers. The CAA provides EPA with broad standard-setting
authority for the CO2 program, with no specific directives
regarding either CO2 standards or CO2 compliance
credits.
EPCA also specifies that NHTSA may not consider the availability of
CAFE credits (for transfer, trade, or direct application) toward
compliance with new standards when establishing the standards
themselves.\403\ Therefore, this analysis, like that presented in the
NPRM, considers 2020 to be the last model year in which carried-forward
or transferred credits can be applied for the CAFE program. Beginning
in model year 2021, today's ``standard setting'' analysis for NHTSA's
program is conducted assuming each fleet must comply with the CAFE
standard separately in every model year.
---------------------------------------------------------------------------
\403\ 49 U.S.C. 32902(h)(3).
---------------------------------------------------------------------------
The ``unconstrained'' perspective acknowledges that these
flexibilities exist as part of the program, and, while not considered
by NHTSA in setting standards, are nevertheless important to consider
when attempting to estimate the real impact of any alternative. Under
[[Page 24304]]
the ``unconstrained'' perspective, credits may be earned, transferred,
and applied to deficits in the CAFE program throughout the full range
of model years in the analysis. The Final Environmental Impact Analysis
(FEIS) accompanying today's final rule, like the corresponding Draft
EIS analysis, presents results of ``unconstrained'' modeling. Also,
because the CAA provides no direction regarding consideration of any
CO2 credit provisions, today's analysis, like the NPRM
analysis, includes simulation of carried-forward and transferred
CO2 credits in all model years.
Some commenters took issue broadly with this treatment of
compliance credits. Michalek and Whitefoot wrote that ``we find this
requirement problematic because the automakers use these flexibilities
as a common means of complying with the regulation, and ignoring them
will bias the cost-benefit analysis to overestimate costs.'' \404\
---------------------------------------------------------------------------
\404\ Michalek, J. and Whitefoot, K., NHTSA-2018-0067-11903, at
10-11.
---------------------------------------------------------------------------
Counter to the above general claim, the CAFE model does provide
means to simulate manufacturers' potential application of some
compliance credits, and both the analysis of CO2 standards
and the NEPA analysis of CAFE standards do make use of this aspect of
the model. As discussed above, NHTSA does not have the discretion to
consider the credit program--in fact, the agency is prohibited by
statute from doing so--in establishing maximum feasible standards.
Further, as discussed below, the agencies also continue to find it
appropriate for the analysis largely to refrain from simulating two of
the mechanisms allowing the use of compliance credits.
The model's approach to simulating compliance decisions accounts
for the potential to earn and use CAFE credits as provided by EPCA/
EISA. The model similarly accumulates and applies CO2
credits when simulating compliance with EPA's standards. Like past
versions, the current CAFE model can be used to simulate credit carry-
forward (a.k.a. banking) between model years and transfers between the
passenger car and light truck fleets but not credit carry-back (a.k.a.
borrowing) from future model years or trading between manufacturers.
Regarding the potential to carry back compliance credits, UCS
commented that, although past versions of the CAFE model had
``considered this flexibility in its approach to multiyear modeling,''
NHTSA had, without explanation, ``abruptly discontinued support of this
method of compliance,'' such that ``manufacturers are generally
incentivized to over comply, regardless of whether carrying forward a
deficit to be compensated by later overcompliance would be a more cost-
effective method of compliance.'' \405\ Citing the potential that
manufacturers could make use of carried back credits in the future, UCS
also stated that ``NHTSA's decision to constrain it in the model is
unreasonable and arbitrary.'' \406\ UCS effectively implies that the
agencies should base standards on analysis that presumes manufacturers
will take full theoretical advantage of provisions allowing credits to
be borrowed.
---------------------------------------------------------------------------
\405\ UCS, NHTSA-2018-0067-12039, Technical Appendix, at 44.
\406\ UCS, op. cit., at 77.
---------------------------------------------------------------------------
The agencies have carefully considered these comments, and while
EPA's decisions regarding CO2 standards can consider the
potential to carry back compliance credits from later to earlier model
years, and NHTSA's ``unconstrained'' evaluation could also do so, past
examples of failed attempts to carry back CAFE credits (e.g., a MY2014
carry back default leading to a civil penalty payment) underscore the
riskiness of such ``borrowing.'' Recent evidence indicates
manufacturers are disinclined to take such risks,\407\ and both
agencies find it reasonable and prudent to refrain from attempting to
simulate such ``borrowing'' in rulemaking analysis.
---------------------------------------------------------------------------
\407\ Section IX, below, reviews data regarding manufacturers'
use of CAFE compliance credit mechanism during MYs 2011-2016, and
shows that the use of ``carry back'' credits is, relative to the use
of other compliance credit mechanisms, too small to discern.
---------------------------------------------------------------------------
Unlike past versions, the NPRM and current versions of CAFE model
provide a basis to specify (in model inputs) CAFE credits available
from model years earlier than those being explicitly simulated. For
example, with this analysis representing model years 2017-2050
explicitly, credits earned in model year 2012 are made available for
use through model year 2017 (given the current five-year limit on
carry-forward of credits). The banked credits are specific to both the
model year and fleet in which they were earned.
In addition to the above-mentioned comments, UCS also cited as
``errors'' that ``the model does not accurately reflect the one-time
exemption from the EPA 5-year credit life for credits earned in the MY
2010-2015 timeframe'' and ``NHTSA assumes that there will be absolutely
no credit trading between manufacturers.''
As discussed below, in the course of updating the analysis fleet
from MY 2016 to MY 2017, the agencies have updated and expanded the
manner in which the model accounts for credits earned prior to MY 2017,
including credits earned as early as MY 2009. In order to increase the
realism with which the model transitions between the early model year
(MYs 2017-2020) and the later years that are the subject of this
action, the agencies have accounted for the potential that some
manufacturers might trade some of these pre-MY 2017 credits to other
manufacturers. However, as with the NPRM, the analysis refrains from
simulating the potential that manufacturers might continue to trade
credits during and beyond the model years covered by today's action.
The agencies remain concerned that any realistic simulation of such
trading would require assumptions regarding which specific pairs of
manufacturers might actually trade compliance credits, and the evidence
to date makes it clear that the credit market is far from fully
``open.'' With respect to the FCA comment cited above, the agencies
also remain concerned that to set standards based on an analysis that
presumes the use of program flexibilities risks making the
corresponding actions mandatory. Some flexibilities--credit carry-
forward (banking) and transfers between fleets in particular--involve
little risk, because they are internal to a manufacturer and known in
advance. As discussed above, credit carry-back involves significant
risk, because it amounts to borrowing against future improvements,
standards, and production volume and mix--and anticipated market demand
for fuel efficient vehicles often fail to materialize. Similarly,
credit trading also involves significant risk, because the ability of
manufacturer A to acquire credits from manufacturer B depends not just
on manufacturer B actually earning the expected amount of credit, but
also on manufacturer B being willing to trade with manufacturer A, and
on potential interest by other manufacturers. Manufacturers' compliance
plans have already evidenced cases of compliance credit trades that
were planned and subsequently aborted, reinforcing the agencies'
judgment that, like credit banking, credit trading involves too much
risk to be included in an analysis that informs decisions about the
stringency of future standards. Nevertheless, recognizing that some
manufacturers have actually been trading credits, the agencies have, as
in the NPRM, included in the sensitivity analysis a case that simulates
``perfect'' trading of compliance credits, focusing
[[Page 24305]]
on CO2 standards to illustrate the hypothetical maximum
potential impact of trading. The FRIA summarizes results of this and
other cases included in the sensitivity analysis.
As discussed in the CAFE model documentation, the model's default
logic attempts to maximize credit carry-forward--that is, to ``hold
on'' to credits for as long as possible. If a manufacturer needs to
cover a shortfall that occurs when insufficient opportunities exist to
add technology in order to achieve compliance with a standard, the
model will apply credits. Otherwise the manufacturer carries forward
credits until they are about to expire, at which point it will use them
before adding technology that is not considered cost-effective. The
model attempts to use credits that will expire within the next three
years as a means to smooth out technology application over time to
avoid both compliance shortfalls and high levels of over-compliance
that can result in a surplus of credits. Although it remains impossible
precisely to predict manufacturer's actual earning and use of
compliance credits, and this aspect of the model may benefit from
future refinement as manufacturers and regulators continue to gain
experience with these provisions, this approach is generally consistent
with manufacturers' observed practices.
NHTSA introduced the CAFE Public Information Center to provide
public access to a range of information regarding the CAFE
program,\408\ including manufacturers' credit balances. However, there
is a data lag in the information presented on the CAFE PIC that may not
capture credit actions across the industry for as much as several
months. Furthermore, CAFE credits that are traded between manufacturers
are adjusted to preserve the gallons saved that each credit
represents.\409\ The adjustment occurs at the time of application
rather than at the time the credits are traded. This means that a
manufacturer who has acquired credits through trade, but has not yet
applied them, may show a credit balance that is either considerably
higher or lower than the real value of the credits when they are
applied. For example, a manufacturer that buys 40 million credits from
Tesla may show a credit balance in excess of 40 million. However, when
those credits are applied, they may be worth only 1/10 as much--making
that manufacturer's true credit balance closer to 4 million than 40
million.
---------------------------------------------------------------------------
\408\ CAFE Public Information Center, http://www.nhtsa.gov/CAFE_PIC/CAFE_PIC_Home.htm (last visited June 22, 2018).
\409\ CO2 credits for EPA's program are denominated
in metric tons of CO2 rather than gram/mile compliance
credits and require no adjustment when traded between manufacturers
or fleets.
---------------------------------------------------------------------------
For the NPRM, the agencies reviewed then-recent credit balances,
estimated the potential that some manufacturers could trade credits,
and developed inputs that make carried-forward credits available in
each of model years 2011-2015, after subtracting credits assumed to be
traded to other manufacturers, adding credits assumed to be acquired
from other manufacturers through such trades, and adjusting any traded
credits (up or down) to reflect their true value for the fleet and
model year into which they were traded.\410\ For today's analysis, an
additional model year's data was available in mid-2019, and the
agencies updated these inputs, as summarized in Table VI-12, Table VI-
13, and Table VI-14. While the CAFE model will transfer expiring
credits into another fleet (e.g., moving expiring credits from the
domestic car credit bank into the light truck fleet), some of these
credits were moved into the initial banks to improve the efficiency of
application and both to reflect better the projected shortfalls of each
manufacturer's regulated fleets and to represent observed behavior. For
context, a manufacturer that produces one million vehicles in a given
fleet, and experiences a shortfall of 2 mpg, would need 20 million
credits, adjusted for fuel savings, to offset the shortfall completely.
---------------------------------------------------------------------------
\410\ The adjustments, which are based upon the CAFE standard
and model year of both the party originally earning the credits and
the party applying them, were implemented assuming the credits would
be applied to the model year in which they were set to expire. For
example, credits traded into a domestic passenger car fleet for MY
2014 were adjusted assuming they would be applied in the domestic
passenger car fleet for MY 2019.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24306]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.097
[GRAPHIC] [TIFF OMITTED] TR30AP20.098
[[Page 24307]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.099
BILLING CODE 4910-59-C
In addition to the inclusion of these existing credit banks, the
CAFE model also updated its treatment of credits in the rulemaking
analysis. EPCA requires that NHTSA set CAFE standards at maximum
feasible levels for each model year without consideration of the
program's credit mechanisms. However, as recent NHTSA CAFE/EPA tailpipe
CO2 emissions rulemakings have evaluated effects of
standards over longer time periods, the early actions taken by
manufacturers required more nuanced representation. Accordingly, the
CAFE model now provides for a setting to establish a ``last year to
consider credits.'' This adjustment is set at the last year for which
new standards are not being considered (MY 2020 in this analysis). This
allows the model to replicate the practical application of existing
credits toward compliance in early years but also to examine the impact
of proposed standards based solely on fuel economy improvements in all
years for which new standards are being considered.
Regarding the model's simulation of manufacturers' potential
earning and application of compliance credits, UCS commented that the
model ``inexplicably lets credits expire'' because ``all technologies
which pay for themselves within the assumed payback period are applied
to all manufacturers, regardless of credit status.'' UCS also claimed
that ``NHTSA did not accurately reflect unique attributes of EPA's
credit bank,'' that ``credits are not traded between manufacturers,''
and that ``NHTSA does not model credit carryback for compliance.''
\411\ Relatedly, as discussed above, UCS attributes modeling outcomes
to the ``effective cost'' metric used to select from among available
fuel-saving technologies.\412\ As discussed in Section VI.B.1, the
agencies expect that manufacturers are likely to improve fuel economy
voluntarily insofar as doing so ``pays back'' economically within a
short period (30 months), and the agencies note that periods of
regulatory stability have, in fact, been marked by CAFE levels
exceeding requirements. As discussed above, the agencies have excluded
simulation of credit trading (except in MYs prior to those under
consideration, aside from an idealized case presented in the
sensitivity analysis) and likewise excluded simulation of potential
``carryback'' provisions. The agencies have excluded modeling these
scenarios not just because of the analytical complexities involved (and
rejecting, for example, the random number generator analysis suggested
by UCS), but also because the agencies agree that the actual provisions
regarding trading and borrowing of compliance credits create too much
risk to be used in the analysis underlying consideration of standards.
However, as discussed above, the agencies have revised the ``metric''
used to prioritize available options to apply fuel-saving technologies.
As discussed below, the agencies have revised model inputs to include
the large quantity of ``legacy'' compliance credits EPA has made
available under its CO2 standards.
---------------------------------------------------------------------------
\411\ UCS, NHTSA-2018-0067-12039, Technical Appendix, at 35-46.
\412\ UCS, NHTSA-2018-0067-12039, Technical Appendix, at 28-30.
---------------------------------------------------------------------------
The CAFE model has also been modified to include a similar
representation of existing credit banks in EPA's CO2
program. While the life of a CO2 credit, denominated in
metric tons of CO2, has a five-year life, matching the
lifespan of CAFE credits, such credits earned in the early MY 2009-2011
years of the EPA program, may be used through MY 2021.\413\ The CAFE
model was not modified to allow
[[Page 24308]]
exceptions to the life-span of compliance credits, and, to reflect
statutory requirements, treated them as if they may be carried forward
for no more than five years, so the initial credit banks were modified
to anticipate the years in which those credits might be needed. MY 2016
was simulated explicitly in the NPRM analysis to prohibit the inclusion
of banked credits in MY 2016 (which could be carried forward from MY
2016 to MY 2021), and thus underestimated the extent to which
individual manufacturers, and the industry as a whole, could rely on
these early credits to comply with EPA standards between MY 2016 and MY
2021. However, as indicated in the NPRM, the final rule's model inputs
updated the analysis fleet's basis to MY 2017, such that these
additional banked credits can be included. The credit banks with which
the simulations in this analysis were conducted are presented in the
following Tables:
---------------------------------------------------------------------------
\413\ In the 2010 rule, EPA placed limits on credits earned in
MY 2009, which expired prior to this rule. However, credits
generated in MYs 2010-2011 may be carried forward, or traded, and
applied to deficits generated through MY 2021.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.100
[[Page 24309]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.101
BILLING CODE 4910-59-C
While the CAFE model does not simulate the ability to trade credits
between manufacturers, it does simulate the strategic accumulation and
application of compliance credits, as well as the ability to transfer
credits between fleets to improve the compliance position of a less
efficient fleet by leveraging credits earned by a more efficient fleet.
The model prefers to hold on to earned compliance credits within a
given fleet, carrying them forward into the future to offset potential
future deficits. This assumption is consistent with observed strategic
manufacturer behavior dating back to 2009.
From 2009 to present, no manufacturer has transferred CAFE credits
into a fleet to offset a deficit in the same year in which they were
earned. This has occurred with credits acquired from other
manufacturers via trade but not with a manufacturer's own credits.
Therefore, the current representation of credit transfers between
fleets--where the model prefers to transfer expiring, or soon-to-be-
expiring credits rather than newly earned credits--is both appropriate
and consistent with observed industry behavior.
This may not be the case for CO2 standards, though it is
difficult to be certain at this point. The CO2 program
seeded the industry with a large quantity of early compliance credits
(earned in MYs 2009-2011) \414\ prior to the existence formal
CO2 standards. Early credits from MYs 2010 and 2011,
however, do not expire until 2021. Thus, for manufacturers looking to
offset deficits, it is more sensible to exhaust credits that were
generated during later model years (which are set to expire within the
next five years), rather than relying on the initial bank of credits
from MYs 2010 and 2011. The first model year for which earned credits
outlive the initial bank is MY 2017, for which final manufacturer
CO2 performance data (and hence, banked credits) has not yet
been released. However, considering that under the CO2
program manufacturers simultaneously comply with passenger car and
light truck fleets, to more accurately represent the CO2
credit system the CAFE model allows (and encourages) intra-year
transfers between regulated fleets for the purpose of simulating
compliance with the CO2 standards.
---------------------------------------------------------------------------
\414\ In response to public comment, EPA eliminated the possible
use of credits earned in MY 2009 for future model years. However,
credits earned in MY 2010 and MY 2011 remain available for use.
---------------------------------------------------------------------------
a) Off-Cycle and A/C Efficiency Adjustments to CAFE and Average
CO2 Levels
In addition to more rigorous accounting of CAFE and CO2
credits, the model now also accounts for air conditioning efficiency
and off-cycle adjustments. NHTSA's program considers those adjustments
in a manufacturer's compliance calculation starting in MY 2017, and the
NPRM version of the model used the adjustments claimed by each
manufacturer in MY 2016 as the starting point for all future years.
Because air conditioning efficiency and off-cycle adjustments are not
credits in NHTSA's program, but rather adjustments to compliance fuel
economy (much like the Flexible Fuel Vehicle adjustments due to phase
out in MY 2019), they may be included under either a ``standard
setting'' or ``unconstrained'' analysis perspective.
The manner in which the CAFE model treats the EPA and CAFE A/C
efficiency and off-cycle credit programs is similar, but the model also
accounts for A/C leakage (which is not part of NHTSA's program). When
determining the compliance status of a
[[Page 24310]]
manufacturer's fleet (in the case of EPA's program, PC and LT are the
only fleet distinctions), the CAFE model weighs future compliance
actions against the presence of existing (and expiring) CO2
credits resulting from over-compliance with earlier years' standards,
A/C efficiency credits, A/C leakage credits, and off-cycle credits.
Another aspect of credit accounting, implemented in the NPRM
version of the CAFE model, involved credits related to the application
of off-cycle and A/C efficiency adjustments, which manufacturers earn
by taking actions such as special window glazing or using reflective
paints that provide fuel economy improvements in real-world operation
but do not produce measurable improvements in fuel consumption on the
2-cycle test.
NHTSA's inclusion of off-cycle and A/C efficiency adjustments began
in MY 2017, while EPA has collected several years' worth of submissions
from manufacturers about off-cycle and A/C efficiency technology
deployment. Currently, the level of deployment can vary considerably by
manufacturer, with several claiming extensive Fuel Consumption
Improvement Values (FCIV) for off-cycle and A/C efficiency
technologies, and others almost none. The analysis of alternatives
presented here (and in the NPRM) does not attempt to project how future
off-cycle and A/C efficiency technology use will evolve or speculate
about the potential proliferation of FCIV proposals submitted to the
agencies. Rather, this analysis uses the off-cycle credits submitted by
each manufacturer for MY 2017 compliance, and, with a few exceptions,
carries these forward to future years. Several of the technologies
described below are associated with A/C efficiency and off-cycle FCIVs.
In particular, stop-start systems, integrated starter generators, and
full hybrids are assumed to generate off-cycle adjustments when applied
to vehicles to improve their fuel economy. Similarly, higher levels of
aerodynamic improvements are assumed to include active grille shutters
on the vehicle, which also qualify for off-cycle FCIVs.
The NPRM analysis assumed that any off-cycle FCIVs that are
associated with actions outside of the technologies discussed in
Section VI.C (either chosen from the pre-approved ``pick list,'' or
granted in response to individual manufacturer petitions) remained at
the levels claimed by manufacturers in MY 2017. Any additional A/C
efficiency and off-cycle adjustments that accrued as the result of
explicit technology application calculated dynamically in each model
year for each alternative. The NPRM version of the CAFE model also
represented manufacturers' credits for off-cycle improvements, A/C
efficiency improvements, and A/C leakage reduction in terms of values
applicable across all model years.
Recognizing that application of these improvements thus far varies
considerably among manufacturers, such that some manufacturers have
opportunities to earn significantly more of the corresponding
adjustments over time, the agencies have expanded the CAFE model's
representation of these credits to provide for year-by-year
specification of the amounts of each type of adjustment for each
manufacturer, denominated in grams CO2 per mile,\415\ as
summarized in the following table:
---------------------------------------------------------------------------
\415\ For estimating their contribution to CAFE compliance, the
grams CO2/mile values in Table VI-1711 are converted to
gallons/mile and applied to a manufacturer's 2-cycle CAFE
performance. When calculating compliance with EPA's CO2
program, there is no conversion necessary (as standards are also
denominated in grams/mile).
\416\ These values are specified in the ``market_ref.xlsx''
input file's ``Credits and Adjustments'' worksheet. The file is
available with the archive of model inputs and outputs posted at
https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24311]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.102
[[Page 24312]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.103
[[Page 24313]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.104
BILLING CODE 4910-59-C
In addition to these refinements to the estimation of the
quantities of adjustments earned over time by each manufacturer, the
agencies revised the
[[Page 24314]]
CAFE model to apply estimates of the corresponding costs. For today's
analysis, the agencies applied estimates developed previously by EPA,
adjusting these values to 2019 dollars. The following table summarizes
inputs through model year 2030:
[GRAPHIC] [TIFF OMITTED] TR30AP20.105
The model currently accounts for any off-cycle adjustments
associated with technologies that are included in the set of fuel-
saving technologies explicitly simulated as part of this proposal (for
example, start-stop systems that reduce fuel consumption during idle or
active grille shutters that improve aerodynamic drag at highway speeds)
and accumulates these adjustments up to the 10 g/mi cap. As a practical
matter, most of the adjustments for which manufacturers are claiming
off-cycle FCIV exist outside of the technology tree, so the cap is
rarely reached during compliance simulation. The agencies have
considered the potential to model their application explicitly.
However, doing so would require data regarding which vehicle models
already possess these improvements as well as the cost and expected
value of applying them to other models in the future. Such data is
currently too limited to support explicit modeling of these
technologies and adjustments.
b) Alternative Fuel Vehicles
When establishing maximum feasible fuel economy standards, NHTSA is
prohibited from considering the availability of alternatively fueled
vehicles,\417\ and credit provisions related to AFVs that significantly
increase their fuel economy for CAFE compliance purposes. Under the
``standard setting'' perspective, these technologies (pure battery
electric vehicles and fuel cell vehicles) \418\ are not available in
the compliance simulation to improve fuel economy. Under the
``unconstrained'' perspective, such as is documented in the DEIS and
FEIS, the CAFE model considers these technologies in the same manner as
other available technologies, and may apply them if they represent
cost-effective compliance pathways. However, under both perspectives,
the analysis continues to include dedicated AFVs that already exist in
the MY 2017 fleet (and their projected future volumes). Also, because
the CAA provides no direction regarding consideration of alternative
fuels, the final rule's analysis includes simulation of the potential
that some manufacturers might introduce new AFVs in response to
CO2 standards. To represent the compliance benefit from such
a response fully, NHTSA modified the CAFE model to include the specific
provisions related to AFVs under the CO2 standards. In
particular, the CAFE model now carries a full representation of the
production multipliers related to electric vehicles, fuel cell
vehicles, plug-in hybrids, and CNG vehicles, all of which vary by year
through MY 2021.
---------------------------------------------------------------------------
\417\ 49 U.S.C. 32902(h).
\418\ Dedicated compressed natural gas (CNG) vehicles should
also be excluded in this perspective but are not considered as a
compliance strategy under any perspective in this analysis.
---------------------------------------------------------------------------
EPCA also provides that CAFE levels may, subject to limitations, be
adjusted upward to reflect the sale of flexible fuel vehicles (FFVs).
Although these adjustments end after model year 2020, the final rule's
analysis, like the NPRM's, includes estimated potential use through MY
2019, as summarized below:
[[Page 24315]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.106
For its part, EPA has provided that manufacturers selling
sufficient numbers of PHEVs, BEVs, and FCVs may, when calculating fleet
average CO2 levels, ``count'' each unit of production as
more than a single unit. The CAFE model accounts for these
``multipliers.'' As for the NPRM, the final rule's analysis applies the
following multipliers:
[GRAPHIC] [TIFF OMITTED] TR30AP20.107
For example, under EPA's current regulation, when calculating the
average CO2 level achieved by its MY 2019 passenger car
fleet, a manufacturer may treat each 1,000 BEVs as 2,000 BEVs. When
calculating the average level required of this fleet, the manufacturer
must use the actual production volume (in this example, 1,000 units).
Similarly, the manufacturer must use the actual production volume when
calculating compliance credit balances.
There were no natural gas vehicles in the baseline fleet, and the
analysis did not apply natural gas technology due to cost
effectiveness. The application of a 2.0 multiplier for natural gas
vehicles for MYs 2022-2026 would have no impact on the analysis because
given the state of natural gas vehicle refueling infrastructure, the
cost to equip vehicles with natural gas tanks, the outlook for
petroleum prices, and the outlook for battery prices, we have little
basis to project more than an inconsequential response to this
incentive in the foreseeable future.
For the final rule's analysis, the CAFE model can be exercised in a
manner that simulates these current EPA requirements, or that simulates
two alternative approaches. The first includes the above-mentioned
multipliers in the calculation of average requirements, and the second
also includes the multipliers in the calculation of credit balances.
The central analysis reflects current regulations. The sensitivity
analysis presented in the FRIA includes a case
[[Page 24316]]
applying multipliers to the calculation of achieved and required
average CO2 levels, and calculation of credit balances.
c) Civil Penalties
Throughout the history of the CAFE program, some manufacturers have
consistently achieved fuel economy levels below applicable standards,
electing instead to pay civil penalties as specified by EPCA. As in
previous versions of the CAFE model, the current version allows the
user to specify inputs identifying such manufacturers and to consider
their compliance decisions as if they are willing to pay civil
penalties for non-compliance with the CAFE program. As with the NPRM,
the civil penalty rate in the current analysis is $5.50 per 1/10 of a
mile per gallon, per vehicle manufactured for sale.
NHTSA notes that treating a manufacturer as if it is willing to pay
civil penalties does not necessarily mean that it is expected to pay
penalties in reality. Doing so merely implies that the manufacturer
will only apply fuel economy technology up to a point, and then stop,
regardless of whether or not its corporate average fuel economy is
above its standard. In practice, the agencies expect that many of these
manufacturers will continue to be active in the credit market, using
trades with other manufacturers to transfer credits into specific
fleets that are challenged in any given year, rather than paying
penalties to resolve CAFE deficits. The CAFE model calculates the
amount of penalties paid by each manufacturer, but it does not simulate
trades between manufacturers. In practice, some (possibly most) of the
total estimated penalties may be a transfer from one OEM to another.
Although EPCA, as amended in 2007 by the Energy Independence and
Security Act (EISA), prescribes these specific civil penalty provisions
for CAFE standards, the Clean Air Act (CAA) does not contain similar
provisions. Rather, the CAA's provisions regarding noncompliance
prohibit sale of a new motor vehicle that is not covered by an EPA
certificate of conformity, and in order to receive such a certificate
the new motor vehicle must meet EPA's Section 202 regulations,
including applicable emissions standards. Therefore, inputs regarding
civil penalties--including inputs regarding manufacturers' potential
willingness to treat civil penalty payment as an economic choice--apply
only to simulation of CAFE standards. On the other hand, some of the
same manufacturers recently opting to pay civil penalties instead of
complying with CAFE standards have also recently led adoption of lower-
GWP refrigerants, and the ``A/C leakage'' credits count toward
compliance only with CO2 standards, not CAFE standards. The
model accounts for this difference between the programs.
When considering technology applications to improve fleet fuel
economy, the model will add technology up to the point at which the
effective cost of the technology (which includes technology cost,
consumer fuel savings, consumer welfare changes, and the cost of
penalties for non-compliance with the standard) is less costly than
paying civil penalties or purchasing credits. Unlike previous versions
of the model, the current implementation further acknowledges that some
manufacturers experience transitions between product lines where they
rely heavily on credits (either carried forward from earlier model
years or acquired from other manufacturers) or simply pay penalties in
one or more fleets for some number of years. The model now allows the
user to specify, when appropriate for the regulatory program being
simulated, on a year-by-year basis, whether each manufacturer should be
considered as willing to pay penalties for non-compliance. This
provides additional flexibility, particularly in the early years of the
simulation. As discussed above, this assumption is best considered as a
method to allow a manufacturer to under-comply with its standard in
some model years--treating the civil penalty rate and payment option as
a proxy for other actions it may take that are not represented in the
CAFE model (e.g., purchasing credits from another manufacturer, carry-
back from future model years, or negotiated settlements with NHTSA to
resolve deficits).
For the NPRM, NHTSA relied on past compliance behavior and
certified transactions in the credit market to designate some
manufacturers as willing to pay CAFE penalties in some model years. The
full set of NPRM assumptions regarding manufacturer behavior with
respect to civil penalties is presented in Table VI-21, which shows all
manufacturers were assumed to be willing to pay civil penalties prior
to MY 2020. This was largely a reflection of either existing credit
balances (which manufacturers will use to offset CAFE deficits until
the credits reach their expiration dates) or inter-manufacturer trades
assumed likely to happen in the near future, based on previous
behavior. The manufacturers in the table whose names appear in bold all
had at least one regulated fleet (of three) whose CAFE was below its
standard in MY 2016. Because the NPRM analysis began with the MY 2016
fleet, and no technology could be added to vehicles that are already
designed and built, all manufacturers could generate civil penalties in
MY 2016. However, once a manufacturer is designated as unwilling to pay
penalties, the CAFE model will attempt to add technology to the
respective fleets to avoid shortfalls.
[[Page 24317]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.108
Several of the manufacturers in Table VI-21 that were presumed to
be willing to pay civil penalties in the early years of the program
have no history of paying civil penalties. However, several of those
manufacturers have either bought or sold credits--or transferred
credits from one fleet to another to offset a shortfall in the
underperforming fleet. As the CAFE model does not simulate credit
trades between manufacturers, providing this additional flexibility in
the modeling avoids the outcome where the CAFE model applies more
technology than needed in the context of the full set of compliance
flexibilities at the industry level. By statute, NHTSA cannot consider
credit flexibilities when setting standards, so most manufacturers
(those without a history of civil penalty payment) are assumed to
comply with their standards through fuel economy improvements for the
model years being considered in this analysis. The notable exception to
this assumption is Fiat Chrysler Automobiles (FCA), which could still
satisfy the requirements of the program through a combination of credit
application and civil penalties through MY 2025 before eventually
complying exclusively through fuel economy improvements in MY 2026.
As mentioned above, the CAA does not provide civil penalty
provisions similar to those provisions specified in EPCA/EISA, and the
above-mentioned corresponding inputs apply only to simulation of
compliance with CAFE standards.
Some stakeholders offering comments related to the analytical
treatment of civil penalties indicated that NHTSA should tend toward
assuming manufacturers will take advantage of this EPCA provision as an
economically attractive alternative to compliance. Other commenters
implied that NHTSA should tend toward not relying on compliance
flexibilities in the analysis used to determine the maximum feasible
stringency of CAFE standards. For example, New York University's
Institute for Policy Integrity (IPI) offered the following comments:
NHTSA assumes that most manufacturers will be unwilling to pay
penalties based in part on the fact that most manufacturers have not
paid penalties in recent years. The Proposed Rule cites the
statutory prohibition on NHTSA considering credit trading as a
reason to assume manufacturers without a history of paying penalties
will comply through technology alone, whatever the cost. But this is
an arbitrary assumption and is in no way dictated by the statute.
NHTSA knows as much, since elsewhere in the proposed rollback, the
agency explains ``EPCA is very clear as to which flexibilities are
not to be considered'' and NHTSA is allowed to consider off-cycle
adjustments because they are not specifically mentioned. But
considering penalties are not mentioned as off-limits for NHTSA in
setting the standards either. Instead, the prohibition focuses on
credit trading and transferring. The penalty safety valve has
existed in EPCA for decades, and Congress clearly would have known
how to add penalties to the list of trading and transferring. The
fact that Congress did not bar NHTSA from considering penalties as a
safety valve means that NHTSA must consider manufacturer's efficient
use of penalties as a cost minimizing compliance option. Besides,
NHTSA does consider penalties for some of the manufacturers making
its statutory justification even less rational.\419\
---------------------------------------------------------------------------
\419\ Institute for Policy Integrity, NHTSA-2018-0067-12213, at
24.
On the other hand, in more general comments about NHTSA's
analytical treatment of program flexibilities, FCA stated that ``when
flexibilities are considered while setting targets, they cease to be
flexibilities and become simply additional technology mandates.'' \420\
---------------------------------------------------------------------------
\420\ FCA, Docket #NHTSA-2018-0067-11943, at 6.
---------------------------------------------------------------------------
NHTSA agrees with IPI that EPCA does not expressly prohibit NHTSA,
when conducting analysis supporting determinations of the maximum
feasible stringency of future CAFE standards, from including
manufacturers' potential tendency to pay civil penalties rather than
complying with those standards. However, EPCA also does not require
NHTSA to include this tendency in its analysis. NHTSA also notes, as
does IPI, that EPCA does prohibit NHTSA from including credit trading,
transferring, or the availability of credits in such
[[Page 24318]]
analysis (although NHTSA interprets this prohibition to apply only to
the model years for which standards are being set). This statutory
difference is logical based on the way credits and penalties function
differently under EPCA. Because credits help manufacturers achieve
compliance with CAFE standards, absent the statutory prohibition,
credits would be relevant to the feasibility of a standard.\421\
Penalties, on the other hand, do not enable a manufacturer to comply
with an applicable standard; penalties are for noncompliance.\422\ When
Congress added credit trading provisions to EPCA in 2007, NHTSA
anticipated that competitive considerations would make manufacturers
reluctant to engage in such trades. Since that time, manufacturers
actually have demonstrated otherwise, although the reliance on
trading--especially between specific pairs of OEMs--appears to vary
widely. At this time, NHTSA considers it most likely that manufacturers
will shift away from paying civil penalties and toward compliance
credit trading. Consequently, for NHTSA to include civil penalty
payment in its analysis would increasingly amount to using civil
penalty payment as an analytical proxy for credit trading. Having
further considered the question, NHTSA's current view is, therefore,
that including civil penalty payment beyond MY 2020 would effectively
subvert EPCA's prohibition against considering credit trading.
Therefore, for today's announcement, NHTSA has modified its analysis to
assume that BMW, Daimler, FCA, JLR, and Volvo would consider paying
civil penalties through MY 2020, and that all manufacturers would apply
as much technology as would be needed in order to avoid paying civil
penalties after MY 2020.
---------------------------------------------------------------------------
\421\ See 49 U.S.C. 32911(b) (``Compliance is determined after
considering credits available to the manufacturer . . . . '').
\422\ See id.
---------------------------------------------------------------------------
3. Technology Effectiveness Values
The next input required to simulate manufacturers' decision-making
processes for the year-by-year application of technologies to specific
vehicles is estimates of how effective each technology would be at
reducing fuel consumption. In the NPRM, the agencies used full-vehicle
modeling and simulation to estimate the fuel economy improvements
manufacturers could make to a fleet of vehicles, considering those
vehicles' technical specifications and how combinations of technologies
interact. Full-vehicle modeling and simulation uses computer software
and physics-based models to predict how combinations of technologies
perform as a full system under defined conditions.
A model is a mathematical representation of a system, and
simulation is the behavior of that mathematical representation over
time. In this analysis, the model is a mathematical representation of
an entire vehicle,\423\ including its individual components such as the
engine and transmission, overall vehicle characteristics such as mass
and aerodynamic drag, and the environmental conditions, such as ambient
temperature and barometric pressure. The agencies simulated the model's
behavior over test cycles, including the 2-cycle laboratory compliance
tests (or 2-cycle tests),\424\ to determine how the individual
components interact. 2-cycle tests are test cycles that are used to
measure fuel economy and emissions for CAFE and CO2
compliance, and therefore are the relevant test cycles for determining
technology effectiveness when establishing standards. In the
laboratory, 2-cycle testing involves sophisticated test and measurement
equipment, carefully controlled environmental conditions, and precise
procedures to provide the most repeatable results possible with human
drivers. Measurements using these structured procedures serve as a
yardstick for fuel economy and CO2 emissions.
---------------------------------------------------------------------------
\423\ Our full vehicle model was composed of sub-models, which
is why the full vehicle model could also be referred to as a full
system model, composed of sub-system models.
\424\ EPA's compliance test cycles are used to measure the fuel
economy of a vehicle. For readers unfamiliar with this process, it
is like running a car on a treadmill following a program--or more
specifically, two programs. The ``programs'' are the ``urban
cycle,'' or Federal Test Procedure (abbreviated as ``FTP''), and the
``highway cycle,'' or Highway Fuel Economy Test (abbreviated as
``HFET''), and they have not changed substantively since 1975. Each
cycle is a designated speed trace (of vehicle speed versus time)
that all certified vehicles must follow during testing. The FTP is
meant roughly to simulate stop and go city driving, and the HFET is
meant roughly to simulate steady flowing highway driving at about 50
mph. For further details on compliance testing, see the discussion
in Section VI.B.3.a)(7).
---------------------------------------------------------------------------
Full-vehicle modeling and simulation was initially developed to
avoid the costs of designing and testing prototype parts for every new
type of technology. For example, if a truck manufacturer has a concept
for a lightweight tailgate and wants to determine the fuel economy
impact for the weight reduction, the manufacturer can use physics-based
computer modeling to estimate the impact. The vehicle, modeled with the
proposed change, can be simulated on a defined test route and under a
defined test condition, such as city or highway driving in warm ambient
temperature conditions, and compared against the baseline reference
vehicle. Full-vehicle modeling and simulation allows the consideration
and evaluation of different designs and concepts before building a
single prototype. In addition, full vehicle modeling and simulation is
beneficial when considering technologies that provide small incremental
improvements. These improvements are difficult to measure in laboratory
tests due to variations in how vehicles are driven over the test cycle
by human drivers, variations in emissions measurement equipment, and
variations in environmental conditions.\425\
---------------------------------------------------------------------------
\425\ Difficulty with controlling for such variability is
reflected, for example, in 40 CFR 1065.210, Work input and output
sensors, which describes complicated instructions and
recommendations to help control for variability in real world (non-
simulated) test instrumentation set up.
---------------------------------------------------------------------------
Full-vehicle modeling and simulation requires detailed data
describing the individual technologies and performance-related
characteristics. Those specifications generally come from design
specifications, laboratory measurements, and other subsystem
simulations or modeling. One example of data used as an input to the
full vehicle simulation are engine maps for each engine technology that
define how much fuel is consumed by the engine technology across its
operating range.
Using full-vehicle modeling and simulation to estimate technology
efficiency improvements has two primary advantages over using single or
limited point estimates. An analysis using single or limited point
estimates may assume that, for example, one fuel economy improving
technology with an effectiveness value of 5 percent by itself and
another technology with an effectiveness value of 10 percent by itself,
when applied together achieve an additive improvement of 15 percent.
Single point estimates generally do not provide accurate effectiveness
values because they do not capture complex relationships among
technologies. Technology effectiveness often differs significantly
depending on the vehicle type (e.g., sedan versus pickup truck) and how
the technology interacts with other technologies on the vehicle, as
different technologies may provide different incremental levels of fuel
economy improvement if implemented alone or in tandem with other
technologies. Any oversimplification of these complex interactions
leads to less accurate and often overestimated effectiveness estimates.
In addition, because manufacturers often implement several fuel-
saving
[[Page 24319]]
technologies simultaneously when redesigning a vehicle, it is difficult
to isolate the effect of individual technologies using laboratory
measurement of production vehicles alone. Modeling and simulation
offers the opportunity to isolate the effects of individual
technologies by using a single or small number of baseline vehicle
configurations and incrementally adding technologies to those baseline
configurations. This provides a consistent reference point for the
incremental effectiveness estimates for each technology and for
combinations of technologies for each vehicle type. Vehicle modeling
also reduces the potential for overcounting or undercounting technology
effectiveness.
An important feature of this analysis is that the incremental
effectiveness of each technology and combinations of technologies be
accurate and relative to a consistent baseline vehicle. The absolute
fuel economy values of the full vehicle simulations are used only to
determine incremental effectiveness and are never used directly to
assign an absolute fuel economy value to any vehicle model or
configuration for the rulemaking analysis.
For this analysis, absolute fuel economy levels are based on the
individual fuel economy values from CAFE compliance data for each
vehicle in the baseline fleet. The incremental effectiveness from the
full vehicle simulations performed in Autonomie, a physics-based full-
vehicle modeling and simulation software developed and maintained by
the U.S. Department of Energy's Argonne National Laboratory, are
applied to baseline fuel economy to determine the absolute fuel economy
of applying the first technology change. For subsequent technology
changes, incremental effectiveness is applied to the absolute fuel
economy level of the previous technology configuration.
For example, if a Ford F150 2-wheel drive crew cab and short bed in
the baseline fleet has a fuel economy value of 30 mpg for CAFE
compliance, 30 mpg will be considered the reference absolute fuel
economy value. A similar full vehicle model in the Autonomie simulation
may begin with an average fuel economy value of 32 mpg, and with
incremental addition of a specific technology X its fuel economy
improves to 35 mpg, a 9.3 percent improvement. In this example, the
incremental fuel economy improvement (9.3 percent) from technology X
would be applied to the F150's 30 mpg absolute value.
For this analysis, the agencies determined the incremental
effectiveness of technologies as applied to the 2,952 unique vehicle
models in the analysis fleet. Although, as mentioned above, full-
vehicle modeling and simulation reduces the work and time required to
assess the impact of moving a vehicle from one technology state to
another, it would be impractical--if not impossible--to build a unique
vehicle model for every individual vehicle in the analysis fleet.
Therefore, as explained further below, vehicle models are built in a
way that maintains similar attributes to the analysis fleet vehicles,
which ensures key components are reasonably represented.
We received a wide array of comments regarding the full-vehicle
modeling and simulation performed for the NPRM, but there was general
agreement that full-vehicle modeling and simulation was the appropriate
method to determine technology effectiveness.\426\ Stakeholders
commented on other areas, such as full vehicle simulation tools,
inputs, and assumptions, and these comments will be discussed in the
following sections. For this final rule, the agencies continued to use
the same full-vehicle simulation approach to estimate technology
effectiveness for technology adoption in the rulemaking timeframe. The
next sections will discuss the details of the explicit input
specifications and assumptions used for the final rule analysis.
---------------------------------------------------------------------------
\426\ See NHTSA-2018-0067-12039; NHTSA-2018-0067-12073. UCS and
AAM both agreed that full vehicle simulation can significantly
improve the estimates of technology effectiveness.
---------------------------------------------------------------------------
a) Why This Rulemaking Used Autonomie Full-Vehicle Modeling and
Simulation To Determine Technology Effectiveness
The NPRM and final rule analysis use effectiveness estimates for
technologies developed using Autonomie, a physics-based full-vehicle
modeling and simulation software developed and maintained by the U.S.
Department of Energy's Argonne National Laboratory.\427\ Autonomie was
designed to serve as a single tool to meet requirements of automotive
engineering throughout the vehicle development process, and has been
under continuous improvement by Argonne for over 20 years. Autonomie is
commercially available and widely used in the automotive industry by
suppliers, automakers, and academic researchers (who publish findings
in peer reviewed academic journals).\428\ DOE and manufacturers have
used Autonomie and its ability to simulate a large number of powertrain
configurations, component technologies, and vehicle-level controls over
numerous drive cycles to support studies on fuel efficiency, cost-
benefit analysis, and carbon dioxide emissions,\429\ and other topics.
---------------------------------------------------------------------------
\427\ More information about Autonomie is available at https://www.anl.gov/technology/project/autonomie-automotive-system-design
(last accessed June 21, 2018). As mentioned in the preliminary
regulatory impact analysis (PRIA) for this rule, the agencies used
Autonomie version R15SP1, the same version used for the 2016 Draft
TAR.
\428\ Rousseau, A. Shidore, N. Karbowski, D. Sharer, ``Autonomie
Vehicle Validation Summary.'' https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/anl-autonomie-vehicle-model-validation-1509.pdf.
\429\ Delorme et al. 2008, Rousseau, A, Sharer, P, Pagerit, S.,
& Das, S. ``Trade-off between Fuel Economy and Cost for Advanced
Vehicle Configurations,'' 20th International Electric Vehicle
Symposium (EVS20), Monaco (April 2005); Elgowainy, A., Burnham, A.,
Wang, M., Molburg, J., & Rousseau, A. ``Well-To-Wheels Energy Use
and Greenhouse Gas Emissions of Plug-in Hybrid Electric Vehicles,''
SAE 2009-01-1309, SAE World Congress, Detroit, April 2009.
---------------------------------------------------------------------------
Autonomie has also been used to provide the U.S. government with
data to make decisions about future research, and is used by DOE for
analysis supporting budget priorities and plans for programs managed by
its Vehicle Technologies Office (VTO), and to support decision making
among competing vehicle technology research and development
projects.\430\ In addition, Autonomie is the primary vehicle simulation
tool used by DOE to support its U.S. DRIVE program, a government-
industry partnership focused on advanced automotive and related energy
infrastructure technology research and development.\431\
---------------------------------------------------------------------------
\430\ U.S. DOE Benefits & Scenario Analysis publications is
available at https://www.autonomie.net/publications/fuel_economy_report.html (last accessed September 11, 2019).
\431\ For more information on U.S. Drive, see https://www.energy.gov/eere/vehicles/us-drive.
---------------------------------------------------------------------------
Autonomie is a MathWorks-based software environment and framework
for automotive control-system design, simulation, and analysis.\432\ It
is designed for rapid and easy integration of models with varying
levels of detail (low to high fidelity), abstraction (from subsystems
to systems and entire architectures), and processes (e.g., calibration,
validation). By building models automatically, Autonomie allows the
quick simulation of many component technologies and powertrain
configurations, and, in this case, to assess the energy consumption of
advanced powertrain technologies. Autonomie simulates subsystems,
[[Page 24320]]
systems, or entire vehicles; evaluates and analyzes fuel efficiency and
performance; performs analyses and tests for virtual calibration,
verification, and validation of hardware models and algorithms;
supports system hardware and software requirements; links to
optimization algorithms; and supplies libraries of models for
propulsion architectures of conventional powertrains as well as hybrid
and electric vehicles.
---------------------------------------------------------------------------
\432\ Halbach, S. Sharer, P. Pagerit, P., Folkerts, C. &
Rousseau, A. ``Model Architecture, Methods, and Interfaces for
Efficient Math-Based design and Simulation of Automotive Control
Systems,'' SAE 2010-01-0241, SAE World Congress, Detroit, April,
2010.
---------------------------------------------------------------------------
With hundreds of pre-defined powertrain configurations along with
vehicle level control strategies developed from dynamometer test data,
Autonomie is a highly capable tool for analyzing advantages and
drawbacks of applying different technology options within each
technology family, including conventional, parallel hybrid, power-split
hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles
(PHEVs), battery electric vehicles (BEV) and fuel cell vehicles (FCVs).
Autonomie also allows users to evaluate the effect of component sizing
on fuel consumption for different powertrain technologies as well as to
define component requirements (e.g., power, energy) to maximize fuel
displacement for a specific application.\433\ To evaluate properly any
powertrain-configuration or component-sizing influence, vehicle-level
control models are critical, especially for electric drive vehicles
like hybrids and plug-in hybrids. Argonne has extensive expertise in
developing vehicle-level control models based on different approaches,
from global optimization to instantaneous optimization, rule-based
optimization, and heuristic optimization.\434\
---------------------------------------------------------------------------
\433\ Nelson, P., Amine, K., Rousseau, A., & Yomoto, H. (EnerDel
Corp.), ``Advanced Lithium-ion Batteries for Plug-in Hybrid-electric
Vehicles,'' 23rd International Electric Vehicle Symposium (EVS23),
Anaheim, CA, (Dec. 2007); Karbowski, D., Haliburton, C., & Rousseau,
A. ``Impact of Component Size on Plug-in Hybrid Vehicles Energy
Consumption using Global Optimization,'' 23rd International Electric
Vehicle Symposium (EVS23), Anaheim, CA, (Dec. 2007).
\434\ Karbowski, D., Kwon, J., Kim, N., & Rousseau, A.,
``Instantaneously Optimized Controller for a Multimode Hybrid
Electric Vehicle,'' SAE paper 2010-01-0816, SAE World Congress,
Detroit, April 2010; Sharer, P., Rousseau, A., Karbowski, D., &
Pagerit, S. ``Plug-in Hybrid Electric Vehicle Control Strategy--
Comparison between EV and Charge-Depleting Options,'' SAE paper
2008-01-0460, SAE World Congress, Detroit (April 2008); and
Rousseau, A., Shidore, N., Carlson, R., & Karbowski, D. ``Impact of
Battery Characteristics on PHEV Fuel Economy,'' AABC08.
---------------------------------------------------------------------------
Autonomie has been developed to consider real-world vehicle metrics
like performance, hardware limitations, utility, and drivability
metrics (e.g., towing capability, shift busyness, frequency of engine
on/off transitions), which are important to producing realistic
estimates of fuel economy and CO2 emission rates. This
increasing realism has, in turn, steadily increased confidence in the
appropriateness of using Autonomie to make significant investment
decisions. Autonomie has also been validated for a number of powertrain
configurations and vehicle classes using Argonne's Advanced Mobility
Technology Laboratory (AMTL) (formerly Advanced Powertrain Research
Facility, or APRF) vehicle test data.\435\
---------------------------------------------------------------------------
\435\ Jeong, J., Kim, N., Stutenberg, K., Rousseau, A.,
``Analysis and Model Validation of the Toyota Prius Prime.'' SAE
2019-01-0369, SAE World Congress, Detroit, April 2019; Kim, N,
Jeong, J. Rousseau, A. & Lohse-Busch, H. ``Control Analysis and
Thermal Model Development of PHEV,'' SAE 2015-01-1157, SAE World
Congress, Detroit, April 2015; Kim, N., Rousseau, A. & Lohse-Busch,
H. ``Advanced Automatic Transmission Model Validation Using
Dynamometer Test Data,'' SAE 2014-01-1778, SAE World Congress,
Detroit, Apr. 14; Lee, D. Rousseau, A. & Rask, E. ``Development and
Validation of the Ford Focus BEV Vehicle Model,'' 2014-01-1809, SAE
World Congress, Detroit, Apr. 14; Kim, N., Kim, N., Rousseau, A., &
Duoba, M. ``Validating Volt PHEV Model with Dynamometer Test Data
using Autonomie,'' SAE 2013-01-1458, SAE World Congress, Detroit,
Apr. 13; Kim, N., Rousseau, A., & Rask, E. ``Autonomie Model
Validation with Test Data for 2010 Toyota Prius,'' SAE 2012-01-1040,
SAE World Congress, Detroit, Apr. 12; Karbowski, D., Rousseau, A,
Pagerit, S., & Sharer, P. ``Plug-in Vehicle Control Strategy--From
Global Optimization to Real Time Application,'' 22th International
Electric Vehicle Symposium (EVS22), Yokohama, (October 2006).
---------------------------------------------------------------------------
Argonne has spent several years developing, applying, and expanding
the means to use distributed computing to exercise its Autonomie full-
vehicle simulation tool over the scale necessary for realistic analysis
to provide data for CAFE and CO2 standards rulemaking. The
NPRM and PRIA detailed how Argonne used Autonomie to estimate the fuel
economy impacts for roughly a million combinations of technologies and
vehicle types.436 437 Argonne developed input parameters for
Autonomie to represent every combination of vehicle, powertrain, and
component technologies considered in this rulemaking. The sequential
addition of more than 50 fuel economy-improving technologies to ten
vehicle types generated more than 140,000 unique technology and vehicle
combinations. Running the Autonomie powertrain sizing algorithms to
determine the appropriate amount of engine downsizing needed to
maintain overall vehicle performance when vehicle mass reduction is
applied and for certain engine technology changes (discussed further,
below) increased the total number of simulations to more than one
million. The result of these simulations is a useful dataset
identifying the impacts of combinations of vehicle technologies on
energy consumption--a dataset that can be referenced as an input to the
CAFE model for assessing regulatory compliance alternatives.
---------------------------------------------------------------------------
\436\ As part of the Argonne simulation effort, individual
technology combinations simulated in Autonomie were paired with
Argonne's BatPAC model to estimate the battery cost associated with
each technology combination based on characteristics of the
simulated vehicle and its level of electrification. Information
regarding Argonne's BatPAC model is available at http://www.cse.anl.gov/batpac/.
\437\ Additionally, the impact of engine technologies on fuel
consumption, torque, and other metrics was characterized using GT
POWER simulation modeling in combination with other engine modeling
that was conducted by IAV Automotive Engineering, Inc. (IAV). The
engine characterization ``maps'' resulting from this analysis were
used as inputs for the Autonomie full-vehicle simulation modeling.
Information regarding GT Power is available at https://www.gtisoft.com/gt-suite-applications/propulsion-systems/gt-power-engine-simulation-software.
---------------------------------------------------------------------------
The following sections discuss the full-vehicle modeling and
simulation inputs and data assumptions, and comments received on the
NPRM analysis. The discussion is necessarily technical, but also
important to understand the agencies' decisions to modify (or not) the
Autonomie analysis for the final rule.
(1) Full-Vehicle Modeling, Simulation Inputs and Data Assumptions
The agencies provided extensive documentation that quantitatively
and qualitatively described the over 50 technologies considered as
inputs to the Autonomie modeling.438 439 These inputs
consisted of engine technologies, transmission technologies, powertrain
electrification, light-weighting, aerodynamic improvements, and tire
rolling resistance improvements.\440\ The PRIA provided an overview of
the sub-models for each technology, including the internal combustion
engine model, automatic transmission model, and others.\441\ The
Argonne NPRM model documentation expanded on these sub-models in detail
to show the interaction of each sub-model input and output.\442\
[[Page 24321]]
For example, as shown in Figure VI-2, the input for Autonomie's driver
model (i.e., the model used to approximate the driving behavior of a
real driver) is vehicle speed, and outputs are accelerator pedal, brake
pedal, and torque demand.
---------------------------------------------------------------------------
\438\ NHTSA-2018-0067-12299. Preliminary Regulatory Impact
Analysis (July 2018).
\439\ NHTSA-2018-0067-0007. Islam, E., S, Moawad, A., Kim, N,
Rousseau, A. ``A Detailed Vehicle Simulation Process To Support CAFE
Standards 04262018--Report'' ANL Autonomie Documentation. Aug 21,
2018. NHTSA-2018-0067-0004. ANL Autonomie Data Dictionary. Aug 21,
2018. NHTSA-2018-0067-0003. ANL Autonomie Summary of Main Component
Assumptions. Aug 21, 2018. NHTSA-2018-0067-0005. ANL Autonomie Model
Assumptions Summary. Aug 21, 2018. NHTSA-2018-0067-1692. ANL BatPac
Model 12 55. Aug 21, 2018.
\440\ SAFE Rule for MY2021-2026 PRIA Chapter 6.2.3 Technology
groups in Autonomie simulations and CAFE model.
\441\ PRIA at 189.
\442\ NHTSA-2018-0067-0007. Islam, E., S, Moawad, A., Kim, N,
Rousseau, A. ``A Detailed Vehicle Simulation Process To Support CAFE
Standards 04262018--Report'' ANL Autonomie Documentation. Aug 21,
2018.
[GRAPHIC] [TIFF OMITTED] TR30AP20.109
Effectiveness inputs for the NPRM and the final rule analysis were
specifically developed to consider many real world and compliance test
cycle constraints, to the extent a computer model could capture them.
Examples include the advanced engine knock model discussed below, in
addition to other constraints like allowing cylinder deactivation to
occur in ways that would not negatively impact noise-vibration-
harshness (NVH), and similarly optimizing the number of engine on/off
events (e.g., from start/stop 12V micro hybrid systems) to balance
between effectiveness and NVH.
One major input used in the effectiveness modeling that the
agencies provided key specifications for in the PRIA are engine fuel
maps that define how an engine equipped with specific technologies
operates over a variety of engine load (torque) and engine speed
conditions. The engine maps used as inputs to the Autonomie modeling
portion of the analysis were developed by starting with a base map and
then modifying that base map, incrementally, to model the addition of
engine technologies. These engine maps, developed using the GT-Power
modeling tool by IAV, were based off real-world engine designs.
Simulated operation of these engines included the application of an IAV
knock model, also developed from real-world engine
data.443 444 Using this process, which incorporated real-
world data, ensured that real-world constraints were considered for
each vehicle type. Although the same type of engine map is used for all
technology classes, the effectiveness varies based on the
characteristics of each vehicle type. For example, a compact car with a
turbocharged engine will have different fuel economy and performance
values than a pickup truck with the same engine technology type. The
engine map specifications are discussed further in Section VI.C.1 of
this preamble and Section VI of FRIA.
---------------------------------------------------------------------------
\443\ Engine knock in spark ignition engines occurs when
combustion of some of the air/fuel mixture in the cylinder does not
result from propagation of the flame front ignited by the spark
plug, but one or more pockets of air/fuel mixture explodes outside
of the envelope of the normal combustion front.
\444\ See IAV material submitted to the docket; IAV_20190430_Eng
22-26 Updated_Docket.pdf,
IAV_Engine_tech_study_Sept_2016_Docket.pdf, IAV_Study for 4 Cylinder
Gas Engines_Docket.pdf.
---------------------------------------------------------------------------
The agencies also provided key details about input assumptions for
various vehicle specifications like transmission gear ratios, tire
size, final drive ratios, and individual component weights.\445\ Each
of these assumptions, to some extent, varied between the ten technology
classes to capture appropriately real-world vehicle specifications like
wheel mass or fuel tank mass. These specific input assumptions were
developed based on the latest test data and current market fleet
information.\446\ The agencies relied on default assumptions developed
by the Autonomie team, based on test data and technical publication
review, for other model inputs required by Autonomie, such as throttle
time response and shifting strategies for different transmission
technologies. The Autonomie modeling tool did not simulate vehicle
attributes determined to have minimal impacts, like whether a vehicle
had a sun roof or hood scoops, as those attributes would have trivial
impact in the overall analysis.
---------------------------------------------------------------------------
\445\ ANL Autonomie Model Assumptions Summary. Aug 21, 2018,
NHTSA-2018-0067-0005. ANL--Summary of Main Component Performance and
Assumptions NPRM. Aug 21, 2018, NHTSA-2018-0067-0003.
\446\ See further details in Section VI.B.1 Analysis Fleet.
---------------------------------------------------------------------------
Because the agencies model ten different vehicle types to represent
the 2,952 vehicles in the baseline fleet, improper assumptions about an
advanced technology could lead to errors in estimating effectiveness.
Autonomie is a sophisticated full-vehicle modeling tool that requires
extensive technology characteristics based on both physical and
intangible data, like proprietary software. With a few technologies,
the agencies did not have publicly available data, but had received
confidential business information confirming such technologies
potential availability in the market during the rulemaking time frame.
For such technologies, including advanced cylinder deactivation, the
agencies adopted a method in the CAFE model to represent the
effectiveness of the technology, and did not explicitly simulate the
technologies in the Autonomie model. For this limited set of
technologies, the agencies determined that effectiveness could
reasonably be represented as a fixed value.\447\ Effectiveness values
for technologies not explicitly simulated in Autonomie are discussed
further in the individual technology sections of this preamble.
---------------------------------------------------------------------------
\447\ For final rule, 9 out of 50 plus technologies use fixed
offset effectiveness values. The total effectiveness of these
technologies cannot be captured on the 2-cycle test or, like ADEAC,
they are a new technology where robust data that could be used as an
input to the technology effectiveness modeling does not yet exist.
Specifically, these nine technologies are LDB, SAX, EPS, IACC, EFR,
ADEAC, DSLI, DSLIAD and TURBOAD.
---------------------------------------------------------------------------
The agencies sought comments on all effectiveness inputs and input
assumptions, including the specific data used to characterize the
technologies,
[[Page 24322]]
such as data to build the technology input, data representing operating
range of technologies, and data for variation among technology inputs.
The agencies also sought comment on the effectiveness values used for
technologies not explicitly defined in Autonomie.
Meszler Engineering Services, commenting on behalf of the Natural
Resources Defense Council, and ICCT questioned the accuracy of the
effectiveness estimates in the Argonne database, and as an example
Meszler analyzed the fuel economy impacts of a 10-speed automatic
transmission relative to a baseline 8-speed automatic transmission,
concluding that the widely ranging effectiveness estimates were
unexpected. ICCT questioned the accuracy of the IAV engine maps that
serve as an input to the Autonomie effectiveness modeling, and asked
whether those could ``reasonably stand as a foundation for automotive
developments and technology combinations'' discussed elsewhere in their
comments. ICCT also questioned whether Autonomie realistically and
validly modeled synergies between technologies, using the effectiveness
values from CEGR and transmissions as an example. Meszler stated that
the agencies have an obligation to validate the Autonomie estimates
before using them to support the NPRM or any other rulemaking. The
agencies also received comments on the specific effectiveness estimates
generated by Autonomie; however, those comments will be discussed in
each individual technology section, below.
Despite these criticisms, Meszler stated that the critiques of the
Autonomie technology database were not meant to imply that the
Autonomie vehicle simulation model used to develop the database was
fundamentally flawed, or that the model could not be used to derive
accurate fuel economy impact estimates. Meszler noted that, as with any
model, estimates derived with Autonomie are only valid for a given set
of modeling parameters and if those parameters are well defined, the
estimates should be accurate and reliable. Conversely, if those
parameters are not well defined, the estimates would be inaccurate and
unreliable. Meszler stated that the agencies must make the full set of
modeling assumptions used for the Autonomie database available for
review and comment.
We agree with Meszler that, in general, when inputs to a model are
inaccurate, output effectiveness results may be too high or too low.
The technology effectiveness estimates from modeling results often vary
with the type of vehicle and the other technologies that are on that
vehicle.\448\ The Autonomie output database consists of permutations of
over 50 technologies for each of the ten technology classes simulated
by the CAFE model. A wide range of effectiveness is expected when going
from a baseline technology to an advanced technology across different
technology classes because there are significant differences in how
much power is required from the powertrain during 2-cycle testing
across the ten vehicle types. This impacts powertrain operating
conditions (e.g., engine speed and load) during 2-cycle testing. Fuel
economy improving technologies have different effectiveness at each of
those operating conditions so vehicles that have higher average power
demands will have different effectiveness than vehicles with lower
average power demands. Further, the differences in effectiveness at
higher power and lower power vary by technology so the overall
relationship is complex. Large-scale full-vehicle modeling and
simulation account for these interactions and complexities.
---------------------------------------------------------------------------
\448\ The PRIA Chapter 6.2.2.1, Table 6-2 and Table 6-3 defined
the characteristics of the reference technology classes that
representative of the analysis fleet.
---------------------------------------------------------------------------
Before conducting any full-vehicle modeling and simulation, the
agencies spent a considerable amount of time and effort developing the
specific inputs used for the Autonomie analysis. The agencies believe
that these technology inputs provide reasonable estimates for the
light-duty vehicle technologies the agencies expect to be available in
the market in the rulemaking timeframe. As discussed earlier, these
inputs vary in effectiveness due to how different vehicles, like
compact cars and pickup trucks, operate on the 2-cycle test and in the
real world. Some technologies, such as 10-speed automatic transmissions
(AT10) relative to 8-speed automatic transmissions (AT8), can and
should have different effectiveness results in the analysis between two
different technology classes.\449\ These unique synergistic effects can
only be taken into account through conducting full-vehicle modeling and
simulation, which the agencies did here.
---------------------------------------------------------------------------
\449\ Separately, the agencies modified specific transmission
modeling parameters for the final rule after additional review,
including a thorough review of public comments, and this review is
discussed in detail in Section VI.C.2.
---------------------------------------------------------------------------
With regards to Meszler's comment that the agencies have an
obligation to validate the Autonomie estimates before using them to
support the NPRM or any other rulemaking, the agencies would like to
point Meszler to the description of the Argonne Autonomie team's robust
process for vehicle model validation that was contained in the
PRIA.\450\ To summarize, the NPRM and final rule analysis leveraged
extensive vehicle test data collected by Argonne National
Laboratory.\451\ Over the past 20 years, the Argonne team has developed
specific instrumentation lists and test procedures for collecting
sufficient information to develop and validate full vehicle models. In
addition, the agencies described the Argonne team's efforts to validate
specific component models as well, such as the advanced automatic
transmission and dual clutch transmission models.\452\
---------------------------------------------------------------------------
\450\ PRIA at 216-7. See also N. Kim, A. Rousseau, E. Rask,
``Autonomie Model Validation with Test Data for 2010 Toyota Prius,''
SAE 2012-01-1040, SAE World Congress, Detroit, Apr12. https://www.autonomie.net/docs/5%20-%20Presentations/Validation/SAE%202012-01-1040.pdf; Vehicle Validation Status, February 2010 https://www.autonomie.net/docs/5%20-%20Presentations/Validation/vehicle_validation_status.pdf; Tahoe HEV Model Development in PSAT,
SAE paper 2009-01-1307, April 2009 https://www.autonomie.net/docs/5%20-%20Presentations/Validation/tahoe_hev.pdf; PHEV Model
Validation, U.S.DOE Merit Review 2008 https://www.autonomie.net/docs/5%20-%20Presentations/Validation/phev_model_validation.pdf ;
PHEV HyMotion Prius model validation and control improvements, 23rd
International Electric Vehicle Symposium (EVS23), Dec. 2007 https://www.autonomie.net/docs/5%20-%20Presentations/Validation/phev_hymotion_prius.pdf; Integrating Data, Performing Quality
Assurance, and Validating the Vehicle Model for the 2004 Prius Using
PSAT, SAE paper 2006-01-0667, April 2006; https://www.autonomie.net/docs/5%20-%20Presentations/Validation/integrating_data.pdf.
\451\ A list of the vehicles that have been tested at the APRF
can be found under http://www.anl.gov/energy-systems/group/downloadable-dynamometer-database.
\452\ Kim, N., Rousseau, N., Lohse-Bush, H. ``Advanced Automatic
Transmission Model Validation Using Dynamometer Test Data,'' SAE
2014-01-1778, SAE World Congress, Detroit, April 2014; Kim, N.,
Lohse-Bush, H., Rousseau, A. ``Development of a model of the dual
clutch transmission in Autonomie and validation with dynamometer
test data,'' International Journal of Automotive Technologies, March
2014, Volume 15, Issue 2, pp 263-71.
---------------------------------------------------------------------------
The agencies also described the process for validating inputs used
to develop the IAV engine maps,453 454 another input to the
Autonomie simulations. As discussed in the PRIA, IAV's engine model
development relied on a collection of sub-models that controlled
independent combustion characteristics such as heat release, combustion
knock, friction, heat flow, and other combustion optimization tools.
These sub-models and other
[[Page 24323]]
computational fluid dynamics models were utilized to convert test data
for use in the IAV engine map development. Specific combustion
parameters, like from test data for the coefficient of variation for
the indicated mean effective pressure (COV of IMEP), which is a common
variable for combustion stability in a spark ignited engine, was used
to assure final engine models were reasonable. The assumptions and
inputs used in the modeling and validation of engine model results
leveraged IAV's global engine database, which included benchmarking
data, engine test data, single cylinder test data and prior modeling
studies, and also technical publications and information presented at
conferences. The agencies referenced in the PRIA that engine maps were
validated with engine dynamometer test data to the maximum extent
possible.\455\ Because the NPRM and the final rule analysis considered
some technologies not yet in production, the agencies relied on
technical publications and engine modeling by IAV to develop and
corroborate inputs and input assumptions where engine dynamometer test
data was not available.
---------------------------------------------------------------------------
\453\ See PRIA at 251.
\454\ See IAV material submitted to the docket; IAV_20190430_Eng
22-26 Updated_Docket.pdf,
IAV_Engine_tech_study_Sept_2016_Docket.pdf, IAV_Study for 4 Cylinder
Gas Engines_Docket.pdf.
\455\ See PRIA at 288.
---------------------------------------------------------------------------
In addition, as described earlier in this section, the full set of
NPRM modeling assumptions used for the Autonomie database were
available for review and comment in the docket for this
rulemaking.\456\ The full set of modeling assumptions used for the
final rule are also available in the docket.\457\
---------------------------------------------------------------------------
\456\ NHTSA-2018-0067-0007. Islam, E., S, Moawad, A., Kim, N,
Rousseau, A., ``A Detailed Vehicle Simulation Process To Support
CAFE Standards 04262018--Report'' ANL Autonomie Documentation. Aug
21, 2018. NHTSA-2018-0067-0004. ANL Autonomie Data Dictionary. Aug
21, 2018. NHTSA-2018-0067-0003. ANL Autonomie Summary of Main
Component Assumptions. Aug 21, 2018. NHTSA-2018-0067-0005. ANL
Autonomie Model Assumptions Summary. Aug 21, 2018. NHTSA-2018-0067-
1692. ANL BatPac Model 12 55. Aug 21, 2018. Preliminary Regulatory
Impact Analysis (July 2018). Posted July 2018 and updated August 23
and October 16, 2018.
\457\ The CAFE Model is available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system with documentation and all inputs and outputs supporting
today's notice.
---------------------------------------------------------------------------
Both ICCT and Meszler also commented on the availability of
technologies within the Autonomie database, with Meszler stating that
with limited exceptions, technologies were not included in the NPRM
CAFE model if they were not included in the simulation modeling that
underlay the Argonne database, and accordingly if a combination of
technologies was not modeled during the development of the Argonne
database, that package (or combination) of technologies was not
available for adoption in the CAFE model. Meszler stated that these
constraints limited the slate of technologies available to respond to
fuel economy standards, and independently expanding the model to
include additional technologies or technology combinations is not
trivial.
ICCT gave specific examples of key efficiency technologies that it
stated Autonomie did not include, like advanced DEAC, VCR, Miller
Cycle, e-boost, and HCCI. ICCT argued that this was especially
problematic as the agencies appeared to have available engine maps from
IAV on advanced DEAC, VCR, Miller Cycle, E-boost (and from advanced
DEAC, VCR, Miller Cycle, E-boost, HCCI from EPA) that Argonne or the
agencies have been unable to or opted not to include in their modeling.
ICCT stated that the agencies must disclose how Autonomie had been
updated to incorporate ``cutting edge'' 2020-2025 automotive
technologies to ensure they reflect available improvements.\458\
---------------------------------------------------------------------------
\458\ ICCT also made the same request of EPA's ALPHA model, and
the agencies' response to that comment is discussed in Section
VI.C.1 Engine Paths, below.
---------------------------------------------------------------------------
The agencies have updated the final rule analysis to include
additional technologies. In the NPRM, the agencies presented the engine
maps for all of the technologies that ICCT listed, except HCCI, and
sought comment on the engine maps, technical assumptions and the
potential use of the technologies for the final rule analysis. Based on
the available technical information and the ICCT and Meszler comments,
for the final rule analysis, VCR, Miller Cycle (VTG), and e-boost (VTGe
with 48V BISG) technologies have been added and included in the
Autonomie modeling and simulations, and advanced DEAC technology has
been added using fixed point effectiveness estimates in the CAFE model
analysis. The agencies disagree with ICCT's assessment of HCCI and do
not believe it will be available for wide-scale application in the
rulemaking timeframe, and therefore have not included it as a
technology. HCCI technology has been in the research phase for several
decades, and the only production applications to date use a highly-
limited version that restricts HCCI combustion to a very narrow range
of engine operating conditions.459 460 461 Additional
discussion of how Autonomie-modeled and non-modeled technologies are
incorporated into the CAFE Model is located in Section VI.B.3.c),
below.
---------------------------------------------------------------------------
\459\ Mazda introduced Skyactiv-X in Europe with a mild hybrid
technology to assist the engine.
\460\ Mazda News. ``Revolutionary Mazda Skyactiv-X engine
details confirmed as sales start,'' May 6, 2019. https://www.mazda-press.com/eu/news/2019/revolutionary-mazda-skyactiv-x-engine-details-confirmed-as-sales-start/. Last accessed Dec. 2, 2019.
\461\ Confer. K. Kirwan, J. ``Ultra Efficient Light-Duty
Powertrain with Gasoline Low-Temperature Combustion.'' DOE Merit
Review. June 9, 2017. https://www.energy.gov/sites/prod/files/2017/06/f34/acs094_confer_2017_o.pdf. Last accessed Dec. 2, 2019.
---------------------------------------------------------------------------
ICCT and Meszler also commented that the agencies overly limited
the availability of several technologies in the NPRM analysis. In
response, the agencies reconsidered the restrictions that were applied
in the NPRM analysis, and agree with the commenters for several
technologies and technology classes. Many technologies identified by
the commenters are now in production for the MY2017 as well as MY2018
and MY2019. The agencies also think that the baseline fleet compliance
data reflects adoption of many of these technologies. For the final
rule analysis, the agencies have expanded the availability of several
technologies. In the CAFE model, the agencies are now allowing parallel
hybrids (SHEVP2) to be adopted with high compression Atkinson mode
engines (HCR0 and HCR1). In addition, as mentioned above, the Autonomie
full-vehicle modeling included Variable Compression Ratio engine (VCR),
Miller Cycle Engine (VTG), E-boost (VTGe) technologies, and cylinder
deactivation technologies (DEAC) to be applied to turbocharged engines
(TURBO1). As these changes relate to the technology effectiveness
modeling, the CAFE model analysis now includes effectiveness estimates
based on full vehicle simulations for all of these technology
combinations.
We disagree with comments stating the agencies should allow every
technology to be available to every vehicle class.\462\ Discussed
earlier in this section, Autonomie models key aspects of vehicle
operation that are most relevant to assessing fuel economy, vehicle
performance and certain aspects of drivability (like EPA 2-cycle tests,
EPA US06 cycle tests, gradability, low speed acceleration time from 0-
to-60 mph, passing acceleration time from 50 to 80 mph, and number of
transmission shifts). However, there are other critical aspects of
vehicle functionality and operation that the agencies considered beyond
those criteria, that cannot necessarily be reflected in the Autonomie
modeling. For example, a pickup truck can be modeled with a
[[Page 24324]]
continuously variable transmission (CVT) and show improvements on the
2-cycle tests. However, pickup trucks are designed to provide high load
towing utility.\463\ CVTs lack the torque levels needed to provide that
towing utility, and would fail mechanically if subject to high load
towing.\464\ The agencies provided discussions of some of these
technical considerations in the PRIA, and explained why the agencies
had limited technologies for certain vehicle classes, such as limiting
CVTs on pickups as in the example above. These and other limitations
are discussed further in the individual technology sections.
---------------------------------------------------------------------------
\462\ NHTSA-2018-0067-11723. NRDC Attachment2 at p. 4.
\463\ SAE J2807. ``Performance Requirements for Determining Tow-
Vehicle Gross Combination Weight Rating and Trailer Weight Rating.''
Feb. 4, 2016.
\464\ PRIA at p. 223 and 340.
---------------------------------------------------------------------------
The agencies also received a variety of comments that conflated
aspects of the Autonomie models with technology inputs and input
assumptions. For example, commenters expressed concern about the
transmission gear set and final drive values used for the NPRM
analysis, or more specifically, that the gear ratios were held constant
across applications.\465\ In this case, both the inputs (gear set and
final drive ratio) and input assumption (ratios held constant) were
discussed by the commenters. Because these comments are actually about
technology inputs to the Autonomie model, for these and similar cases,
the agencies are addressing the comments in the individual technology
sections which discuss the technology inputs and input assumptions that
impact the effectiveness values for those technologies.
---------------------------------------------------------------------------
\465\ NHTSA-2018-0067-11873. Comments from Roush Industries,
Attachment 1, at p. 14-15. NHTSA-2018-0067-11873. Comments from
CARB, at p.110.
---------------------------------------------------------------------------
For the NPRM analysis, the agencies prioritized using inputs that
were based on data for identifiable technology configurations and that
reflected practical real world constraints. The agencies provided
detailed information on the NPRM analysis inputs and input assumptions
in the NPRM Preamble, PRIA and Argonne model documentation for engine
technologies, transmission technologies, powertrain electrification,
light-weighting, aerodynamic improvements, tire rolling resistance
improvements, and other vehicle technologies. Comments and the
agencies' assessment of comments for each technology are discussed in
the individual technology sections below. Through careful consideration
of the comments, the agencies have updated analytical inputs associated
with several technologies, and as discussed above, have included
several advanced technologies for which technical information was
included in the NPRM. However, for most technologies, the agencies have
determined that the technology inputs and input assumptions that were
used in the NPRM analysis remain reasonable and the best available for
the final rule analysis.
(2) How The Agencies Defined Different Vehicle Types in Autonomie
As described in the NPRM, Argonne produced full-vehicle models and
ran simulations for many combinations of technologies, on many types of
vehicles, but it did not simulate literally every single vehicle model/
configuration in the analysis fleet because it would be impractical to
assemble the requisite detailed information--much of which would likely
only be provided on a confidential basis--specific to each vehicle
model/configuration and because the scale of the simulation effort
would correspondingly increase by orders of magnitude. Instead, Argonne
simulated 10 different vehicle types, corresponding to the five
``technology classes'' generally used in CAFE analysis over the past
several rulemakings, each with two performance levels and corresponding
vehicle technical specifications (e.g., small car, small performance
car, pickup truck, performance pickup truck, etc.).
Technology classes are a means of specifying common technology
input assumptions for vehicles that share similar characteristics.
Because each vehicle technology class has unique characteristics, the
effectiveness of technologies and combinations of technologies is
different for each technology class. Conducting Autonomie simulations
uniquely for each technology class provides a specific set of
simulations and effectiveness data for each technology class. Like the
Draft TAR analysis, there are separate technology classes for compact
cars, midsize cars, small SUVs, large SUVs, and pickup trucks. However,
new for the NPRM analysis and carried into this final rule analysis,
each of those vehicle types has been split into ``low'' (or
``standard'') performance and a ``high'' performance versions, which
represent two classes with similar body styles but different levels of
performance attributes (for a total of 10 technology classes). The
separate technology classes for high performance and low performance
vehicles better account for performance diversity across the fleet.
NHTSA directed Argonne to develop a vehicle assumptions database to
capture vehicle attributes that would comprise the full vehicle models.
For each vehicle technology class, representative vehicle attributes
and characteristics were identified from publicly available information
and automotive benchmarking databases like A2Mac1,\466\ Argonne's
Downloadable Dynamometer Database (D\3\),\467\ and EPA compliance and
fuel economy data,\468\ EPA's guidance on the cold start penalty on 2-
cycle tests.\469\ The resulting vehicle assumptions database consists
of over 100 different attributes like vehicle frontal area, drag
coefficient, fuel tank weight, transmission housing weight,
transmission clutch weight, hybrid vehicle component weights, and
weights for components that comprise engines and electric machines,
tire rolling resistance, transmission gear ratios and final drive
ratio. Each of the 10 different vehicle types was assigned a set of
these baseline attributes and characteristics, to which combinations of
fuel-saving technologies were added as inputs for the Autonomie
simulations. For example, the characteristics of the MY 2016 Honda Fit
were considered along with a wide range of other compact cars to
identify representative characteristics for the Autonomie simulations
for the base compact car technology class. The simulations determined
the fuel economy achieved when applying each combination of
technologies to that vehicle type, given its baseline characteristics.
---------------------------------------------------------------------------
\466\ A2Mac1: Automotive Benchmarking. (Proprietary data).
Retrieved from https://a2mac1.com.
\467\ Downloadable Dynamometer Database (D\3\). ANL Energy
Systems Division. https://www.anl.gov/es/downloadable-dynamometer-database. Last accessed Oct. 31, 2019.
\468\ Data on Cars used for Testing Fuel Economy. EPA Compliance
and Fuel Economy Data. https://www.epa.gov/compliance-and-fuel-economy-data/data-cars-used-testing-fuel-economy. Last accessed Oct.
31, 2019.
\469\ EPA PD TSD at p.2-265--2-266.
---------------------------------------------------------------------------
For each vehicle technology class and for each vehicle attribute,
Argonne estimated the attribute value using statistical distribution
analysis of publicly available data and data obtained from the A2Mac1
benchmarking database.\470\ Some
[[Page 24325]]
vehicle attributes were also based on test data and vehicle
benchmarking, like the cold-start penalty for the FTP test cycle and
vehicle electrical accessories load. The analysis of vehicle attributes
used in the NPRM was discussed in the Argonne model documentation,\471\
and values for each vehicle technology class were provided with the
NPRM for public review.\472\
---------------------------------------------------------------------------
\470\ A2Mac1 is subscription-based benchmarking service that
conducts vehicle and component teardown analyses. Annually, A2Mac1
removes individual components from production vehicles such as oil
pans, electric machines, engines, transmissions, among the many
other components. These components are weighed and documented for
key specifications which is then available to their subscribers.
\471\ NHTSA-2018-0067-0007, at 131. Islam, E., S, Moawad, A.,
Kim, N, Rousseau, A., ``A Detailed Vehicle Simulation Process To
Support CAFE Standards 04262018--Report'' ANL Autonomie
Documentation. Aug 21, 2018.
\472\ NHTSA-2018-0067-0003. ANL Autonomie Summary of Main
Component Assumptions. Aug 21, 2018.
---------------------------------------------------------------------------
The agencies did not believe it was appropriate to assign one
single engine mass for each vehicle technology class in the NPRM
analysis. To account for the difference in weight for different engine
types, Argonne performed a regression analysis of engine peak power
versus weight, based on attribute data taken from the A2Mac1
benchmarking database. For example, to account for weight of different
engine sizes like 4-cylinder versus 8-cylinder, Argonne developed a
relationship curve between peak power and engine weight based on the
A2Mac1 benchmarking data. For the NPRM analysis, this relationship was
used to estimate mass for all engine types regardless of technology
type (e.g., variable valve lift and direct injection). Secondary weight
reduction associated with changes in engine technology was applied by
using this linear relationship between engine power and engine weight
from the A2Mac1 benchmarking database. When a vehicle in the analysis
fleet with an 8-cylinder engine adopted a more fuel efficient 6-
cylinder engine, the total vehicle weight would reflect the updated
engine weight with two less cylinders based on the peak power versus
engine weight relationship. The impact of engine mass reduction on
effectiveness is accounted for directly in the Autonomie simulation
data through the application of the above relationship. Engine mass
reduction through downsizing is, therefore, appropriately not included
as part of vehicle mass reduction technology that is discussed in
Section VI.C.4 because doing so would result in double counting the
impacts. As discussed further below, for the final rule the agencies
improved upon the precision of engine weights by creating two curves to
separately represent naturally aspirated engine designs and
turbocharged engine designs.
In addition, certain attributes were held at constant levels within
each technology class to maintain vehicle functionality, performance
and utility including noise, vibration, and harshness (NVH), safety,
performance and other utilities important for customer satisfaction.
For example, in addition to the vehicle performance constraints
discussed in Section VI.B.3.a)(6), the analysis does not allow the
frontal area of the vehicle to change, in order to maintain utility
like ground clearance, head-room space, and cargo space, and a cold-
start penalty is used to account for fuel economy degradation for
heater performance and emissions system catalyst light-off.\473\ This
allows us to capture the discrete improvement in technology
effectiveness while maintaining vehicle attributes that are important
vehicle utility, consumer acceptance and compliance with criteria
emission standards, and considering these constraints similar to how
manufacturers do in the real world.
---------------------------------------------------------------------------
\473\ The catalyst light-off is the temperature necessary to
initiate the catalytic reaction and this energy is generated from
engine.
---------------------------------------------------------------------------
The agencies sought comment on the analytical approach used to
determine vehicle attributes and characteristics for the Autonomie
modeling. In response, the agencies received a wide variety of comments
on vehicle attributes ranging from discussions of performance increase
from technology adoption (e.g., if a vehicle adopting an electrified
powertrain improved its time to accelerate from 0-60 mph), to comments
on vehicle attributes not modeled in Autonomie, like heated seats and
cargo space.
Toyota and the Alliance commented that the inclusion of performance
vehicle classes addressed the market reality that some consumers will
purchase vehicles for their performance attributes and will accept the
corresponding reduction in fuel economy. Furthermore, Toyota commented
that some gain in performance is more realistic, and that ``dedicating
all powertrain improvements to fuel efficiency is inconsistent with
market reality.'' Toyota ``supports the agencies' inclusion of
performance classes in compliance modeling where a subset of certain
models is defined to have higher performance and a commensurate
reduction in fuel efficiency.'' \474\ Also, in support of the addition
of performance vehicle classes, the Alliance commented that ``vehicle
categories have been increased to 10 to better recognize the range of
0-60 performance characteristics within each of the 5 previous
categories, in recognition of the fact that many vehicles in the
baseline fleet significantly exceeded the previously assumed 0-60
performance metrics. This provides better resolution of the baseline
fleet and more accurate estimates of the benefits of technology.''
\475\
---------------------------------------------------------------------------
\474\ Toyota, Attachment 1, Docket No. NHTSA-2018-0067-12098, at
p. 6.
\475\ Alliance of Automobile Manufacturers, Attachment ``Full
Comment Set,'' Docket No. NHTSA-2018-0067-12073, at p.135.
---------------------------------------------------------------------------
UCS commented that the CAFE model incorporates technology
improvements to each vehicle by applying the effectiveness improvement
of the average vehicle in the technology class, leading to discrete
``stepped'' effectiveness levels for technologies across the different
vehicle types. UCS stated that in contrast, the OMEGA model takes into
account a vehicle's performance characteristics through response-
surface modeling based on relative deviation from the class average
modeled in ALPHA.\476\
---------------------------------------------------------------------------
\476\ NHTSA-2018-0067-12039, at p.24.
---------------------------------------------------------------------------
Although differences between the ALPHA and Autonomie models are
discussed in more detail below, for the NPRM vehicle simulation
analysis the agencies expanded the number of vehicle classes from the
five classes used in the Draft TAR to ten classes, to represent better
the diversity of vehicle characteristics across the fleet. Each of
these ten vehicle technology classes are empirically built from
benchmarking data and other information from various sources, amounting
to hundreds of vehicle characteristics data points to develop each
vehicle class. The agencies expand on these vehicle classes and
characteristics in Section VI.B.3.(a)(2) Vehicle Types in Autonomie and
Section VI.B.3.(a)(3) How Vehicle Models are Built in Autonomie and
Optimized for Simulation. The agencies believe that the real-world data
used to define vehicle characteristics for each of the ten vehicle
classes, in addition to the ten vehicle technology classes themselves,
ensures the analysis reasonably accounts for the diversity in vehicle
characteristics across the fleet.
The agencies believe that UCS's characterization of how technology
improvements are applied in the analysis is a misleading
oversimplification. While the analysis approach in the final rule uses
a representative effectiveness value, the value is not linked solely to
the vehicle technology class, as the UCS implies. The entire technology
combination, or technology key, which includes the vehicle technology
class, is used to
[[Page 24326]]
determine the value for the platform being considered. Within each
vehicle class, the interactions between the added technology and the
full vehicle system (including other technologies and substantial road
load characteristics) are considered in the effectiveness values
calculated for each technology during compliance modeling. As discussed
under each of the technology pathways sections, the effectiveness for
most technologies is reported as a range rather than a single value.
The range exists because the effectiveness for each technology is
adjusted based on the technologies it is coupled with and the major
road load characteristics of the full vehicle system. This approach, in
combination with using the baseline vehicle's initial performance
values as a starting point for performance improvement, results in a
widely variable level of improvement for the system, dependent on
individual vehicle platform characteristics. As a result, the
application of a response-surface approach would likely result in
minimal improvement in accuracy for the Autonomie and CAFE model
analysis approach.
For the final rule analysis, the agencies used the same process to
obtain the vehicle attributes and characteristics for the vehicle
technology classes. Data was acquired from publicly available sources,
Argonne D\3\, EPA compliance and fuel economy data, and A2mac1
benchmarking data. Accordingly, the attributes and characteristics of
the modeled vehicles reflect actual vehicles that meet customer
expectations and automakers' capabilities to manufacture the vehicles.
In addition, for the final rule, the agencies improved the NPRM
analysis by updating some of the attribute values to account for
changes in the fleet. For example, the agencies have updated vehicle
electrical accessory load on the test cycle to reflect higher
electrical loads associated with contemporary vehicle features.
(3) How This Rulemaking Builds Vehicle Models for Autonomie and
Optimize Them for Simulation
Before any simulation is initiated in Autonomie, Argonne must
``build'' a vehicle by assigning reference technologies and initial
attributes to the components of the vehicle model representing each
technology class.\477\ The reference technologies are baseline
technologies that represent the first step on each technology pathway
used in the analysis. For example, a compact car is built by assigning
it a baseline engine, a baseline 6-speed automatic transmission (AT6),
a baseline level of aerodynamic improvement (AERO0), a baseline level
of rolling resistance improvement (ROLL0), a baseline level of mass
reduction technology (MR0), and corresponding attributes from the
Argonne vehicle assumptions database like individual component
weights.\478\ A baseline vehicle will have a unique starting point for
the simulation and a unique set of assigned inputs and attributes,
based on its technology class.
---------------------------------------------------------------------------
\477\ For the NPRM analysis, Chapter 8 Vehicle-Sizing Process in
the ANL Model Documentation had discussed this process in detail.
Further discussion of this process is located in Chapter 8 of the
ANL Model Documentation for this final rule.
\478\ See Section VI.A.7.
---------------------------------------------------------------------------
The next step in the process is to run a powertrain sizing
algorithm that ensures the built vehicle meets or exceeds defined
performance metrics, including low-speed acceleration (i.e., time
required to accelerate from 0-60 mph), high-speed passing acceleration
(time required to accelerate from 50-80 mph), gradeability (e.g. the
ability of the vehicle to maintain constant 65 miles per hour speed on
a six percent upgrade), and towing capacity. Together, these
performance criteria are widely used by industry as metrics to quantify
vehicle performance attributes that consumers observe and that are
important for vehicle utility and customer satisfaction.
In the compact car example used above, the agencies assigned an
initial specific engine design and engine power, transmission, AERO,
ROLL, and MR technologies, and other attributes like vehicle weight. If
the built vehicle does not meet all the performance criteria in the
first iteration, then the engine power is increased to meet the
performance requirement. This increase in power is from higher engine
displacement, which could involve an increase in number of cylinders,
leading to an increase in the engine weight. The iterative process
continues to check whether the compact car with updated engine power,
and corresponding updated engine weight, meets its defined performance
metrics. The loop stops once all the metrics are met, and at this
point, a compact car technology class vehicle model becomes ready for
simulation. For further discussion of the vehicle performance metrics,
see Section VI.B.3.(a).
Autonomie then adopts a single fuel saving technology to the
baseline vehicle model, keeping everything else the same except for
that one technology and the attributes associated with it. For example,
the model would apply an 8-speed automatic transmission in place of the
baseline 6-speed automatic transmission, which would lead to either an
increase or decrease in the total weight of the vehicle based on the
technology class assumptions. At this point, Autonomie confirms whether
performance metrics are met for this new vehicle model through the
previously discussed sizing algorithm. Once a technology has been
assigned to the vehicle model and the resulting vehicle meets its
performance metrics, those vehicle models will be used as inputs to the
full vehicle simulations. So, in the example of the 6-speed to 8-speed
automatic transmission technology update, the agencies now have the
initial ten vehicle models (one for each technology class), plus the
ten new vehicle models with the updated 8-speed automatic transmission,
which adds up to 20 different vehicle models for simulation. This
permutation process is conducted for each of the over 50 technologies
considered, and for all ten technology classes, which results in more
than one million optimized vehicle models.
Figure VI-3 shows the process for building vehicles in Autonomie
for simulation.
[[Page 24327]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.110
Some of the technologies require extra steps for optimization
before the vehicle models are built for simulation; for example, the
sizing and optimization process is more complex for the electrified
vehicles (i.e., HEVs, PHEVs) compared to vehicles with internal
combustion engines, as discussed further, below. Throughout the vehicle
building process, the following items are considered for optimization:
Vehicle weight is decreased or increased in response to
switching from one type of technology to another for the technologies
for which the agencies consider weight, such as different engine and
transmission types;
Vehicle performance is decreased or increased in response
to the addition of mass reduction technologies when switching from one
vehicle model to another vehicle model for the same engine;
Vehicle performance is decreased or increased in response
to the addition of a new technology when switching from one vehicle
model to another vehicle model for the same hybrid electric machine;
and
Electric vehicle battery size is decreased or increased in
response to the addition of mass, aero and/or tire rolling resistance
technologies when switching from one vehicle model to another vehicle
model.
Every time a vehicle adopts a new technology, the vehicle weight is
updated to reflect the new component weight. For some technologies, the
direct weight change is easy to assess. For example, in the NPRM the
agencies designated weights for transmissions so, when a vehicle is
updated to a higher geared transmission, the weight of the original
transmission is replaced with the corresponding transmission weight
(e.g., the weight of a vehicle moving from a 5-speed automatic
transmission to an 8-speed automatic transmission will be updated based
on the 8-speed transmission weight).
For other technologies, like engine technologies, assessing the
updated vehicle weight is much more complex. Discussed earlier,
modeling a change in engine technology involves both the new technology
adoption and a change in power (because the reduction in vehicle weight
leads to lower engine loads, and a resized engine). When a new engine
technology is adopted on a vehicle the agencies account for the
associated weight change to the vehicle based on the earlier discussed
regression analysis of weight versus power. For the NPRM engine weight
regression analysis, the agencies considered 19 different engine
technologies that consisted of unique components to achieve fuel
economy improvements. This regression analysis is technology agnostic
by taking the approach of using engine peak power versus engine weight
because it removed biases to any specific engine technology in the
analysis. Although the agencies do not estimate the specific weight for
each individual engine technology, such as VVT and SGDI, this process
provides a reasonable estimate of the weight differences among engine
technologies.
[[Page 24328]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.111
For the final rule analysis, the agencies used the same process to
assign initial weights to the original 19 engines, plus the added
engines. However, the agencies improved upon precision of the weights
by creating two separate curves separately to represent naturally
aspirated engine designs and turbocharged engine designs.\479\ This
update resulted in two benefits. First, small naturally aspirated 4-
cylinder engines that adopted turbocharging technology reflected the
increased weight of associated components like ducting, clamps, the
turbocharger itself, a charged air cooler, wiring, fasteners, and a
modified exhaust manifold. Second, larger cylinder count engines like
naturally aspirated 8-cylinder and 6-cylinder engines that adopted
turbocharging and downsized technologies would have lower weight due to
having fewer engine cylinders. For example, a naturally aspirated 8-
cylinder engine that adopts turbocharging technology when downsized to
a 6-cylinder turbocharged engine appropriately reflects the added
weight of turbocharging components, and the lower weight of fewer
cylinders.
---------------------------------------------------------------------------
\479\ ANL Model Documentation for the final rule analysis,
Chapter 5.2.9 Engine Weight Determination.
---------------------------------------------------------------------------
As with conventional vehicle models, electrified vehicle models
were built from the ground up. For the NPRM analysis, Argonne used data
from the A2mac1 database and vehicle test data to define different
attributes like weights and power. Argonne used one electric motor
specific power for each type of hybrid and electric vehicle.\480\ For
MY2017, the U.S. market has an expanded number of available hybrid and
electric vehicle models. To capture appropriately the improvements for
electrified vehicles for the final rule analysis, the agencies applied
the same regression analysis process that considers electric motor
weight versus electric motor power for vehicle models that have adopted
electric motors. Benchmarking data for hybrid and electric vehicles
from the A2Mac1 database was analyzed to develop a regression curve of
electric motor peak power versus electric motor weight.\481\
---------------------------------------------------------------------------
\480\ NHTSA-2018-0067-0005. ANL Autonomie Model Assumptions
Summary. Aug 21, 2018. Non_Vehicle_Attributes tab. Specific power
for PS and P2 HEVs was set to 2750 watts/kg, plug-in HEVs were set
to 375 watts/kg, and electric vehicles were set to 1400 watts/kg.
\481\ ANL Model Documentation for the final rule analysis,
Chapter 5.2.10 Electric Machines System Weight.
---------------------------------------------------------------------------
(4) How Autonomie Sizes Powertrains for Full Vehicle Simulation
The agencies maintain performance neutrality of the full vehicle
simulation analysis by resizing engines, electric machines, and hybrid
electric vehicle battery packs at specific incremental technology
steps. To address product complexity and economies of scale, engine
resizing is limited to specific incremental technology changes that
would typically be associated with a major vehicle or engine
redesign.\482\ Manufacturers have repeatedly told the agencies that the
high costs for redesign and the increased manufacturing complexity that
would result from resizing engines for small technology changes
preclude them from doing so. It would be unreasonable and unaffordable
to resize powertrains for every unique combination of technologies, and
exceedingly so for every unique combination of technologies across
every vehicle model due to the extreme manufacturing complexity that
would be required to do so. The agencies reiterated in the NPRM that
the analysis should not include engine resizing with the application of
every technology or for combinations of technologies that drive small
performance changes so that the analysis better reflects what is
feasible for manufacturers.\483\
---------------------------------------------------------------------------
\482\ See 83 FR 43027 (Aug. 24, 2018).
\483\ For instance, a vehicle would not get a modestly bigger
engine if the vehicle comes with floor mats, nor would the vehicle
get a modestly smaller engine without floor mats. This example
demonstrates small levels of mass reduction. If manufacturers
resized engines for small changes, manufacturers would have
dramatically more part complexity, potentially losing economies of
scale.
---------------------------------------------------------------------------
When a powertrain does need to be resized, Autonomie attempts to
mimic manufacturers' development approaches to the extent possible.
Discussed earlier, the Autonomie vehicle building process is initiated
by building a baseline vehicle model with a baseline engine,
transmission, and other baseline vehicle technologies. This baseline
vehicle model (for each technology class) is sized to meet a specific
set of
[[Page 24329]]
performance criteria, including acceleration and gradeability.
The modeling also accounts for the industry practice of platform,
engine, and transmission sharing to manage component complexity and the
associated costs.\484\ At a vehicle refresh cycle, a vehicle may
inherit an already resized powertrain from another vehicle within the
same engine-sharing platform that adopted the powertrain in an earlier
model year. In the Autonomie modeling, when a new vehicle adopts fuel
saving technologies that are inherited, the engine is not resized (the
properties from the baseline reference vehicle are used directly and
unchanged) and there may be a small change in vehicle performance. For
example, in Figure VI-3, Vehicle 2 inherits Eng01 from Vehicle 1 while
updating the transmission. Inheritance of the engine with new
transmission may change performance. This example illustrates how
manufacturers generally manage manufacturing complexity for engines,
transmissions, and electrification technologies.
---------------------------------------------------------------------------
\484\ Ford EcoBoost Engines are shared across ten different
models in MY2019. https://www.ford.com/powertrains/ecoboost/. Last
accessed Nov. 05, 2019.
---------------------------------------------------------------------------
Autonomie implements different powertrain sizing algorithms
depending on the type of powertrain being considered because different
types of powertrains contain different components that must be
optimized.\485\ For example, the conventional powertrain resizing
considers the reference power of the conventional engine (e.g., Eng01,
a basic VVT engine, is rated at 108 kilowatts and this is the starting
reference power for all technology classes) against the power-split
hybrid (SHEVPS) resizing algorithm that must separately optimize engine
power, battery size (energy and power), and electric motor power. An
engine's reference power rating can either increase or decrease
depending on the architecture, vehicle technology class, and whether it
includes other advanced technologies.
---------------------------------------------------------------------------
\485\ ANL Model Documentation for the final rule Analysis,
Chapter 8.3.1 Conventional-Vehicle Sizing Algorithm; Chapter 8.3.2
Split-HEV Sizing Algorithm; 8.3.4 Blended PHEV sizing Algorithm;
8.3.5 Voltec PHEV (Extended Range) Vehicle Sizing Algorithm; Chapter
8.3.6 BEV Sizing Algorithm.
---------------------------------------------------------------------------
Performance requirements also differ depending on the type of
powertrain because vehicles with different powertrain types may need to
meet different criteria. For example, a plug-in hybrid electric vehicle
(PHEV) powertrain that is capable of traveling a certain number of
miles on its battery energy alone (referred to as all-electric range,
or AER, or as performing in electric-only mode) is also sized to ensure
that it can meet the performance requirements of a US06 cycle in
electric-only mode.
The powertrain sizing algorithm is an iterative process that
attempts to optimize individual powertrain components at each step. For
example, the sizing algorithm for conventional powertrains estimates
required power to meet gradeability and acceleration performance and
compares it to the reference engine power for the technology class. If
the power required to meet gradeability and acceleration performance
exceeds the reference engine power, the engine power is updated to the
new value. Similarly, if the reference engine power exceeds the
gradeability and acceleration performance power, it will be decreased
to the lower power rating. As the change in power requires a change
design of the engine, like increasing displacement (e.g., going from a
5.2-liter to 5.6-liter engine, or vice versa) or increasing cylinder
count (e.g., going from an I4 to a V6 or vice versa), the engine weight
will also change. The new engine power is used to update the weight of
the engine.
Next, the conventional powertrain sizing algorithm enters an
acceleration algorithm loop to verify low-speed acceleration
performance (time it takes to go from 0 mph to 60 mph). In this step,
Autonomie adjusts engine power to maintain a performance attribute for
the given technology class and updates engine weight accordingly. Once
the performance criteria are met, Autonomie ends the low-speed
acceleration performance algorithm loop and enters a high-speed
acceleration (time it takes to go from 50 mph to 80 mph) algorithm
loop. Again, Autonomie might need to adjust engine power to maintain a
performance attribute for the given technology, and it exits this loop
once the performance criteria have been met. At this point, the sizing
algorithm is complete for the conventional powertrain based on the
designation for engine type, transmissions type, aero type, mass
reduction technology and low rolling resistance technology.
Figure VI-5 below shows the sizing algorithm for conventional
powertrains.
[[Page 24330]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.112
Depending on the type of powertrain considered, the sizing
algorithms may also size to meet different performance criteria in
different order. The powertrain sizing algorithms for electrified
vehicles are considerably more complex, and are discussed in further
detail in Section VI.C.3, below.
(5) How the Agencies Considered Maintaining Vehicle Attributes
For this rulemaking analysis, consistent with past CAFE and
CO2 rulemakings, the agencies have analyzed technology
pathways manufacturers could use for compliance that attempt to
maintain vehicle attributes, utility, and performance. Using this
approach allows the agencies to assess costs and benefits of potential
standards under a scenario where consumers continue to get the similar
vehicle attributes and features, other than changes in fuel economy.
The purpose of constraining vehicle attributes is to simplify the
analysis and reduce variance in other attributes that consumers value
across the analyzed regulatory alternatives. This allows for a more
streamlined accounting of costs and benefits by not requiring the
values of other vehicle attributes that trade off with fuel economy.
Several examples of vehicle attributes, utility and performance
that could be impacted by adoption of fuel economy improving technology
include the following.
[[Page 24331]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.113
Consequences for the agencies not fully considering or accounting
for potential changes in vehicle attributes, utility, and performance
are degradation in vehicle attributes, utility, and performance that
lead to consumer acceptance issues without accounting for the
corresponding costs and/or not accounting for the costs of technology
designs that maintain vehicle attributes, utility, and performance. The
agencies incorporated changes in the NPRM analysis and that are carried
into this final rule that address deficiencies in past analyses,
including the Draft TAR and Proposed Determination analyses. These
changes were discussed in the NPRM and are repeated in the discussion
of individual technologies in this Preamble, the FRIA, and supporting
documents. The following are several examples of technologies that did
not maintain vehicle attributes, utility, and performance in the Draft
TAR and Proposed Determination analyses.
For the EPA Draft TAR and Proposed Determination analyses, HCR
engine and downsized and turbocharged engine technologies effectiveness
was estimated using Tier 2 certification fuel, which has a higher
octane rating compared to regular octane fuel.486 487 This
does not maintain functionality because consumers would incur higher
costs for using premium fuel in order to achieve the modeled fuel
economy improvements, compared to baseline engines that were replaced,
which operated on lower cost regular octane fuel. By not maintaining
the fuel octane functionality and vehicle attributes, the EPA Draft TAR
and Proposed Determination analyses applied higher effectiveness for
these technologies than could be achieved had regular octane fuel been
assumed for the HCR and downsized turbocharged engines. The Draft TAR
and Proposed Determination analyses also did not account for the higher
costs that would be incurred by consumers to pay for high octane fuel.
These issues were addressed in the
[[Page 24332]]
NPRM and this final rule analysis, and account for some of the
effectiveness and cost differences between the Draft TAR/Proposed
Determination and the NPRM/final rule.\488\
---------------------------------------------------------------------------
\486\ Tier 2 fuel has an octane rating of 93. Typical regular
grade fuel has an octane rating of 87 ((R+M)/2 octane.
\487\ EPA Proposed Determination at 2-209 to 2-212.
\488\ For more details, see Section VI.C.1 Engine Paths.
---------------------------------------------------------------------------
Another example is mass reduction technology. As background, the
agencies characterize mass reduction as either primary mass reduction
or secondary mass reduction. Primary mass reduction involves reducing
mass of components that can be done independently of the mass of other
components. For example, the mass of a hood (e.g., replacing a steel
hood with an aluminum hood) or reducing the mass of a seat are examples
of primary mass reduction because each can be implemented
independently. When there is a significant level of primary mass
reduction, other components that are designed based on the mass of
primary components, may be redesigned and have lower mass. An example
of secondary mass reduction is the brake system. If the mass of primary
components is reduced sufficiently, the resulting lighter weight
vehicle could maintain braking performance and attributes, and safety
with a lighter weight brake system. Mass reduction in the brake system
is secondary mass reduction because it requires primary mass reduction
before it can be incorporated. For the EPA Draft TAR and Proposed
Determination analyses, secondary mass reduction was applied
exclusively based on cost, with no regard to whether sufficient primary
mass reduction was applied concurrently. The analyses did not account
for the degraded functionality of the secondary components and systems
and also understated the costs for lower levels of mass reduction.\489\
These issues were addressed in the NPRM and this final rule analysis,
and account for some of the cost differences between the Draft TAR/
Proposed Determination and the NPRM/final rule.
---------------------------------------------------------------------------
\489\ For more details, see Section VI.C.4 Mass Reduction.
---------------------------------------------------------------------------
The agencies note that for some technologies it is not reasonable
or practicable to match exactly the baseline vehicle's attributes,
utility, and performance. For example, when engines are resized to
maintain acceleration performance, if the agencies applied a criterion
that allowed no shift in performance whatsoever, there would be an
extreme proliferation of unique engine displacements. Manufacturers
have repeatedly and consistently told the agencies that the high costs
for redesign and the increased manufacturing complexity that would
result from resizing engines for small technology changes preclude them
from doing so. It would be unreasonable and unaffordable to resize
powertrains for every unique combination of technologies, and
exceedingly so for every unique combination technologies across every
vehicle model due to the extreme manufacturing complexity that would be
required to do so.\490\ For the NPRM and final rule analyses, engine
resizing is limited to specific incremental technology changes that
would typically be associated with a major vehicle or engine redesign
to address product complexity and economies of scale considerations.
The EPA Draft TAR and Proposed Determination analyses adjusted the
effectiveness of every technology combination assuming performance
could be held constant for every combination, and the analysis did not
recognize or account for the extreme complexity nor the associated
costs for that impractical assumption. The NPRM and final rule analyses
account for these real-world practicalities and constraints, and doing
so explains some of the effectiveness and cost differences between the
Draft TAR/Proposed Determination and the NPRM/final rule.
---------------------------------------------------------------------------
\490\ For more details, see Section VI.B.3.a)(6) Performance
Neutrality.
---------------------------------------------------------------------------
The subsections for individual technologies discuss the technology
assumptions and constraints that were considered to maintain vehicle
attributes, utility, and performance as closely as possible. The
agencies believe that any minimal remaining differences, which may
directionally either improve or degrade vehicle attributes, utility and
performance are small enough to have de minimis impact on the analysis.
(6) How the Agencies Considered Performance Neutrality
The CAFE model examines technologies that can improve fuel economy
and reduce CO2 emissions. An improvement in efficiency can
be realized by improving the powertrain that propels the vehicle (e.g.,
replacing a 6-cylinder engine with a smaller, turbocharged 4-cylinder
engine), or by reducing the vehicle's loads or burdens (e.g., lowering
aerodynamic drag, reducing vehicle mass and/or rolling resistance).
Either way, these changes reduce energy consumption and create a range
of choices for automobile manufacturers. At the two ends of the range,
the manufacturer can choose either:
(A) To design a vehicle that does same the amount of work as before
but uses less fuel.
For example, a redesigned pickup truck would receive a turbocharged
V6 engine in place of the outgoing V8. The pickup would offer no
additional towing capacity, acceleration, larger wheels and tires,
expanded infotainment packages, or customer convenience features, but
would achieve a higher fuel economy rating (and correspondingly lower
CO2 emissions).
(B) To design a vehicle that does more work and uses the same
amount of fuel as before.
For example, a redesigned pickup truck would receive a turbocharged
V6 engine in place of the outgoing V8, but with engine efficiency
improvements that allow the same amount of fuel to do more work. The
pickup would offer improved towing capacity, improved acceleration,
larger wheels and tires, an expanded (heavier) infotainment package,
and more convenience features, while maintaining (not improving) the
fuel economy rating of the previous year's model.
In other words, automakers weigh the trade-offs between vehicle
performance/utility and fuel economy, and they choose a blend of these
attributes to balance meeting fuel economy and emissions standards and
suiting the demands of their customers.
Historically, vehicle performance has improved over the years. The
average horsepower is the highest that it has ever been; all vehicle
types have improved horsepower by at least 49 percent compared to the
1975 model year, and pickup trucks have improved by 141 percent.\491\
Since 1978, the 0-60 acceleration time of vehicles has improved by 39-
47 percent depending on vehicle type.\492\ Also, to gain consumer
acceptance of downsized turbocharged engines, manufacturers have stated
they often offer an increase in performance.\493\ Fuel economy has also
improved, but the horsepower and acceleration trends show that not 100
percent of technological improvements have been applied to fuel
savings. While future trends are uncertain, the past trends suggest
vehicle performance is unlikely to decrease, as it seems reasonable to
assume that customers
[[Page 24333]]
will at a minimum demand vehicles that offer the same utility as
today's fleet.
---------------------------------------------------------------------------
\491\ The 2018 EPA Automotive Trends Report (EPA-420-R-19-002
March 2019) https://www.epa.gov/automotive-trends/download-automotive-trends-report.
\492\ The 2018 EPA Automotive Trends Report (EPA-420-R-19-002
March 2019) https://www.epa.gov/automotive-trends/download-automotive-trends-report.
\493\ Alliance of Automobile Manufacturers, Attachment
``Comment,'' Docket No. EPA-HQ-OAR-2015-0827-4089, at p. 122.
---------------------------------------------------------------------------
For this rulemaking analysis, consistent with past CAFE and
CO2 rulemakings, the agencies have analyzed technology
pathways manufacturers could use for compliance that attempt to
maintain vehicle attributes, utility and performance. NHTSA's analysis
in the Draft TAR used the same approach for performance neutrality as
was used for the NPRM and is being carried into this final rule. This
approach is described throughout this section and further in FRIA
Section VI. For the Draft TAR and Proposed Determination, the EPA
analyses used an approach that maintained 0-60 mph acceleration time
for every technology package. However, that approach did not account
for the added development, manufacturing, assembly and service parts
complexity and associated costs that would be incurred by manufacturers
to produce the substantial number of engine variants that would be
required to achieve those CO2 improvements.\494\ Using the
NPRM approach, which is carried into this final rule, allows the
agencies to assess costs and benefits of potential standards under a
scenario where consumers continue to get the same vehicle attributes
and features, other than changes in fuel economy (approaching the
scenario in example ``A'' above). This approach also eliminates the
need to assess the value of changes in vehicle attributes and features.
As discussed later in this section, while some small level of
performance increase is unavoidable when conducting this type of
analysis, the added technology results almost exclusively in improved
fuel economy. This allows the cost of these technologies to reflect
almost entirely the cost of compliance with standards with nearly
neutral vehicle performance.
---------------------------------------------------------------------------
\494\ Each variant would require a unique engine displacement,
requiring unique internal engine components, such as crankshaft,
connecting rods and others.
---------------------------------------------------------------------------
The CAFE model maintains the initial performance and utility levels
of the analysis vehicle fleet, while considering real world constraints
faced by manufacturers.
To maintain performance neutrality when applying fuel economy
technologies, it is first necessary to characterize the performance
levels of each of the nearly 3000 vehicle models in the MY 2017
baseline fleet. As discussed in Section VI.B.1.b) Assigning Vehicle
Technology Classes, above, each individual vehicle model in the
analysis fleet was assigned to one of ten vehicle ``technology
classes''--the class that is most similar to the vehicle model. The
technology classes include five standard class vehicles (compact car,
midsize car, small SUV, midsize SUV, pickup) plus five ``performance''
versions of these same body styles.\495\ Each vehicle class has a
unique set of attributes and characteristics, including vehicle
performance metrics, that describe the typical characteristics of the
vehicles in that class.
---------------------------------------------------------------------------
\495\ Separate technology classes were created for high
performance and low performance vehicles to better account for
performance diversity across the fleet.
---------------------------------------------------------------------------
The analysis used four criteria to characterize vehicle performance
attributes and utility:
Low-speed acceleration (time required to accelerate from 0-60
mph)
High-speed acceleration (time required to accelerate from 50-
80 mph)
Gradeability (the ability of the vehicle to maintain constant
65 miles per hour speed on a six percent upgrade)
Towing capacity
Low-speed and high-speed acceleration target times are typical of
current production vehicles and range from 6 to 10 seconds depending on
the vehicle class; for example, the midsize SUV performance class has a
low- and high-speed acceleration target of 7 seconds.\496\ The
gradeability criterion requires that the vehicle, given its attributes
of weight, engine power, and transmission gearing, be capable of
maintaining a minimum of 65 mph while going up a six percent grade. The
towing criterion, which is applicable only to the pickup truck and
performance pickup truck vehicle technology classes, is the same as the
gradeability requirement but adds an additional payload/towing mass
(3,000 lbs. for pickups, or 4,350 lbs for performance pickups) to the
vehicle, essentially making the vehicle heavier.
---------------------------------------------------------------------------
\496\ Note, for all vehicle classes, the low and high-speed
acceleration targets use the same value. See section VI.B.1.b)(1)
Assigning Vehicle Technology Classes for a list of low-speed
acceleration target by vehicle technology class.
---------------------------------------------------------------------------
In addition, to maintain the capabilities of certain electrified
vehicles in the 2017 baseline fleet, the analysis required that those
vehicles be capable of achieving the accelerations and speeds of
certain standard driving cycles. The agencies use the US06 ``aggressive
driving'' cycle and the UDDS ``city driving'' cycle to ensure that core
capabilities of BEVs and PHEVs, such as driving certain speeds and/or
distances in electric-only mode, are maintained. In addition to the
four criteria discussed above, the following performance criteria are
applied to these electrified vehicles:
Battery electric vehicles (BEV) are sized to be capable of
completing the US06 ``aggressive driving'' cycle.
Plug-in hybrid vehicles with 50 mile all-electric range
(PHEV50) are sized to be capable of completing the US06 ``aggressive
driving'' cycle in electric-only mode.
Plug-in hybrid vehicles with 20 mile all-electric range
(PHEV20) are sized to be capable of completing the UDDS ``city
driving'' cycle in electric-only (charge depleting) mode.\497\
---------------------------------------------------------------------------
\497\ PHEV20's are blended-type plug-in hybrid vehicles, which
are capable of completing the UDDS cycle in charge depleting mode
without assistance from the engine. However, under higher loads,
this charge depleting mode may use supplemental power from the
engine.
---------------------------------------------------------------------------
Together, these performance criteria are widely used by industry as
metrics to quantify vehicle performance attributes that consumers
observe and that are important for vehicle utility and customer
satisfaction.\498\
---------------------------------------------------------------------------
\498\ Conlon, B., Blohm, T., Harpster, M., Holmes, A. et al.,
``The Next Generation ``Voltec'' Extended Range EV Propulsion
System,'' SAE Int. J. Alt. Power. 4(2):2015, doi:10.4271/2015-01-
1152. Kapadia, J., Kok, D., Jennings, M., Kuang, M., et al.,
``Powersplit or Parallel--Selecting the Right Hybrid Architecture,''
SAE Int. J. Alt. Power. 6(1):2017, doi:10.4271/2017-01-1154. Islam,
E., A. Moawad, N. Kim, and A. Rousseau, 2018a, An Extensive Study on
Vehicle Sizing, Energy Consumption and Cost of Advance Vehicle
Technologies, Report No. ANL/ESD-17/17, Argonne National Laboratory,
Lemont, Ill., Oct 2018.
---------------------------------------------------------------------------
When certain fuel-saving technologies are applied that affect
vehicle performance to a significant extent, such as replacing a pickup
truck's V8 engine with a turbocharged V6 engine, iterative resizing of
the vehicle powertrain (engine, electric motors, and/or battery) is
performed in the Autonomie simulation such that the above performance
criteria is maintained. For example, if the aforementioned engine
replacement caused an improvement in acceleration, the engine may be
iteratively resized until vehicle acceleration performance is shifted
back to the initial target time for that vehicle technology class. For
the low and high-speed acceleration criteria, engine resizing
iterations continued until the acceleration time was within plus or
minus 0.2 seconds of the target time,499 500 which is judged
to balance
[[Page 24334]]
reasonably the precision of engine resizing with the number of
simulation iterations needed to achieve performance within the 0.2
second window, and the associated computer resources and time required
to perform the iterative simulations. Engine resizing is explained
further in Section VI.B.3.a)(4) How Autonomie Sizes Powertrains for
Full Vehicle Simulation and the Argonne Model Documentation for the
final rule analysis.
---------------------------------------------------------------------------
\499\ For example, if a vehicle has a target 0-60 acceleration
time of 6 seconds, a time within 5.8-6.2 seconds was accepted.
\500\ With the exception of a few performance electrified
vehicle types which, based on observations in the marketplace, use
different criteria to maintain vehicle performance without battery
assist. Performance PHEV20, and Performance PHEV50 resize to the
performance of a conventional six-speed automatic (CONV 6AU).
Performance SHEVP2, engines/electric-motors were resized if the 0-60
acceleration time was worse than the target, but not resized if the
acceleration time was better than the target time.
---------------------------------------------------------------------------
The Autonomie simulation resizes until the least capable of the
performance criteria is met, to ensure the pathways do not degrade any
of the vehicle performance metrics. It is possible that as one
criterion target is reached after the application of a specific
technology or technology package, other criteria may be better than
their target values. For example, if the engine size is decreased until
the low speed acceleration target is just met, it is possible that the
resulting engine size would cause high speed acceleration performance
to be better than its target.\501\ Or, a PHEV50 may have an electric
motor and battery appropriately sized to operate in all electric mode
through the repeated accelerations and high speeds in the US06 driving
cycle, but the resulting motor and battery size enables the PHEV50
slightly to over-perform in 0-60 acceleration, which utilizes the power
of both the electric motor and combustion engine.
---------------------------------------------------------------------------
\501\ The Autonomie simulation databases include all of the
estimated performance metrics for each combination of technology as
modeled.
---------------------------------------------------------------------------
To address product complexity and economies of scale, engine
resizing is limited to specific incremental technology changes that
would typically be associated with a major vehicle or engine
redesign.\502\ Manufacturers have repeatedly and consistently told the
agencies that the high costs for redesign and the increased
manufacturing complexity that would result from resizing engines for
small technology changes preclude them from doing so. It would be
unreasonable and unaffordable to resize powertrains for every unique
combination of technologies, and exceedingly so for every unique
combination technologies across every vehicle model due to the extreme
manufacturing complexity that would be required to do so. Engine
displacements are further described in Section VI.C.1 Engine Paths.
---------------------------------------------------------------------------
\502\ See 83 FR 43027 (Aug. 24, 2018).
---------------------------------------------------------------------------
To address this issue, and consistent with past rulemakings, the
NPRM simulation allowed engine resizing when mass reductions of 7.1
percent, 10.7 percent, 14.2 percent (and 20 percent for the final rule
analysis) were applied to the vehicle curb weight,\503\ and when one
powertrain architecture was replaced with another architecture during a
redesign cycle.\504\ At its refresh cycle, a vehicle may also inherit
an already resized powertrain from another vehicle within the same
engine-sharing platform. The analysis did not re-size the engine in
response to adding technologies that have smaller effects on vehicle
performance. For instance, if a vehicle's curb weight is reduced by 3.6
percent (MR1), causing the 0-60 mile per hour time to improve slightly,
the analysis would not resize the engine. The criteria for resizing
used for the analysis better reflects what is feasible for
manufacturers to do.\505\
---------------------------------------------------------------------------
\503\ These correspond, respectively, to reductions of 10%, 15%,
20%, and 28.2% of the vehicle glider mass. For more detail on glider
mass calculation, see section VI.C.4 Mass Reduction.
\504\ Some engine and accessory technologies may be added to an
engine without an engine architecture change. For instance,
manufacturers may adapt, but not replace engine architectures to
include cylinder deactivation, variable valve lift, belt-integrated
starter generators, and other basic technologies. However, switching
from a naturally aspirated engine to a turbo-downsized engine is an
engine architecture change typically associated with a major
redesign and radical change in engine displacement.
\505\ For instance, a vehicle would not get a modestly bigger
engine if the vehicle comes with floor mats, nor would the vehicle
get a modestly smaller engine without floor mats. This example
demonstrates small levels of mass reduction. If manufacturers
resized engines for small changes, manufacturers would have
dramatically more part complexity, potentially losing economies of
scale.
---------------------------------------------------------------------------
Automotive manufacturers have commented that the CAFE model's
consideration of the constraints faced in relation to vehicle
performance and economies of scale are realistic.
Industry associations and individual manufacturers widely supported
the use of the performance metrics used in the NPRM analysis, the use
of standard and higher performance technology classes, and the
representation in the analysis of the real-world manufacturing
complexity constraints and criteria for powertrain redesign.
The Alliance of Automobile Manufacturers (Alliance), Ford, and
Toyota stated that the inclusion of additional performance metrics such
as gradeability are appropriate. Specifically in support of the
gradeability performance criteria, the Alliance commented that
``performance metrics related to vehicle operation in top gear are just
as critical to customer acceptance as are performance metrics such as
0-60 mph times that focus on performance in low-gear ranges.'' \506\
The Alliance also commented specifically on the relationship between
gradeability and downsized engines, stating that as ``engine downsizing
levels increase, top-gear gradeability becomes more and more
important,'' and further that the consideration of gradeability ``helps
prevent the inclusion of small displacement engines that are not
commercially viable and that would artificially inflate fuel savings.''
\507\
---------------------------------------------------------------------------
\506\ Alliance of Automobile Manufacturers, Attachment ``Full
Comment Set,'' Docket No. NHTSA-2018-0067-12073, at 139.
\507\ Alliance of Automobile Manufacturers, Attachment ``Full
Comment Set,'' Docket No. NHTSA-2018-0067-12073, at 135.
---------------------------------------------------------------------------
Ford and Toyota similarly commented in support of the CAFE model's
consideration of multiple performance criteria. Ford stated that this
model ``takes a more realistic approach to performance modeling'' and
``better replicates OEM attribute-balancing practices.'' Ford stated
furthermore that ``OEMs must ensure that each individual performance
measure--and not an overall average--meets its customer's
requirements,'' and that, in contrast, previous analyses did ``not
align with product planning realities.'' \508\ Toyota commented in
support of including gradeability as a performance metric ``to avoid
underpowered engines and overestimated fuel savings.'' \509\
---------------------------------------------------------------------------
\508\ Ford, Attachment 1, Docket No. NHTSA-2018-0067-11928, at
8.
\509\ Toyota, Attachment 1, Docket No. NHTSA-2018-0067-12098, at
6.
---------------------------------------------------------------------------
Toyota and the Alliance commented that the inclusion of performance
vehicle classes addressed the market reality that some consumers will
purchase vehicles for their performance attributes and will accept the
corresponding reduction in fuel economy. Furthermore, Toyota commented
that most consumers consider more than just fuel economy when
purchasing a vehicle, and that ``dedicating all powertrain improvements
to fuel efficiency is inconsistent with market reality.'' Toyota
``supports the agencies' inclusion of performance classes in compliance
modeling where a subset of certain models is defined to have higher
performance and a commensurate reduction in fuel efficiency.'' \510\
Also in support of the addition of performance vehicle classes, the
Alliance commented that ``vehicle categories have been increased to 10
to better recognize the range of 0-60 performance
[[Page 24335]]
characteristics within each of the 5 previous categories, in
recognition of the fact that many vehicles in the baseline fleet
significantly exceeded the previously assumed 0-60 performance metrics.
This provides better resolution of the baseline fleet and more accurate
estimates of the benefits of technology.'' \511\
---------------------------------------------------------------------------
\510\ Toyota, Attachment 1, Docket No. NHTSA-2018-0067-12098, at
6.
\511\ Alliance of Automobile Manufacturers, Attachment ``Full
Comment Set,'' Docket No. NHTSA-2018-0067-12073, at 135.
---------------------------------------------------------------------------
Toyota also commented in support of various real-world
manufacturing complexity constraints employed in the analysis for
powertrain redesigns. Toyota commented that model parameters such as
redesign cycles and engine sharing across vehicle models place a more
realistic limit on the number of engines and transmissions that a
manufacturer is capable of introducing. Toyota also commented in
support of the constraints that the CAFE model placed on engine
resizing, stating that ``there are now more realistic limits placed on
the number of engines and transmissions in a powertrain portfolio which
better recognizes [how] manufacturers must manage limited engineering
resources and control supplier, production, and service costs.
Technology sharing and inheritance between vehicle models tends to
limit the rate of improvement in a manufacturer's fleet.'' Toyota
pointed out that this is in contrast to previous analyses in which
resizing was too unconstrained, which created an ``unmanageable number
of engine configurations within a vehicle platform'' and spawned cases
where ``engine downsizing and power reduction sometimes exceeded limits
beyond basic acceleration requirements needed for vehicle safety and
customer satisfaction.'' \512\
---------------------------------------------------------------------------
\512\ Toyota, Attachment 1, Docket No. NHTSA-2018-0067-12098, at
6.
---------------------------------------------------------------------------
The above comments from the Alliance, Ford, and Toyota support the
methodologies the agencies employed to conduct a performance neutral
analysis. These methodologies helped to ensure that multiple
performance criteria, including gradeability, are all individually
accounted for and maintained when a vehicle powertrain is resized, and
that real-world manufacturing complexity constraints are factored in to
the agencies' analysis of feasible pathways manufacturers could take to
achieve compliance with CAFE standards. The agencies continue to
believe this is a reasonable approach for the aforementioned reasons.
Environmental advocacy groups and CARB criticized the CAFE model's
engine resizing constraints and how they affected the acceleration
performance criteria.
CARB, The International Council on Clean Transportation (ICCT), the
Union of Concerned Scientists (UCS), and the American Council for an
Energy-Efficient Economy (ACEEE) commented that the CAFE model was not
performance neutral, allowing an improvement in performance which
reduced the effectiveness of applied fuel-saving technologies and/or
increased the cost of compliance. Specifically, ACEEE stated that there
appeared to be a shortfall in the fuel economy effectiveness of
technology packages, potentially resulting from the effectiveness being
``consumed'' by additional vehicle performance rather than improvement
of fuel economy. Several of these same commenters conducted analyses
attempting to quantify the magnitude of these changes in vehicle
performance for various vehicle technology classes.
CARB commented on the performance shift of several vehicle types.
Analyzing the 0-60 acceleration for the medium car non-performance
technology class and looking at all cases with resized engines, CARB
claimed that ``effectively half of the simulations resulted in improved
performance.'' \513\ Focusing on electrified vehicles in that same
technology class, CARB stated that ``the data from the Argonne
simulations shows that 76 of the 88 strong electrified packages
(including P2HPV, SHEVPS, BEV, FCEV, PHEV), where Argonne purposely
resized the system to maintain performance neutrality, resulted in
notably faster 0 to 60 mph acceleration times and passing times.''
Specifically regarding parallel hybrid electric vehicles (SHEVP2), CARB
stated that all modeled packages resulted in improved performance.\514\
UCS commented that the NPRM analysis allowed too much change in vehicle
performance, stating that ``while some performance creep may be
reasonable'' many performance values show ``an overlap between
performance and non-performance vehicles'' within the compact car
technology class.\515\
---------------------------------------------------------------------------
\513\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 180. Note that the target acceleration
time for medium car non-performance is in fact 9.0 seconds, as
indicated in ANL documentation, but was incorrectly reported as 9.4s
in NPRM table II-7 in the NPRM.
\514\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 186.
\515\ Union of Concerned Scientists, Attachment 2, Docket No.
NHTSA-2018-0067-12039, at 24.
---------------------------------------------------------------------------
The agencies carefully considered these comments. For the NPRM
analysis, the SHEVP2 engines/electric-motors were resized if the 0-60
acceleration time was worse than the target, but not resized if the
acceleration time was better than the target. This approach maintained
vehicle performance with a depleted battery (without electric assist)
in order to maintain fully the performance and utility characteristics
under all conditions, and improved performance when electric assist was
available (when the battery is not depleted), such as during the 0-60
mph acceleration. The agencies found that this resulted in some
parallel hybrid vehicles having improved 0-60 acceleration times. This
approach was initially chosen for the NPRM because the resulting level
of improved performance was consistent with observations of how
industry had applied SHEVP2 technology. However, in assessing the CARB
comment, the agencies balanced the NPRM approach for SHEVP2 performance
with the agencies' criteria of maintaining vehicle functionality and
performance when technology is applied. Both could not be fully
achieved under all conditions for the case of the SHEVP2.
The agencies concluded it is reasonable to maintain performance
including electric assist when SHEVP2 technology is applied to a
standard (non-performance) vehicle, and therefore the analysis for the
final rule allows upsizing and downsizing of the parallel hybrid
powertrain (SHEVP2) using the 0.2 seconds window around the
target.\516\ For performance vehicles, the agencies concluded that it
remains reasonable to maintain vehicle performance with a depleted
battery (without electric assist) in order to maintain fully the
performance characteristics under all conditions, and continued to use
the NPRM methodology.
---------------------------------------------------------------------------
\516\ To represent marketplace trends better, the performance
class of SHEVP2's allow acceleration time below 0.2 seconds less
than the target, and PHEV20's and PHEV50's inherit combustion engine
size from the conventional powertrain they are replacing. Further
discussion of resizing targets can be found in Chapter 8 of the ANL
Model Documentation for the final rule analysis.
---------------------------------------------------------------------------
The refinement for the standard performance SHEVP2 resolved the
electrified packages issue identified by CARB, and also addressed most
of the change in performance in the overall fleet, including with
compact cars as mentioned by UCS. As explained further below, the
agencies assessed performance among the alternatives for the final rule
analysis. That assessment showed that, with the final rule refinements,
245 out of 255 total resized vehicles (96 percent of vehicles) in the
medium non-performance class (same
[[Page 24336]]
class focused on by CARB), had 0-60 mph acceleration times within the
plus-or-minus 0.2 second window (8.8 to 9.2 seconds).\517\ The only
vehicles outside the window were certain strong electrified vehicles
which exceeded 0-60 the acceleration target as a result of achieving
other performance criteria, such as the US06 driving cycles in all-
electric-mode.\518\
---------------------------------------------------------------------------
\517\ This includes 135 strong electrified vehicles.
\518\ As noted earlier, electrified vehicles had to be capable
of successfully completing UDDS or US06 driving cycles in all-
electric mode, and in some cases the resulting motor size produced
improved acceleration times.
---------------------------------------------------------------------------
The assessment also showed that for the small car class (mentioned
by UCS) the acceleration times of performance and non-performance
vehicles do not go beyond each other's targets. For example, the
vehicle in the small car class with the very best 0-60 mph time and a
conventional powertrain achieves an 8.38 second 0-60 mph time, which is
slower than the performance small car baseline of 8 seconds. This
vehicle had multiple incremental technologies applied, including for
example aerodynamic improvements, and has not reached the threshold for
engine resizing.\519\ After engine resizing, the ``fastest''
conventional small car has a 0-60 mph time of 9.9 seconds, only 0.1
seconds from the target of 10 seconds.\520\
---------------------------------------------------------------------------
\519\ Discussion of engine resizing can be found in Section
VI.B.3.a)(5).
\520\ See NPRM Autonomie simulation database for Small cars,
Docket ID NHTSA-2018-0067-1855.
---------------------------------------------------------------------------
CARB also commented on the improvement of ``passing times,'' or 50-
80 mph high-speed acceleration times. As stated above, an improvement
in one or more of the performance criteria is an expected outcome when
using the rulemaking analysis methodology that resizes powertrains such
that there is no degradation in any of the performance metrics.
Consistent with past rulemakings, the agencies do not believe it is
appropriate for the rulemaking analysis to show pathways that degrade
vehicle performance or utility for one or more of the performance
criteria, as doing so would adversely impact functional capability of
the vehicle and could lead to customer dissatisfaction. The agencies
agree there is very small increase in passing performance for some
technology combinations, and believe this is an appropriate outcome.
High-speed acceleration is rarely the least-capable performance
criteria.
CARB, ICCT, UCS, and H-D Systems (HDS), in an attempt to identify a
potential cause for changes in performance, commented that the CAFE
model should have placed fewer constraints on engine resizing. CARB and
ICCT commented that engine resizing should have been allowed even at
low levels of mass reduction. Comments from CARB, UCS, HDS, and ICCT
stated that engine resizing should also have been allowed for other
incremental technologies, and within their comments they conducted
performance analysis of non-resized cases.
CARB claimed that requiring a minimum of 7.1 percent curb weight
reduction before engine resizing is a constraint that ``limits the
optimization of the technologies being applied.'' \521\ UCS stated that
``a significant share of the benefit of a few percent reduction in mass
has gone towards improved performance rather than improved fuel
economy, leaving a substantial benefit of mass reduction underutilized
and/or uncounted.'' \522\ ICCT also commented that ``when vehicle
lightweighting is deployed at up to a 7 percent mass reduction, the
engine is not resized even though less power would be needed for the
lighter vehicle, meaning any such vehicles inherently are higher
performance.'' \523\
---------------------------------------------------------------------------
\521\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 178. Note, a 7.1% curb weight reduction
equates to the agencies' third level of mass reduction (MR3);
additional discussion of engine resizing for mass reduction can be
found in Section VI.B.3.a)(4) Autonomie Sizes Powertrains for Full
Vehicle Simulation] and in the ANL Model Documentation for the final
rule analysis.
\522\ Union of Concerned Scientists, Attachment 2, Docket No.
NHTSA-2018-0067-12039, at 11.
\523\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-50.
---------------------------------------------------------------------------
UCS and HDS commented on the lack of resizing for technologies
other than mass reduction, with HDS stating that ``the Agencies
incorrectly limited the efficacy of technologies that reduce tractive
load because their modeling does not re-optimize engine performance
after applying these technologies.'' \524\ CARB also commented that the
lack of resizing when a BISG or CISG system is added ``results in a
less than optimized system that does not take full advantage of the
mild hybrid system.'' Similarly, ICCT noted a case in which a Dodge RAM
``did not apply engine downsizing with the BISG system on that truck,
so there are also significant performance benefits that should be
accounted for, meaning that for constant-performance the fuel
consumption reduction would be even greater.'' \525\
---------------------------------------------------------------------------
\524\ H-D Systems, Attachment 1, Docket No. NHTSA-2018-0067-
12395, at 4. For reference, technologies that reduce tractive road
load include mass reduction, aerodynamic drag reduction, and tire
rolling resistance reduction.
\525\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-24.
---------------------------------------------------------------------------
CARB further commented on the performance improvement in cases
without engine resizing by stating that ``94 percent of the packages
modeled result in improved performance,'' and that for these non-
resized cases that were actually adopted by a vehicle in the
simulation, ``fewer than 20 percent maintained baseline performance
with gains of 2 percent or less in acceleration time.'' \526\ Referring
specifically to non-resized electrified vehicles, CARB also stated that
``44,878 of the 53,818 packages, or greater than 83 percent, result in
improved performance.'' \527\ CARB also commented that engine sharing
across different vehicles within a platform, which in some cases may
constrain resizing for a member of that platform, should not dictate
that these engines must remain identical in all aspects, and that
``this overly restrictive sharing of identical engines newly imposed in
the CAFE Model is not consistent with today's industry practices and
results in less optimal engine sizing and causes a systematic
overestimation of technology costs to meet the existing standards.''
\528\
---------------------------------------------------------------------------
\526\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 183.
\527\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 187.
\528\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 185.
---------------------------------------------------------------------------
The agencies note broadly, in response to these comments, that when
conducting an analysis which balances performance neutrality against
the realities faced by manufacturers, such as manufacturing complexity,
economies of scale, and maintaining the full range of performance
criteria, it is inevitable to observe at least some minor shift in
vehicle performance. For example, if a new transmission is applied to a
vehicle, the greater number of gear ratios helps the engine run in its
most efficient range which improves fuel economy, but also helps the
engine to run in the optimal ``power band'' which improves performance.
Thus, the technology can provide both improved fuel economy and
performance. Another example is applying a small amount of mass
reduction that improves both fuel economy and performance by a small
amount. Resizing the engine to maintain performance in these examples
would require a unique engine displacement that is only slightly
different than the baseline engine. While engine resizing in these
incremental cases could have some small benefit to fuel economy, the
[[Page 24337]]
gains may not justify the costs of producing unique niche engines for
each combination of technologies. If manufacturers were to produce
marginally downsized engines to complement every small increment of
mass reduction or technology, the resulting large number of engine
variants that would need to be manufactured would cause a substantial
increase in manufacturing complexity, and require significant changes
to manufacturing and assembly plants and equipment.\529\ The high costs
would be economically infeasible.
---------------------------------------------------------------------------
\529\ For example, each unique engine would require unique
internal components such as crankshafts, pistons, and connecting
rods, as well as unique engine calibrations for each displacement.
Assembly plants would need to stock and feed additional unique
engines to the stations where engines are dressed and inserted into
vehicles.
---------------------------------------------------------------------------
Also, as noted in the NPRM, the 2015 NAS report stated that ``[f]or
small (under 5 percent [of curb weight]) changes in mass, resizing the
engine may not be justified, but as the reduction in mass increases
(greater than 10 percent [of curb weight]), it becomes more important
for certain vehicles to resize the engine and seek secondary mass
reduction opportunities.'' \530\ In consideration of both the NAS
report and comments received from manufacturers, the agencies
determined it would be reasonable to allow allows engine resizing upon
adoption of 7.1 percent, 10.7 percent, 14.2 percent, and 20 percent
curb weight reduction, but not at 3.6 percent and 5.3 percent.\531\
Resizing is also allowed upon changes in powertrain type or the
inheritance of a powertrain from another vehicle in the same platform.
The increments of these higher levels of mass reduction, or complete
powertrain changes, more appropriately match the typical engine
displacement increments that are available in a manufacturer's engine
portfolio.
---------------------------------------------------------------------------
\530\ National Research Council. 2011. Assessment of Fuel
Economy Technologies for Light-Duty Vehicles. Washington, DC--The
National Academies Press. http://nap.edu/12924.
\531\ These curb weight reductions equate to the following
levels of mass reduction as defined in the analysis: MR3, MR4, MR5
and MR6, but not MR1 and MR2; additional discussion of engine
resizing for mass reduction can be found in Section VI.B.3.a)(6)
Autonomie Sizes Powertrains for Full Vehicle Simulation.
---------------------------------------------------------------------------
The agencies point to the comments from manufacturers, discussed
further above, which support the agencies' assertion that the CAFE
model's resizing constraints are appropriate. As discussed previously,
Toyota commented that this approach better considers the constraints of
engineering resources and manufacturing costs and results in a more
realistic number of engines and transmissions.\532\ The Alliance also
commented on the benefit of constraining engine resizing, stating that
``the platform and engine sharing methodology in the model better
replicates reality by making available to each manufacturer only a
finite number of engine displacements, helping to prevent
unrealistically `over-optimized' engine sizing.'' \533\
---------------------------------------------------------------------------
\532\ Toyota, Attachment 1, Docket No. NHTSA-2018-0067-12098, at
6.
\533\ Alliance of Automobile Manufacturers, Attachment ``Full
Comment Set,'' Docket No. NHTSA-2018-0067-12073, at 140.
---------------------------------------------------------------------------
Another comment from CARB stated that engine resizing ``was only
simulated for cases where those levels of mass reduction were applied,
in the absence of virtually all other technology or efficiency
improvements.'' \534\ The agencies do not agree that resizing should be
simulated in all cases which involve small incremental technologies. In
the final rule analysis, vehicles can have engines resized at four (out
of six) levels of mass reduction technology, during a vehicle redesign
cycle which changes powertrain architecture, and by inheritance during
a vehicle refresh cycle. As discussed previously, the application of
small incremental technologies such as reductions in aerodynamic drag
or rolling resistance does not justify the high cost and complexity of
producing additional varieties of engine sizes. Accordingly, for each
curb weight reduction level of 7.1 percent or above and for each
vehicle technology class, Autonomie sized a baseline engine by running
a simulation of a vehicle without incremental technologies applied;
then, those baseline engines were inherited by all other simulations
using the same levels of curb weight reduction, which also added any
variety of incremental technologies.\535\ For further clarification, in
any case in which a vehicle adopts a 7.1 percent or more curb weight
reduction, no matter what other technologies were already present or
are added to the vehicle in conjunction with the mass reduction, that
vehicle will receive an engine which has been appropriately sized for
the newly applied mass reduction level.\536\ This can be observed in
the Autonomie simulation databases by tracking the ``EngineMaxPower''
column (not the ``VehicleSized'' column).
---------------------------------------------------------------------------
\534\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 178.
\535\ In the Autonomie simulation database files, the
simulations which establish baseline sized engines are marked
``yes'' in the ``VehicleSized'' column, and the subsequent
simulations which use this engine and add other incremental
technologies are marked ``inherited.'' For a list of Autonomie
simulation database files, see Table VI-4 Autonomie Simulation
Database Output Files in Section VI.A.7 Structure of Model Inputs
and Outputs.
\536\ For example, if a vehicle possesses MR2, AERO1, and ROLL1
and subsequently adopts MR3, AERO1, ROLL2, the vehicle will adopt
the lower engine power level associated with MR3. As a counter
example, if a vehicle possesses MR3, ROLL1, and AERO1 and
subsequently adopts MR3, ROLL1, AERO2, the engine will not be
resized and it will retain the power level associated with MR3.
---------------------------------------------------------------------------
Finally, ICCT claimed that the agencies did not sufficiently report
performance-related vehicle information. ICCT commented that the output
files did not show data on ``engine displacement, the maximum power of
each engine, the maximum torque of each engine, the initial and final
curb weight of each vehicle (in absolute terms), and estimated 0-60 mph
acceleration.'' ICCT claimed that because this data was not found, the
agencies are ``showing that they have not even attempted to analyze
accurately the future year fleet for their performance'' and that ``the
agencies are intentionally burying a critical assumption, whereby their
future fleet has not been appropriately downsized, and it therefore has
greatly increased utility and performance characteristics.'' \537\
---------------------------------------------------------------------------
\537\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-74.
---------------------------------------------------------------------------
In fact, for the NPRM, and again for this final rule, the agencies
did analyze vehicle performance and have made the data available to the
public. An indication of the actual engine displacement change is
available by noting the displacements used in Automonie simulation
database for each of the technology states. The displacements reported
in Autonomie are used by the full-vehicle-simulation within the
Autonomie model, and while they do not directly represent each specific
vehicle's actual engine sizes, they do fully reflect the relative
change in engine size that is applied to each vehicle. It is the
relative change in engine size that is relevant for the analysis.
Similarly, the vehicle power and torque used by the full vehicle
simulations are reported in the Autonomie simulation databases; their
values and relative change across an engine resizing event can be
observed. Initial and final curb weights for the analysis fleet are
reported in Vehicles Report output file column titled ``CW Initial''
and ``CW,'' respectively. The time required for 0-60 mph acceleration
is reported in the Autonomie simulation database files. A detailed
description of the engine resizing methodology is available in the
Argonne Model
[[Page 24338]]
Documentation, which explains how vehicle characteristics are used to
calculate powertrain size.\538\ These data and information that are
available in the Autonomie and CAFE model documentation provide the
information needed to analyze performance, and in fact, this is
evidenced by the statements of numerous commenters discussed in this
section. The agencies have conducted their own performance analysis,
which is discussed further below, using the same data documentation
mentioned here.
---------------------------------------------------------------------------
\538\ See Chapter 8 of the ANL Model documentation for the final
rule analysis.
---------------------------------------------------------------------------
Updates to the CAFE model have minimized performance shift over the
simulated model years, and have eliminated performance differences
between simulated standards.
The Autonomie simulation updates, discussed previously, were
included in the final rule analysis, and have resulted in average
performance that is similar across the regulatory alternatives. Because
the regulatory analysis compares differences in impacts among the
alternatives, the agencies believe that having consistent performance
across the alternatives is an important aspect of performance
neutrality. If the vehicle fleet had performance gains which varied
significantly depending on the alternative, performance differences
would impact the comparability of the simulations. Using the NPRM CAFE
model data, the agencies analyzed the sales-weighted average 0-60
performance of the entire simulated vehicle fleet for MYs 2016 and
2029, and identified that the Augural standards had 4.7 percent better
0-60 mph acceleration time compared to the NPRM preferred alternative,
which had no changes in standards in MYs 2021-2026.\539\ This
assessment confirmed the observations of the various commenters. With
the refinements that were incorporated for the final rule, similar
analysis showed that the Augural standards had a negligible 0.1 percent
difference in 0-60 mph acceleration time compared to the NPRM preferred
alternative.\540\
---------------------------------------------------------------------------
\539\ The agencies' analysis matched all MY 2016 and MY 2029
vehicles in the NPRM Vehicles Report output file, under both the
Augural standards and preferred alternative, with the appropriate 0-
60 mph acceleration time from the NPRM Autonomie simulation
databases. This was done by examining each vehicle's assigned
technologies, finding the Autonomie simulation with the
corresponding set of technologies, and extracting that simulation's
0-60 mph acceleration time. This process effectively assigned a 0-60
time to every vehicle in the fleet for four scenarios: (1) MY 2016
under augural standards, (2) MY 2016 under the preferred
alternative, (3) MY 2029 under augural standards, and (4) MY 2029
under the preferred alternative. For each scenario, an overall
fleet-wide weighted average 0-60 time was calculated, using each
vehicle's MY2016 sales volumes as the weight. For more information,
see the FRIA Section VI.
\540\ This updated analysis used the FRM CAFE Model Vehicles
Report output file and the FRM Autonomie simulation databases. The
final rule analysis introduced an updated MY 2017 fleet as a
starting point, replacing the NPRM 2016MY fleet. For more
information, see the FRIA Chapter VI.
---------------------------------------------------------------------------
The updates applied to the final rule Autonomie simulations also
resulted in further minimizing the performance change across model
years. As the agencies attempted to minimize this performance shift
occurring ``over time,'' it was also acknowledged that a small increase
would be expected and would be reasonable. This increase is attributed
to the analysis recognizing the practical constraints on the number of
unique engine displacements manufacturers can implement, and therefore
not resizing powertrains for every individual technology and every
combination of technologies when the performance impacts are small.
Perfectly equal performance with 0 percent change would not be
achievable while accounting for these real world resizing constraints.
The performance analysis in the 2011 NAS report shared a similar view
on performance changes, stating that ``truly equal performance involves
nearly equal values . . . within 5 percent.'' \541\ In response to
comments, using NPRM CAFE model data, the agencies analyzed the sales-
weighted average 0-60 performance of the entire simulated vehicle
fleet, and identified that the performance increase from MYs 2016 and
2029 was 7.5 percent under Augural Standards and 3.1 percent under the
NPRM preferred alternative standards. The agencies conducted a similar
analysis using final rule data and found the performance increase over
time from MYs 2017 to 2029 was 3.9 percent for Augural Standards and
4.0 percent for the NPRM preferred alternative standards. The agencies
determined this change in performance is reasonable and note it is
within the 5 percent bound in discussed by NAS in its 2011 report.
---------------------------------------------------------------------------
\541\ National Research Council. 2011. Assessment of Fuel
Economy Technologies for Light-Duty Vehicles. Washington, DC--The
National Academies Press, at 62. http://nap.edu/12924.
---------------------------------------------------------------------------
This assessment shows that for the final rule analysis, performance
is neutral across regulatory alternatives and across the simulated
model years allowing for fair, direct comparison among the
alternatives.
(7) How the Agencies Simulated Vehicle Models on Test Cycles
After vehicle models are built for every combination of
technologies and vehicle classes represented in the analysis, Autonomie
simulates their performance on test cycles to calculate the
effectiveness improvement of the fuel-economy-improving technologies
that have been added to the vehicle. Discussed earlier, the agencies
minimize the impact of potential variation in determining effectiveness
by using a series of tests and procedures specified by federal law and
regulations under controlled conditions.
Autonomie simulates vehicles in a very similar process as the test
procedures and energy consumption calculations that manufacturers must
use for CAFE and CO2 compliance.542 543 544
Argonne simulated each vehicle model on several test procedures to
evaluate effectiveness. For vehicles with conventional powertrains and
micro hybrids, Autonomie simulates the vehicles on EPA 2-cycle test
procedures and guidelines.\545\ For mild and full hybrid electric
vehicles and FCVs, Autonomie simulates the vehicles using the same EPA
2-cycle test procedure and guidelines, and the drive cycles are
repeated until the initial and final state of charge are within a SAE
J1711 tolerance. For PHEVs, Autonomie simulates vehicles in similar
procedures and guidelines as SAE J1711.\546\ For BEVs Autonomie
simulates vehicles in similar procedures and guidelines as SAE
J1634.\547\
---------------------------------------------------------------------------
\542\ EPA, ``How Vehicles are Tested.'' https://www.fueleconomy.gov/feg/how_tested.shtml. Last accessed Nov 14,
2019.
\543\ ANL model documentation for final rule Chapter 6. Test
Procedures and Energy Consumption Calculations.
\544\ EPA Guidance Letter. ``EPA Test Procedures for Electric
Vehicles and Plug-in Hybrids.'' Nov. 14, 2017. https://www.fueleconomy.gov/feg/pdfs/EPA%20test%20procedure%20for%20EVs-PHEVs-11-14-2017.pdf. Last accessed Nov. 7, 2019.
\545\ 40 CFR part 600.
\546\ PHEV testing is broken into several phased based on SAE
J1711. Charge-Sustaining on the City cycle, Charge-Sustaining on the
HWFET cycle, Charge-Depleting on the City and HWFET cycles.
\547\ SAE J1634. ``Battery Electric Vehicle Energy Consumption
and Range Test Procedure.'' July 12, 2017.
---------------------------------------------------------------------------
b) Selection of One Full-Vehicle Modeling and Simulation Tool
The NPRM described tools that the agencies previously used to
estimate technology effectiveness. For the analysis supporting the 2012
final rule for MYs 2017 and beyond, the agencies used technology
effectiveness estimates from EPA's lumped parameter model (LPM). The
LPM was calibrated using data from vehicle simulation work performed by
Ricardo Engineering.\548\
[[Page 24339]]
The agencies also used full vehicle simulation modeling data from
Autonomie vehicle simulations performed by Argonne for mild hybrid and
advanced transmission effectiveness estimates.549 550
---------------------------------------------------------------------------
\548\ Response to Peer Review of: Ricardo Computer Simulation of
Light-Duty Vehicle Technologies for Greenhouse Gas Emission
Reduction in the 2020-2025 Timeframe, EPA-420-R-11-021 (December
2011), available at https://nepis.epa.gov/Exe/ZyPDF.cgi/P100D5BX.PDF?Dockey=P100D5BX.PDF.
\549\ Joint TSD: Final Rulemaking for 2017-2025 Light-Duty
Vehicle Greenhouse Emission Standards and Corporate Average Fuel
Economy Standards. August 2012. EPA-420-R-12-901.3.3.1.3 Argonne
National Laboratory Simulation Study p. 3-69.
\550\ Moawad, A. and Rousseau, A., ``Impact of Electric Drive
Vehicle Technologies on Fuel Efficiency,'' Energy Systems Division,
Argonne National Laboratory, ANL/ESD/12-7, August 2012.
---------------------------------------------------------------------------
For the 2016 Draft TAR analysis, EPA and NHTSA used two different
full system simulation programs for complementary but separate
analyses. NHTSA used Argonne's Autonomie tool, described in detail
above, with engine map inputs developed by IAV using GT-Power in 2014,
and updated in 2016.551 552 553 Argonne, in coordination
with NHTSA, developed a methodology for large scale simulation using
Autonomie and distributed computing, thus overcoming one of the
challenges to full vehicle simulation that the NAS committee outlined
in its 2015 report and implementing a recommendation that the agencies
use full-vehicle simulation to improve the analysis method of
estimating technology effectiveness.\554\ EPA used a limited number of
full-vehicle simulations performed using its ALPHA model, an EPA-
developed full-vehicle simulation model,\555\ to calibrate the LPM,
used to estimate technology effectiveness. EPA also used the same
modeling approach for its Proposed Determination analysis.\556\
---------------------------------------------------------------------------
\551\ GT-Power Engine Simulation Software. https://www.gtisoft.com/gt-suite-applications/propulsion-systems/gt-power-engine-simulation-software/. Last accessed Oct. 10, 2019.
\552\ 2016 Draft TAR Engine Maps by IAV Automotive Engineering
using GT-Power. https://www.nhtsa.gov/staticfiles/rulemaking/pdf/cafe/IAV_EngineMaps_Details.xlsx. Lass accessed Oct. 10, 2019.
\553\ NHTSA-2018-0067-0003. ANL--Summary of Main Component
Performance Assumptions NPRM.
\554\ See National Research Council. 2015. Cost, Effectiveness,
and Deployment of Fuel Economy Technologies for Light-Duty Vehicles.
Washington, DC: The National Academies Press [hereinafter ``2015 NAS
Report''] at p. 263, available at https://www.nap.edu/catalog/21744/cost-effectiveness-and-deployment-of-fuel-economy-technologies-for-light-duty-vehicles (last accessed June 21, 2018). See also A.
Moawad, A. Rousseau, P. Balaprakash, S. Wild, ``Novel Large Scale
Simulation Process to Support DOT's CAFE Modeling System,''
International Journal of Automotive Technology (IJAT), Paper No.
220150349, Nov 2015; Pagerit, S., Sharper, P., Rousseau, A., Sun, Q.
Kropinski, M. Clark, N., Torossian, J., Hellestrand, G., ``Rapid
Partitioning, Automatic Assembly and Multicore Simulation of
Distributed Vehicle Systems.'' ANL, General Motors, EST Embedded
Systems Technology. 2015. https://www.autonomie.net/docs/5%20-%20Presentations/VPPC2015_ppt.pdf. Last accessed Dec. 9, 2019.
\555\ See Lee, B., S. Lee, J. Cherry, A. Neam, J. Sanchez, and
E. Nam. 2013. Development of Advanced Light-Duty Powertrain and
Hybrid Analysis Tool. SAE Technical Paper 2013-01-0808. doi:
10.4271/2013-01-0808.
\556\ Proposed Determination on the Appropriateness of the Model
Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions Standards
under the Midterm Evaluation, EPA-420-R-16-020 (November 2016),
available at https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100Q3DO.pdf; Final Determination on the
Appropriateness of the Model Year 2022-2025 Light-Duty Vehicle
Greenhouse Gas Emissions Standards under the Midterm Evaluation,
EPA-420-R-17-001 (January 2017), available at https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100QQ91.pdf.
---------------------------------------------------------------------------
In the subsequent August 2017 Request for Comment on
Reconsideration of the Final Determination of the Mid-Term Evaluation
of Greenhouse Gas Emissions Standards for MY 2022-2025 Light-Duty
Vehicles, the agencies requested comments on whether EPA should use
alternative methodologies and modeling, including the Autonomie full-
vehicle simulation tool and DOT's CAFE model, for the analysis that
would accompany its revised Final Determination.\557\ As discussed in
the NPRM, stakeholders questioned the efficacy of the combined outputs
and assumptions of the LPM and ALPHA,\558\ especially as the tools were
used to evaluate increasingly heterogeneous combinations of
technologies in the vehicle fleet.\559\
---------------------------------------------------------------------------
\557\ 82 FR 39551 (Aug. 21, 2017).
\558\ 83 FR 43022 (``At NHTSA-2016-0068-0082, p. 49, FCA
provided the following comments, ``FCA believes EPA is
overestimating the benefits of technology. As the LPM is calibrated
to those projections, so too is the LPM too optimistic.'' FCA also
shared the chart, `LPM vs. Actual for 8 Speed Transmissions.' '').
\559\ 83 FR 43022 (referencing Automotive News ``CAFE math gets
trickier as industry innovates'' (Kulisch), March 26, 2018.).
---------------------------------------------------------------------------
More specifically, the Auto Alliance noted that their previous
comments to the midterm evaluation, in addition to comments from
individual manufacturers, highlighted multiple concerns with EPA's
ALPHA model that were unresolved, but addressed in Autonomie.\560\
First, the Alliance expressed concern over ALPHA modeling errors
related to road load reductions, stating that an error derived from how
mass and coast-down coefficients were updated when mass, tire and aero
improvements were made resulted in benefits overstated by 3 percent to
11 percent for all vehicle types. Next, the Alliance repeated its
concern that EPA should consider top-gear gradeability as one of its
performance metrics to maintain functionality, noting that EPA had
acknowledged the industry's comments in the Proposed Determination,
``but generally dismissed the auto industry concerns.'' Additional
analysis by EPA in its Response to Comments document did not allay the
Alliance's concerns,\561\ as the Alliance concluded that ``[c]onsistent
with the National Academy of Sciences recommendation from 2011, EPA
should monitor gradeability to ensure minimum performance.''
---------------------------------------------------------------------------
\560\ EPA-HQ-OAR-2015-0827-9194, at p. 36-44.
\561\ The Alliance noted that in higher-gear-count
transmissions, like 8-speed automatics, modeled by ALPHA with an
expanded ratio spread to achieve fuel economy, are concerning for
gradeability. Additionally, infinite engine downsizing along with
expanded ratio spread transmission, in real world gradeability may
cause further deteriorate as modeled in ALPHA, which leads to
inflated effectiveness values for powertrains that would not meet
customer demands.
---------------------------------------------------------------------------
Furthermore, the Alliance stated that ALPHA vehicle technology
walks provided in response to manufacturer comments on the Proposed
Determination did not correctly predict cumulative effectiveness when
compared to technologies in real world applications. The Alliance
stated that many of the individual technologies and assumptions used by
ALPHA overestimated technology effectiveness and were derived from
questionable sources. As an example, the Alliance referenced an engine
map used by EPA to represent the Honda L15B7 engine, where the engine
map data was collected by ``(1) taking a picture of an SAE document
containing an image of the engine map, and then (2) `digitizing' the
image by `tracing image contours' '' (citing EPA's ALPHA
documentation). The Alliance could not definitively state whether the
``digitization'' process, lack of detail in the source image, or
another factor were the reasons that some regions of overestimated
efficiency were observed in the engine map, but concluded that ``the
use of this map should be discontinued within ALPHA,'' and ``any
analysis conducted with it is highly questionable.'' Based on these
concerns and others, the Alliance recommended that Autonomie be used to
inform the downstream cost optimization models (i.e., the CAFE model
and/or OMEGA).
Global Automakers argued that NHTSA's CAFE model, which
incorporates data from Autonomie simulations, provided a more
transparent and discrete step through each of the modeling
scenarios.\562\ Global pointed out that the LPM is ``of particular
concern due to its simplified technology projection processes,'' and it
``propagates fundamentally flawed
[[Page 24340]]
content into the ALPHA and OMEGA models and therefore cannot accurately
assess the efficacy of fuel economy technologies.'' Global did note
that EPA ``plans to abandon its reliance on LPM in favor of another
modeling approach,'' referring to the RSE,\563\ but stated that ``EPA
must provide stakeholders with adequate time to evaluate the updated
modeling approach, ensure it is analytically robust, and provide
meaningful feedback.'' Global Automakers concluded that EPA's engine
mapping and tear-down analyses have played an important role in
generating publicly-available information, and stated that the data
should be integrated into the Autonomie model.
---------------------------------------------------------------------------
\562\ EPA-HQ-OAR-2015-0827-9728, at 14.
\563\ See Moskalik, A., Bolon, K., Newman, K., and Cherry, J.
``Representing GHG Reduction Technologies in the Future Fleet with
Full Vehicle Simulation,'' SAE Technical Paper 2018-01-1273, 2018,
doi:10.4271/2018-01-1273. Since 2018, EPA has employed vehicle-
class-specific response surface equations automatically generated
from a large number of ALPHA runs to more readily apply large-scale
simulation results, which eliminated the need for manual calibration
of effectiveness values between ALPHA and the LPM.
---------------------------------------------------------------------------
On the other hand, other stakeholders commented that EPA's ALPHA
modeling should continue to be used, for procedural reasons like,
``[i]t would appear arbitrary for EPA now, after five years of modeling
based on ALPHA, to declare it can no longer use its internally
developed modeling tools and must rely solely on the Autonomie model,''
and ``[t]he ALPHA model is inextricably built into the regulatory and
technical process. It will require years of new analysis to replace the
many ALPHA and OMEGA modeling inputs and outputs that permeate the
entire rulemaking process, should EPA suddenly decide to change its
models.'' \564\ Commenters also cited technical reasons to use ALPHA,
like EPA's progress benchmarking and validating the ALPHA model to over
fifteen various MY 2013-2015 vehicles,\565\ and that technologies like
the ``Atkinson 2'' engine technology were not considered in NHTSA's
compliance modeling.\566\ Commenters also cited that ALPHA was created
to be publicly available, open-sourced, and peer-reviewed, ``to allow
for transparency to both automakers and public stakeholders, without
hidden and proprietary aspects that are present in commercial modeling
products.'' \567\
---------------------------------------------------------------------------
\564\ EPA-HQ-OAR-2015-9826, at 39-40.
\565\ EPA-HQ-OAR-2015-9826, at 40.
\566\ EPA-HQ-OAR-2015-9197, at 28.
\567\ EPA-HQ-OAR-2015-9826, at 38.
---------------------------------------------------------------------------
The agencies described in the NPRM that after having reviewed
comments about whether EPA should use alternative methodologies and
modeling, and after having considered the matter fully, the agencies
determined it was reasonable and appropriate to use Autonomie for full-
vehicle simulation.\568\ The agencies stated that nothing in Section
202(a) of the Clean Air Act (CAA) mandated that EPA use any specific
model or set of models for analysis of potential CO2
standards for light duty vehicles. The agencies also distinguished the
models and the inputs used to populate them; specifically, comments
presented as criticisms of the models, such as ``Atkinson 2'' engine
technology not considered in the compliance modeling, actually
concerned model inputs.\569\
---------------------------------------------------------------------------
\568\ 83 FR 43001.
\569\ 83 FR 43002.
---------------------------------------------------------------------------
With regards to modeling technology effectiveness, the agencies
concluded that, although the CAFE model requires no specific approach
to developing effectiveness inputs, the National Academy of Sciences
recommended, and stakeholders have commented, that full-vehicle
simulation provides the best balance between realism and practicality.
As stated above, Argonne has spent several years developing, applying,
and expanding means to use distributed computing to exercise its
Autonomie full-vehicle simulation tool at the scale necessary for
realistic analysis of technologies that could be used to comply with
CAFE and CO2 standards, and this scalability and related
flexibility (in terms of expanding the set of technologies to be
simulated) makes Autonomie well-suited for developing inputs to the
CAFE model.
In response to the NPRM, the Auto Alliance commented that NHTSA's
modeling and analysis tools are superior to EPA's, noting that NHTSA's
tools have had a significant lead in their development.\570\ The
Alliance pointed out that Autonomie was developed from the beginning to
address the complex task of combining two power sources in a hybrid
powertrain, while EPA's ALPHA model had not been validated or used to
simulate hybrid powertrains. While both models are physics-based
forward looking vehicle simulators, the Alliance commented that
Autonomie is fully documented with available training, while ALPHA
``has not been documented with any instructions making it difficult for
users outside of EPA to run and interpret the model.'' The Alliance
also mentioned specific improvements in the Autonomie simulations since
the Draft TAR, including expanded performance classes to better
consider vehicle performance characteristics, the inclusion of
gradeability as a performance metric, as recommended by the NAS, the
inclusion of new fuel economy technologies, and the removal of unproven
technologies.
---------------------------------------------------------------------------
\570\ NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
The Alliance, Global Automakers, and other automakers writing
separately all stated that the agencies should use one simulation and
modeling tool for analysis.571 572 The Alliance stated that
since both the Autonomie and ALPHA modeling systems answer essentially
the same questions, using both systems leads to inconsistencies and
conflicts, and is inefficient and counterproductive.
---------------------------------------------------------------------------
\571\ NHTSA-2018-0067-12073; NHTSA-2018-0067-12032. Comments of
the Association of Global Automakers, Inc. on the Safer Affordable
Fuel-Efficient Vehicles Rule Docket ID Numbers: NHTSA-2018-0067 and
EPA-HQ-OAR-2018-0283 October 26, 2018.
\572\ NHTSA-2018-0067-11943. FCA Comments on The Safer
Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Years 2021-
2026 Passenger Cars and Light Trucks Notice of Proposed Rulemaking.
---------------------------------------------------------------------------
The agencies agree with the Alliance that the fully developed and
validated Autonomie model fulfills the agencies' analytical needs for
full-vehicle modeling and simulation. The agencies also agree that it
is counterintuitive to have two separate models conducting the same
work.
Some commenters stated that broadly, EPA was required to conduct
its own technical analysis and rely on its own models to do so.\573\
Those comments are addressed in Section IV.
---------------------------------------------------------------------------
\573\ NHTSA-2018-0067-12000; NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
Regarding the merits of EPA's models, and based on previous inputs
and assumptions used to populate those models, ICCT commented that
``[b]ased on the ICCT's global analysis of vehicle regulations, the
EPA's physics-based ALPHA modeling offers the most sophisticated and
thorough modeling of the applicable technologies that has ever been
conducted.'' ICCT listed several reasons for this, including that the
EPA modeling is based on systematic modeling of technologies and their
synergies; it was built and improved upon by extensive modeling by and
with Ricardo (an engineering consulting firm); it incorporated National
Academies input at multiple stages; it has included many peer reviews
at many stages of the modeling and the associated technical reports
published by engineers in many technical journal articles and
conference proceedings; and EPA's Draft TAR analysis, which used ALPHA,
used state-of-the-art engine maps based on benchmarked high-efficiency
engines. ICCT concluded
[[Page 24341]]
that ``[d]espite these rigorous advances in vehicle simulation
modeling, it appears that the agencies have inexplicably abandoned this
approach, expressly disregarding the EPA benchmarked engines, ALPHA
modeling, and all its enhancements since the last rulemaking.''
The hallmarks ICCT lists regarding the ALPHA modeling are equally
applicable to Autonomie.\574\ Autonomie is also based on systematic
modeling of technologies and their synergies when combined as packages.
The U.S. Department of Energy created Autonomie, and over the past two
decades, helped to develop and mature the processes and inputs used to
represent real-world vehicles using continuous feedback from the tool's
worldwide user base of vehicle manufacturers, suppliers, government
agencies, and other organizations. Moreover, using Autonomie brings the
agencies closer to the NAS Committee's stated goal of ``full system
simulation modeling for every important technology pathway and for
every vehicle class.'' \575\ While the NAS Committee originally thought
that full vehicle simulation modeling would not be feasible for the
thousands of vehicles in the analysis fleets because the technologies
present on the vehicles might differ from the configurations used in
the simulation modeling,\576\ Argonne has developed a process to
simulate explicitly every important technology pathway for every
vehicle class. Moreover, although separate from the Autonomie model
itself, the Autonomie modeling for this rulemaking incorporated other
NAS committee recommendations regarding full vehicle simulation inputs
and input assumptions, including using engine-model-generated maps
derived from a validated baseline map in which all parameters except
the new technology of interest are held constant.\577\
---------------------------------------------------------------------------
\574\ See Theo LeSieg, Ten Apples Up On Top! (1961), at 4-32.
\575\ 2015 NAS Report at 358.
\576\ 2015 NAS Report at 359.
\577\ NAS Recommendation 2.1.
---------------------------------------------------------------------------
As discussed further below and in VI.C.1 Engine Paths, this is one
reason why the IAV maps were used instead of the EPA maps, and the
agencies instead referenced EPA's engine maps to corroborate the
Autonomie effectiveness results. The IAV maps are engine-model-
generated maps derived from a validated baseline map in which all
parameters except the new technology of interest are held constant.
While EPA's engine maps benchmarking specific vehicles' engines
incorporate multiple technologies, for example including improvements
in engine friction and reduction in accessory parasitic loads,
comparisons presented in Section VI.C.1 showed that engine maps
developed by IAV, while not exactly the same, are representative of
EPA's engine benchmarking data.
In addition, both ALPHA and Autonomie have been used to support
analyses that have been published in technical journal articles and
conference proceedings, but those analyses differ fundamentally because
of the nature of the tools. ALPHA was developed as a tool to be used by
EPA's in-house experts.\578\ As EPA stated in the ALPHA model peer
review,\579\ ``ALPHA is not intended to be a commercial product or
supported for wide external usage as a development tool.'' \580\
Accordingly, EPA experts have published several peer-reviewed journal
articles using ALPHA and have presented the results of those papers at
conference proceedings.\581\
---------------------------------------------------------------------------
\578\ ALPHA Peer Review, at 4-1.
\579\ ICCT's comments intimate that ALPHA has been peer reviewed
at many stages of the modeling; although EPA has published several
peer-reviewed technical papers, the ALPHA model itself has been
subject to one peer review. See Peer Review of ALPHA Full Vehicle
Simulation Model, available at https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf.
\580\ ALPHA Peer Review, at 4-2.
\581\ See, e.g., Dekraker, P., Kargul, J., Moskalik, A., Newman,
K. et al., ``Fleet-Level Modeling of Real World Factors Influencing
Greenhouse Gas Emission Simulation in ALPHA,'' SAE Int. J. Fuels
Lubr. 10(1):2017, doi:10.4271/2017-01-0899.
---------------------------------------------------------------------------
To explore ICCT's comments on the importance of peer review
further, it is important to take the actual substantive content of the
ALPHA peer review into account.\582\ One reviewer raised significant
questions over the availability of ALPHA documentation, stating
``[t]here is an overall lack of detail on key technical features that
are new in the model,'' and ``[w]e were not able to find any
information on how the model handles component weight changes.''
Reviewers also raised questions related to model readiness, stating
``[a]ccording to the documentation review, ALPHA's stop/start modeling
appears to be very simplistic.'' Moreover, when running ALPHA
simulations, the reviewer noted the results ``strongly suggest that the
model has errors in the underlying equations or coding with respect to
all of the load reductions.'' Also, one reviewer said the following of
ALPHA: ``A specific simulation runtime is significantly high, more than
10 mins. without providing any indication to the user progress made so
far. A fairly more complicated model such as Autonomie available even
with enhanced capabilities is significantly faster today.'' \583\
---------------------------------------------------------------------------
\582\ EPA. ``Peer Review of ALPHA Full Vehicle Simulation
Model.'' EPA-420-R-16-013. October 2016. https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf. Last accessed Nov 18, 2019.
\583\ Peer Review of ALPHA Full Vehicle Simulation Model, at C-
4, available at https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf.
---------------------------------------------------------------------------
The peer reviewer's assessment of Autonomie as a more complicated
model with enhanced capabilities is not surprising, given Autonomie's
history of development. Autonomie is a commercial tool with more than
275 worldwide organizational users, including vehicle manufacturers,
suppliers, government agencies, and nonprofit organizations having
licensed and used Autonomie. Both Autonomie's creators and user base
unaffiliated with Argonne have published over 100 papers, including
peer-reviewed papers in journals, related to Autonomie validation and
other studies.584 585 One could even argue that the tool has
been continuously peer reviewed by these thousands of experts over the
past two decades.
---------------------------------------------------------------------------
\584\ At least 15 peer-reviewed papers authored by ANL experts
have been referenced throughout this Section, and others can be
found at SAE International's website, https://www.sae.org/, using
the search bar for ``Autonomie.''
\585\ See, e.g., Haupt, T., Henley, G., Card, A., Mazzola, M. et
al., ``Near Automatic Translation of Autonomie-Based Power Train
Architectures for Multi-Physics Simulations Using High Performance
Computing,'' SAE Int. J. Commer. Veh. 10(2):483-488, 2017, https://doi.org/10.4271/2017-01-0267; Samadani, E., Lo, J., Fowler, M.,
Fraser, R. et al., ``Impact of Temperature on the A123 Li-Ion
Battery Performance and Hybrid Electric Vehicle Range,'' SAE
Technical Paper 2013-01-1521, 2013, https://doi.org/10.4271/2013-01-1521.
---------------------------------------------------------------------------
In fact, in responding to a peer review comment on the ALPHA
model's underlying equations and coding with respect to road load
reductions, EPA noted that Autonomie had been used as a reference
system simulation tool to validate ALPHA model results.\586\
---------------------------------------------------------------------------
\586\ Peer Review of ALPHA Full Vehicle Simulation Model, at 4-
14 and 4-15, available at https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf.
---------------------------------------------------------------------------
Outside of formal peer-reviewed studies, Autonomie has been used by
organizations like ICCT to support policy documents, position briefs,
and white papers assessing the potential of future efficiency
technologies to meet potential regulatory requirements,\587\
[[Page 24342]]
just as the agencies did in this rulemaking.
---------------------------------------------------------------------------
\587\ See, e.g., Oscar Delgado and Nic Lutsey, Advanced Tractor-
Trailer Efficiency Technology Potential in the 2020-2030 Timeframe
(April 2015), available at https://theicct.org/sites/default/files/publications/ICCT_ATTEST_20150420.pdf; Ben Sharpe, Cost-
Effectiveness of Engine Technologies for a Potential Heavy-Duty
Vehicle Fuel Efficiency Regulation in India (June 2015), available
at https://theicct.org/sites/default/files/publications/ICCT_position-brief_HDVenginetech-India_jun2015.pdf; Ben Sharpe and
Oscar Delgado, Engines and tires as technology areas for efficiency
improvements for trucks and buses in India (working paper published
March 2016), available at https://theicct.org/sites/default/files/publications/ICCT_HDV-engines-tires_India_20160314.pdf.
---------------------------------------------------------------------------
Similarly to ICCT, UCS stated that in contrast to Autonomie, ALPHA
had been thoroughly peer-reviewed and is constantly being updated to
reflect the latest technology developments based on work performed by
the National Vehicle and Fuel Emissions Laboratory.\588\ UCS also
stated that because EPA has direct control over the model and its
interface to OMEGA, EPA can better ensure that the inputs into OMEGA
reflect the most up-to-date data, unlike the Autonomie work, which
effectively has to be ``locked in'' before it can be deployed in the
CAFE model. UCS also stated that ALPHA is based on the GEM model (used
to simulate compliance with heavy-duty vehicle regulations) which was
been updated with feedback from heavy-duty vehicle manufacturers and
suppliers, and in fact, ``NHTSA has such confidence in the GEM model
that they accept its simulation-based results as compliance with the
heavy-duty fuel economy regulations.''
---------------------------------------------------------------------------
\588\ NHTSA-2018-0067-12039 (UCS).
---------------------------------------------------------------------------
Again, the agencies believe that it is important to note that
Autonomie not only meets, but also exceeds, UCS' listed metrics.
Autonomie's models, sub-models, and controls are constantly being
updated to reflect the latest technology developments based on work
performed by Argonne National Laboratory's Advanced Mobility Technology
Laboratory (AMTL) (formerly Advanced Powertrain Research Facility, or
ARPF).589 590 The Autonomie validation has included nine
validation studies with accompanying reports for software, six
validation studies and reports for powertrains, nine validation studies
and reports for advanced components, ten validation studies and reports
for advanced controls, and overall model validation using test data
from over 50 vehicles.\591\
---------------------------------------------------------------------------
\589\ See NPRM PRIA. The agencies cited a succinctly-summarized
presentation of Autonomie vehicle validation procedures based on
AMTL test data in the NPRM ANL modeling documentation and PRIA
docket for stakeholders to review at NHTSA-2018-0067-1972 and NHTSA-
2018-0067-0007.
\590\ Jeong, J., Kim, N., Stutenberg, K., Rousseau, A.,
``Analysis and Model Validation of the Toyota Prius Prime,'' SAE
2019-01-0369, SAE World Congress, Detroit, April 2019; Kim, N,
Jeong, J., Rousseau, A. & Lohse-Busch, H. ``Control Analysis and
Thermal Model Development of PHEV,'' SAE 2015-01-1157, SAE World
Congress, Detroit, April 15; Kim, N., Rousseau, A. & Lohse-Busch, H.
``Advanced Automatic Transmission Model Validation Using Dynamometer
Test Data,'' SAE 2014-01-1778, SAE World Congress, Detroit, Apr.
14.; Lee, D. Rousseau, A. & Rask, E. ``Development and Validation of
the Ford Focus BEV Vehicle Model,'' 2014-01-1809, SAE World
Congress, Detroit, Apr. 14; Kim, N., Kim, N., Rousseau, A., & Duoba,
M. ``Validating Volt PHEV Model with Dynamometer Test Data using
Autonomie,'' SAE 2013-01-1458, SAE World Congress, Detroit, Apr.
13.; Kim, N., Rousseau, A., & Rask, E. ``Autonomie Model Validation
with Test Data for 2010 Toyota Prius,'' SAE 2012-01-1040, SAE World
Congress, Detroit, Apr. 12; Karbowski, D., Rousseau, A, Pagerit, S.,
& Sharer, P. ``Plug-in Vehicle Control Strategy--From Global
Optimization to Real Time Application,'' 22th International Electric
Vehicle Symposium (EVS22), Yokohama, (October 2006).
\591\ Rousseau, A. Moawad, A. Kim, Namdoo. ``Vehicle System
Simulation to Support NHTSA CAFE standards for the Draft Tar.''
https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/anl-nhtsa-workshop-vehicle-system-simulation.pdf. Last accessed Nov 20, 2019.
---------------------------------------------------------------------------
In fact, using Autonomie, which has validated data based on test
data from over 50 vehicles, alleviates other stakeholder concerns about
the level of model validation in past analyses. For example, Global
Automakers expressed concerns about whether the effectiveness values
used in past EPA analysis, generated from ALPHA full-vehicle model
simulations, were properly validated, stating that ``[a]lthough EPA
claims that the LPM was calibrated based on thorough testing and
modeling with the ALPHA model, the materials provided with the Proposed
and Final Determination only cover 18 percent of the projected vehicle
fleet with regards to specific combinations of powertrain technology
presented by EPA in the MY 2025 OMEGA pathway. It is unclear how EPA
calibrated the LPM for the remaining 82 percent of the projected
vehicles. EPA's failure to publicly share the data for such a large
percentage of vehicles raises questions about the quality of data.''
\592\ While simple modeled parameters like single dimensional linear
systems, such as engine dynamometer torque measurements can be
validated through other models,\593\ full vehicle systems are complex
multi-dimensional non-linear systems that need to be developed with
multiple data sets, and validated with other fully independent data
sets. Autonomie's models and sub-models have undergone extensive
validation that has proven the models' agreement with empirical data
and the principles of physics.
---------------------------------------------------------------------------
\592\ Docket ID EPA-HQ-OAR-2015-0827-9728. Global later repeated
that ``only 18% of all vehicle data used as inputs to the ALPHA
modeling was made available in the EPA's public sources. Additional
data had to be specifically requested subsequent to the publication
of the Draft TAR and Proposed Determination. This lack of publicly
available data highlights transparency concerns, which Global
Automakers has raised on several previous occasions.''
\593\ Section 89.307 Dynamometer calibration.
---------------------------------------------------------------------------
In addition, the agencies disagree with UCS' comment that EPA's
direct control over its effectiveness modeling and interface to OMEGA
results in a more up-to-date analysis. Argonne's participation in
developing inputs for the rulemaking analysis allowed the agencies
access to vehicle benchmarking data from more vehicles than if the
agencies were limited by their own resources, and access to the Argonne
staff's extensive experience based on direct coordination with vehicle
manufacturers, suppliers, and researchers that all actively use
Autonomie for their own work. In addition to Autonomie's continuous
updates to incorporate the latest fuel-economy-improving technologies,
discussed throughout this section, the data supplied to and generated
by Autonomie for use in the CAFE model was continuously updated during
the analysis process. This is just one part of the iterative quality
assurance (QA) and quality check (QC) process that the agencies
developed when Argonne's large-scale simulation modeling based in
Autonomie was first used for the Draft TAR.
In addition to Argonne's team constantly updating Autonomie,
Argonne's use of high performance computing (HPC) allowed for constant
update of the analysis during the rulemaking process. Argonne's HPC
platform allows a full set of simulations--over 750,000 modeled
vehicles that incorporate over 50 different fuel-economy-improving
technologies--to be simulated in one week. Subsets of the simulations
can be re-run should issues come up during QA/QC in a day or less.
Tools like the internet and high performance computers have allowed the
agencies to evaluate technology effectiveness with up-to-date inputs
without the proximity of the computers and the people running them
working as a detriment the analysis.
Finally, GEM, ALPHA, and Autonomie were all developed in the MATLAB
computational environment as forward-looking physics-based vehicle
models. Just as ALPHA has roots in GEM, created in 2010 to accompany
the agencies' heavy-duty vehicle CO2 emissions and fuel
consumption standards, Autonomie has its origins in the software PSAT,
developed over 20 years ago. While this information is useful, as
implied by UCS' comment, the origin of the software was less important
than the capabilities the software could provide for today's analysis.
NHTSA's acceptance of GEM
[[Page 24343]]
results for compliance with heavy-duty fuel economy regulations had no
bearing on the decision to use Autonomie to assess the effectiveness of
light-duty fuel economy and CO2 improving technologies. GEM
was developed to serve as the compliance model for heavy-duty
vehicles,\594\ and GEM serves that limited scope very well.
---------------------------------------------------------------------------
\594\ Newman, K., Dekraker, P., Zhang, H., Sanchez, J. et al.,
``Development of Greenhouse Gas Emissions Model (GEM) for Heavy- and
Medium-Duty Vehicle Compliance,'' SAE Int. J. Commer. Veh.
8(2):2015, doi:10.4271/2015-01-2771.
---------------------------------------------------------------------------
UCS did comment that full vehicle simulation could significantly
improve the estimates of technology effectiveness, but thought it
critical that the process be as open and transparent as possible. UCS
pointed to ALPHA results published in peer-reviewed journals as an
example of how transparency has provided the ALPHA modeling effort with
significant and valuable feedback, and contrasted what they
characterized as Autonomie's ``black box'' approach, which they stated
``does not lend itself to similar dialog, nor does it make it easy to
assess the validity of the results.'' Specifically, UCS stated that it
is ``impossible to verify, replicate, or alter the work done by
Autonomie due to the expensive nature of the tools used and lack of
open source or peer-reviewed output.'' In contrast, UCS stated that
EPA's ALPHA model has been thoroughly peer reviewed, and is readily
``downloadable, editable, and accessible to anyone with a MATLAB
license.''
The agencies responses on the merits of how ALPHA and Autonomie
were peer-reviewed are discussed above. Regarding UCS' comment that it
is impossible to verify, replicate, or alter the work done by
Autonomie, the agencies disagree. All inputs, assumptions, model
documentation--including of component models and individual control
algorithms--and outputs for the NPRM Autonomie modeling were submitted
to the docket for review.\595\ Commenters were able to provide a robust
analysis of Autonomie's technology effectiveness inputs, input
assumptions, and outputs, as shown by their comments on specific
vehicle technology effectiveness assumptions, discussed throughout this
section and in the individual technology sections below.
---------------------------------------------------------------------------
\595\ NHTSA-2018-0067-1855. ANL Autonomie Compact Car Vehicle
Class Results. Aug 21, 2018. NHTSA-2018-0067-1856. ANL Autonomie
Performance Compact Car Vehicle Class Results. Aug 21, 2018. NHTSA-
2018-0067-1494. ANL Autonomie Midsize Car Vehicle Class Results. Aug
21, 2018. NHTSA-2018-0067-1487. ANL Autonomie Performance Pick-Up
Truck Vehicle Class Results. Aug 21, 2018. NHTSA-2018-0067-1663. ANL
Autonomie Performance Midsize Car Vehicle Class Results. Aug 21,
2018. NHTSA-2018-0067-1486. ANL Autonomie Small SUV Vehicle Class
Results. Aug 21, 2018 NHTSA-2018-0067-1662. ANL Autonomie
Performance Midsize SUV Vehicle Class Results. Aug 21, 2018. NHTSA-
2018-0067-1661. ANL Autonomie Pickup Truck Vehicle Class Results.
Aug 21, 2018. NHTSA-2018-0067-1485. ANL Autonomie Small Performance
SUV Vehicle Class Results. Aug 21, 2018 NHTSA-2018-0067-1492. ANL
Autonomie Midsize SUV Vehicle Class Results. Aug. 21, 2018. NHTSA-
2018-0067-0005. ANL Autonomie Model Assumptions Summary. Aug 21,
2018. NHTSA-2018-0067-0003. ANL Autonomie Summary of Main Component
Assumptions. Aug 21, 2018. NHTSA-2018-0067-0007. Islam, E. S,
Moawad, A., Kim, N, Rousseau, A. ``A Detailed Vehicle Simulation
Process To Support CAFE Standards 04262018--Report'' ANL Autonomie
Documentation. Aug 21, 2018. NHTSA-2018-0067-0004. ANL Autonomie
Data Dictionary. Aug 21, 2018. NHTSA-2018-0067-1692. ANL BatPac
Model 12 55. Aug 21, 2018. NHTSA-2018-0067-12299. Preliminary
Regulatory Impact Analysis (July 2018). Posted July 2018 and updated
August 23 and October 16, 2018.
---------------------------------------------------------------------------
The agencies also disagree with UCS' assessment of Autonomie as
``expensive.'' While Autonomie is a commercial product, the biggest
financial barrier to entry for both ALPHA and Autonomie is the same: A
MathWorks license.596 597 Regardless, Argonne has made the
version of Autonomie used for this final rule analysis available upon
request, including the individual runs used to generate each technology
effectiveness estimate.\598\
---------------------------------------------------------------------------
\596\ Autonomie. Frequently Asked Questions. ``Which version of
matlab can I use?'' https://www.autonomie.net/faq.html#faq2. Last
accessed Nov. 19, 2019.
\597\ EPA ALPHA v2.2 Technology Walk Samples. ``Running this
version of ALPHA requires Matlab/Simulink with StateFlow 2016b.''
https://www.epa.gov/regulations-emissions-vehicles-and-engines/advanced-light-duty-powertrain-and-hybrid-analysis-alpha.
\598\ Argonne Nationally Laboratory. Autonomie License
Information. https://www.autonomie.net/asp/LicenseRequest.aspx. Last
accessed Nov, 18, 2019.
---------------------------------------------------------------------------
Next, ICCT supplanted its statement that the agencies
``inexplicably'' abandoned ALPHA, commenting that the agencies'
explanation and justification for relying on Autonomie rather than
ALPHA failed to discuss ALPHA in detail, and the agencies did not
compare and contrast the two models. ICCT continued, ``the EPA cannot
select its modeling tool arbitrarily, yet it appeared that the EPA has
whimsically shifted from an extremely well-vetted, up-to-date,
industry-grade modeling tool to a less-vetted, academic-grade framework
with outdated inputs without even attempt to scrutinize the change.''
ICCT also stated that the agencies are legally obligated to acknowledge
and explain when they change position, and ``cannot simply ignore that
EPA previously concluded that the ALPHA modeling accurately projected
real-world effects of technologies and technology packages.''
The agencies disagree that a more in-depth discussion of ALPHA was
required in the NPRM. In acknowledging the transition to using
Autonomie for effectiveness modeling and the CAFE model for analysis of
regulatory alternatives,\599\ the agencies described several analytical
needs that using a single analysis from the CAFE model--with inputs
from the Autonomie tool--addressed. These included that Autonomie
produced realistic estimates of fuel economy levels and CO2
emission rates through consideration of real-world constraints, such as
the estimation and consideration of performance, utility, and
drivability metrics (e.g., towing capability, shift busyness, frequency
of engine on/off transitions).\600\ That EPA previously concluded the
ALPHA modeling accurately projected real-world effects of technologies
and technology packages has no bearing on Autonomie's ability to
fulfill the analytical needs that the agencies articulated in the NPRM,
including that Autonomie also accurately projects real-world effects of
technologies and technology packages.
---------------------------------------------------------------------------
\599\ 83 FR 43000 (Aug. 24, 2018).
\600\ 83 FR 43001 (Aug. 24, 2018).
---------------------------------------------------------------------------
The agencies also disagree with ICCT's characterization of ALPHA as
``an extremely well-vetted, up-to-date, industry-grade modeling tool''
and Autonomie as a ``less-vetted, academic-grade framework with
outdated inputs.'' Again, Autonomie has been used by government
agencies, vehicle manufacturers (and by agencies and manufacturers
together in the collaborative government-industry partnership U.S.
DRIVE program), suppliers, and other organizations because of its
ability to simulate many powertrain configurations, component
technologies, and vehicle-level controls over numerous drive cycles.
Characterizing ALPHA as an ``industry-grade modeling tool'' contravenes
EPA's own description of its tool--an in-house vehicle simulation model
used by EPA, not intended to be a commercial product.\601\
---------------------------------------------------------------------------
\601\ See, e.g., Overview of ALPHA Model, https://www.epa.gov/regulations-emissions-vehicles-and-engines/advanced-light-duty-powertrain-and-hybrid-analysis-alpha; ALPHA Effectiveness Modeling:
Current and Future Light-Duty Vehicle & Powertrain Technologies
(Jan. 20, 2016), available at https://www.epa.gov/sites/production/files/2016-10/documents/alpha-model-sae-govt-ind-mtg-2016-01-20.pdf
(``ALPHA is not a commercial product (e.g. there are no user
manuals, tech support hotlines, graphical user interfaces, or full
libraries of components).''). See also Peer Review of ALPHA Full
Vehicle Simulation Model, available at https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf. While ALPHA peer reviewers found the
model to be a ``fairly simple transparent model . . . [t]he model
execution requires an expert MatLab/Simulink user since no user-
friendly interface currently exists.'' Indeed, EPA noted in response
to this comment that ``[a]s with any internal tool, EPA does not
have the need for a ``user-friendly interface'' like one that would
normally accompany a commercial product which is available for
purchase and fully supported for wide external usage.''
---------------------------------------------------------------------------
[[Page 24344]]
That characterization also contravenes documentation from the
automotive industry indicating that manufacturers consider ALPHA to
generate overly optimistic effectiveness values, to be unrepresentative
of real-world constraints, and a difficult tool to
use.602 603 The Alliance commented to the MTE
reconsideration that ``[p]revious comments from the Alliance and
individual manufacturers to the MTE docket have highlighted multiple
concerns with EPA's ALPHA model. Many of these concerns remain
unresolved.'' \604\ Furthermore, the Alliance commented that ALPHA
``has not been documented with any instructions making it difficult for
users outside of EPA to run and interpret the model.'' \605\ Global
Automakers further stated that the ``lack of publicly available data
[related to inputs used in the ALPHA modeling] highlights transparency
concerns, which Global Automakers has raised on several previous
occasions.'' \606\ In fact, both the Alliance of Automobile
Manufacturers and Global Automakers, the two trade organizations that
represent the automotive industry, concluded that Autonomie should be
used to generate effectiveness inputs for the CAFE model.\607\
---------------------------------------------------------------------------
\602\ See EPA-HQ-OAR-2015-0827-10125, at 7. As part of their
assessment that known technologies could not meet the original MY
2022-2025 standards, Toyota noted that the ALPHA conversion of
Toyota's MY 2015 to MY 2025 performance ``appears to yield overly
optimistic results because the powertrain efficiency curves
represent best-case targets and not the average vehicle, the imposed
performance constraints are unmarketable, and the generated credits
are out of sync with product cadence and design cycles.'' See also
NHTSA-2018-0067-12431, at 7. More recently, Toyota stated in their
comments to the NPRM that ``Toyota's position [on the efficacy of
the OMEGA and LPM models] has been clearly represented by comments
previously submitted by the Alliance of Automobile Manufacturers,
Global Automakers, and Novation Analytics. Those comments identify
the LPM and OMEGA models as sources of inaccuracy in EPA technology
evaluations and provide suggested improvements. Neither model is
transparent, intuitive, or user friendly.''
\603\ EPA-HQ-OAR-2015-0827-9194.
\604\ EPA-HQ-OAR-2015-0827-9194, at 33.
\605\ EPA-HQ-OAR-2015-0827-9194.
\606\ EPA-HQ-OAR-2015-0827-9728.
\607\ EPA-HQ-OAR-2015-0827-9163 at 5. (``EPA should abandon the
lumped-parameter model and instead use NHTSA's Autonomie and Volpe
models to support the Revised Final Determination.''). See also EPA-
HQ-OAR-2015-0827-9728 at 15 (stating the EPA's engine mapping and
tear down analyses ``should be integrated into the Autonomie model,
which then feeds into the Volpe modeling process.''); EPA-HQ-OAR-
2015-0827-9194 at 33.
---------------------------------------------------------------------------
In addition, Autonomie contains up-to-date sub-models to represent
the latest electrification and advanced transmission and advanced
engine technologies. As summarized by the Alliance, ``Autonomie was
developed from the start to address the complex task of combining 2
power sources in a hybrid powertrain.'' \608\ Autonomie has
continuously improved over the years by adopting new technologies into
its modeling framework. Even a small sampling of SAE papers shows how
Autonomie has been validated to simulate the latest fuel-economy-
improving technologies like hybrid vehicles and PHEVs.\609\
---------------------------------------------------------------------------
\608\ Alliance, Docket ID NHTSA-2018-0067-12073 at 135.
\609\ Jeong, J., Kim, N., Stutenberg, K., Rousseau, A.,
``Analysis and Model Validation of the Toyota Prius Prime,'' SAE
2019-01-0369, SAE World Congress, Detroit, April 2019; Kim, N,
Jeong, J. Rousseau, A. & Lohse-Busch, H. ``Control Analysis and
Thermal Model Development of PHEV,'' SAE 2015-01-1157.
---------------------------------------------------------------------------
Moreover, Autonomie effectively considers other real-world
constraints faced by the automotive industry. Vehicle manufacturers and
suppliers spend significant time and effort to ensure technologies are
incorporated into vehicles in ways that will balance consumer
acceptance for attributes such as driving quality,\610\ noise-
vibration-harshness (NVH), and meeting other regulatory mandates, like
EPA's and CARB's On-Board Diagnostics (OBD) requirements,\611\ and
EPA's and CARB's criteria exhaust emissions standards.\612\ The
implementation of new fuel economy improving technologies have at times
raised consumer acceptance issues.\613\ As discussed earlier, there are
diminishing returns for modeling every vehicle attribute and tradeoff,
as each takes time and incurs cost; however, Autonomie sub-models are
designed to account for a number of the key attributes and tradeoffs,
so the resulting effectiveness estimates reflect these real world
constraints.
---------------------------------------------------------------------------
\610\ An example of a design requirement is accommodating the
``lag'' in torque delivery due to the spooling of a turbine in a
turbocharged downsized engine. This affects real-world vehicle
performance, as well as the vehicle's ability to shift during normal
driving and test cycles.
\611\ EPA adopted and incorporated by reference current OBD
regulations by the California ARB, effective for MY 2017, that cover
all vehicles except those in the heavier fraction of the heavy-duty
vehicle class.
\612\ Tier 3 emission standards for light-duty vehicles were
proposed in March 2013 78 FR 29815 (May 21, 2013) and signed into
law on March 3, 2014 79 FR 23413 (June 27, 2014). The Tier 3
standards--closely aligned with California LEV III standards--are
phased-in over the period from MY2017 through MY2025. The regulation
also tightens sulfur limits for gasoline.
\613\ Atiyeh, C. ``What you need to know about Ford's PowerShift
Transmission Problems'' Car and Driver. July 11, 2019. https://www.caranddriver.com/news/a27438193/ford-powershift-transmission-problems/.
---------------------------------------------------------------------------
Furthermore, aside from the fact that Autonomie represents the
structural state-of-the-art in full-vehicle modeling and simulation,
Autonomie can be populated with any inputs that could be populated in
the ALPHA model.\614\ The agencies chose to use specific inputs for
this rulemaking because, as discussed further in Sections VI.C below,
they best represent the technologies that manufacturers could
incorporate in the rulemaking timeframe, in a way that balanced
important concerns like consumer acceptance. Some other examples of how
Autonomie inputs have been updated with the latest vehicle technology
data specifically for this analysis include test data incorporated from
both Argonne and NHTSA-sponsored vehicle benchmarking, including an
updated automatic transmission skip-shifting feature,\615\ additional
application of cylinder deactivation for turbocharged downsized
engines, and as discussed above, new modeling and simulation that
includes variable compression ratio and Miller Cycle engines.
---------------------------------------------------------------------------
\614\ For example, Autonomie used the HCR1 and HCR2 engine maps
used as inputs to ALHPA in the Draft TAR and Proposed Determination.
\615\ NHTSA Benchmarking, ``Laboratory Testing of a 2017 Ford F-
150 3.5 V6 EcoBoost with a 10-speed transmission.'' DOT HS 812 520.
---------------------------------------------------------------------------
Finally, ICCT commented that the agencies must conduct a systematic
comparison of the Autonomie modeling system and ALPHA modeling in
several respects, including the differences in technical inputs and
resulting efficiency estimates, to explain how the choice of model
altered the regulatory technology penetration and compliance cost
estimations, and the differences in modeling methodologies, including
regarding the relative level of experience of the teams conducting the
effectiveness modeling, to demonstrate that the choice to use Autonomie
was not ``due to convenience and easier access by the NHTSA research
team, rather than for any technical improvement.'' ICCT stated that
without performing this comparison, ``it otherwise appears that the
agencies switched from a better-vetted model and system of inputs with
more recent input data to a less-vetted model and system of inputs as a
way to bury many dozens of changes without transparency or expert
assessment (as illustrated in the
[[Page 24345]]
above errors and invalidated data on individual technologies).'' Each
issue is discussed below in turn.
First, regarding technical inputs, technology pathways, and
resulting outputs, ICCT stated that the agencies must compare (1)
whether the models have been routinely strengthened by incorporating
cutting edge 2020-2025 automotive technologies to ensure they reflect
the available improvements; (2) every efficiency technology in the 2016
Draft TAR and original EPA TSD and Proposed and Final Determination
analysis against the NPRM; (3) all the major technology package
pathways (i.e., all combinations with high uptake in the Adopted and
Augural 2025 standards) in the current NPRM versus the 2016 Draft TAR
and the 2016 TSD and original Final Determination analysis; (4) each of
the major 2025 technology package synergies; (5) the modeling work of
EPA's, Ricardo's, and Argonne's 2014-2018 model year engine
benchmarking and modeling of top engine and transmission models; and
``defend why they appear to have chosen to dismiss the superior and
better vetted technology modeling approach.''
ICCT stated that the agencies must make these comparisons because,
``[o]therwise, it seems obvious that the agencies have subjectively
decided to use the modeling that increases the modeled cost, providing
further evidence of a high degree of bias without an objective
accounting of the methodological differences and the sensitivity of the
results to their new decision.'' Moreover, ICCT stated that ``[b]ecause
ALPHA is the dominant, preferred, and better-vetted modeling and was
used in the original Proposed and Final Determination, the agencies are
responsible for assessing and describing how the use of the ALPHA
modeling would result in a different regulatory result for their
analysis of the 2017-2025 adopted [CO2] and Augural CAFE
standards.''
The agencies do not believe that it is necessary to conduct a
retrospective comparison of ALPHA/LPM and Autonomie effectiveness for
every technology in the Draft TAR and Proposed Determination to the
NPRM and final rule analyses, between the two models for technologies
and packages used in the NPRM and final rule analysis, or to explain
where and why Autonomie provided different results from ALPHA and the
LPM, to assess and describe how the use of the ALPHA modeling would
result in a different regulatory result of CAFE and CO2
standards, per ICCT's request. While it is anticipated that different
values will be produced using different tools in an analysis, it is not
appropriate to select the tool for use based on preferred results. The
selection of an analysis tool should be based on an evaluation of the
tool's capabilities and appropriateness for the analysis task. The
analysis tool should support the full extent of the analysis and
support the level of input and output resolution required. To compare
the output of the two models for the purpose of selecting a tool for
the analysis would likely be biased and disingenuous to the purpose of
the analysis. In this case, Autonomie was selected for this analysis
for the reasons discussed throughout this section, and accordingly the
agencies believe that it was reasonable to consider effectiveness
estimates developed with Autonomie.
That said, comparison of how the tools behave is discussed here to
further support the agencies' decision process. To demonstrate, in
addition to everything discussed previously in this section,
differences in how each model handles powertrain systems modeling with
specific examples are discussed below as a reference, and differences
between the agencies' approaches to effectiveness modeling for specific
technologies is discussed in Section VI.C where appropriate. While the
improved approach to estimating technology effectiveness estimates
certainly impacted the regulatory technology penetration, compliance
cost estimates, and ``major 2025 technology packages and synergies,''
how technologies are applied in the compliance modeling and the
associated costs of the technologies is equally as important to
consider when examining factors that might impact the regulatory
analysis; that consideration goes beyond the scope of simply
considering which full vehicle simulation model better performs the
functions required of this analysis.
The agencies have discussed updates to the technologies considered
in the Autonomie modeling throughout this section, in addition to
Autonomie's models and sub-models that control advanced technologies
like hybrid and electrified powertrains. Autonomie's explicit models,
sub-models, and controls for hybrid and electric vehicles have been
continuously validated over the past several years,\616\ as Autonomie
was developed from the beginning to address the complex task of
combining two power sources in a hybrid powertrain.
---------------------------------------------------------------------------
\616\ Karbowski, D., Kwon, J., Kim, N., & Rousseau, A.,
``Instantaneously Optimized Controller for a Multimode Hybrid
Electric Vehicle,'' SAE paper 2010-01-0816, SAE World Congress,
Detroit, April 2010; Sharer, P., Rousseau, A., Karbowski, D., &
Pagerit, S. ``Plug-in Hybrid Electric Vehicle Control Strategy--
Comparison between EV and Charge-Depleting Options,'' SAE paper
2008-01-0460, SAE World Congress, Detroit (April 2008); and
Rousseau, A., Shidore, N., Carlson, R., & Karbowski, D. ``Impact of
Battery Characteristics on PHEV Fuel Economy,'' AABC08; Jeong, J.,
Kim, N., Stutenberg, K., Rousseau, A., ``Analysis and Model
Validation of the Toyota Prius Prime,'' SAE 2019-01-0369, SAE World
Congress, Detroit, April 2019; Kim, N, Jeong, J. Rousseau, A. &
Lohse-Busch, H. ``Control Analysis and Thermal Model Development of
PHEV,'' SAE 2015-01-1157, SAE World Congress, Detroit, April 15;
Lee, D. Rousseau, A. & Rask, E. ``Development and Validation of the
Ford Focus BEV Vehicle Model,'' 2014-01-1809, SAE World Congress,
Detroit, Apr. 14; Kim, N., Kim, N., Rousseau, A., & Duoba, M.
``Validating Volt PHEV Model with Dynamometer Test Data using
Autonomie,'' SAE 2013-01-1458, SAE World Congress, Detroit, Apr.
13.; Kim, N., Rousseau, A., & Rask, E. ``Autonomie Model Validation
with Test Data for 2010 Toyota Prius,'' SAE 2012-01-1040, SAE World
Congress, Detroit, Apr. 12; Karbowski, D., Rousseau, A, Pagerit, S.,
& Sharer, P. ``Plug-in Vehicle Control Strategy--From Global
Optimization to Real Time Application,'' 22th International Electric
Vehicle Symposium (EVS22), Yokohama, (October 2006).
---------------------------------------------------------------------------
Also regarding the inputs to both models, as highlighted in Section
VI.C.3.a), and discussed above, inputs and assumptions for the ALPHA
modeling used for the EPA Draft TAR and Proposed Determination analysis
were projected from benchmarking testing. While it is straightforward
to measure engine fuel consumption and create an engine fuel map, it is
extremely challenging to identify the specific technologies and levels
of technologies present on a benchmarking engine. Attributing changes
in the overall engine fuel consumption to the individual engine
technologies that make up the complete engine involves significant
uncertainty.
The fixed-point model approach used by the ALPHA model does not
develop an effectiveness function and assigns a single value to a
technology. The single value is derived from benchmark testing, which
often does not isolate the effect of a single technology from the
effects of other technologies on the tested vehicle. To isolate a
single technology's effect for use in fixed point modeling properly,
the agencies would need to benchmark multiple versions of a single
vehicle, carefully controlling changes to the vehicles' fuel efficiency
technologies. This process would need to be repeated for a large
portion of the vehicle fleet and would require significant funding and
thousands of lab hours to complete. Without this level of data, fixed-
point effectiveness estimates tend to be too high, as they are unable
to account for synergetic effects of multiple technologies.
Specifically, when EPA benchmarks vehicles like the 2018 Toyota Camry,
the resulting fuel map captures the benefits of many
[[Page 24346]]
technologies associated with that engine. This data can be helpful when
developing controls and validating component operations in modeling,
but it is inaccurate to conclude that fuel consumption is directly
related to individual engine technologies, such as lubrication and
friction reduction, and geometric improvements in efficiency.
Contrasted, the NPRM and final rule Autonomie analyses selected
specific base engine maps and applied technologies incrementally, both
individually and in known combinations, to better isolate the impacts
of the technologies. As discussed above, this also implemented NAS
Recommendation 2.1, to use engine-model-generated maps in the full
vehicle simulations derived from a validated baseline map in which all
parameters except the new technology of interest are held
constant.\617\ While the different methods are valid for different
purposes, the method selected for the analysis presented today was more
useful for measuring the incremental effectiveness increments as
opposed to the absolute values of technology effectiveness, e.g., that
could be measured by benchmarking a technology package.
---------------------------------------------------------------------------
\617\ 2015 NAS Report at p. 82.
---------------------------------------------------------------------------
To provide an example of another difference in behavior between the
simulation tools, a comparison between ALPHA and Autonomie
transmissions shifting behavior was conducted. The comparison
highlighted the differences in how each simulation tool approaches
transmission shift logic. The ALPHA simulation tool used ALPHAShift.
ALPHAShift is an optimization algorithm that uses numerous vehicle
characteristics to find a best shifting strategy. The primary inputs
for the algorithm includes the fuel consumption (or cost) map for the
vehicle engine.\618\ Although a public version of ALPHA is available
for evaluation, the ALPHAShift algorithm used by the tool is hard coded
with fixed values.619 620 This is an issue, because despite
peer reviewed documentation on how to tune the algorithm,\621\ no
documentation of how the algorithm logic works is available for review.
This is confounding for the use of the software, particularly when the
observed behavior of the model departs from expected behavior. Figure
VI-6 below shows simulated gear shift (left) versus actual gear shift
(right), demonstrating an unexpected shift to neutral before shifting
to the requested gear.
---------------------------------------------------------------------------
\618\ Newman, K., Kargul, J., and Barba, D., ``Development and
Testing of an Automatic Transmission Shift Schedule Algorithm for
Vehicle Simulation,'' SAE Int. J. Engines 8(3):2015, doi:10.4271/
2015-01-1142.
\619\ Aymeric, R. Islam, E. S. ``Analysis of EPA's ALPHA Shift
Model--ALPHAShift.'' ANL. March 9, 2020.
\620\ ALPHA v2.2 Technology Walk Samples. EPA. January 2017.
https://www.epa.gov/sites/production/files/2017-01/alpha-20170112.zip. Last Accessed March 9, 2020.
\621\ Newman, K., Kargul, J., and Barba, D., ``Development and
Testing of an Automatic Transmission Shift Schedule Algorithm for
Vehicle Simulation,'' SAE Int. J. Engines 8(3):2015, doi:10.4271/
2015-01-1142.
---------------------------------------------------------------------------
By contrast, and discussed further in VI.C.2 Transmission Paths,
Autonomie uses a fully documented algorithm to develop a best shifting
strategy for each unique vehicle configuration. The algorithm develops
shifting strategies unique to each individual vehicle based on gear
ratio, final drive ratio, engine BSFC and other vehicle
characteristics. This is one example of model behavior, in addition to
the availability of more transparency on this behavior for greater
stakeholder review, that led the agencies to determine it was
reasonable and appropriate to use Autonomie for this analysis.
---------------------------------------------------------------------------
\622\ ALPHA v2.2 Technology Walk Samples. Jan. 12, 2017. https://www.epa.gov/sites/production/files/2017-01/alpha-20170112.zip. Last
accessed Dec 9, 2019.
[GRAPHIC] [TIFF OMITTED] TR30AP20.114
Regarding the technical expertise of the team conducting the
---------------------------------------------------------------------------
effectiveness modeling, ICCT commented:
[T]he agencies should also disclose how much commercial business
is conducted by the Ricardo, IAV, and Argonne Autonomie teams that
underpin the modeling of EPA and NHTSA, respectively, including how
much related research they have done for auto industry clients over
the past ten years. We mention this because we strongly suspect that
Ricardo, upon which EPA built its ALPHA model, has done at least an
order of magnitude (in number of projects, person-hours, and budget)
more work with and for the automotive industry than the IAV and
Autonomie teams have in direct work for
[[Page 24347]]
automotive industry clients. A conventional government procurement
effort that competitively vets potential research expert teams would
presumably have selected for such automotive industry credentials
and experience, yet it appears that the agencies are wholly
deferring to Autonomie's less rigorous research-grade modeling
framework and data due to convenience and easier access by the NHTSA
research team, rather than for any technical improvement, and this
is to the detriment of showing clear understanding of real-world
automotive engineering developments (as demonstrated by many
erroneous technology combination results throughout these comments).
First, NHTSA follows Federal Acquisition Regulation (FAR) to award
contracts and Interagency Agreements (IAAs),\623\ and any awarded
contracts and IAAs must follow the FAR requirements. Importantly, FAR
3.101-1 includes key aspects of conduct and ethics that NHTSA must
follow in awarding a contract or IAA:
---------------------------------------------------------------------------
\623\ Federal Acquisition Regulation (FAR). https://www.acquisition.gov/.
Government business shall be conducted in a manner above
reproach and, except as authorized by statute or regulation, with
complete impartiality and with preferential treatment for none.
Transactions relating to the expenditure of public funds require the
highest degree of public trust and an impeccable standard of
conduct. The general rule is to avoid strictly any conflict of
interest or even the appearance of a conflict of interest in
Government-contractor relationships. While many Federal laws and
regulations place restrictions on the actions of Government
personnel, their official conduct must, in addition, be such that
they would have no reluctance to make a full public disclosure of
their actions.\624\
---------------------------------------------------------------------------
\624\ FAR 3.101-1.
While some factors are more relevant than others in considering
whether to award a contract or enter into an IAA, the amount of work
that an organization has performed, characterized by projects, person-
hours, and budget, is only one of a multitude of factors that is
considered (if it is even considered at all--an agency might not
request this information and an organization might decline to provide
it because of contractual clauses or to protect commercial business
interests) when assessing whether an organization meets the agency's
needs for a specific task. Other factors, such as the federal budget,
also set boundaries for the scope of work that can be performed under
any competitive government procurement effort.
As discussed throughout this section, the team at Argonne National
Laboratory behind Autonomie has developed and refined a state-of-the-
art tool that is used by the automotive industry, government agencies,
and research or other nongovernmental institutions around the world.
The tool has been and continues to be validated to production vehicles,
and updated to include models, sub-models, and controls representing
the state-of-the-art in fuel economy improving technology. To the
extent that ICCT believes that ``research done for auto industry
clients,'' ``work with and for the automotive industry,'' and
``automotive industry credentials and experience,'' are metrics upon
which to base this type of important decision, the agencies point ICCT
to the statements from the automotive industry, above, recommending
Autonomie be used for technology effectiveness modeling.
ICCT concluded that ``[w]hile the agencies are in their process of
conducting a proper vetting of their NPRM's foundational Autonomie-
based modeling, we recommend that they rely on what appears to be the
superior and better vetted technology modeling approach with more
thorough and state-of-the-art advanced powertrain systems modeling and
engine maps from the EPA ALPHA modeling.''
The agencies properly vetted the Autonomie modeling and decided
that Autonomie represented a reasonable and appropriate tool to provide
technology effectiveness estimates for this rulemaking. To the extent
that commenters' concerns were more about the effectiveness results
than the tools used to model technology effectiveness, modeling updates
detailed in the Section VI.B.3.c), below, address those comments. While
some commenters may still be dissatisfied with Autonomie's technology
effectiveness estimates, the agencies believe that the refinement of
inputs and input assumptions, and associated explanation of why those
refinements are appropriate and reasonable, have appropriately
addressed comments on these issues. Importantly, none of these
refinements have led either agency to reconsider using Autonomie for
this rulemaking analysis.
Additional discussion of the agencies' decision to rely on one set
of modeling tools for this rulemaking is located in Section VI.A of
this preamble.
c) Technology Effectiveness Values Implementation in the CAFE Model
While the Autonomie model produces a large amount of information
about each simulation run--for a single technology combination, in a
single technology class--the CAFE model only uses two elements of that
information: Battery costs and fuel consumption on the city and highway
cycles. The agencies combine the fuel economy information from the two
cycles to produce a composite fuel economy for each vehicle, on each
fuel. Plug-in hybrids, being the only dual-fuel vehicles in the
Autonomie simulation, require efficiency estimates of operation on both
gasoline and electricity--as well as an estimate of the utility factor,
or the number of miles driven on each fuel. The fuel economy
information for each technology combination, for each technology class,
is converted into a single number for use in the CAFE model.
As described in greater detail below, each Autonomie simulation
record represents a unique combination of technologies, and the
agencies create a technology ``key'' or technology state vector that
describes all the technology content associated with a record. The 2-
cycle fuel economy of each combination is converted into fuel
consumption (gallons per mile) and then normalized relative to the
starting point for the simulations. In each technology class, the
combination with the lowest technology content is the VVT (only)
engine, with a 5-speed transmission, no electrification, and no body-
level improvements (mass reduction, aerodynamic improvements, or low
rolling resistance tires). This is the reference point (for each
technology class) for all the effectiveness estimates in the CAFE
model. The improvement factors that the model uses are a given
combination's fuel consumption improvement relative to the reference
vehicle in its technology class.
For the majority of the technologies analyzed within the CAFE
Model, the fuel economy improvements were derived from the database of
Autonomie's detailed full-vehicle modeling and simulation results. In
addition to the technologies found in the Autonomie simulation
database, the CAFE modeling system also incorporated a handful of
technologies that were required for CAFE modeling, but were not
explicitly simulated in Autonomie. The total effectiveness of these
technologies either could not be captured on the 2-cycle test, or there
was no robust data that could be used as an input to the full-vehicle
modeling and simulation, like with emerging technologies such as
advanced cylinder deactivation (ADEAC). These additional technologies
are discussed further in Sections VI.B.3 Technology Effectiveness and
individual technologies sections. For calculating fuel economy
improvements attributable to these additional technologies, the model
used defined fuel consumption improvement factors that are constant
[[Page 24348]]
across all technology combinations in the database and scale
multiplicatively when applied together. The Autonomie-simulated and
additional technologies were then externally combined, forming a single
dataset of simulation results (referred to as the vehicle simulation
database, or simply, database), which may then be utilized by the CAFE
modeling system.
To incorporate the results of the combined database of Autonomie-
simulated and additional technologies, while still preserving the basic
structure of the CAFE Model's technology subsystem, it was necessary to
translate the points in this database into corresponding locations
defined by the technology pathways. By recognizing that most of the
pathways are unrelated, and are only logically linked to designate the
direction in which technologies are allowed to progress, it is possible
to condense the paths into a smaller number of groups based on the
specific technology. In addition, to allow for technologies present on
the Basic Engine and Dynamic Road Load (DLR--i.e., MASS, AERO, and
ROLL) paths to be evaluated and applied in any given combination, a
unique group was established for each of these technologies.
As such, the following technology groups are defined within the
modeling system: Engine cam configuration (CONFIG), VVT engine
technology (VVT), VVL engine technology (VVL), SGDI engine technology
(SGDI), DEAC engine technology (DEAC), non-basic engine technologies
(ADVENG), transmission technologies (TRANS), electrification and
hybridization (ELEC), low rolling resistance tires (ROLL), aerodynamic
improvements (AERO), mass reduction levels (MR), EFR engine technology
(EFR), electric accessory improvement technologies (ELECACC), LDB
technology (LDB), and SAX technology (SAX). The combination of
technologies along each of these groups forms a unique technology state
vector and defines a unique technology combination that corresponds to
a single point in the database for each technology class evaluated
within the modeling system.
As an example, a technology state vector describing a vehicle with
a SOHC engine, variable valve timing (only), a 6-speed automatic
transmission, a belt-integrated starter generator, rolling resistance
(level 1), aerodynamic improvements (level 2), mass reduction (level
1), electric power steering, and low drag brakes, would be specified as
``SOHC; VVT; AT6; BISG; ROLL10; AERO20; MR1; EPS; LDB.'' \625\ By
assigning each unique technology combination a state vector such as the
one in the example, the CAFE Model can then assign each vehicle in the
analysis fleet an initial state that corresponds to a point in the
database.
---------------------------------------------------------------------------
\625\ In the example technology state vector, the series of
semicolons between VVT and AT6 correspond to the engine technologies
which are not included as part of the combination, while the gap
between MR1 and EPS corresponds to EFR and the omitted technology
after LDB is SAX. The extra semicolons for omitted technologies are
preserved in this example for clarity and emphasis, and will not be
included in future examples.
---------------------------------------------------------------------------
Once a vehicle is assigned (or mapped) to an appropriate technology
state vector (from one of approximately three million unique
combinations, which are defined in the vehicle simulation database as
CONFIG; VVT; VVL; SGDI; DEAC; ADVENG; TRANS; ELEC; ROLL; AERO; MR; EFR;
ELECACC; LDB; SAX), adding a new technology to the vehicle simply
represents progress from a previous state vector to a new state vector.
The previous state vector simply refers to the technologies that are
currently in use on a vehicle. The new state vector, however, is
computed within the modeling system by adding a new technology to the
combination of technologies represented by the previous state vector,
while simultaneously removing any other technologies that are
superseded by the newly added one.
For example, consider the vehicle with the state vector described
as: SOHC; VVT; AT6; BISG; ROLL10; AERO20; MR1; EPS; LDB. Assume the
system is evaluating PHEV20 as a candidate technology for application
on this vehicle. The new state vector for this vehicle is computed by
removing SOHC, VVT, AT6, and BISG technologies from the previous state
vector,\626\ while also adding PHEV20, resulting in the following:
PHEV20; ROLL10; AERO20; MR1; EPS; LDB.
---------------------------------------------------------------------------
\626\ For more discussion of how the CAFE Model handles
technology supersession, see Section VI.A.7.
---------------------------------------------------------------------------
From here, it is relatively simple to obtain a fuel economy
improvement factor for any new combination of technologies and apply
that factor to the fuel economy of a vehicle in the analysis fleet. The
formula for calculating a vehicle's fuel economy after application of
each successive technology represented within the database is defined,
simply put, as the difference between the fuel economy improvement
factor associated with the technology state vector before application
of a candidate technology, and after the application of a candidate
technology.\627\ This is applied to the original compliance fuel
economy value for a discrete vehicle in the MY 2017 analysis fleet, as
discussed previously in Section VI.B.3 Technology Effectiveness.
---------------------------------------------------------------------------
\627\ For more discussion of how the CAFE Model calculates a
vehicle's fuel economy where the vehicle switches from one type of
fuel to another, for example, from gasoline operation to diesel
operation or from gasoline operation to plug-in hybrid/electric
vehicle operation, see Section VI.A CAFE Model.
---------------------------------------------------------------------------
The fuel economy improvement factor is defined in a way that
captures the incremental improvement of moving between points in the
database, where each point is defined uniquely as a combination of up
to 15 distinct technologies describing, as mentioned above, the
engine's cam configuration, multiple distinct combinations of engine
technologies, transmission, electrification type, and various vehicle
body level technologies.
Unlike the preceding versions of the modeling system, the current
version of the CAFE Model relies entirely on the vehicle simulation
database for calculating fuel economy improvements resulting from all
technologies available to the system. The fuel economy improvements are
derived from the factors defined for each unique technology combination
or state vector. Each time the improvement factor for a new state
vector is added to a vehicle's existing fuel economy, the factor
associated with the old technology combination is entirely removed. In
that sense, application of technologies obtained from the Autonomie
database is ``self-correcting'' within the model. As such, special-case
adjustments defined by the previous version of the model are not
applicable to this one.
Meszler Engineering Services, commenting on behalf of Natural
Resources Defense Council, commented that ``[w]ith very limited
exception, technology is not included in the NPRM CAFE model if it was
not included in the simulation modeling that underlies the Argonne
database,'' citing the ``add-on'' technologies and technologies with
fixed effectiveness values.\628\ Meszler continued, ``[t]his same
limitation controls the coupling of technologies, and by extension the
definition of the CAFE model technology pathways. If a combination of
technologies were not modeled during the development of the Argonne
database, that package (or combination) of technologies is not
available for adoption in the CAFE model. Both of these design
constraints serve to limit the slate of technologies available to
respond to fuel economy
[[Page 24349]]
standards. The slate of available technologies is basically constrained
to those included in NHTSA's research activity. If a technology or
technology combination was not in the NHTSA research planning process,
it is not available in the model.'' Finally, Meszler stated that
``because of the constrained model architecture and the reliance on the
Argonne database for impact estimates, independently expanding the
model to include additional technologies or technology combinations is
not trivial.''
---------------------------------------------------------------------------
\628\ NHTSA-2018-0067-11723, at 4-5.
---------------------------------------------------------------------------
We agree that expanding the database to include new technologies is
not trivial. However, it is possible. The set of available technologies
is part of the model code, and the code is made public upon each
release of the model. Many commenters made modifications to the model
code, conducted additional tests of their own, and presented their
results to the agencies in the form of public comments before the end
of the public comment period. A user could add the new technology,
identify the associated engineering restrictions that determine
combinations for which that technology should not be considered, and
add the relevant rows (representing possible technology combinations
that include the new technology) in the database (which exists locally
on every computer that runs the model). An enterprising user could also
take an existing technology along a given path and replace the
efficiency values with new values--presumably from their own full
vehicle simulations for each technology combination that contains the
technology in question. Given the length of time and computing power
required to simulate vehicle fuel economy on the test cycle for every
possible combination that could be considered by the CAFE model, using
a pre-defined database that represents a large ensemble of simulated
technology combinations is preferable to the alternative of fully
integrating a vehicle simulation model that would be required to run in
real-time during the compliance simulation to evaluate the
effectiveness of every combination considered (not just applied) by the
model.
4. Technology Costs
In the proposal, the agencies estimated present and future costs
for fuel-saving technologies, taking into consideration the type of
vehicle, or type of engine if technology costs vary by application.
These cost estimates are based on three main inputs. First, the
agencies estimated direct manufacturing costs (DMCs), or the component
and labor costs of producing and assembling the physical parts and
systems, with estimated costs assuming high volume production. DMCs
generally do not include the indirect costs of tools, capital
equipment, financing costs, engineering, sales, administrative support
or return on investment. Second, the agencies accounted for these
indirect costs via a scalar markup of direct manufacturing costs (the
retail price equivalent, or RPE). Finally, costs for technologies may
change over time as industry streamlines design and manufacturing
processes. The agencies therefore estimated potential cost improvements
with learning effects (LE). The retail cost of equipment in any future
year is estimated to be equal to the product of the DMC, RPE, and LE.
Considering the retail cost of equipment, instead of merely direct
manufacturing costs, is important to account for the real-world price
effects of a technology, as well as market realities. Absent a
government mandate, motor vehicle manufacturers will not undertake
expensive development and production efforts to implement technologies
without realistic prospects of consumers being willing to pay enough
for such technology to allow for the manufacturers to recover their
investment.
a) Direct Manufacturing Costs
Direct manufacturing costs (DMCs) are the component costs of the
physical parts and systems that make up a complete vehicle. The
analysis used agency-sponsored tear-down studies of vehicles and parts
to estimate the DMCs of individual technologies, in addition to
independent tear-down studies, other publications, and confidential
business information. In the simplest cases, the agency-sponsored
studies produced results that confirmed third-party industry estimates,
and aligned with confidential information provided by manufacturers and
suppliers. In cases with a large difference between the tear-down study
results and credible independent sources, study assumptions were
scrutinized, and sometimes the analysis was revised or updated
accordingly.
Due to the variety of technologies and their applications, and the
cost and time required to conduct detailed tear-down analyses, the
agencies did not sponsor teardown studies for every technology. In
addition, many fuel-saving technologies were considered that are pre-
production, or sold in very small pilot volumes. For those
technologies, a tear-down study could not be conducted to assess costs
because the product is not yet in the marketplace for evaluation. In
these cases, the agencies relied upon third-party estimates and
confidential information from suppliers and manufacturers were relied
upon; however, there are some common pitfalls with relying on
confidential business information to estimate costs. The agencies and
the source may have had incongruent or incompatible definitions of
``baseline.'' The source may have provided DMCs at a date many years in
the future, and assumed very high production volumes, important caveats
to consider for agency analysis. In addition, a source, under no
contractual obligation to the agencies, may provide incomplete and/or
misleading information. In other cases, intellectual property
considerations and strategic business partnerships may have contributed
to a manufacturer's cost information and could be difficult to account
for in the model as not all manufacturer's may have access to
proprietary technologies at stated costs. The agencies carefully
evaluated new information in light of these common pitfalls, especially
regarding emerging technologies.
Specifically, the analysis used third-party, forward-looking
information for advanced cylinder deactivation and variable compression
ratio engines. While these cost estimates may be preliminary (as is the
case with many emerging technologies prior to commercialization), the
agencies consider them to be reasonable estimates of the likely costs
of these technologies.
While costs for fuel-saving technologies reflect the best estimates
available today, technology cost estimates will likely change in the
future as technologies are deployed and as production is expanded. For
emerging technologies, the best information available at the time of
the analysis was utilized, and cost assumptions will continue to be
updated for any future analysis. Below, discussion of each category of
technologies (e.g., engines, transmissions, electrification) summarizes
comments on corresponding direct cost estimates, and reviews estimates
the agencies have applied for today's analysis.
Indirect Costs
As discussed above, direct costs represent the cost associated with
acquiring raw materials, fabricating parts, and assembling vehicles
with the various technologies manufacturers are expected to use to meet
future CAFE and CO2 standards. They include materials,
labor, and variable energy costs required to produce and assemble the
vehicle. However, they do not
[[Page 24350]]
include overhead costs required to develop and produce the vehicle,
costs incurred by manufacturers or dealers to sell vehicles, or the
profit manufacturers and dealers make from their investments. All of
these items contribute to the price consumers ultimately pay for the
vehicle. These components of retail prices are illustrated in Table VI-
23 below.
[GRAPHIC] [TIFF OMITTED] TR30AP20.115
In addition to direct manufacturing costs, the agencies estimated
and considered indirect manufacturing costs. To estimate indirect
costs, direct manufacturing costs are multiplied by a factor to
represent the average price for fuel-saving technologies at retail.
In the Draft TAR and preceding CAFE and safety rulemaking analyses,
NHTSA relied on a factor, referred to as the retail price equivalent
(RPE), to account for indirect manufacturing costs. The RPE accounts
for indirect costs like engineering, sales, and administrative support,
as well as other overhead costs, business expenses, warranty costs, and
return on capital considerations. In the Draft TAR (and subsequent
Determination) as well as the 2012 rulemaking analysis, EPA applied an
``Indirect Cost Multiplier'' (ICM) approach that it first applied in
the 2010 rulemaking regarding standards for MYs 2012-2016, which also
accounted for indirect manufacturing costs, albeit in a different way
than the RPE approach.
Some commenters recommended the agencies rely on the ICM approach
for the current rulemaking, citing EPA's prior peer review and use of
this approach.\629\ Others supported the agencies' reliance on the RPE
approach, citing the National Research Council's observations in 2015
that the ICM approach lacks an empirical basis.\630\ The agencies have
carefully considered these comments, and conclude that while the ICM
approach has conceptual merit, its application requires a range of
specific estimates, and data to support such estimates is scant and, in
some cases, nonexistent. The agencies have, therefore, applied the RPE
approach for this final rule, as in the NPRM analysis and other
rulemaking analyses. The following sections discuss both approaches in
detail to explain why the RPE approach was chosen for this final rule.
---------------------------------------------------------------------------
\629\ See, e.g., ICCT, NHTSA-2018-0067-11741, Attachment 3, at
I-83. See also CFA, NHTSA-2018-0067-12005, Attachment B, at p.189.
\630\ See, e.g., Alliance, NHTSA-2018-0067-12073, at 143. See
also National Research Council, ``Cost, Effectiveness, and
Deployment of Fuel Economy Technologies for Light-Duty Vehicles,''
2015, available at https://www.nap.edu/catalog/21744/cost-effectiveness-and-deployment-of-fuel-economy-technologies-for-lightduty-vehicles (``. . . the empirical basis for such multipliers
is still lacking, and, since their application depends on expert
judgment, it is not possible for to determine whether the Agencies'
ICMs are accurate or not'').
---------------------------------------------------------------------------
(1) Retail Price Equivalent
Historically, the method most commonly used to estimate indirect
costs of producing a motor vehicle has been the retail price equivalent
(RPE). The RPE markup factor is based on an examination of historical
financial data contained in 10-K reports filed by manufacturers with
the Securities and Exchange Commission (SEC). It represents the ratio
between the retail price of motor vehicles and the direct costs of all
activities that manufacturers engage in, including the design,
development, manufacturing, assembly,
[[Page 24351]]
and sales of new vehicles, refreshed vehicle designs, and modifications
to meet safety or fuel economy standards.
Figure VI-7 indicates that for more than three decades, the retail
price of motor vehicles has been, on average, roughly 50 percent above
the direct cost expenditures of manufacturers. This ratio has been
remarkably consistent, averaging roughly 1.5 with minor variations from
year to year over this period. At no point has the RPE markup exceeded
1.6 or fallen below 1.4.\631\ During this time frame, the average
annual increase in real direct costs was 2.5 percent, and the average
annual increase in real indirect costs was also 2.5 percent. Figure VI-
7 illustrates the historical relationship between retail prices and
direct manufacturing costs.\632\
---------------------------------------------------------------------------
\631\ Based on data from 1972-1997 and 2007. Data were not
available for intervening years, but results for 2007 seem to
indicate no significant change in the historical trend.
\632\ Rogozhin, A., Gallaher, M., & McManus, W., 2009,
Automobile Industry Retail Price Equivalent and Indirect Cost
Multipliers. Report by RTI International to Office of Transportation
Air Quality. U.S. Environmental Protection Agency, RTI Project
Number 0211577.002.004, February, Research Triangle Park, N.C.
Spinney, B.C., Faigin, B., Bowie, N., & St. Kratzke, 1999, Advanced
Air Bag Systems Cost, Weight, and Lead Time analysis Summary Report,
Contract NO. DTNH22-96-0-12003, Task Orders--001, 003, and 005.
Washington, DC, U.S. Department of Transportation.
---------------------------------------------------------------------------
An RPE of 1.5 does not imply that manufacturers automatically mark
up each vehicle by exactly 50 percent. Rather, it means that, over
time, the competitive marketplace has resulted in pricing structures
that average out to this relationship across the entire industry.
Prices for any individual model may be marked up at a higher or lower
rate depending on market demand. The consumer who buys a popular
vehicle may, in effect, subsidize the installation of a new technology
in a less marketable vehicle. But, on average, over time and across the
vehicle fleet, the retail price paid by consumers has risen by about
$1.50 for each dollar of direct costs incurred by manufacturers.
[GRAPHIC] [TIFF OMITTED] TR30AP20.116
It is also important to note that direct costs associated with any
specific technology will change over time as some combination of
learning and resource price changes occurs. Resource costs, such as the
price of steel, can fluctuate over time and can experience real long-
term trends in either direction, depending on supply and demand.
However, the normal learning process generally reduces direct
production costs as manufacturers refine production techniques and seek
out less costly parts and materials for increasing production volumes.
By contrast, this learning process does not generally influence
indirect costs. The implied RPE for any given technology would thus be
expected to grow over time as direct costs decline relative to indirect
costs. The RPE for any given year is based on direct costs of
technologies at different stages in their learning cycles, and which
may have different implied RPEs than they did in previous years. The
RPE averages 1.5 across the lifetime of technologies of all ages, with
a lower average in earlier years of a technology's life, and, because
of learning effects on direct costs, a higher average in later years.
The RPE has been used in all NHTSA safety and most previous CAFE
rulemakings to estimate costs. The National Academy of Sciences
recommends RPEs of 1.5 for suppliers and 2.0 for in-house production be
used to estimate total costs. The Alliance of Automobile Manufacturers
also advocates these values as appropriate markup factors for
estimating costs of technology changes. An RPE of 2.0 has also been
adopted by a coalition of environmental and research groups (NESCCAF,
ICCT, Southwest Research Institute, and TIAX-LLC) in a report on
reducing heavy truck emissions, and 2.0 is recommended by the U.S.
Department of Energy for estimating the cost of hybrid-electric and
automotive fuel cell costs ((see Vyas et al. (2000) in Table VI-24,
below).
Table VI-24 below lists other estimates of the RPE. Note that all
RPE estimates vary between 1.4 and 2.0, with most in the 1.4 to 1.7
range.
[[Page 24352]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.117
The RPE has thus enjoyed widespread use and acceptance by a variety
of governmental, academic, and industry organizations. The RPE has been
the most commonly used basis for indirect cost markups in regulatory
analyses. However, as noted above, the RPE is an aggregate measure
across all technologies applied by manufacturers and is not technology
specific. A more detailed examination of these technologies is possible
through an alternative measure, the indirect cost multiplier, which was
developed to focus more specifically on technologies used to meet CAFE
and CO2 standards.
---------------------------------------------------------------------------
\633\ Duleep, K.G. ``2008 Analysis of Technology Cost and Retail
Price.'' Presentation to Committee on Assessment of Technologies for
Improving Light Duty Vehicle Fuel Economy, January 25, Detroit, MI.;
Jack Faucett Associates, September 4, 1985. Update of EPA's Motor
Vehicle Emission Control Equipment Retail Price Equivalent (RPE)
Calculation Formula. Chevy Chase, MD--Jack Faucett Associates;
McKinsey & Company, October 2003. Preface to the Auto Sector Cases.
New Horizons--Multinational Company Investment in Developing
Economies, San Francisco, CA.; NRC (National Research Council),
2002. Effectiveness and Impact of Corporate Average Fuel Economy
Standards, Washington, DC--The National Academies Press; NRC, 2011.
Assessment of Fuel Economy Technologies for Light Duty Vehicles.
Washington, DC--The National Academies Press; Sierra Research, Inc.,
November 21, 2007, Study of Industry-Average Mark-Up Factors used to
Estimate Changes in Retail Price Equivalent (RPE) for Automotive
Fuel Economy and Emissions Control Systems, Sacramento, CA--Sierra
Research, Inc.; Vyas, A. Santini, D., & Cuenca, R. 2000. Comparison
of Indirect Cost Multipliers for Vehicle Manufacturing. Center for
Transportation Research, Argonne National Laboratory, April.
Argonne, Ill.
---------------------------------------------------------------------------
(2) Indirect Cost Multiplier
A second approach to accounting for indirect costs is the indirect
cost multiplier (ICM). ICMs specifically evaluate the components of
indirect costs likely to be affected by vehicle modifications
associated with environmental regulation. EPA developed the ICM concept
to enable the application of markups more specific to each technology.
For example, the indirect cost implications of using tires with better
rolling resistance would not be the same as those for developing an
entire new hybrid vehicle technology, which would require far more R&D,
capital investment, and management oversight. With more than 80
different technologies available to incrementally achieve fuel economy
improvements,\634\ a wide range of indirect cost effects might be
expected. ICMs attempt to isolate only those indirect costs that would
have to change to develop a specific technology. Thus, for example, if
a company were to hire additional staff to sell vehicles equipped with
fuel economy improving technology, or to search the technology
requirements of new CO2 or CAFE standards, the cost of these
staff would be included in ICMs. However, if these functions were
accomplished by existing staff, they would not be included. For
example, if an executive who normally devoted 10 percent of his time to
fuel economy standards compliance were to devote 50 percent of his time
in response to new more stringent requirements, his salary would not be
included in ICMs because he would be paid the same salary regardless of
whether he devoted his time to addressing CAFE requirements, developing
new performance technologies, or improving the company's market share.
ICMs thus do not account for the diverted resources required for
manufacturers to meet these standards, but rather for the net change in
costs manufacturers might experience because of hiring additional
personal or acquiring additional assets or services.
---------------------------------------------------------------------------
\634\ There are roughly 40 different basic unique technologies,
but variations among these technologies roughly double the possible
number of different technology applications.
---------------------------------------------------------------------------
For past rulemakings EPA developed both short-term and long-term
ICMs. Long-term ICMs are lower than short-term ICMs. This decline
reflects the belief that many indirect costs will decline over time.
For example, research is initially required to develop a new technology
and apply it throughout the vehicle fleet, but a lower level of
research will be required to improve, maintain, or adapt that new
technology to subsequent vehicle designs.
While the RPE was derived from data in financial statements
(reflecting real-world operating and financial results), no similar
data sources were available to estimate ICMs. ICMs are based on the
RPE, broken into its components, as shown in Table VI-25. Adjustment
factors were then developed for those components, based on the
complexity and time frame of low-, medium-, and high-complexity
technologies. The adjustment factors were developed from two panels of
engineers with background in the automobile industry. Initially, a
group of engineers met and developed an estimate of ICMs for three
different technologies. This ``consensus'' panel examined one low
complexity technology, one medium complexity technology, and one high
complexity technology, with the initial intent of using these
technologies to represent ICM factors for all technologies falling in
those categories. At a later date, a second panel was convened to
examine three more technologies (one low, one medium, and one high
complexity), using a modified Delphi approach to estimate indirect cost
effects. The results from the second panel identified the same pattern
as those of the original report--the indirect cost multipliers increase
with the
[[Page 24353]]
complexity of the technology and decrease over time. The values derived
in process are higher than those in the RPE/IC Report by values ranging
from 0.09 (that is, the multiplier increased from 1.20 to 1.29) to 0.19
(the multiplier increased from 1.45 to 1.64). This variation may be due
to differences in the technologies used in each panel. The results are
shown in Figure VI-8, together with the historical average RPE.
[GRAPHIC] [TIFF OMITTED] TR30AP20.118
In subsequent CAFE and CO2 analyses for MYs 2011, as
well as for the 2012-2016 rulemaking, a simple average of the two
resulting ICMs in the low and medium technology complexity categories
was applied to direct costs for all unexamined technologies in each
specific category. For high complexity technologies, the lower
consensus-based estimate was used for high complexity technologies
currently being produced, while the higher modified Delphi-based
estimate was used for more advanced technologies, such as plug-in
hybrid or electric vehicles, which had little or no current market
penetration. Note that ICMs originally did not include profit or
``return on capital,'' a fundamental difference from the RPE. However,
prior to the 2012-2016 CAFE analysis, ICMs were modified to include
provision for return on capital.
(3) Application of ICMs in the 2017-2025 Analysis
For the model year 2017-2025 rulemaking analysis, NHTSA and EPA
revisited technologies evaluated by EPA staff and reconsidered their
method of application. The agencies were concerned that averaging
consensus and modified Delphi ICMs might not be the most accurate way
to develop an estimate for the larger group of unexamined technologies.
Specifically, there was concern that some technologies might not be
representative of the larger groups they were chosen to represent.
Further, the agencies were concerned that the values developed under
the consensus method were not subject to the same analytical discipline
as those developed from the modified Delphi method. As a result, the
agencies relied primarily on the modified Delphi-based technologies to
establish their revised distributions. Thus, for the MY 2017-2025
analysis, the agencies used the following basis for estimating ICMs:
All low complexity technologies were estimated to equal
the ICM of the modified Delphi-based low technology-passive aerodynamic
improvements.
All medium complexity technologies were estimated to equal
the ICM of the modified Delphi-based medium technology-engine turbo
downsizing.
Strong hybrids and non-battery plug-in hybrid electric
vehicles (PHEVs) were estimated to equal the ICM of the high complexity
consensus-based high technology-hybrid electric vehicle.
PHEVs with battery packs and full electric vehicles were
estimated to equal the ICM of the high complexity modified Delphi-based
high technology-plug-in hybrid electric vehicle.
In addition to shifting the proxy basis for each technology group,
the agencies reexamined each technology's complexity designation in
light of the examined technologies that would serve as the basis for
each group. The resulting designations together with the associated
proxy technologies are shown in Table VI-25.
[[Page 24354]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.119
Many basic technologies noted in Table VI-25 have variations
sharing the same complexity designation and ICM estimate. Table VI-26
lists each technology used in the CAFE model together with their ICM
category and the year through which the short-term ICM would be
applied. Note that the number behind each ICM category designation
refers to the source of the ICM estimate, with 1 indicating the
consensus panel and 2 indicating the modified Delphi panel.
BILLING CODE 4910-59-P
[[Page 24355]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.120
[[Page 24356]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.121
[[Page 24357]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.122
[[Page 24358]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.123
BILLING CODE 4910-59-C
An additional adjustment was made to ICMs to account for the fact
that they were derived from the RPE analysis for a specific year
(2007). The agencies believed it would be more appropriate to base ICMs
on the expected long-term average RPE rather than that of one specific
year. To account for this, ICMs were normalized to an average RPE
multiplier level of 1.5.
Table VI-27 lists values of ICMs by technology category used in the
previous MY 2017-2025 rulemaking. As noted previously, the Low 1 and
Medium 1 categories, which were derived using the initial consensus
panel, are not used. Short-term values applied to CAFE technologies
thus range from 1.24 for Low complexity technologies, 1.39 for Medium
complexity technologies, 1.56 for High1 complexity technologies, and
1.77 for High2 complexity technologies. When long-term ICMs are applied
in the year following that noted in the far-right column of Table VI-
27, these values will drop to 1.19 for Low, 1.29 for Medium, 1.35 for
High1 and 1.50 for High2 complexity technologies.
[GRAPHIC] [TIFF OMITTED] TR30AP20.124
Note that ICMs for warranty costs are listed separately in Table
VI-27. This was done because warranty costs are treated differently
than other indirect costs. In some previous analyses (prior to MY 2017-
2025), learning was applied directly to total costs. However, the
agencies believe learning curves are more appropriately applied only to
direct costs, with indirect costs established up front based on the ICM
and held constant while direct costs are reduced by learning.
Warranties are an exception to this because warranty costs involve
future replacement of defective parts, and the cost of these parts
would reflect the effect of learning. Warranty costs were thus treated
as being subject to learning along with direct costs.\635\
---------------------------------------------------------------------------
\635\ Note that warranty costs also involve labor costs for
installation. This is typically done at dealerships, and it is
unlikely labor costs would be subject to learning curves that affect
motor vehicle parts or assembly costs. However, the portion of these
costs that is due to labor versus that due to parts is unknown, so
for this analysis, learning is applied to the full warranty cost.
---------------------------------------------------------------------------
The effect of learning on direct costs, together with the eventual
substitution of lower long-term ICMs, causes the effective markup from
ICMs to differ from the initial ICM on a yearly basis. An example of
how this occurs is provided in Table VI-28.\636\ This table, which was
originally developed for the MY 2017-2025 analysis, traces the effect
of learning on direct costs and its implications for both total costs
and the ICM-based markup. Direct costs are assigned a value
(proportion) of 1 to facilitate analysis on the same basis as ICMs (in
an ICM markup factor, the proportion of direct costs is represented by
1 while the proportion of indirect costs is represented by the fraction
of 1 to the right of the decimal.) Table VI-28 examines the effects of
these factors on turbocharged downsized engines, one of the more
prevalent CAFE technologies.
---------------------------------------------------------------------------
\636\ Table VI-22 illustrates the learning process from the base
year consistent with the direct cost estimate obtained by the
agencies. It is a mature technology well into the flat portion of
the learning curve. Note that costs were actually applied in this
rulemaking example beginning with MY 2017.
---------------------------------------------------------------------------
[[Page 24359]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.125
The second column of Table VI-28 lists the learning schedule
applied to turbocharged downsized engines. Turbocharged downsized
engines are a mature technology, so the learning schedule captures the
relatively flat portion of the learning curve occurring after larger
decreases have already reduced direct costs. The cost basis for
turbocharged downsized engines in the analysis was effective in 2012,
so this is the base year for this calculation when direct costs are set
to 1. The third column shows the progressive decline in direct costs as
the learning schedule in column 2 is applied to direct costs. Column 4
contains the value of all indirect costs except warranty. Turbocharged
downsized engines are a medium-complexity technology, so this value is
taken from the Medium2 row of Table VI-27. The initial value in 2012 is
the short-term value, which is used through 2018. During this time,
these indirect costs are not affected by learning, and they remain
constant. Beginning in 2019, the long-term ICM from Table VI-27 is
applied.
The fifth column contains warranty costs. As previously mentioned,
these costs are considered to be affected by learning like direct
costs, so they decline steadily until the long-term ICM is applied in
2019, at which point they drop noticeably before continuing their
gradual decline. In the sixth column, direct and indirect costs are
totaled. Results indicate a decline in total costs of roughly 30
percent during this 14-year period. The last column shows the effective
ICM-based markup, which is derived by dividing total costs by direct
costs. Over this period, the ICM-based markup rose from the initial
short-term ICM level of 1.39 to 1.45 in 2018. It then declined to 1.35
in 2019 when the long-term ICM was applied to the 2019 direct cost.
Over the remaining years, it gradually rises back up to 1.41 as
learning continues to degrade direct costs.
There are thus two somewhat offsetting processes affecting total
costs derived from ICMs. The first is the learning curve, which reduces
direct costs, which raises the effective ICM-based markup. As noted
previously, learning reflects learned efficiencies in assembly methods
as well as reduced parts and materials costs. The second is the
application of a long-term ICM, which reduces the effective ICM-based
markup. This represents the reduced burden needed to maintain new
technologies once they are fully developed. In this case, the two
processes largely offset one another and produce an average real ICM
over this 14-year period that roughly equals the original short-term
ICM.
Figure VI-9 illustrates this process for each of the 4 technologies
used to represent the universe of fuel economy and CO2
improving technologies. As with the turbocharged engines, aerodynamic
improvements and mild hybrid vehicles show a gradual increase in the
effective ICM-based markup through the point where the long-term ICM is
applied. At that time, the ICM-based markup makes an abrupt decline
before beginning a gradual rise. The decline due to application of
long-term ICMs is particularly pronounced in the case of the mild
hybrid--even more so than for the advanced hybrid. The advanced hybrid
ICM behaves somewhat differently because it is shown through its
developing stages when more radical learning is applied, but only every
few years. This produces a significant step-up in ICM levels concurrent
with each learning
[[Page 24360]]
application, followed by a sharp decline when the long-term ICM is
applied. After that, it begins a gradual rise as more moderate learning
is applied to reflect its shift to a mature technology. Note that as
with the turbocharged downsized engine example above, for the
aerodynamic improvements and mild hybrid technologies, the offsetting
processes of learning and long-term ICMs result in an average ICM over
the full time frame that is roughly equal to the initial short-term
ICM. However, the advanced hybrid ICM rose to a level significantly
higher than the initial ICM. This is a direct function of the rapid
learning schedule applied in the early years to this developing
technology. Brand new technologies might thus be expected to have
effective lifetime ICM markups exceeding their initial ICMs, while more
mature technologies are more likely to experience ICMs over their
remaining life span that more closely approximate their initial ICMs.
[GRAPHIC] [TIFF OMITTED] TR30AP20.126
ICMs for these 4 technologies would drive the indirect cost markup
rate for the analysis. However, the effect on total costs is also a
function of the relative incidence of each of the 50+ technologies
shown in Table VI-26 which are assumed to have ICMs similar to one of
these 4 technologies. The net effect on costs of these ICMs is also
influenced by the learning curve appropriate to each technology,
creating numerous different and unique ICM paths. The average ICM
applied by the model is also a function of each technology's direct
cost and because ICMs are applied to direct costs, the measured
indirect cost is proportionately higher for any given ICM when direct
costs are higher. The average ICM applied to the fleet for any given
model year is calculated as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.127
where:
D = direct cost of each technology
A = application rate for each technology
ICM = average ICM applied to each technology
and n = 1, 2 . . . . 88
The CAFE model predicts technology application rates assuming
manufacturers will apply technologies
[[Page 24361]]
to meet standards in a logical fashion based on estimated costs and
benefits. The application rates will thus be different for each model
year and for each alternative scenario examined. For the MY 2017-2025
FRIA, to illustrate the effects of ICMs on total technology costs,
NHTSA calculated the weighted average ICM across all technologies for
the preferred alternative.\637\ This was done separately for each
vehicle type and then aggregated based on predicted sales of each
vehicle type used in the model. Results are shown in Table VI-29.
---------------------------------------------------------------------------
\637\ For each alternative, this rulemaking examined numerous
scenarios based on different assumptions, and these assumptions
could influence the relative frequency of selection of different
technologies, which in turn could affect the average ICM. The
scenario examined here assumed a 3 percent discount rate, a 1-year
payback period, real world application of expected civil penalties,
and reflects expected voluntary over-compliance by manufacturers.
[GRAPHIC] [TIFF OMITTED] TR30AP20.128
The ICM-based markups in Table VI-29 were derived in a manner
consistent with the way the RPE is measured, that is, they reflect
combined influences of direct cost learning and changes in indirect
cost requirements weighted by both the incidence of each technology's
adaptation and the relative direct cost of each technology. The results
indicate generally higher ICMs for passenger cars than for light
trucks. This is a function of the technologies estimated to be adopted
for each respective vehicle type, especially in later years when
hybrids and electric vehicles become more prevalent in the passenger
car fleet. The influence of these advanced vehicles is driven primarily
by their direct costs, which greatly outweigh the costs of other
technologies. This results in the application of much more weight to
their higher ICMs. This is most notable in MYs 2024 and 2025 for
passenger cars, when electric vehicles begin to enter the fleet. The
average ICM increased 0.013 in 2024 primarily because of these
vehicles. It immediately dropped 0.017 in 2025 because both an
additional application of steep (20 percent) learning is applied to the
direct cost of these vehicles (which reduces their relative weight),
and the long-term ICM becomes effective in that year (which decreases
the absolute ICM factor). Both influences occur one year after these
vehicles begin to enter the fleet because of CAFE requirements.
ICMs also change over time, again, reflecting the different mix of
technologies present during earlier years but that are often replaced
with more expensive technologies in later years. Across all model
years, the wide-ranging application of diverse technologies required to
meet CAFE and CO2 standards produced an average ICM-based
markup (or RPE equivalent) of approximately 1.34, applying only 67
percent of the indirect costs found in the RPE and implying total costs
11 percent below those predicted by the RPE-based calculation.
(4) Uncertainty
As noted above, the RPE and ICM assign different markups over
direct manufacturing costs, and thus imply different total cost
estimates for CAFE and CO2 technologies. While there is a
level of uncertainty associated with both markups, this uncertainty
stems from different issues. The RPE is derived from financial
statements and is thus grounded in historical data. Although
compilation of this data is subject to some level of interpretation,
the two independent researchers who derived RPE estimates from these
financial reports each reached essentially identical conclusions,
placing the RPE at roughly 1.5. All other estimates of the RPE fall
between 1.4 and 2.0, and most are between 1.4 and 1.7. There is thus a
reasonable level of consistency among researchers that RPEs are 1.4 or
greater. In addition, the RPE is a measure of the cumulative effects of
all operations manufacturers undertake in the course of producing their
vehicles, and is thus not specific to individual technologies, nor of
CAFE or CO2 technologies in particular. Because this
provides only a single aggregate measure, using the RPE multiplier
results in the application of a common incremental markup to all
technologies. This assures the aggregate cost effect across all
technologies is consistent with empirical data, but it does not allow
for indirect cost discrimination among different technologies or over
time. Because it is applied across all changes, this implies the markup
for some technologies is likely to be understated, and for others it is
likely to be overstated.
By contrast, the ICM process derives markups specific to several
CAFE and CO2 technologies, but these markups
[[Page 24362]]
have no basis in empirical data. They are based on informed judgment of
a panel of engineers with auto industry experience regarding cost
effects of a small sample (roughly 8 percent) of the 50+ technologies
applied to achieve compliance with CAFE and CO2 standards.
Uncertainty regarding ICMs is thus based both on the accuracy of the
initial assessments of the panel on the examined technologies and on
the assumption that these 4 technologies are representative of the
remaining technologies that were not examined. Both agencies attempted
to categorize these technologies in the most representative way
possible. However, while this represented the best judgment of EPA and
NHTSA's engineering staffs at that time, the actual effect on indirect
costs remains uncertain for most technologies. As with RPEs, this means
that even if ICMs were accurate for the specific technologies examined,
indirect cost will be understated for some technologies and overstated
for others.
There was considerable uncertainty demonstrated in the ICM panel's
assessments, as illustrated by the range of estimates among the 14
modified Delphi panel members surrounding the central values reported
by the panel. These ranges are shown in Table VI-30 and Figure VI-10,
Figure VI-11, and Figure VI-12 below. For the low complexity
technology, passive aerodynamic improvements, panel responses ranged
from a low of basically no indirect costs (1.001 short term and 1.0
long term), to a high of roughly a 40 percent markup (1.434 and 1.421).
For the medium complexity technology, turbo charged and downsized
engines, responses ranged from a low estimate implying almost no
indirect cost (1.018 and 1.011), to a high estimate implying that
indirect costs for this technology would roughly equal the average RPE
(1.5) for all technologies (1.527 and 1.445). For the high complexity
technology, plug-in hybrid electric vehicles, responses ranged from a
low estimate that these vehicles would require significantly less
indirect cost than the average RPE (1.367 and 1.121) to a high estimate
implying they would require more indirect costs than the average RPE
(2.153 and 1.691). There was considerable diversity of opinion among
the panel members.\638\ This is apparent in Figure VI-10, Figure VI-11,
and Figure VI-12, which show the 14 panel members' final estimates for
short-term ICMs as scatter plots.
---------------------------------------------------------------------------
\638\ Sample confidence intervals, which mitigate the effect of
outlying opinions, indicate a less extreme but still significant
range of ICMs. Applying mean ICMs helps mitigate these potential
differences, but there is clearly a significant level of uncertainty
regarding indirect costs. A t-distribution is used to estimate
confidence intervals because of the small sample size (14 panel
members).
[GRAPHIC] [TIFF OMITTED] TR30AP20.129
[[Page 24363]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.130
[GRAPHIC] [TIFF OMITTED] TR30AP20.131
[[Page 24364]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.132
Although these results were based on modified Delphi panel
techniques, it is apparent the goal of the Delphi process, an eventual
consensus or convergence of opinion among panel experts, was not
achieved. Given this lack of consensus and the divergence of ICM-based
results from the only available empirical measure (the RPE), there is
considerable uncertainty that current ICM estimates provide a realistic
basis of estimating indirect costs. ICMs have not been validated
through a direct accounting of actual indirect costs for individual
technologies, and they produce results that conflict with the only
available empirical evidence of indirect cost markups. Further, they
are intended to represent indirect costs specifically associated with
the most comprehensive redesign effort ever undertaken by the auto
industry, with virtually every make/model requiring ground-up design
modifications to comply. This includes entirely new vehicle design
concepts, extensive material substitution, and complete drivetrain
redesigns, all of which require significant research efforts and
assembly plant redesign. Under these circumstances, one might expect
indirect costs to equal or possibly increase above the historical
average, but not to decrease, as implied by estimated ICMs. For
regulations, such as the CAFE and CO2 emission standards
under consideration, that drive changes to nearly every vehicle system,
the overall average indirect costs should align with the RPE value.
Applying RPE to the cost for each technology assures that alignment.
In the 2015 NAS study, the Committee stated a conceptual agreement
with the ICM method because ICM takes into account design challenges
and the activities required to implement each technology. However,
although endorsing ICMs as a concept, the NAS Committee stated ``the
empirical basis for such multipliers is still lacking, and, since their
application depends on expert judgment, it is not possible to determine
whether the Agencies' ICMs are accurate or not.'' \639\ NAS also stated
``the specific values for the ICMs are critical because they may affect
the overall estimates of costs and benefits for the overall standards
and the cost effectiveness of the individual technologies.'' \640\ The
Committee encouraged continued research into ICMs given the lack of
empirical data for them to evaluate ICMs used by the agencies in past
analyses. On balance, and considering the relative merits of both
approaches for realistically estimating indirect costs, the agencies
consider the RPE method to be a more reliable basis for estimating
indirect costs.
---------------------------------------------------------------------------
\639\ National Research Council of the National Academies
(2015). Cost, Effectiveness, and Deployment of Fuel Economy
Technologies for Light-Duty Vehicles. https://www.nap.edu/resource/21744/deps_166210.pdf.
\640\ Ibid.
---------------------------------------------------------------------------
(5) Using RPE To Evaluate Indirect Costs in This Analysis
To ensure overall indirect costs in the analysis align with the
historical RPE value, the primary analysis has been developed based on
applying the RPE value of 1.5 to each technology. As noted previously,
the RPE is the ratio of aggregate retail prices to aggregate direct
manufacturing costs. The ratio already reflects the mixture of learned
costs of technologies at various stages of maturity. Therefore, the RPE
is applied directly to the learned direct cost for each technology in
each year. This was previously done in the MY 2017-2025 FRIA for the
preferred alternative for that rulemaking, used in the above analysis
of average ICMs. Results are shown in Table VI-31.
Recognizing there is uncertainty in any estimate of indirect costs,
a sensitivity analyses of indirect costs has also been conducted by
applying a lower RPE value as a proxy for the ICM approach. This value
was derived from a direct comparison of incremental technology costs
determined in the MY 2017-2025 FRIA.\641\ This analysis is summarized
in Table VI-31 below. From this table, total costs were estimated to be
roughly 18 percent lower using ICMs compared to the RPE. As previously
mentioned, there are two different reasons for these differences. The
first is the direct effect of applying a higher retail markup. The
second is an indirect effect resulting from the influence these
differing markups have on the order of the selection of technologies in
the CAFE model, which can change as different direct cost levels
interact with altered retail markups, shifting their relative overall
effectiveness.
---------------------------------------------------------------------------
\641\ See Table 5-9a in Final Regulatory Impact Analysis,
Corporate Average Fuel Economy for MY 2017-MY 2025 Passenger Cars
and Light Trucks.
---------------------------------------------------------------------------
The relative effects of ICMs may vary somewhat by scenario, but in
this case, the application of ICMs produces total
[[Page 24365]]
technology cost estimates roughly 18 percent lower than those that
would result from applying a single RPE factor to all technologies, or,
conversely, the RPE produces estimates that averaged 21 percent higher
than the ICM. Under the CAFE model construct, which will apply an
alternate RPE to the same base technology profile to represent ICMs,
this implies an RPE equivalent of 1.24 would produce similar net
impacts [1.5/(1 + x) = 1.21, x = 0.24]. This value is applied for the
ICM proxy estimate. Additional values were also examined over a range
of 1.1-2.0. The results, as well as the reference case using the 1.5
RPE, are summarized in Table VI-32.
[GRAPHIC] [TIFF OMITTED] TR30AP20.133
[GRAPHIC] [TIFF OMITTED] TR30AP20.134
Several responders submitted comments on the issue of indirect
costs. The International Council on Clean Transportation (ICCT) stated
that ``The agencies abandoned their previously-used indirect cost
multiplier method for estimating total costs, which was vetted with
peer review, and more complexly handled differing technologies with
different supply chain and manufacturing aspects. The agencies have, at
this point, opted to use a simplistic retail price equivalent method,
which crudely assumes all technologies have a 50 percent markup from
the direct manufacturing technology cost. We recommend the agencies
revert back to the previously-used and better substantiated ICM
approach.'' \642\
---------------------------------------------------------------------------
\642\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
A private commenter, Thomas Stephens, noted that ``In Section II.
Technical Foundation for NPRM Analysis, under 1. Data Sources and
Processes for Developing Individual Technology Assumptions, the
agencies state that indirect costs are estimated using a Retail Price
Equivalent (RPE) factor. Concerns with RPE factors and the difficulty
of accounting for differences in indirect costs of different
technologies when using this approach were identified by the EPA
(Rogozhin et al., Using indirect cost multipliers to estimate the total
cost of adding new technology in the automobile industry, International
Journal of Production Economics 124, 360-368, 2010), which suggested
using indirect cost (IC) multipliers instead of RPE factors. The EPA
developed and updated IC multipliers for relevant vehicle technologies
with automotive industry input and review. The agencies should consider
using these IC multipliers to estimate indirect manufacturing costs
instead of RPE factors.'' \643\
---------------------------------------------------------------------------
\643\ NHTSA-2018-0067-12067.
---------------------------------------------------------------------------
By contrast, the Alliance of Automobile Manufacturers (The
Alliance) ``supports the use of retail
[[Page 24366]]
price equivalents in the compliance cost modeling to estimate the
indirect costs associated with the additional added technology required
to meet a given future standard. The alternative indirect cost
multiplier (``ICM'') approach is not sufficiently developed for use in
rulemaking. As noted by the National Research Council, the indirect
cost multipliers previously developed by EPA have not been validated
with empirical data.\644\ Furthermore, in reference to the memorandum
documenting the development of ICMs previously used by EPA, Exponent
Failure Analysis Associates found that,
---------------------------------------------------------------------------
\644\ Cost, Effectiveness, and Development of Fuel Economy
Technologies for Light-Duty Vehicles, pages 248-49, National
research Council, the National Academies Press (2015).
---------------------------------------------------------------------------
Past Toyota Comments on Atkinson-Cycle Benefits Have Addressed Only
Those Derived From Variable Valve Timing
In response to these comments the agencies continue to find the RPE
approach preferable to the ICM approach, at least at this stage in the
development ICM estimates, for the reasons discussed both above and
previously in the NPRM. The agencies note that the concerns are not
with the concept of ICMs, but rather with the judgment-based values
suggested for use as ICMs, which have not been validated, and which
conflict with the empirically derived RPE value. The agencies will
continue to monitor any developments in ICM methodologies as part of
future rulemakings.
c) Stranded Capital Costs
Past analyses accounted for costs associated with stranded capital
when fuel economy standards caused a technology to be replaced before
its costs were fully amortized. The idea behind stranded capital is
that manufacturers amortize research, development, and tooling expenses
over many years, especially for engines and transmissions. The
traditional production life-cycles for transmissions and engines have
been a decade or longer. If a manufacturer launches or updates a
product with fuel-saving technology, and then later replaces that
technology with an unrelated or different fuel-saving technology before
the equipment and research and development investments have been fully
paid off, there will be unrecouped, or stranded, capital costs.
Quantifying stranded capital costs accounts for such lost investments.
In the Draft TAR and NPRM analyses, only a few technologies for a
few manufacturers were projected to have stranded capital costs. As
more technologies are included in this analysis, and as the CAFE model
has been expanded to account for platform and engine sharing and
updated with redesign and refresh cycles, accounting for stranded
capital has become increasingly complex. Separately, manufacturers may
be shifting their investment strategies in ways that may alter how
stranded capital calculations were traditionally considered. For
example, some suppliers sell similar transmissions to multiple
manufacturers. Such arrangements allow manufacturers to share in
capital expenditures, or amortize expenses more quickly.
Manufacturers share parts on vehicles around the globe, achieving
greater scale and greatly affecting tooling strategies and costs. Given
these trends in the industry and their uncertain effect on capital
amortization, and given the difficulty of handling this uncertainty in
the CAFE model, this analysis does not account for stranded capital.
The agencies' analysis continues to rely on the CAFE model's explicit
year-by-year accounting for estimated refresh and redesign cycles, and
shared vehicle platforms and engines, to moderate the cadence of
technology adoption and thereby limit the implied occurrence of
stranded capital and the need to account for it explicitly. The
agencies will monitor these trends to assess the role of stranded
capital moving forward.
d) Cost Learning
Manufacturers make improvements to production processes over time,
which often result in lower costs. ``Cost learning'' reflects the
effect of experience and volume on the cost of production, which
generally results in better utilization of resources, leading to higher
and more efficient production. As manufacturers gain experience through
production, they refine production techniques, raw material and
component sources, and assembly methods to maximize efficiency and
reduce production costs. Typically, a representation of this cost
learning, or learning curves, reflect initial learning rates that are
relatively high, followed by slower learning as additional improvements
are made and production efficiency peaks. This eventually produces an
asymptotic shape to the learning curve, as small percent decreases are
applied to gradually declining cost levels. These learning curve
estimates are applied to various technologies that are used to meet
CAFE standards.
For the NPRM and this final rule, the agencies estimated cost
learning by considering methods established by T.P. Wright \645\ and
later expanded upon by J.R. Crawford. Wright, examining aircraft
production, found that every doubling of cumulative production of
airplanes resulted in decreasing labor hours at a fixed percentage.
This fixed percentage is commonly referred to as the progress rate or
progress ratio, where a lower rate implies faster learning as
cumulative production increases. J.R. Crawford expanded upon Wright's
learning curve theory to develop a single unit cost model,\646\ that
estimates the cost of the nth unit produced given the following
information is known: (1) Cost to produce the first unit; (2)
cumulative production of n units; and (3) the progress ratio.
---------------------------------------------------------------------------
\645\ Wright, T.P., Factors Affecting the Cost of Airplanes.
Journal of Aeronautical Sciences, Vol. 3 (1936), pp. 124-125.
Available at http://www.uvm.edu/pdodds/research/papers/others/1936/wright1936a.pdf.
\646\ Crawford, J.R., Learning Curve, Ship Curve, Ratios,
Related Data, Burbank, California-Lockheed Aircraft Corporation
(1944).
---------------------------------------------------------------------------
As pictured in Figure VI-13, Wright's learning curve shows the
first unit is produced at a cost of $1,000. Initially cost per unit
falls rapidly for each successive unit produced. However, as production
continues, cost falls more gradually at a decreasing rate. For each
doubling of cumulative production at any level, cost per unit declines
20 percent, so that 80 percent of cost is retained. The CAFE model uses
the basic approach by Wright, where cost reduction is estimated by
applying a fixed percentage to the projected cumulative production of a
given fuel economy technology.
[[Page 24367]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.135
The analysis accounts for learning effects with model year-based
cost learning forecasts for each technology that reduce direct
manufacturing costs over time. The agencies evaluated the historical
use of technologies, and reviewed industry forecasts to estimate future
volumes for the purpose of developing the model year-based technology
cost learning curves.
The following section discusses the agencies' development of model
year-based cost learning forecasts, including how the approach has
evolved from the 2012 rulemaking for MY 2017-2025 vehicles, and how the
progress ratios were developed for different technologies considered in
the analysis. Finally, the agencies discuss how these learning effects
are applied in the CAFE Model.
(1) Time Versus Volume-Based Learning
For the 2012 joint CAFE/CO2 rulemaking, the agencies
developed learning curves as a function of vehicle model year.\647\
Although the concept of this methodology is derived from Wright's
cumulative production volume-based learning curve, its application for
CAFE and CO2 technologies was more of a function of time.
More than a dozen learning curve schedules were developed, varying
between fast and slow learning, and assigned to each technology
corresponding to its level of complexity and maturity. The schedules
were applied to the base year of direct manufacturing cost and
incorporate a percentage of cost reduction by model year declining at a
decreasing rate through the technology's production life. Some newer
technologies experience 20 percent cost reductions for introductory
model years, while mature or less complex technologies experience 0-3
percent cost reductions over a few years.
---------------------------------------------------------------------------
\647\ CAFE 2012 Final Rule, NHTSA DOT, 77 FR 62624.
---------------------------------------------------------------------------
In their 2015 report to Congress, the National Academy of Sciences
(NAS) recommended the agencies should ``continue to conduct and review
empirical evidence for the cost reductions that occur in the automobile
industry with volume, especially for large-volume technologies that
will be relied on to meet the CAFE/GHG standards.'' \648\
---------------------------------------------------------------------------
\648\ Cost, Effectiveness, and Deployment of Fuel Economy
Technologies for Light-Duty Vehicles, National Research Council of
the National Academies (2015), available at https://www.nap.edu/resource/21744/deps_166210.pdf.
---------------------------------------------------------------------------
In response, the agencies have incorporated statically projected
cumulative volume production data of fuel economy improving
technologies, representing an improvement over the previously used
time-based method. Dynamic projections of cumulative production are not
feasible with current CAFE model capabilities, so one set of projected
cumulative production data for most vehicle technologies was developed
for the purpose of determining cost impact. For many technologies
produced and/or sold in the U.S., historical cumulative production data
was obtained to establish a starting point for learning schedules.
Groups of similar technologies or technologies of similar complexity
may share identical learning schedules.
The slope of the learning curve, which determines the rate at which
cost reductions occur, has been estimated using research from an
extensive literature review and automotive cost tear-down reports (see
below). The slope of the learning curve is derived from the progress
ratio of manufacturing automotive and other mobile source technologies.
(2) Deriving the Progress Ratio Used in This Analysis
Learning curves vary among different types of manufactured
products. Progress ratios can range from 70 to 100 percent, where 100
percent indicates no learning can be achieved.\649\ Learning effects
tend to be greatest in operations where workers often touch the
product, while effects are less substantial in operations consisting of
more automated processes. As automotive manufacturing plant processes
become increasingly automated, a progress ratio towards the higher end
would seem more suitable. The agencies incorporated findings from
automotive cost-teardown studies with EPA's literature review of
learning-related studies to estimate a progress ratio used to determine
learning schedules of fuel economy improving technologies.
---------------------------------------------------------------------------
\649\ Martin, J., ``What is a Learning Curve?'' Management and
Accounting Web, University of South Florida, available at: https://www.maaw.info/LearningCurveSummary.htm.
---------------------------------------------------------------------------
EPA's literature review examined and summarized 20 studies related
to learning in manufacturing industries and mobile source
manufacturing.\650\
[[Page 24368]]
The studies focused on many industries, including motor vehicles,
ships, aviation, semiconductors, and environmental energy. Based on
several criteria, EPA selected five studies providing quantitative
analysis from the mobile source sector (progress ratio estimates from
each study are summarized in Table VI-33, below). Further, those
studies expand on Wright's Learning Curve function by using cumulative
output as a predictor variable, and unit cost as the response variable.
As a result, EPA determined a best estimate of 84 percent as the
progress ratio in mobile source industries. However, of those five
studies, EPA at the time placed less weight on the Epple et al. (1991)
study, because of a disruption in learning due to incomplete knowledge
transfer from the first shift to introduction of a second shift at a
North American truck plant. While learning may have decelerated
immediately after adding a second shift, the agencies note that unit
costs continued to fall as the organization gained experience operating
with both shifts. The agencies now recognize that disruptions are an
essential part of the learning process and should not, in and of
themselves, be discredited. For this reason, the analysis uses a re-
estimated average progress ratio of 85 percent from those five studies
(equally-weighted).
---------------------------------------------------------------------------
\650\ Cost Reduction through Learning in Manufacturing
Industries and in the Manufacture of Mobile Sources, United States
Environmental Protection Agency (2015). Prepared by ICF
International and available at https://19january2017snapshot.epa.gov/sites/production/files/2016-11/documents/420r16018.pdf.
[GRAPHIC] [TIFF OMITTED] TR30AP20.136
In addition to EPA's literature review, this progress ratio
estimate was informed based on NHTSA's findings from automotive cost-
teardown studies. NHTSA routinely performs evaluations of costs of
previously issued Federal Motor Vehicle Safety Standards (FMVSS) for
new motor vehicles and equipment. NHTSA's engages contractors to
perform detailed engineering ``tear-down'' analyses for representative
samples of vehicles, to estimate how much specific FMVSS add to the
weight and retail price of a vehicle. As part of the effort, cost and
production volume are examined for automotive safety technologies. In
particular, the agency estimated costs from multiple cost tear-down
studies for technologies with actual production data from the Cost and
weight added by the Federal Motor Vehicle Safety Standards for MY 1968-
2012 passenger cars and LTVs (2017).\656\
---------------------------------------------------------------------------
\651\ Argote, L., Epple, D., Rao, R. D., & Murphy, K., The
acquisition and depreciation of knowledge in a manufacturing
organization--Turnover and plant productivity, Working paper,
Graduate School of Industrial Administration, Carnegie Mellon
University (1997).
\652\ Benkard, C. L., Learning and Forgetting--The Dynamics of
Aircraft Production, The American Economic Review, Vol. 90(4), pp.
1034-54 (2000).
\653\ Epple, D., Argote, L., & Devadas, R., Organizational
Learning Curves--A Method for Investigating Intra-Plant Transfer of
Knowledge Acquired through Learning by Doing, Organization Science,
Vol. 2(1), pp. 58-70 (1991).
\654\ Epple, D., Argote, L., & Murphy, K., An Empirical
Investigation of the Microstructure of Knowledge Acquisition and
Transfer through Learning by Doing, Operations Research, Vol. 44(1),
pp. 77-86 (1996).
\655\ Levitt, S. D., List, J. A., & Syverson, C., Toward an
Understanding of Learning by Doing--Evidence from an Automobile
Assembly Plant, Journal of Political Economy, Vol. 121 (4), pp. 643-
81 (2013).
\656\ Simons, J. F., Cost and weight added by the Federal Motor
Vehicle Safety Standards for MY 1968-2012 Passenger Cars and LTVs
(Report No. DOT HS 812 354). Washington, DC--National Highway
Traffic Safety Administration (November 2017), at pp. 30-33.
---------------------------------------------------------------------------
NHTSA chose five vehicle safety technologies with sufficient data
to estimate progress ratios of each, because these technologies are
large-volume technologies and are used by almost all vehicle
manufacturers. Table VI-34 below includes these five technologies and
yields an average progress rate of 92 percent:
[GRAPHIC] [TIFF OMITTED] TR30AP20.137
[[Page 24369]]
For a final progress ratio used in the CAFE model, the five
progress rates from EPA's literature review and five progress rates
from NHTSA's evaluation of automotive safety technologies results were
averaged. This resulted in an average progress rate of approximately 89
percent. Equal weight was placed on progress ratios from all 10
sources. More specifically, equal weight was placed on the Epple et al.
(1991) study, because disruptions have more recently been recognized as
an essential part in the learning process, especially in an effort to
increase the rate of output. Further discussion of how the progress
ratios were derived for this analysis is located in FRIA Section 9.
ICCT commented that the choice to use safety technology as a model
for fuel efficiency led to lower learning rates in the NPRM analysis
compared to prior analyses.\657\ ICCT stated that safety technologies
were chosen for the NPRM because they are used by almost every
manufacturer, in contrast to fuel efficiency technologies, where not
every manufacturer will use them, particularly when they are first
introduced. ICCT stated that to show the impact of changing learning
rates, the agencies should run a sensitivity analysis using the
learning rates in the TAR, as well as EPA's learning rates in its Final
Determination. ICCT concluded that ``[w]ithout doing so and without
conducting a peer review of the change in approach, it appears clear
the agencies have decided to switch to a new costing method that
affects all future costs, but without any significant research
justification, vetting, or review.''
---------------------------------------------------------------------------
\657\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
The agencies' selection of a progress rate of 0.89 is based on an
average of findings across research and literature reviews conducted by
NHTSA and EPA. The EPA cited rates were derived from five studies
selected from a sample of 20 transportation modal learning studies that
were examined by an EPA contractor, ICF International.\658\ One of
these 5 studies (Benkard (2000) examines learning in the commercial
aircraft industry, which the author notes has many unique features that
influence marginal costs. It also has the lowest progress rate. The
agencies note that EPA regulates all mobile sources, and while the
inclusion of non-passenger vehicle studies in their report was
justified, it may have biased the estimate of learning attributable to
the motor vehicle industry. Notably, nearly all of the other studies
included in the ICF International study found progress rates higher
than the 0.84 rate selected by the authors at that time. In reviewing
the ICF study, NHTSA found many other studies not included in the
report, including many specific to the motor vehicle and environmental
technology industries. Over 90 percent of those studies indicated
higher progress ratios than ICF recommended.\659\ The agencies' current
approach includes a broader and more representative sample of these
studies rather than the narrow sample selected by ICF.
---------------------------------------------------------------------------
\658\ Cost Reduction through Learning in Manufacturing
Industries and in the Manufacture of Mobile Sources. United States
Environmental Protection Agency. Prepared by ICF International and
available at: https://19january2017snapshot.epa.gov/sites/production/files/2016-11/documents/420r16018.pdf.
\659\ See, for example, progress ratios of multiple technologies
referenced in The Carbon Productivity Challenge: Curbing Climate
Change and Sustaining Economic Growth, McKinsey Climate Change
Special Initiative, McKinsey Global Institute, June 2008 (quoting
from UC Berkeley Energy Resource Group, Navigant Consulting) and
Technology Innovation for Climate Mitigation and its Relation to
Government Policies, Edward S. Rubin, Carnegie Mellon University,
Presentation to the UNFCCC Workshop on Climate Change Mitigation,
Bonn, Germany, June 19, 2004.
---------------------------------------------------------------------------
The agencies do not agree that safety technologies are adopted by
all manufacturers at an early stage. Most safety technologies are
initially offered as options or standard equipment on only a small
segment of the vehicle fleet, typically luxury vehicles. After a number
of years, these technologies may be adopted on less expensive vehicles,
and eventually they will become required equipment on all vehicles, but
the production process is gradual, as it is with fuel efficiency
technologies. FMVSS are necessarily established as performance
standards--and automakers are free to develop or choose from existing
technologies to achieve such performance requirements--much like
automakers can develop or choose from a number of established fuel
efficiency technologies to achieve fuel economy requirements. Further,
the derivation of progress ratios is based on the concept of a doubling
of cumulative production, not time. Therefore, even if production
continues at a different pace, it should not disqualify non-fuel
efficiency studies. Moreover, the derivation of the progress ratio used
in the TAR and Final Determination document were not confined to fuel
efficiency technologies. In fact, as noted above, they even included at
least one entirely unrelated study of the aircraft industry.
Finally, the agencies note that the previous learning schedules
used in the TAR and EPA's Final Determination were only developed
through 2025, whereas this final rule projects learning through 2050.
The previous learning schedules are thus not directly compatible with
the analysis conducted in this Final Rule, making a sensitivity
analysis problematic.
(3) Obtaining Appropriate Baseline Years for Direct Manufacturing Costs
To Create Learning Curves
Direct manufacturing costs for each fuel economy improving
technology were obtained from various sources, as discussed above. To
establish a consistent basis for direct manufacturing costs in the
rulemaking analysis, each technology cost is adjusted to MY 2018
dollars. For each technology, the DMC is associated with a specific
model year, and sometimes a specific production volume, or cumulative
production volume. The base model year is established as the MY in
which direct manufacturing costs were assessed (with learning factor of
1.00). With the aforementioned data on cumulative production volume for
each technology and the assumption of a 0.89 progress ratio for all
automotive technologies, the agencies can solve for an implied cost for
the first unit produced. For some technologies, the agencies used
modestly different progress ratios to match detailed cost projections
if available from another source (for instance, batteries for plug-in
hybrids and battery electric vehicles).
This approach produced reasonable estimates for technologies
already in production, and some additional steps were required to set
appropriate learning rates for technologies not yet in production.
Specifically, for technologies not yet in production in MY 2017 (the
baseline analysis fleet), the cumulative production volume in MY 2017
is zero, because manufacturers have not yet produced the technologies.
For pre-production cost estimates in the NPRM, the agencies often
relied on confidential business information sources to predict future
costs. Many sources for pre-production cost estimates include
significant learning effects, often providing cost estimates assuming
high volume production, and often for a timeframe late in the first
production generation or early in the second generation of the
technology. Rapid doubling and re-doubling of a low cumulative volume
base with Wright's learning curves can provide unrealistic cost
estimates. In addition, direct manufacturing cost projections can vary
depending on the initial production volume assumed. Accordingly, the
agencies carefully examined direct costs with learning, and made
adjustments to the starting point for those technologies on the
learning curve to better align
[[Page 24370]]
with the assumptions used for the initial direct cost estimate.
(4) Cost Learning as Applied in the CAFE Model
For the NPRM analysis, the agencies updated the manner in which
learning effects apply to costs. In the Draft TAR analysis, the
agencies had applied learning curves only to the incremental direct
manufacturing costs or costs over the previous technology on the
technology tree. In practice, two things were observed: (1) If the
incremental direct manufacturing costs were positive, technologies
could not become less expensive than their predecessors on the
technology tree, and (2) absolute costs over baseline technology
depended on the learning curves of root technologies on the technology
tree. For the NPRM and final rule analysis, the agencies applied
learning effects to the incremental cost over the null technology state
on the applicable technology tree. After this step, the agencies
calculated year-by-year incremental costs over preceding technologies
on the tech tree to create the CAFE model inputs. As discussed below,
for the final rule, the agencies revised the CAFE model to replace
incremental cost estimates with absolute estimates, each specified
relative to the null technology state on the applicable technology
tree. This change facilitated quality assurance and is expected to make
cost inputs more transparently relatable to detailed model output.
Likewise, this change made it easier to apply learning curves in the
course of developing inputs to the CAFE model.
The agencies grouped certain technologies, such as advanced
engines, advanced transmissions, and non-battery electric components
and assigned them to the same learning schedule. While these grouped
technologies differ in operating characteristics and design, the
agencies chose to group them based on their complexity, technology
integration, and economies of scale across manufacturers. The low
volume of certain advanced technologies, such as hybrid and electric
technologies, poses a significant issue for suppliers and prevents them
from producing components needed for advanced transmissions and other
technologies at more efficient high scale production. The technology
groupings were carried over from the NPRM analysis for the final rule
analysis.\660\ Like the NPRM, this final rule analysis uses the same
groupings that considers market availability, complexity of technology
integration, and production volume of the technologies that can be
implemented by manufacturers and suppliers. For example, technologies
like ADEAC and VCR are grouped together; these technologies were not in
production or were only in limited introduction in MY 2017, and are
planned to be introduced in limited production by a few manufacturers.
The details of these technologies are discussed in Section VI.C.
---------------------------------------------------------------------------
\660\ See PRIA Chapter 6 for technology groupings.
---------------------------------------------------------------------------
In addition, for the final rule, as discussed in Section VI.A.4
Compliance Simulation, the agencies expanded model inputs to extend the
explicit simulation of technology application through MY 2050, in
response to comments on the NPRM. Accordingly, the agencies updated the
learning curves for each technology group to cover MYs through 2050.
For MYs 2017-2032, the agencies expect incremental improvements in all
technologies, particularly in electrification technologies because of
increased production volumes, labor efficiency, improved manufacturing
methods, specialization, network building, and other factors. While
these and other factors contribute to continual cost learning, the
agencies believe that many fuel economy improving technologies
considered in this rule will approach a flat learning level by the
early 2030s. Specifically, older and less complex internal combustion
engine technologies and transmissions will reach a flat learning curve
sooner when compared to electrification technologies, which have more
opportunity for improvement. For batteries and non-battery
electrification components, the agencies estimated a steeper learning
curve that will gradually flatten after MY 2040. For a more detailed
discussion of the electrification learning curves used for the final
rule analysis, see Section VI.C.3.e) Electrification Costs. The
following Table VI-35 and Table VI-36 show the learning curve schedules
for CAFE model technologies for MYs 2017-2033 and MYs 2034-2050.
BILLING CODE 4910-59-P
[[Page 24371]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.138
[[Page 24372]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.139
[[Page 24373]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.140
[[Page 24374]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.141
BILLING CODE 4910-59-C
Each technology in the CAFE Model is assigned a learning schedule
developed from the methodology explained previously. For example, the
[[Page 24375]]
following chart shows learning rates for several technologies
applicable to midsize sedans, demonstrating that while the agencies
estimate that such learning effects have already been almost entirely
realized for engine turbocharging (a technology that has been in
production for many years), the agencies estimate that significant
opportunities to reduce the cost of the greatest levels of mass
reduction (e.g., MR5) remain, and even greater opportunities remain to
reduce the cost of batteries for HEVs, PHEVs, BEVs. In fact, for
certain advanced technologies, the agencies determined that the results
predicted by the standard learning curves progress ratio was not
realistic, based on unusual market price and production relationships.
For these technologies, the agencies developed specific learning
estimates that may diverge from the 0.89 progress rate. As shown in
Figure VI-14, these technologies include: Turbocharging and downsizing
level 1 (TURBO1), variable turbo geometry electric (VTGE), aerodynamic
drag reduction by 15 percent (AERO15), mass reduction level 5 (MR5), 20
percent improvement in low-rolling resistance tire technology over the
baseline, and battery integrated starter/generator (BISG).
[GRAPHIC] [TIFF OMITTED] TR30AP20.142
(5) Potential Future Approaches to Considering Cost Learning in the
CAFE Model
As discussed above, cost inputs to the CAFE model incorporate
estimates of volume-based learning. As an alternative approach, the
agencies have considered modifications to the CAFE model that would
calculate degrees of volume-based learning dynamically, responding to
the model's application of affected technologies. While it is intuitive
that the degree of cost reduction achieved through experience producing
a given technology should depend on the actual accumulated experience
(i.e., volume) producing that technology, such dynamic implementation
in the CAFE model is thus far infeasible. Insufficient data have been
available regarding manufacturers' historical application of specific
technology. Further, insofar as the agencies' estimates of underlying
direct manufacturing costs already make some assumptions about volume
and scale, insufficient information is currently available to determine
how to dynamically adjust these underlying costs. It should be noted
that if learning responds dynamically to volume, and volume responds
dynamically to learning, an internally consistent model solution would
likely require iteration of the CAFE model to seek a stable solution
within the model's representation of multiyear planning. As discussed
below, the CAFE model now supports iteration to balance vehicle
[[Page 24376]]
cost and fuel economy changes with corresponding changes in sales
volumes, but, this iteration is not yet implemented in a manner that
would necessarily support the balance of learning effects on a
multiyear basis. The agencies invited comment on the issue, seeking
data and methods that would provide the basis for a practicable
approach to doing so. Having reviewed comments on cost learning
effects, the agencies conclude it remains infeasible to calculate
degrees of volume-based learning in a manner that responds dynamically
to modeled technology application. The agencies will continue to
examine this issue for future development.
e) Cost Accounting
The CAFE model applied for the NPRM analysis used an incremental
approach to specifying technology cost estimates, such that the cost
for any given technology was specified as an incremental value,
relative to the technology immediately preceding on the relevant
technology pathway. For example, the cost of a 7-speed transmission was
specified as an amount beyond the cost of a 6-speed transmission. This
approach necessitated careful dynamic accounting for the progressive
application of the technology as the model worked on a step-by-step
basis to ``build'' a technology solution. As discussed in the
corresponding model documentation, the model included complex logic to
``back out'' some of these costs carefully when, for example, replacing
a conventional powertrain with a hybrid-electric system.\661\
---------------------------------------------------------------------------
\661\ The CAFE Model is available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system with documentation and all inputs and outputs supporting
today's notice.
---------------------------------------------------------------------------
To facilitate specification of detailed model inputs and review of
detailed model outputs, today's CAFE model replaces incremental cost
inputs with absolute cost inputs, such that the estimated cost of each
technology is specified relative to a common reference point for the
relevant technology pathway. For example, the cost of the above-
mentioned 7-speed transmission is specified relative to a 4-speed
transmission, as is the cost of every other transmission technology.
This change in the structure of cost inputs does not, by itself, change
model results, but it does make the connection between these inputs and
corresponding outputs more transparent. Model documentation
accompanying today's analysis presents details of the updated structure
for model cost inputs.
5. Other Inputs to the Agencies' Analysis
CAFE Model input files described above defining the analysis fleet
and the fuel-saving technologies to be included in the analysis span
more than a million records, but deal with a relatively discrete range
of subjects (e.g., what vehicles are in the fleet, what are the key
characteristics of those vehicles, what fuel-saving technologies are
expected to be available, and how might adding those technologies
impact vehicles' fuel economy levels and costs). The CAFE Model makes
use of a considerably wider range of other types of inputs, and most of
these are contained in other model input files. The nature and function
of many of these inputs remains unchanged relative to the model and
input files applied for the analysis documented in the proposal that
preceded today's notice. The CAFE Model documentation accompanying
today's notice lists and describes all model inputs, and explains how
inputs are used by the model. Many commenters addressed not only the
model's function and design, but also specific inputs. Most input
values are discussed either above (e.g., the preceding subsection
addresses specific inputs regarding technology costs) or below, in
subsections discussing specific economic, energy, safety, and
environmental factors. The remainder of this subsection provides an
overview of the scope of different model input files. The overview is
organized based on CAFE Model file types, as in the model
documentation.
a) Market Data File
The ``Market Data'' file contains the detailed description--
discussed above--of the vehicle models and model configurations each
manufacturer produces for sale in the U.S. The file also contains a
range of other inputs that, though not specific to individual vehicle
models, may be specific to individual manufacturers. The file contains
a set of specific worksheets, as follows:
``Manufacturers'' worksheet: Lists specific manufacturers,
indicates whether manufacturers are expected to prefer paying CAFE
fines to applying technologies that would not be cost-effective,
indicates what ``payback period'' defines buyers' willingness to pay
for fuel economy improvements, enumerates CAFE and CO2
credits banked from model years prior to those represented explicitly,
and indicates how sales ``multipliers'' are to be applied when
simulating compliance with CO2 standards.
``Credits and Adjustments'' worksheet: Enumerates estimates--
specific to each manufacturer and fleet--of expected CO2 and
CAFE adjustments reflecting improved AC efficiency, reduced AC
refrigerant leakage, improvements to ``off cycle'' efficiency, and
production of flexible fuel vehicles (FFVs). The model applies AC
refrigerant leakage adjustments only to CO2 levels, and
applies FFV adjustments only to CAFE levels.
``Vehicles'' worksheet: Lists vehicle models and model
configurations each manufacturer produces for sale in the U.S.;
identifies shared vehicle platforms; indicates which engine and
transmission is present in each vehicle model configuration; specifies
each vehicle model configuration's fuel economy level, production
volume, and average price; specifies several engineering
characteristics (e.g., curb weight, footprint, and fuel tank volume);
assigns each vehicle model configuration to a regulatory class,
technology class, engine class, and safety class; specifies schedules
on which specific vehicle models are expected to be redesigned and
freshened; specifies how much U.S. labor is involved in producing each
vehicle model/configuration; and indicates whether specific
technologies are already present on specific vehicle model
configurations, or, due to engineering or product planning
considerations, should be skipped.
``Engines'' worksheet: Identifies specific engines used by each
manufacturer and for each engine, lists a unique code (referenced by
the engine code specified for each vehicle model configuration and
identifies the fuel(s) with which the engine is compatible, the
valvetrain design (e.g., DOHC), the engine's displacement, cylinder
configuration and count, and the engine's aspiration type (e.g.,
naturally aspirated, turbocharged). The worksheet also indicates
whether specific technologies are already present on specific engines,
or, due to engineering or product planning considerations, should be
skipped.
``Transmissions'' worksheet: Similar to the Engines worksheet,
identifies specific transmissions used by each manufacturer and for
each transmission, lists a unique code (referenced by the transmission
code specified for each vehicle model configuration and identifies the
type (e.g., automatic or CVT) and number of forward gears. Also
indicates whether specific technologies are already present or, due to
engineering or product planning considerations, should be skipped.
[[Page 24377]]
b) Technologies File
The Technologies file identifies about six dozen technologies to be
included in the analysis, indicates when and how widely each technology
can be applied to specific types of vehicles, provides most of the
inputs involved in estimating what costs will be incurred, and provides
some of the inputs involved in estimating impacts on vehicle fuel
consumption and weight. The file contains the following types of
worksheets:
``Parameters'' worksheet: Not to be confused with the
``Parameters'' file discussed below, this worksheet in the Technologies
file indicates, for each technology class, the share of the vehicle's
curb weight represented by the ``glider'' (the vehicle without the
powertrain).
``Technologies'' worksheet: For each named technology, specifies
the share of the entire fleet to which the technology may be
additionally applied in each model year.
Technology Class worksheets: In a separate worksheet for each of
the 10 technology classes discussed above (and an additional 2--not
used for this analysis--for heavy-duty pickup trucks and vans),
identifies whether and how soon the technology is expected to be
available for wide commercialization, specifies the percentage of miles
a vehicle is expected to travel on a secondary fuel (if applicable, as
for plug-in hybrid electric vehicles), indicates a vehicle's expected
electric power and all-electric range (if applicable), specifies
expected impacts on vehicle weight, specifies estimates of costs in
each model year (and factors by which electric battery costs are
expected to be reduced in each model year), specifies any estimates of
maintenance and repair cost impacts, and specifies any estimates of
consumers' willingness to pay for the technology.
Engine Type worksheets: In a separate worksheet for each of 28
initial engine types identified by cylinder count, number of cylinder
banks, and configuration (DOHC, unless identified as OHV or SOHC),
specifies estimates of costs in each model year, as well as any
estimates of impacts on maintenance and repair costs.
c) Parameters File
The ``Parameters'' file contains inputs spanning a range of
considerations, such as economic and labor utilization impacts, vehicle
fleet characteristics, fuel prices, scrappage and safety model
coefficients, fuel properties, and emission rates. The file contains a
set of specific worksheets, as follows:
Economic Values worksheet: Specifies a variety of inputs, including
social and consumer discount rates to be applied, the ``base year'' to
which to discount social benefits and costs (i.e., the reference years
for present value analysis), discount rates to be applied to the social
cost of CO2 emissions, the elasticity of highway travel with
respect to per-mile fuel costs (also referred to as the rebound
effect), the gap between test (for certification) and on-road (aka real
world) fuel economy, the fixed amount of time involved in each refuel
event, the share of the tank refueled during an average refueling
event, the value of travel time (in dollars per hour per vehicle), the
estimated average number of miles between mid-trip EV recharging events
(separately for 200 and 300-mile EVs), the rate (in miles of capacity
per hour of charging) at which EV batteries are recharged during such
events, the values (in dollars per vehicle-mile) of congestion and
noise costs, costs of vehicle ownership and operation (e.g., sales
tax), economic costs of oil imports, estimates of future macroeconomic
measures (e.g., GDP), and rates of growth in overall highway travel
(separately for low, reference, and high oil prices).
Vehicle Age Data worksheet: Specifies nominal average survival
rates and annual mileage accumulation for cars, vans and SUVs, and
pickup trucks. These inputs are used only for displaying estimates of
avoided fuel savings and CO2 emissions while the model is
operating. Calculations reported in model output files reflect, among
other things, application of the scrappage model.
Fuel Prices worksheet: Separately for gasoline, E85, diesel,
electricity, hydrogen, and CNG, specifies historical and estimated
future fuel prices (and average rates of taxation). Includes values
reflecting low, reference, and high estimates of oil prices.
Scrappage Model Values worksheet: Specifies coefficients applied by
the scrappage model, which the CAFE Model uses to estimate rates at
which vehicles will be scrapped (removed from service) during the
period covered by the analysis.
Historic Fleet Data worksheet: For model years not simulated
explicitly (here, model years through 2016), and separately for cars,
vans and SUVs, and pickup trucks, specifies the initial size (i.e.,
number new vehicles produced for sale in the U.S.) of the fleet, the
number still in service in the indicated calendar year (here, 2016),
the relative shares of different fuel types, and the average fuel
economy achieved by vehicles with different fuel types, and the
averages of horsepower, curb weight, fuel capacity, and price (when
new).
Safety Values worksheet: Specifies coefficients used to estimate
the extent to which changes in vehicle mass impact highway safety. Also
specifies statistical value of highway fatalities, the share of
incremental risk (of any additional driving) internalized by drivers,
rates relating the cost of damages from non-fatal losses to the cost of
fatalities, and rates relating the occurrence of non-fatal injuries to
the occurrence of fatalities.
Fatality Rates worksheet: Separately for each model year from 1975-
2050, and separately for each vehicle age (through 39 years) specifies
the estimated nominal number of fatalities incurred per billion miles
of travel by which to offset fatalities.
Credit Trading Values worksheet: Specifies whether various
provisions related to compliance credits are to be simulated (currently
limited to credit carry-forward and transfers), and specifies the
maximum number of years credits may be carried forward to future model
years. Also specifies statutory (for CAFE only) limits on the quantity
of credit that may be transferred between fleets, and specifies amounts
of lifetime mileage accumulation to be assumed when adjusting the value
of transferred credits. Also accommodates a setting indicating the
maximum number of model years to consider when using expiring credits.
Employment Values worksheet: Specifies the estimated average
revenue OEMs and suppliers earn per employee, the retail price
equivalent factor applied in developing technology costs, the average
quantity of annual labor (in hours) per employee, a multiplier to apply
to U.S. final assembly labor utilization in order to obtain estimated
direct automotive manufacturing labor, and a multiplier to be applied
to all labor hours.
Fuel Properties worksheet: Separately for gasoline, E85, diesel,
electricity, hydrogen, and CNG, specifies energy density, mass density,
carbon content, and tailpipe SO2 emissions (grams per unit
of energy).
Fuel Import Assumptions worksheet: Separately for gasoline, E85,
diesel, electricity, hydrogen, and CNG, specifies the extent to which
(a) changes in fuel consumption lead to changes in net imports of
finished fuel, (b) changes in fuel consumption lead to changes in
domestic refining output, (c) changes in domestic refining output lead
to changes in domestic crude oil production, and (d) changes in
domestic refining output lead to changes in net imports of crude oil.
[[Page 24378]]
Emissions Health Impacts worksheet: Separately for NOX,
SO2 and PM2.5 emissions, separately for upstream
and vehicular emissions, and for each of calendar years 2016, 2020,
2025, and 2030, specifies estimates of various health impacts, such as
premature deaths, acute bronchitis, and respiratory hospital
admissions.
Carbon Dioxide Emission Costs worksheet: For each calendar year
through 2080, specifies low, average, and high estimates of the social
cost of CO2 emissions, in dollars per metric ton.
Accommodates analogous estimates for CH4 and N2O.
Criteria Pollutant Emission Costs worksheet: Separately for
NOX, SO2 and PM2.5 emissions,
separately for upstream and vehicular emissions, and for each of
calendar years 2016, 2020, 2025, and 2030, specifies social costs on a
per-ton basis.
Upstream Emissions (UE) worksheets: Separately for gasoline, E85,
diesel, electricity, hydrogen, and CNG, and separately for calendar
years 2017, 2020, 2025, 2030, 2035, 2040, 2045, and 2050, and
separately for various upstream processes (e.g., petroleum refining),
specifies emission factors (in grams per million BTU) for each included
criteria pollutant (e.g., NOX) and toxic air contaminant
(e.g., benzene).
Tailpipe Emissions (TE) worksheets: Separately for gasoline and
diesel, for each of model years 1975-2050, for each vehicle vintage
through age 39, specifies vehicle tailpipe emission factors (in grams
per mile) for CO, VOC, NOX, PM2.5,
CH4, N2O, acetaldehyde, acrolein, benzene,
butadiene, formaldehyde, and diesel PM10.
d) Scenarios File
The CAFE Model represents each regulatory alternative as a discrete
scenario, identifying the first-listed scenario as the baseline
relative to which impacts are to be calculated. Each scenario is
described in a worksheet in the Scenarios input file, with standards
and related provisions specified separately for each regulatory class
(passenger car or light truck) and each model year. Inputs specify the
standards' functional forms and defining coefficients in each model
year. Multiplicative factors and additive offsets are used to convert
fuel economy targets to CO2 targets, the two being directly
mathematically related by a linear transformation. Additional inputs
specify minimum CAFE standards for domestic passenger car fleets,
determine whether upstream emissions from electricity and hydrogen are
to be included in CO2 compliance calculations, specify the
governing rates for CAFE civil penalties, specify estimates of the
value of CAFE and CO2 credits (for CAFE Model operating
modes applying these values), specify how flexible fuel vehicles (FFVs)
and PHEVs are to be accounted for in CAFE compliance calculations,
specific caps on adjustments reflecting improvements to off-cycle and
AC efficiency and emissions, specify any estimated amounts of average
Federal tax credits earned by HEVs, PHEVs, BEVs, and FCVs. The
worksheets also accommodate some other inputs, such those as involved
in analyzing standards for heavy-duty pickups and vans, not used in
today's analysis.
e) ``Run Time'' Settings
In addition to inputs contained in the above-mentioned files, the
CAFE Model makes use of some settings selected when operating the
model. These include which standards (CAFE or CO2) are to be
evaluated; what model years the analysis is to span; when technology
application is to begin; what ``effective cost'' mode is to be used
when selecting among technologies; whether use of compliance credits is
to be simulated and, if so, until what model year; whether dynamic
economic models are to be exercised and, if so, how many sales model
iterations are to be undertaken and using what price elasticity;
whether low, average, or high estimates are to be applied for fuel
prices, the social cost of carbon, and fatality rates; by how much to
scale benefits to consumers; and whether to report an implicit
opportunity cost.
f) Simulation Inputs
As mentioned above, the CAFE Model makes use of databases of
estimates of fuel consumption impacts and, as applicable, battery costs
for different combinations of fuel saving technologies. For today's
analysis, the agencies developed these databases using a large set of
full vehicle and accompanying battery cost model simulations developed
by Argonne National Laboratory. To be used as files provided separately
from the model and loaded every time the model is executed, these
databases are prohibitively large, spanning more than a million records
and more than half a gigabyte. To conserve space and speed model
operation, the agencies have integrated the databases into the CAFE
Model executable file. When the model is run, however, the databases
are extracted and placed in an accessible location on the user's disk
drive. The databases, each of which is in the form of a simple (if
large) text file, are as follows:
``FE1_Adjustments.csv:'' This is the main database of fuel
consumption estimates. Each record contains such estimates for a
specific indexed (using a multidimensional ``key'') combination of
technologies for each of the technology classes in the Market Data and
Technologies files. Each estimate is specified as a percentage of the
``base'' technology combination for the indicated technology class.
``FE2_Adjustments.csv:'' Specific to PHEVs, this is a database of
fuel consumption estimates applicable to operation on electricity,
specified in the same manner as those in the main database.
``Battery_Costs.csv:'' Specific to technology combinations
involving vehicle electrification (including 12V stop-start systems),
this is a database of estimates of corresponding base costs (before
learning effects) for batteries in these systems.
g) On Road Fuel Economy and CO2 Emissions Gap
Rather than rely on the compliance values of fuel economy for
either historical vehicles or vehicles that go through the full
compliance simulation, the model applies an ``on-road gap'' to
represent the expected difference between fuel economy on the
laboratory test cycle and fuel economy under real-world operation. In
other words, all of the reported physical impacts analysis (including
emissions impacts) are based on actual real world fuel consumption and
emissions, not on values based on 2-cycle fuel economy ratings and
CO2 emission rates, nor on regulatory incentives such as
sales multipliers that treat a single vehicle as two vehicles, or that
set aside emissions resulting from generation of electricity to power
electric vehicles. This was a topic of interest in the recent peer
review of the CAFE model. While the model currently allows the user to
specify an on-road gap that varies by fuel type (gasoline, E85, diesel,
electricity, hydrogen, and CNG), it does not vary over time, by vehicle
age, or by technology combination. It is possible that the ``gap''
between laboratory fuel economy and real-world fuel economy has changed
over time, that fuel economy changes as a vehicle ages, or that
specific combinations of fuel-saving technologies have a larger
discrepancy between laboratory and real-world fuel economy than others.
For today's analysis, and considering data EPA collects from
manufacturers regarding vehicles' fuel economy and CO2 as
tested for both fuel economy and emissions compliance and for vehicle
fuel economy and emissions labeling
[[Page 24379]]
(labeling making use of procedures spanning a wider range of real-world
vehicle operating conditions), the agencies have determined that the
future gap is, at this time, best estimated using the same values
applied for the analysis documented in the NPRM. The agencies will
continue to assess such test data and any other available data
regarding real-world fuel economy and emissions and, as warranted, will
revise methods and inputs representing the gap between laboratory and
real-world fuel economy and CO2 emissions in future
rulemakings. The sensitivity analysis summarized in the FRIA
accompanying the final rule includes cases representing narrower and
wider gaps.
C. The Model Applies Technologies Based on a Least-Cost Technology
Pathway to Compliance, Given the Framework Above
The CAFE model, discussed in detail above, is designed to simulate
compliance with a given set of CAFE or tailpipe CO2
emissions standards for each manufacturer that sells vehicles in the
United States. For the final rule analysis, the model began with a
representation of the MY 2017 vehicle model offerings for each
manufacturer that included the specific engines and transmissions on
each model variant, observed sales volumes, and all fuel economy
improving technology that is already present on those vehicles. From
there the model added technology, in response to the standards being
considered, in a way that minimized the cost of compliance and
reflected many real-world constraints faced by automobile
manufacturers. The model addressed fleet year-by-year compliance,
taking into consideration vehicle refresh and redesign schedules and
shared platforms, engines, and transmissions among vehicles.
The agencies evaluated a wide array of technologies manufacturers
could use to improve the fuel economy of new vehicles, in both the
immediate future and during the timeframe of this rulemaking, to meet
the fuel economy and CO2 standards. The agencies evaluated
costs for these technologies, and looked at how costs may change over
time. The agencies also considered how fuel-saving technologies may be
used on many types of vehicles (ranging from small cars to trucks) and
how the technologies may perform in improving fuel economy and
CO2 emissions in combination with other technologies. With
cost and effectiveness estimates for technologies, the agencies
forecast how manufacturers may respond to potential standards and can
estimate the associated costs and benefits related to technology and
equipment changes. This assists the assessment of technological
feasibility and is a building block for the consideration of economic
practicability of the standards.
The agencies described in the NPRM that the characterization of
current and anticipated fuel-saving technologies relied on portions of
the analysis presented in the Draft TAR, in addition to new information
that had been gathered and developed since conducting that analysis,
and the significant, substantive input that was received during the
Draft TAR comment period.\662\ The Draft TAR considered many
technologies previously assessed in the 2012 final rule; \663\ in some
cases, manufacturers have nearly universally adopted a technology in
today's new vehicle fleet (for example, electric power steering), but
in other cases, manufacturers only occasionally use a technology in
today's new vehicle fleet (like turbocharged engines). For a few
technologies considered in the 2012 rulemaking, manufacturers began
implementing the technologies but have since largely pivoted to other
technologies due to consumer acceptance issues (for instance,
drivability and performance feel issues associated with some dual
clutch transmissions without a torque converter) or limited commercial
success.
---------------------------------------------------------------------------
\662\ 83 FR 43021-22 (Aug. 24, 2018).
\663\ 77 FR 62624 (Oct. 15, 2012).
---------------------------------------------------------------------------
In some cases, EPA and NHTSA presented different analytical
approaches in the Draft TAR. However, for the NPRM and final rule
analysis, the agencies harmonized their analytical approach to use one
set of effectiveness values (developed with one tool), one set of cost
assumptions, and one set of assumptions about the limitations of some
technologies. To develop these assumptions, the agencies evaluated many
sources of data, in addition to many stakeholder comments received on
the Draft TAR. The preferred approach was to harmonize on sources and
methodologies that were data-driven and reproducible for independent
verification, produced using tools utilized by OEMs, suppliers, and
academic institutions, and using tools that could support both CAFE and
CO2 analysis. As the agencies noted in the NPRM, a single
set of assumptions also facilitated and focused public comment by
reducing burden on stakeholders who sought to review all of the
supporting documentation surrounding the analysis.
The agencies also identified a preference to use values developed
from careful review of commercialized technologies; however, in some
cases for technologies that are new, and are not yet for sale in any
vehicle, the analysis relied on information from other sources,
including CBI and third-party research reports and publications. The
agencies strived to keep the technology analysis as current as possible
in light of the ongoing technology development and implementation in
the automotive industry. Additional emerging technologies added for the
final rule analysis are described in further detail, below.
The agencies' process to develop effectiveness assumptions is
described in detail in Section VI.B.3 Technology Effectiveness, and
summarized here. The NPRM and final rule analysis modeled combinations
of more than 50 fuel economy-improving technologies across 10 vehicle
types (an increase from five vehicle types in NHTSA's Draft TAR
analysis). Only 10 vehicle technology classes were used because large
portions of the production volume in the analysis fleet have similar
specifications, especially in highly competitive segments. For
instance, many mid-sized sedans, small SUVs, and large SUVs coalesce
around similar specifications, respectively. Baseline simulations have
been aligned around these modal specifications. Parametrically
combining these technologies generated more than 100,000 unique
combinations per vehicle class. Multiplying the unique technology
combinations by the 10 technology classes resulted in the simulation of
more than one million individual full-vehicle system models. Modeling
was also conducted to determine appropriate levels of engine downsizing
required to maintain baseline vehicle performance when advanced mass
reduction technology or advanced engine technology were applied.
Performance neutrality is discussed in detail in VI.B.3.
Some baseline vehicle assumptions used in the simulation modeling
were updated since the Draft TAR based on public comments, and further
assessment of the NPRM and final rule analysis fleets. The agencies
updated assumptions about curb weight, as well as technology properties
like baseline rolling resistance, aerodynamic drag coefficients, and
frontal areas. Many of the assumptions are aligned with published
research from the Department of Energy and other independent
[[Page 24380]]
sources.\664\ Additional transmission technologies and more levels of
aerodynamic technologies than NHTSA presented in the Draft TAR analysis
were also added for the analysis. Having additional technologies in the
model allowed the agencies to assign baselines and estimate fuel-
savings opportunities with more precision.
---------------------------------------------------------------------------
\664\ See, e.g., Islam, E., A. Moawad, N. Kim, and A. Rousseau,
2018a, An Extensive Study on Vehicle Sizing, Energy Consumption and
Cost of Advance Vehicle Technologies, Report No. ANL/ESD-17/17,
Argonne National Laboratory, Lemont, Ill., Oct 2018. https://www.autonomie.net/pdfs/ANL_BaSce_FY17_Report_10042018.pdf. Last
accessed March 18, 2020; Pannone, G. ``Technical Analysis of Vehicle
Load Reduction Potential for Advanced Clean Cars,'' April 29, 2015.
Available at https://www.arb.ca.gov/research/apr/past/13-313.pdf.
Last accessed December 28, 2019.
---------------------------------------------------------------------------
To develop technology cost assumptions, the agencies estimated
present and future costs for fuel-saving technologies, taking into
consideration the type of vehicle, or type of engine if technology
costs vary by application. Since the 2012 final rule, many cost
assessments, including tear down studies, were funded and completed,
and presented as part of the Draft TAR analysis. These studies
evaluated transmissions, engines, hybrid technologies, and mass
reduction.\665\ The NPRM and final rule analyses use the 2016 Draft
TAR's cost estimates for many technologies. In addition to those
studies, the analysis also leveraged research reports from other
organizations to assess costs.\666\ Consistent with past analyses, this
analysis used BatPaC to provide estimates for future battery costs for
hybrids, plug-in hybrids, and electric vehicles, taking into account
the different battery design characteristics and taking into account
the size of the battery for different applications.\667\ The agencies
also updated technology costs for the NPRM to 2016 dollars, because, as
in many cases, technology costs were estimated several years ago, and
since then have further updated technology costs to 2018 dollars for
the final rule.
---------------------------------------------------------------------------
\665\ FEV prepared several cost analysis studies for EPA on
subjects ranging from advanced 8-speed transmissions to belt
alternator starter, or Start/Stop systems. NHTSA also contracted
with Electricore and EDAG on teardown studies evaluating mass
reduction. The 2015 NAS report on fuel economy technologies for
light-duty vehicles also evaluated the agencies' technology costs
developed based on these teardown studies, and the technology costs
used in this proposal were updated accordingly.
\666\ For example, the agencies relied on reports from the
Department of Energy's Office of Energy Efficiency & Renewable
Energy's Vehicle Technologies Office. More information on that
office is available at https://www.energy.gov/eere/vehicles/vehicle-technologies-office. Other agency reports that were relied on for
technology or other information are referenced throughout the NPRM
and accompanying PRIA, and this final rule and the accompanying
FRIA.
\667\ For instance, battery electric vehicles with high levels
of mass reduction may use a smaller battery than a comparable
vehicle with less mass reduction technology and still deliver the
same range on a charge. See, e.g., Ward, J. & Gohlke, D. & Nealer,
Rachael. (2017). The Importance of Powertrain Downsizing in a
Benefit-Cost Analysis of Vehicle Lightweighting. JOM. 69.
---------------------------------------------------------------------------
Cost and effectiveness values were estimated for each technology
included in the analysis. As mentioned above, more than 50 technologies
were considered in the NPRM and final rule analyses, and the agencies
evaluated many combinations of these technologies in many applications.
In the NPRM, the agencies identified overarching potential issues in
assessing technology effectiveness and cost, including:
Baseline vehicle technology level assessed as too low, or
too high. Compliance information was extensively reviewed and
supplemented with available literature on the vehicle models considered
in the analysis fleet. Manufacturers could also review the baseline
technology assignments for their vehicles, and the analysis
incorporates feedback received from manufacturers.
Technology costs too low or too high. Tear down cost
studies, CBI, literature, and the 2015 NAS study information were
referenced to estimate technology costs. In cases where one technology
appeared to exceed all other technologies on cost and effectiveness,
information was acquired from additional sources to confirm or reject
assumptions. Cost assumptions for emerging technologies were reassessed
in cases where new information became available.
Technology effectiveness too high or too low in
combination with other vehicle technologies. Technology effectiveness
was evaluated using the Autonomie full-vehicle simulation modeling,
taking into account the impact of other technologies on the vehicle and
the vehicle type. Inputs and modeling for the analysis took into
account laboratory test data for production and some pre-production
technologies, technical publications, manufacturer and supplier CBI,
and simulation modeling of specific technologies. Evaluating recently
introduced production products to inform the technology effectiveness
models of emerging technologies was preferred; however, some
technologies that are not yet in production were considered using CBI.
Simulation modeling used carefully chosen baseline configurations to
provide a consistent, reasonable reference point for the incremental
effectiveness estimates.
Vehicle performance not considered or applied in an
infeasible manner. Performance criteria, including low speed
acceleration (0-60 mph time), high speed acceleration (50-80 mph time),
towing, and gradeability (six percent grade at 65 mph) were also
considered. In the simulation modeling, resizing was applied to achieve
the same performance level as the baseline for the least capable
performance criteria but only with significant design changes. The
analysis struck a balance by employing a frequency of engine downsizing
that took product complexity and economies of scale into account.
Availability of technologies for production application
too soon or too late. A number of technologies were evaluated that are
not yet in production. CBI was gathered on the maturity and timing of
these technologies and the cadence at which manufacturers could adopt
these technologies.
Product complexity and design cadence constraints too low
or too high. Product platforms, refresh and redesign cycles, shared
engines, and shared transmissions were also considered in the analysis.
Product complexity and the cadence of product launches were matched to
historical values for each manufacturer.
Customer acceptance under estimated or over estimated.
Resale prices for hybrid vehicles, electric vehicles, and internal
combustion engine vehicles were evaluated to assess consumer
willingness to pay for those technologies. The analysis accounts for
the differential in the cost for those technologies and the amount
consumers have actually paid for those technologies. Separately, new
dual-clutch transmissions and manual transmissions were applied to
vehicles already equipped with these transmission architectures.
The agencies sought comments on all assumptions for fuel economy
technology costs, effectiveness, availability, and applicability to
vehicles in the fleet.
Several commenters compared the technology effectiveness and cost
estimates from prior rulemaking actions to the NPRM, some commenting
that the NPRM analysis represented a better balance of input from all
stakeholders regarding the potential costs and benefits of future fuel
economy improving technologies,\668\ and some commenting that the NPRM
analysis represented a step back from the Draft TAR and EPA's Proposed
Determination in terms of both the analysis itself and the resulting
conclusions about the level of technology required to meet the
[[Page 24381]]
augural standards.\669\ Specifically, while some commenters stated that
the Draft TAR and subsequent EPA midterm review documents had recently
concluded that augural standards were achievable with very low levels
of electrification based on currently available information on
technology effectiveness and cost,\670\ other commenters reiterated
that conventional gasoline powertrains alone were insufficient to
achieve post-2021 model year targets.\671\
---------------------------------------------------------------------------
\668\ See, e.g., NHTSA-2018-0067-11928.
\669\ See, e.g., NHTSA-2018-0067-11873.
\670\ See, e.g., NHTSA-2018-0067-11969.
\671\ See, e.g., NHTSA-2018-0067-12150.
---------------------------------------------------------------------------
Generally, the automotive industry supported the agencies' NPRM
analysis over previous analyses. In addition to the automotive
industry's support of the agencies' use of one modeling tool for
analysis, discussed in Section IV, above, the industry also commented
in support of specific technology effectiveness, cost, and adoption
assumptions used in the updated analysis.
The Alliance commented in support of the NPRM modeling approach,
and referenced important technology-specific features of the modeling
process, including ``The acknowledgement and application of real-world
limitations on technology application including a limit on the number
of engine displacements available to any one manufacturer, application
of shared platforms, engines, and transmissions, and the reality that
improvements and redesigns of components are not only extended across
vehicles but sometimes constrained in implementation opportunity to
common vehicle redesign cycles; recognition of the need for
manufacturers to follow ``technology'' pathways that retain capital and
implementation expertise, such as specializing in one type of engine or
transmission instead of following an unconstrained optimization that
would cause manufacturers to leap to unrelated technologies and show
overly optimistic costs and benefits; the application of specific
instead of generic technology descriptions that allow for the above-
mentioned real-world constraints; [and] the need to accommodate for
intellectual property rights in that not all technologies will be
available to all manufacturers.'' \672\
---------------------------------------------------------------------------
\672\ NHTSA-2018-0067-12073, at 9.
---------------------------------------------------------------------------
More specifically, the Alliance commented that the analysis
appropriately restricted the application of some technologies, like the
application of low rolling resistance tires on performance vehicles,
and limited aerodynamic improvements for trucks and minivans.\673\
Similarly, the Alliance commented in support of the decision to exclude
HCR2 technology from the analysis, citing previous comments stating
that ``the inexplicably high benefits ascribed to this theoretical
combination of technologies has not been validated by physical
testing.''
---------------------------------------------------------------------------
\673\ NHTSA-2018-0067-12073, at 134.
---------------------------------------------------------------------------
Ford commented more broadly that ``[t]he previous analyses
performed by the Agencies too often selected technology benefits from
the high-end of the forecasted range, and cost from the lower-end, in
part because deference was given to supplier or other third-party
claims over manufacturers' estimates.'' \674\ Ford noted that,
``[m]anufacturer estimates, while viewed as conservative by some, are
informed by years of experience integrating new technologies into
vehicle systems in a manner that avoids compromising other important
attributes (NVH, utility, safety, etc.),'' continuing that ``[t]he need
to preserve these attributes often limits the actualized benefit of a
new technology, an effect insufficiently considered in projections from
most non-OEM sources.'' Ford concluded, as mentioned above, that the
NPRM analysis better balanced these considerations.
---------------------------------------------------------------------------
\674\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
Toyota commented that the discrepancy between the automotive
industry and prior regulatory assessments stemmed from ``agency
modeling relying on overly optimistic assumptions about technology cost
effectiveness and deployment rates.'' \675\ Toyota pointed to a prior
analysis that projected compliance for Toyota's MY 2025 lineup using
the ALPHA model as an example of how ``the agency's analysis failed to
account for customer requirements (cost, power, weight-adding options,
etc.) that erode optimal fuel economy, and normal business
considerations that govern the pace of technology deployment.'' In
contrast, Toyota stated that the ``[m]odeled technology cost,
effectiveness, and compliance pathways in the proposed rulemaking rely
on more recent data as well as more realistic assumptions about the
level of technology already on the road today, the pace of technology
deployment, and trade-offs between vehicle efficiency and customer
requirements.''
---------------------------------------------------------------------------
\675\ NHTSA-2018-0067-12150.
---------------------------------------------------------------------------
Honda, in its feedback on the models used in the standard setting
process, commented that ``the current version of the CAFE model is
reasonably accurate in terms of technology efficiency, cost, and
overall compliance considerations, and reflects a notable improvement
over previous agency modeling efforts conducted over the past few
years.'' \676\
---------------------------------------------------------------------------
\676\ NHTSA-2018-0067-11818.
---------------------------------------------------------------------------
FCA commented in recognition of the CAFE model improvements over
the Draft TAR version, but noted they ``continue to believe that the
cost and benefits used as inputs to the model are overly optimistic.''
\677\ FCA used its updated Jeep Wrangler Unlimited and Ram 1500 pickup
models as examples of vehicles that ``provide real life examples of the
costs and benefits that can be achieved with fuel and weight saving
technology;'' however, ``after all of the real world concerns such as
emissions, drivability, OBD, and fuels are considered, the benefits
observed remain less than those derived by the Autonomie model and used
as inputs to the Volpe model.''
---------------------------------------------------------------------------
\677\ NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
Conversely, environmental groups, consumer groups, and some States
and localities commented that the Draft TAR and subsequent EPA analyses
were more representative of the current state of vehicle technologies.
These groups all generally commented, in different terms, that the NPRM
analysis technology effectiveness was understated and technology costs
were overstated, and additional constraints the agencies placed on the
analysis, like excluding technologies already in production or
constraining technology pathways, also helped lead to that result.\678\
---------------------------------------------------------------------------
\678\ NHTSA-2018-0067-11873; NHTSA-2018-0067-11984.
---------------------------------------------------------------------------
ICCT commented that the agencies ``ignored their own rigorous 2015-
2017 technological assessment, and have adopted a series of invalid and
unsupportable decisions which artificially constrain the availability
and dramatically under-estimate levels of effectiveness of many
different fuel economy improvement and GHG-reduction technologies and
unreasonably increase modeled compliance costs.'' \679\ ICCT also
commented that the agencies ignored, suppressed, dismissed, or
restricted the use of work done to update technologies and technology
cost and effectiveness assessments since the 2012 final rule for MYs
2017-2025. ICCT stated that the ``invalid high cost result [of the
modeled augural standards in 2025] was created by the agencies by
making many dozens of unsupported changes in the technology
effectiveness and availability inputs, the technology cost inputs, and
the technology package constraints.''
[[Page 24382]]
ICCT stated that ``the agencies failed to capture the latest available
information and, as a result, their assessment incorrectly and
artificially overstates technology costs.''
---------------------------------------------------------------------------
\679\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
CARB commented that the agencies did not present sufficient new
evidence to change previous technical findings, specifically in regards
to conventional vehicle technologies.\680\ CARB stated that instead of
relying on new information, as had been asserted as justification for
the proposal, the analysis was based on older data that did not reflect
current technology. Accordingly, CARB pointed out that previous
analysis by the agencies projected far less need for electrification
than what was required in the proposal, stating that the underlying
cause is a reduction in the assumed cumulative improvements for what
advanced gasoline technology is able to achieve.
---------------------------------------------------------------------------
\680\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
A coalition of States and Cities similarly commented that ``[t]he
Agencies' conclusions regarding the technology necessary to meet the
2025 standards and the cost of that technology run counter to the
evidence before the agency, diverge from prior factual findings without
explanation and without transparency as to the source of data relied
on, and are unsupported by any reasoned analysis. Such analysis bears
many hallmarks of an arbitrary and capricious action.'' \681\
---------------------------------------------------------------------------
\681\ NHTSA-2018-0067-11735 (citing State Farm, 463 U.S. at 43;
Fox Television, 556 U.S. at 515; Humane Soc. of U.S. v. Locke, 626
F.3d 1040, 1049 (9th Cir. 2010)).
---------------------------------------------------------------------------
Roush Industries, commenting on behalf of CARB, commented that
``the 2018 PRIA projected average costs for technology implementation
to achieve the existing standards to be significantly overstated and in
conflict with the 2016 Draft TAR cost estimates generated by the
Agencies only two years earlier.'' \682\ Roush commented that the Draft
TAR analyses of cost and incremental fuel economy improvement necessary
to achieve the augural standards was consistent with Roush's own
estimates and other published data.
---------------------------------------------------------------------------
\682\ NHTSA-2018-0067-11984.
---------------------------------------------------------------------------
Similarly, H-D Systems (HDS), commenting on behalf of the
California DOJ, commented that ``the estimates in the 2016 TAR on
technology cost and effectiveness still represent the correct estimates
based on the latest available data.'' \683\ HDS, in its analysis of the
costs of technologies to meet different potential standards between the
Draft TAR and the NPRM, noted that ``costs for most conventional (i.e.,
non-electric) drivetrain technologies were similar in both reports in
that costs were within +5% of the average of the costs from the two
reports. The only exception was the cost estimate for the High CR
second generation Atkinson cycle or HCR2 engine which was estimated to
be much more expensive. Due to differences in nomenclature,
transmission technology costs could not be directly compared but were
similar at the highest efficiency level. In contrast, cost of hybrid
technology was estimated to be much higher in the PRIA and were 200 to
250% higher for strong hybrids. Costs of drag reduction, rolling
resistance reduction and auxiliary system technologies were also quite
similar but the cost of mass reduction was substantially higher in the
PRIA by a factor of 2 to 3. Costs of engine friction reduction appear
not to be included in the cost computation for the PRIA although the
technology appears to be integrated into some of the engine technology
packages analyzed in the PRIA to estimate effectiveness.''
---------------------------------------------------------------------------
\683\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
CFA commented that ``[t]he overarching discussion of technology
developments that introduces the NHTSA analysis is fundamentally flawed
and infects the entire proposal,'' taking issue with the NPRM statement
that ``some options considered in the original order for the National
Program ha[d] not worked out as EPA/NHTSA anticipated.'' \684\ CFA
commented that the agencies failed to note that some technology options
have performed better than anticipated, and ``the fact that some
technologies have done better than expected is a basis for increasing
the standards, not in the context of a mid-term review that was
supposed to tweak the long-term program.''
---------------------------------------------------------------------------
\684\ NHTSA-2018-0067-12005.
---------------------------------------------------------------------------
NCAT commented that the ``inflation of projected technology costs
does not appear to be attributable primarily to the projected cost of
any given technology, but rather to modeling constraints on the
application of such technologies to vehicles. Many of these constraints
appear to be arbitrary and NHTSA's departure from prior analyses in
these respects is not adequately supported.'' \685\
---------------------------------------------------------------------------
\685\ NHTSA-2018-0067-11969.
---------------------------------------------------------------------------
Environmental groups and States also commented that the agencies
either should reincorporate all the Draft TAR or the EPA Proposed and
Final Determination analyses' technologies, technology effectiveness
values, and technology costs into the analysis, and/or compare the
final rule analysis with those prior analyses to show how the updated
assumptions changed the results from those prior analyses.
For example, ICCT commented that ``[f]or the agencies to conduct a
credible regulatory assessment they must remove all the technology
availability constraints, re-incorporate and make available the full
portfolio of technology options as was available in EPA's analysis for
the original 2017 Final Determination, and include at least 15 g/mile
CO2 for off-cycle credits by 2025, to credibly reflect the
real-world technology developments in the auto industry.'' \686\ ICCT
also stated that ``[t]he agencies need to identify each and every
technology cost input used in their modeling, and provide a clear
engineering and evidence based justification for why that cost differs
from the costs employed in the extremely well documented and well
justified Draft TAR and in EPA's 2016 TSD and 2017 Final Determination,
taking into account the above discussion of significant new evidence
developed since those prior estimates were made. Absent such disclosure
and justification, the default assumption needs to be that the prior
costs estimated based on the most recent data are more appropriate than
the estimates used for the proposal.''
---------------------------------------------------------------------------
\686\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
In addition, groups of commenters were equally split on the ability
of technologies to meet different compliance targets. For example, the
Alliance commented that ``the only technologies that have demonstrated
the improvements necessary to meet the MY 2025 standards are strong
hybrids, plug-in electric vehicles, and fuel cell electric vehicles.
The Agencies' analysis for this Proposed Rule predict the need for
significant growth in sales of electrified vehicles, a finding
consistent with third-party analyses.'' \687\ In contrast, UCS
commented that electrified powertrains ``are not especially relevant
for the MY 2022-2025 regulations.'' \688\
---------------------------------------------------------------------------
\687\ NHTSA-2018-0067-Alliance at 15.
\688\ NHTSA-2018-0067-UCS at 23.
---------------------------------------------------------------------------
The agencies are aware that the prior analyses concluded that
compliance with the augural standards could largely be met through
advances in gasoline vehicle technologies, and with only very low
levels of strong hybrids and electric vehicles. As the agencies stated
in the NPRM, consistent with both agencies' statutes, the proposal was
entirely de novo, based on an entirely new analysis reflecting the best
and most up-to-date information available to the agencies at the time
of this rulemaking.\689\ As discussed in Section IV, Section VI.B, and
further below, the NPRM and final rule analyses reflect updates to
[[Page 24383]]
technology effectiveness estimates, technology costs, and the
methodology for applying technologies to vehicles that the agencies
believed better represent the state of technology and the associated
costs compared to prior analyses, that result in pathways to compliance
that look both similar and different to those in prior analyses.
---------------------------------------------------------------------------
\689\ 83 FR 42897.
---------------------------------------------------------------------------
That said, several of the effectiveness and cost values used in the
NPRM and final rule analysis were directly carried over from the 2012
rule for MYs 2017-2025, Draft TAR, and EPA Midterm Evaluation
analyses.\690\ Several others were carried over from the 2015 NAS
report,\691\ which the agencies heavily relied upon in past analyses
even if specific cost or effectiveness values were not used. Different
technology effectiveness estimates, cost estimates, or adoption
constraints were employed where the agencies had information, from
technical reports, manufacturers, or other stakeholders, indicating
that a technology could or could not be feasibly adopted in the
rulemaking timeframe, or a technology could or could not be adopted in
the way that the agencies had previously modeled it. Notably, most
differences in pathways to compliance are attributable to only a few
significant differences between this rulemaking analysis and prior
rulemaking analyses.
---------------------------------------------------------------------------
\690\ See, e.g., PRIA at 449, 451, 452, 453, 458.
\691\ See, e.g., PRIA at 358-360.
---------------------------------------------------------------------------
For example, as discussed in Section VI.B.3 Technology
Effectiveness and Modeling and Section VI.C.1 Engine Paths, in the EPA
Draft TAR and Proposed Determination analyses, effectiveness of HCR
engine technologies and downsized turbocharged engine technologies were
estimated using Tier 2 certification fuel. Tier 2 certified fuel has a
higher octane rating compared to regular octane
fuel.692 693 694 As summarized by EPA in the PD TSD, ``EPA's
estimate of effectiveness for gasoline-fueled engines and engine
technologies was based on Tier 2 Indolene fuel although protection for
operation in-use on Tier 3 gasoline (87 AKI E10) was included in the
analysis of engine technologies considered both within the Draft TAR
and Proposed Determination. Additionally, in the technology assessment
for this Proposed Determination, EPA has considered the required engine
sizing and associated effectiveness adjustments when performance
neutrality is maintained on 87AKI gasoline typical of real-world use.''
\695\
---------------------------------------------------------------------------
\692\ Draft TAR at 5-228.
\693\ Tier 2 fuel has an octane rating of 93. Typical regular
grade fuel has an octane rating of 87 ((R+M)/2 octane.
\694\ EPA Proposed Determination TSD at 2-209 to 2-212.
\695\ EPA Proposed Determination TSD at 2-210.
---------------------------------------------------------------------------
NHTSA's effectiveness analysis for the Draft TAR used some engine
maps also developed using premium octane gasoline. However, at the time
NHTSA stated the agency would ensure all future engine model
development will be performed with regular grade octane gasoline.\696\
Commenters like Ford stated the effectiveness estimates for turbo
downsized engine packages were too high, in part because of the use of
high octane fuel. However they also commented in appreciation of
NHTSA's acknowledgement that any subsequent analysis would be based on
fuel at an appropriate octane level, as they stated the impact of the
change needed to be reflected in future analyses.\697\
---------------------------------------------------------------------------
\696\ Draft TAR at 5-504, 5-512.
\697\ Ford Motor Company Response to the Draft TAR September 26,
2016 NHTSA-2016-0068-0048, at 4.
---------------------------------------------------------------------------
Engine specifications used to create the engine maps for the NPRM
and the final rule analysis were developed using Tier 3 fuel to assure
the engines were capable of operating on real world regular octane (87
pump octane = (R+M/2)). The process was similar to what manufacturers
must do to ensure engines have acceptable noise, vibration, harshness,
drivability, performance, and will not fail prematurely when operated
on regular octane fuel. This eliminated the need for any adjustments
that were applied in the 2016 Draft TAR and PD TSD to account for Tier
2 to Tier 3 fuel properties. This accounts for some of the
effectiveness and cost differences for engine technologies between the
Draft TAR/Proposed Determination and the NPRM/final rule. For more
details, see Section VI.C.1 Engine Paths.
The agencies believe ICCT's and other commenters' assertions that
the engine maps should reflect Tier 2 fuel and not be updated for Tier
3 fuel would ignore these important considerations, and would provide
engine maps that could not achieve the fuel economy improvements unless
operated on high octane fuel. Therefore, the agencies determined that
engine maps developed for the Draft TAR and EPA Proposed Determination
that were based on Tier 2 fuel should not be used for the NPRM and
final rule analyses for these technical reasons.
As another related example, the agencies described that prior
analyses had relied heavily on the availability of the HCR2 (or ATK2)
``future'' Atkinson Cycle engine as a cost-effective pathway to
compliance for stringent alternatives, but many engine experts
questioned its technical feasibility and near-term commercial
practicability.\698\ The agencies explained that EPA staff began
theoretical development of this conceptual engine with a best-in-class
2.0L Atkinson cycle engine and then increased the efficiency of the
engine map further, through the theoretical application of additional
technologies in combination, including cylinder deactivation, engine
friction reduction, and cooled exhaust gas recirculation. While the
potential of such an engine is interesting, nevertheless the engine
remains entirely speculative. No production HCR2/ATK2 engine, as
outlined in the EPA SAE paper,\699\ has ever been commercially
produced. Furthermore, the engine map has not been validated with
hardware, bench data, or even on a prototype level (as no such engine
exists to test to validate the engine map).
---------------------------------------------------------------------------
\698\ 83 FR 43038.
\699\ Schenk, C. and Dekraker, P., ``Potential Fuel Economy
Improvements from the Implementation of cEGR and CDA on an Atkinson
Cycle Engine,'' SAE Technical Paper 2017-01-1016, 2017. Available at
https://doi.org/10.4271/2017-01-1016.
---------------------------------------------------------------------------
Vehicle manufacturers also commented on EPA's effectiveness
assumptions and estimates of HCR2/ATK2 model's future penetration
levels in the Draft TAR, stating ``[t]he effectiveness values for the
`futured' ATK2 package--projected at 40% penetration in 2025MY and
includes cooled exhaust gas recirculation (CEGR) and cylinder
deactivation (DEAC)--are too high, primarily due to overtly-optimistic
efficiencies in the base engine map, insufficient accounting of CEGR
and DEAC integration losses, and no accounting of the impact of 91RON
Tier 3 test fuel,'' and that ``44% fleet-wide penetration of ATK2 in
2025MY is unrealistic given the limited number of powertrain refresh
cycles available before 2025MY. In addition, it is unreasonable to
assume that OEMs already heavily invested in different high-efficiency
powertrain pathways (e.g., turbo-downsizing) would be able to commit
the immense resources needed to reach these high ATK2 penetration
levels in such a short time.'' \700\
---------------------------------------------------------------------------
\700\ Ford Motor Company Response to the Draft TAR September 26,
2016 NHTSA-2016-0068-0048, at 4.
---------------------------------------------------------------------------
Accordingly, the agencies decided to not include HCR2 technology in
the NPRM and final rule analysis. The engine model was not used because
no observable physical demonstration of the speculative technology
combination model has yet been created. Further,
[[Page 24384]]
many questions remain about the model's practicability as specified,
especially in high load, low engine speed operating conditions. The
HCR2 model combines multiple technologies to provide cumulative
estimate of benefits without consideration the practical interaction of
technologies. This approach runs contrary to the modeling approach
attempted in the NPRM and final rule analysis. The approach the
agencies tried to follow restricted models to adding discrete advanced
technologies. This approach allowed an accounting of synergetic
effects, identified incremental benefits, and increased the precision
of cost estimates.
As another example, further discussed in Section VI.B.1 Analysis
Fleet, the agencies had traditionally taken different approaches to
assigning baseline road load reduction technology assignments. For
analyzing baseline levels of mass reduction in an analysis fleet, NHTSA
had developed for the Draft TAR a regression model to summarize a
vehicle's weight savings using a relative performance approach and
accounting for vehicle content, using cost curves developed from
teardown studies of a MY 2011 Honda Accord and MY 2014 Chevrolet
Silverado pickup truck. EPA developed its own methodology that
classified vehicles based on weight reductions from a MY 2008 vehicle,
compared to the MY 2014 version of the same vehicle, using a cost curve
from a tear-down study of a MY 2010 Toyota Venza. In the EPA's mass
reduction technology costing approach, a cost reduction was applied
when mass reduction 1 technology was applied to a system at mass
reduction 0 technology level. NHTSA's approach, used in the NPRM and
final rule analysis, set baseline mass reduction assignments so costs
of implementing mass reduction technologies are fully applied as
vehicle platforms move along the mass reduction technology path.
The agencies also included additional advanced powertrain
technologies and other vehicle-level technologies in the technology
pathways between the Draft TAR and NPRM, and between the NPRM and final
rule. However, manufacturers and suppliers have repeatedly told the
agencies that there are diminishing returns to increasing the
complexity of advanced gasoline engines, including in the amount of
fuel efficiency benefit that they can provide. For example, Toyota
commented, in response to the EPA SAE paper benchmarking the 2018 Camry
with the 2.5L Atkinson-cycle engine and ``futuring'' midsize exemplar
vehicles based on the generated engine map,\701\ that although EPA's
addition of cylinder deactivation to the hypothetical 2025 exemplar
vehicle is technically possible and would provide some fuel economy and
CO2 benefit, the primary function of cylinder deactivation
is to reduce engine pumping losses which the Atkinson cycle and EGR
already accomplish on the 2018 Camry.\702\ Toyota concluded, ``The
overlapping and redundant measures to reduce engine pumping losses
would add costs with diminishing efficiency returns.'' Similarly,
BorgWarner commented that they ``do not expect that variable
compression ratio (VCR) or homogeneous charge compression ignition
(HCCI) will see broad application in the short term, if ever. While
each of these technologies can offer marginal efficiency gains at some
engine speed-load conditions, the use of down-sized boosted engines
with 8-10 speed transmissions makes it possible to run engines at near
optimum conditions and effectively minimizes gains from VCR or HCCI.
VCR mechanisms result in additional mass, cost and complexity, and true
HCCI has yet to be demonstrated in a production vehicle. The agencies
do not believe that OEMs will judge these technologies to be cost
effective.'' \703\
---------------------------------------------------------------------------
\701\ Kargul, J., Stuhldreher, M., Barba, D., Schenk, C. et al.,
``Benchmarking a 2018 Toyota Camry 2.5-Liter Atkinson Cycle Engine
with Cooled-EGR,'' SAE Technical Paper 2019-01-0249, 2019,
doi:10.4271/2019-01-0249.
\702\ NHTSA-2018-0067-12431, at 8.
\703\ NHTSA-2018-0067-11895.
---------------------------------------------------------------------------
So, while previous analyses may have shown pathways to compliance
with increasingly complex advanced gasoline engines, the NPRM and final
rule analyses more appropriately reflect that the most complex gasoline
engine technologies will account for a smaller share of manufacturers'
products during the rulemaking timeframe. However, despite this fact,
the NPRM and final rule analysis include more advanced powertrain
technologies than previous analyses, in part to account for important
considerations like intellectual property and the fact that some
manufacturers have already started down the path of incorporating a
certain advanced engine technology in their product portfolio, and that
abrupt switching to another advanced engine technology would result in
unrealistic stranding of capital costs. In addition, greater precision
in how cumulative technologies applied to engines, as estimated through
the Autonomie effectiveness modeling, appropriately reflects the
diminishing returns to efficiency benefits that those advanced engines
can provide. Moreover, as identified by a wide range of commenters,
battery costs are projected to fall in the rulemaking timeframe to a
point where, in the compliance modeling, it becomes more cost effective
to add electrification technologies to vehicles than to apply other
advanced gasoline engine technologies.
Finally, the agencies declined to incorporate some information and
data for the NPRM or final rule central analysis for reasons discussed
in the following sections. In general, the data produced by agencies or
submitted by commenters failed to isolate effectiveness impacts of
individual technologies (or in some cases a combination of two or
several technologies). The data included effects from additional
unaccounted and undocumented technologies. Because the effectiveness
improvement measured or claimed resulted from more than just the
reported sources, the actual effectiveness of the technology or
technologies is obfuscated and easily under or over predicted. Using
effectiveness values generated in this manner carries a high risk of
double counting effectiveness and undercounting costs.
In many cases, this problem exists where data or information is
based on laboratory testing or on-road testing of production vehicles
or components including engines and transmissions. Production vehicles
and components usually include multiple technology improvements from
one redesign to the next, and rarely incorporate just a single
technology change. Furthermore, technology improvements on production
vehicles in some cases cannot be readily observed, such as the level of
mechanical friction in an engine, and isolation and identification of
the improvement attributable to each technology would be impractical
given the costs and time required to do so. That said, in some cases,
where possible to do so, the agencies used the data or information from
production vehicles to corroborate information from the Autonomie
simulations. However, the agencies declined to apply that data or
information directly in the analysis if the effectiveness improvement
attributable to a particular technology could not be isolated.
The agencies made these updates from prior analyses not, as some
commenters have suggested, to ``artificially overstate technology
costs,'' \704\ or to ``ignore the knowledge and expertise of the EPA
engineering
[[Page 24385]]
and compliance staff,'' \705\ ``so that the model in many instances
selects more expensive, less fuel efficient technology while excluding
less expensive and more efficient alternatives,'' \706\ but because the
updates reflected the agencies' reasonable assessment of the current
state of vehicle technologies and their costs, and the state of future
vehicle technologies and costs in the rulemaking timeframe.
---------------------------------------------------------------------------
\704\ NHTSA-2018-0067-11741 at 7.
\705\ NHTSA-2018-0067-11741 at I-23.
\706\ NHTSA-2018-0067-12123.
---------------------------------------------------------------------------
Separate from the decision to update assumptions used for the NPRM
analysis from prior analyses, the agencies did refine some technology
effectiveness and cost assumptions from the NPRM to this final rule
analysis. In addition to being appropriate for technical reasons, this
should address some commenters' overarching concerns about understated
technology effectiveness and overstated technology costs. For example,
several commenters noted that the costs of BISG/CISG systems were
higher for small Cars/SUVs and medium cars than for medium SUVs and
pickup trucks, which the Alliance and FCA described as ``implausible''
and ``misaligned with industry understanding,'' and which ICCT
described as ``contrary to basic engineering logic, which holds that a
system which would be smaller and have lower energy and power
requirements would be less expensive, not more.'' \707\ The agencies
agree, and have made changes to address this issue, as described in
Section VI.C.3.a) Electrification.
---------------------------------------------------------------------------
\707\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
After considering comments, the agencies also added several engine
technologies and technology combinations for the final rule analysis.
These included a basic high compression ratio Atkinson cycle engine, a
variable compression ratio engine, a variable turbo geometry engine,
and a variable turbo geometry with electric assist engine (VTGe). The
NPRM discussed and provided engine maps for each of these technologies.
The agencies also added new technology combinations including diesel
engines with cylinder deactivation, turbocharged engines with advanced
cylinder deactivation, diesel engines paired with manual transmissions,
and diesel engines paired with 12-volt start-stop technology.
Transmission revisions included updating the effectiveness of 6-speed
automatic transmissions, applying updated shift logic for 10-speed
automatic transmissions, and increasing the gear span for efficient 10-
speed automatic transmissions. Mass reduction technology was expanded
to include up to 20 percent curb weight reduction, compared with up to
10 percent for the NPRM. These changes, and the comments upon which
they were based, are described in further detail in the following
sections.
1. Engine Paths
The internal combustion (IC) engine is a heat engine that converts
chemical energy in a fuel into mechanical energy. Chemical energy of
the fuel is first converted to thermal energy by means of combustion or
oxidation with air inside the engine. This thermal energy raises the
temperature and pressure of the gases within the engine, and the high-
pressure gas then expands against the internal mechanisms of the
engine. This expansion is converted by the mechanical linkages of the
engine to a rotating crankshaft, which is the output of the engine. The
crankshaft, in turn, is connected to a transmission to transmit the
rotating mechanical energy to the desired final use, particularly the
propulsion of vehicles.
IC engines can be categorized in a number of different ways
depending upon which technologies are designed into the engine: By type
of ignition (e.g., spark ignition or compression ignition), by engine
cycle (e.g., Otto cycle or Atkinson cycle), by valve actuation (e.g.,
overhead valve (OHV), single overhead camshaft (SOHC), or dual overhead
camshaft (DOHC)), by basic design (e.g., reciprocating or rotary), by
configuration and number of cylinders (e.g., inline four-cylinder (I4)
or V-shaped six-cylinder (V6)), by air intake (e.g., forced induction
(turbo or super charging) or naturally aspirated), by method of fuel
delivery (e.g., port injection or direction injection), by fuel type
(e.g., gasoline or diesel), by application (e.g., passenger car or
light truck),or by type of cooling (e.g., air-cooled or water-cooled).
For each combination of technologies among the various categories,
there is a theoretical maximum efficiency for all engines within that
set. There are various metrics that can be used to compare engine
efficiency, and the four metrics the agencies use or discuss in this
preamble are:
Brake specific fuel consumption (BSFC), which is the mass
of fuel consumed per unit of work output (amount of fuel used to
produce power);
Brake thermal efficiency (BTE), which is the total fuel
energy released per unit of work output (percentage of fuel used to
produce power);
Fuel consumption (gallons per mile), which looks at the
gallons of fuel consumed per unit of work output (mile travelled); and
Fuel economy (in MPG), which is the amount of work output
(miles travelled) per unit (gallon) of fuel consumed.
When comparing the efficiency of IC engines, it is important to
identify the metric(s) used and the test cycle for the measurement
because results vary widely when engines operate over different test
cycles. Two-cycle fuel economy tests used to certify vehicles'
compliance with the CAFE standards tend to overestimate the average
fuel economy motorists will typically achieve during on-road
operation.\708\ In the NPRM and for this final rule analysis, the
agencies considered technology effectiveness for the 2-cycle test
procedures and AC and off-cycle test procedures to evaluate how
technologies could be applied for manufacturers to comply with
standards. The agencies also considered real world operation beyond
these test procedures when considering IC engine technologies in order
to assure the technologies were configured and specified in a manner
that could be used in real world vehicle applications.
---------------------------------------------------------------------------
\708\ 77 FR 62988.
---------------------------------------------------------------------------
a) Fuel Octane
As mentioned in other sections of the Preamble, the agencies go to
great lengths to ensure engine technologies considered for potential
compliance pathways are feasible for real-world implementation and
effectiveness. An important facet of this evaluation are both the fuels
that are used for efficiency testing and also the fuels that consumers
may purchase in the marketplace.
In the NPRM, the agencies included a general overview of fuel
octane (stability) level, including levels currently available, and the
potential impact of fuel octane on engines developed for the U.S.
market.\709\ The agencies described that a typical, overarching goal of
optimal spark-ignited engine design and operation is to maximize the
greatest amount of energy from the fuel available, without manifesting
detrimental impacts to the engine over expected operating conditions.
Design factors, such as compression ratio, intake and exhaust value
control specifications, and combustion chamber and piston
characteristics, among others, are all impacted by the octane of the
fuel consumers are anticipated to use.\710\
---------------------------------------------------------------------------
\709\ PRIA at 253.
\710\ In addition, PRIA Chapter 6 contains a brief discussion of
fuel properties, octane levels used for engine simulation and in
real-world testing, and how octane levels can impact performance
under these test conditions.
---------------------------------------------------------------------------
[[Page 24386]]
The agencies also discussed potential challenges associated with
octane levels available currently, and how those octane levels may play
a role in potential vehicle fuel efficiency improvements. Vehicle
manufacturers typically develop their engines and engine control system
calibrations based on the fuel available to consumers. In many cases,
manufacturers may recommend a fuel grade for best performance and to
prevent potential damage. In some cases, manufacturers may require a
specific fuel grade for both best performance, to achieve advertised
power ratings, and/or to prevent potential engine damage.
Consumers, though, may or may not choose to follow the
manufacturer's recommendation or requirement for a specific fuel grade
for their vehicle. As such, vehicle manufacturers often choose to
employ engine control strategies for scenarios where the consumer uses
a lower than recommended, or required, fuel octane level, as a way to
mitigate potential engine damage over the life of a vehicle. These
strategies limit the extent to which some efficiency improving engine
technologies can be implemented, such as increased compression ratio
and intake system and combustion chamber designs that increase burn
rates and rate of in-cylinder pressure rise. If the minimum octane
level available in the market were higher (especially the current sub-
octane regular grade in the mountain states), vehicle manufacturers
might not feel compelled to design vehicles sub-optimally to
accommodate such blends.
When knock (also referred to as detonation) is encountered during
engine operation, at the most basic level, non-turbocharged engines can
adjust the timing of the spark that ignites the fuel, as well as the
amounts of fuel injected at each intake stroke (``fueling''). In
turbocharged applications, knocking is typically controlled by
adjusting boost levels along with spark timing and/or the amount of
fuel injected. Past rulemakings discussed other techniques that may be
employed to allow higher compression ratios, including optimizing spark
timing, and adding of cooled exhaust gas recirculation (EGR).
Regardless of the type of spark-ignition engine or technology employed,
efforts to reduce or prevent knock with the lower-octane fuels that are
available in the market result in the loss of potential power output,
creating a ``knock-limited'' constraint on performance and efficiency.
The agencies noted that despite limits imposed by available fuel
grades, manufacturers continue to make progress in extracting more
power and efficiency from spark-ignited engines. Production engines are
safely operating with regular 87 AKI fuel with compression ratios and
boost levels once viewed as only possible with premium fuel. According
to the Department of Energy, the average gasoline octane level has
remained fundamentally flat starting in the early 1980's and decreased
slightly starting in the early 2000s. During this time, however, the
average compression ratio for the U.S. fleet has increased from 8.4 to
10.52, a more than 20 percent increase. As explained by the Department
of Energy, ``[t]here is some concern that in the future, auto
manufacturers will reach the limit of technological increases in
compression ratios without further increases in the octane of the
fuel.'' \711\ As such, manufacturers are still limited by the fuel
grades available to consumers and the need to safeguard the durability
of their products for all of the available fuels; thus, the potential
improvement in the design of spark-ignition engines continues to be
overshadowed by the fuel grades available to consumers.
---------------------------------------------------------------------------
\711\ Fact of the Week, Fact #940: August 29, 2016 Diverging
Trends of Engine Compression Ratio and Gasoline Octane Rating, U.S.
Department of Energy, https://www.energy.gov/eere/vehicles/fact-940-august-29-2016-diverging-trends-engine-compression-ratio-and-gasoline-octane (last visited Mar. 21, 2018).
---------------------------------------------------------------------------
EPA and NHTSA also described ongoing research and positions from
automakers and advocacy groups on fuel octane levels, including
comments received during past agency rulemakings and on the 2016 Draft
TAR regarding the potential for increasing octane levels in the U.S.
market. The agencies described arguments for adjusting to octane
levels, including making today's premium grade the base grade of fuel
available, which could enable low cost design changes to improve fuel
economy and reduce tailpipe CO2 emissions. Challenges
associated with this approach include the increased cost to consumers
who drive vehicles designed for current regular octane grade fuel, who
would not benefit from the use of the higher cost higher-octane fuel.
The costs of such a transition to higher-octane fuel would be high and
persist well into the future, since unless current regular octane fuel
were unavailable in the North American market, manufacturers would be
effectively unable to redesign their engines to operate on higher-
octane fuel. In addition, the full benefits of such a transition would
not be realized until vehicles with such redesigned engines were
produced for a sufficient number of model years largely to replace the
current on-road vehicle fleet. The transition to net positive benefits
would take many years.
The agencies also described input received from renewable fuel
industry stakeholders and from the automotive industry supporting high-
octane gasoline fuel blends to enable fuel economy and CO2
improving technologies such as higher compression ratio engines.
Stakeholders suggested that mid-level (e.g., E30) high-octane ethanol
blends should be considered and that EPA should consider requiring that
mid-level blends be made available at service stations. Stakeholders
supporting higher-octane blends suggested that higher-octane gasoline
could provide auto manufacturers with more flexibility to meet more
stringent standards by enabling opportunities for use of lower tailpipe
CO2 emitting technologies (e.g., higher compression ratio
engines, improved turbocharging, optimized engine combustion).
The agencies sought additional comment in the NPRM on various
aspects of current fuel octane levels and how fuel octane could play a
role in the future. More specifically, the agencies sought comment on
how increasing fuel octane levels could have an impact on product
offerings and engine technologies, as well as what improvements to fuel
economy and tailpipe CO2 emissions could result from higher-
octane fuels. The agencies sought comment on an ideal octane level for
mass-market consumption, and whether there were downsides with
increasing the available octane levels and, potentially, eliminating
lower-octane fuel blends. EPA also requested comment on whether and how
EPA could require the production and use of higher-octane gasoline
consistent with Title II of the Clean Air Act.
The agencies received numerous, wide-ranging comments in response
to the NPRM discussion, and some direct responses to the agencies'
requests for comments. The commenters included fuel producers,
individual vehicle manufactures, environmental groups, vehicle
suppliers, fuel advocacy groups, and agricultural organizations, among
others. Commenters provided a broad range of comments ranging from
explication of the many challenges to increasing available octane
levels, to claims of the substantial efficiency
[[Page 24387]]
increases that could be easily obtained by requiring higher-octane
levels.
Several ethanol industry stakeholders commented in support of
requiring higher-octane fuels using mid-level ethanol blends. The High-
Octane, Low Carbon (HOLC) Alliance commented that it believes ``NHTSA
and EPA have a critical opportunity to cost-effectively ensure progress
in fuel efficiency and CO2 emissions standards. Scientific
experts agree that high-octane, low-carbon fuel can yield greater fuel
economy and emissions benefits when paired with internal combustion
engines (ICEs). But, to realize such benefits, automobile manufacturers
require approval sooner rather than later to such fuels. Alternatively,
automobile manufacturers will be limited in their ability to maximize
the environmental performance of their vehicles until non-liquid fuel
engines become more readily available. In finalizing the Proposed Rule,
the HOLC Alliance strongly urges EPA and NHTSA to establish a pathway
forward toward incentivizing the production and adoption of higher-
octane, lower carbon fuels. By doing so, EPA and NHTSA can continue to
incrementally increase CO2 and fuel economy standards,
respectively.'' \712\
---------------------------------------------------------------------------
\712\ HOLC Alliance, Detailed Comments, EPA-HQ-OAR-2018-0283-
4196.
---------------------------------------------------------------------------
Renewable Fuels Associations (RFA) commented that ``it strongly
believes vehicles and fuels must be considered together as integrated
systems. As EPA has recognized in the past, a `systems approach enables
emission reductions that are both technologically feasible and cost
effective beyond what would be possible looking at vehicle and fuel
standards in isolation.' Because ethanol-based high-octane low-carbon
fuel blends would enable cost-effective gains in fuel economy and
carbon dioxide reductions, the agencies should take steps to support
[high-octane low-carbon] fuels in the final SAFE rule.'' \713\
---------------------------------------------------------------------------
\713\ RFA, Detailed Comments, EPA-HQ-OAR-2018-0283-4409.
---------------------------------------------------------------------------
RFA cited several studies indicating benefits are available from
raising the floor of fuel octane levels currently available, and,
particularly, ``[t]he results from the studies reviewed generally
support a main conclusion that splash blending ethanol is a highly
effective means of raising the octane rating of gasoline and enabling
low-cost efficiencies and reduced emissions in modern spark-ignition
engines.'' \714\ In addition, National Corn Growers Association stated
that, ``[w]ithout a change in fuel, automakers are reaching the limits
on the efficiency gains that can be achieved with technology changes.''
\715\
---------------------------------------------------------------------------
\714\ RFA, Detailed Comments, EPA-HQ-OAR-2018-0283-4409.
\715\ National Corn Growers Association, https://www.ncga.com/file/1621/NCGA%20Comments20Docket%20No.%20EPA-HQ-OAR-2018-0283%20and%20NHTSA-2018-0067.pdf.
---------------------------------------------------------------------------
The National Corn Growers Association, in conjunction with
associated corn growing and agricultural groups, pointedly stated the
EPA should, ``[s]et a minimum fuel octane level of 98 RON and phase out
low octane fuels as new optimized vehicles enter the market in MY
2023,'' and concluded that approving a ``midlevel ethanol blend vehicle
certification fuel would enable automakers to expedite design and
testing of optimized vehicles for use with this new fuel.'' \716\
---------------------------------------------------------------------------
\716\ National Corn Growers Association, https://www.ncga.com/file/1621/NCGA%20Comments%20Docket%20No.%20EPA-HQ-OAR-2018-0283%20and%20NHTSA-2018-0067.pdf.
---------------------------------------------------------------------------
The 25x25 Alliance commented that ``to meet the dual goals of
greater fuel efficiency and reduced GHG emissions, the utilization of
higher compression spark ignition internal combustion engines will be
essential. Increasing engine compression improves thermal efficiency.
However, as compression increases, higher-octane fuels will be needed
to prevent engine knock. Automakers and advocacy groups have expressed
support for increases to fuel octane levels for the US market. Ethanol
with its octane rating of 113 offers engine knock resistance at a lower
cost than any other octane booster in gasoline. In addition, ethanol's
lower direct and life-cycle GHG emissions as compared to gasoline are
well documented. For this reason, a fuel produced from a mixture of
ethanol and gasoline and used in conjunction with advanced high
compression engines presents itself as a technology pathway capable of
complying with new CAFE/GHG standards.'' They continue, ``HOLC
supporters recognize numerous barriers and other associated regulatory
hurdles must be resolved before HOLC ethanol fuels are adopted at large
scale. . . 25x25 believes it is imperative that the vehicle and fuel be
treated as a comprehensive system. To date CAFE/GHG standards have
largely focused on vehicle engine technology. Advanced engine vehicles
perform best in concert with fuels of suitable properties and
composition to optimally enable and power them.'' \717\
---------------------------------------------------------------------------
\717\ 25x25 Alliance, Detailed Comments, EPA-HQ-OAR-2018-0283-
4210.
---------------------------------------------------------------------------
The American Coalition for Ethanol (ACE) commented that ``high-
octane blends comprised of 25 to 30 percent ethanol would help bring
down the cost for consumers compared to the premium-priced octane level
advocated by oil refiners. Ethanol has a blending octane rating of
nearly 113 and trades at a steep discount to gasoline. In many
wholesale markets today, ethanol costs at least 60 cents per gallon
less than gasoline. Ethanol delivers the highest octane at the lowest
cost, allowing automakers to benefit by continuing to develop high-
compression engine technologies and other product offerings to achieve
efficiency improvements and reduced emissions. The ideal way to
transition from today's legacy fleet to new vehicles with advanced
engine technologies designed to run optimally on a high-octane fuel is
to utilize FFVs as bridge vehicles that can provide immediate demand
for mid-level ethanol blends.'' \718\
---------------------------------------------------------------------------
\718\ ACE, Detailed Comments, EPA-HQ-OAR-2018-0283-4033.
---------------------------------------------------------------------------
Growth Energy commented that with a mid-level ethanol blend,
automakers not only get higher-octane that they can use to optimize
engines and gain further fuel efficiency, they will also see a fuel
that has demonstrably lower carbon dioxide emissions.\719\ The Illinois
Corn Growers' Association et al., commented that ``NHTSA and EPA must
adapt the existing regulatory structure to reflect the specific
characteristics of mid-level blend fuels. Working together, the ethanol
industry, automakers, EPA and NHTSA can bring about, during the period
covered by the SAFE program, a new generation of high efficiency
internal combustion engines optimized to take advantage of this new
fuel's unique properties.'' \720\
---------------------------------------------------------------------------
\719\ Growth Energy, Detailed Comments, EPA-HQ-OAR-2010-0799-
9540-A2.
\720\ Comment removed because it contains copyrighted data,
Illinois Corn Growers Association, et al., https://www.regulations.gov/document?D=EPA-HQ-OAR-2018-0283-4198.
---------------------------------------------------------------------------
Ethanol industry commenters provided comment on several EPA actions
they believe would be necessary to support higher-octane mid-level fuel
blends:
Set a minimum fuel octane level and phase out low-octane
fuels as new optimized vehicles enter the market;
Approve a high-octane, mid-level ethanol blend vehicle
certification fuel;
Correct the fuel economy formula by updating the R-Factor
to be at or nearly ``1'' to reflect documented operation of modern
engine technology;
Extend a RVP waiver of 1 psi to all gasoline containing at
least 10 percent ethanol;
Adopt the Argonne National Laboratory GREET model to
determine updated lifecycle carbon emissions for ethanol;
[[Page 24388]]
Establish meaningful credits to automakers to incentivize
transition to higher-octane fuel vehicles and continue to support flex-
fuel vehicles; and
Provide equal treatment to vehicle technologies that
reduce carbon emissions.
The Clean Fuels Development Coalition, et al. suggested that, ``the
`ideal octane level' to optimize LDV performance, fuel efficiency, and
reduce harmful emissions and consumer costs is 98-100 RON produced with
E30+ `clean octane.' '' \721\ Concurrently, the HOLC Alliance and ACE,
among others, also supported that 98 to 100 RON would be ideal octane
levels for the nation.\722\
---------------------------------------------------------------------------
\721\ Clean Fuels Development Coalition, et al., Detailed
Comments, NHTSA-2018-0067-11988.
\722\ HOLC Alliance, Detailed Comments, EPA-HQ-OAR-2018-0283-
4196; ACE, Detailed Comments, EPA-HQ-OAR-2018-0283-4033.
---------------------------------------------------------------------------
BorgWarner, a supplier to major automobile manufacturers, commented
that ``[f]uel octane is a limiting factor in the selection of
compression ratio for all spark-ignition engines and the amount of
boost for turbocharged engines. Higher-octane is particularly effective
for using higher compression ratios with boosted engines,'' and stated
that ``[t]here is substantial merit to raising the minimum octane
required because current fuel pricing penalizes consumers for using
higher-octane fuel. A base octane of 95 RON would be consistent with
Europe. This would allow consistent development of engines for the
broader US-EU market. Prior to the introduction of ethanol into
gasoline, the base blend for regular fuel was typically 92 RON.
Addition of 10% ethanol to this base blend gave 95 RON regular, so the
base blend would be reformulated to retain the 92 RON at a lower cost.
Returning to the previous base blend would be cost effective to the
consumer.'' \723\
---------------------------------------------------------------------------
\723\ BorgWarner, Detailed Comments, EPA-HQ-OAR-2018-0283-4174.
---------------------------------------------------------------------------
Auto manufacturers also provided comment on the topic of higher-
octane fuels. The Alliance of Automobile Manufacturers (the Auto
Alliance) commented that it ``has long advocated for the availability
of cost-effective, higher-octane fuel. The Alliance also believes the
Agencies should require a transition to a higher minimum-octane
gasoline (minimum 95-98 RON). There are several ways to produce higher-
octane grade gasoline, such as expanding the ethanol availability, but
the Alliance does not promote any sole or particular pathway.'' \724\
The Alliance reiterated its position regarding fuel octane levels
where, ``[t]he Alliance has long supported two goals regarding the
octane (anti-knock) properties of gasoline: (1) The availability of
cost effective higher-octane fuels, greater than 95 Research Octane
Number (RON) and (2) the immediate elimination of subgrade fuel less
than 87 anti-knock index (AKI).'' The Alliance also noted that ``[t]he
higher-octane fuel that is available today is sold as a premium grade.
To support future engine technologies, the approach taken with today's
premium fuel option would not be expected to provide an attractive
value proposition to the customer; therefore, a new higher minimum-
octane gasoline, 95-98 RON, is needed to achieve anticipated
performance.''
---------------------------------------------------------------------------
\724\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Ford Motor Company agreed with the Auto Alliance's collective
comments on fuel octane level and added specific support to raising
minimum octane levels, stating that ``Ford concurs with those comments
and supports increasing the marketplace octane rating in the U.S. to a
minimum of 95 Research Octane Number (RON).'' Ford also generally
supported the agencies' fuel octane discussion in terms of impacts to
vehicle performance, where ``[h]igher octane gasoline enables
opportunities for the use of key energy-efficient technologies,
including: Higher compression ratio engines, lighter and smaller
engines, improved turbocharging, optimized engine combustion phasing/
timing, and low temperature combustion strategies. All of these
technologies paired with higher-octane gasoline permit smaller engines
to meet the demands of the consumer while at the same time providing
higher overall efficiencies.'' \725\
---------------------------------------------------------------------------
\725\ Ford, Detailed Comments, EPA-HQ-OAR-2018-0283-5691.
---------------------------------------------------------------------------
Volkswagen commented ``[t]here may be several potential ways to
achieve a high-octane fuel that may be more costly to the vehicle than
others. Achieving an E10 high-octane fuel may mean a different hardware
set than on E20 or E30 high-octane fuel. Elimination of sub-grades of
market fuel (less than 87AKI) quickly is very important. If current 87
AKI and 85 AKI fuels remain in the market for backward compatibility
(such as if an E30 were chosen as the high-octane fuel of the future),
a robust method at the fuel dispensing station and incorporated into
the fueling station equipment to prevent mis-fueling is necessary.
However, an E10 high-octane pathway might have far fewer compatibility
problems and might bring extra fuel economy to the drivers of those
current vehicles.'' \726\
---------------------------------------------------------------------------
\726\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
The agencies also received comments from the petroleum industry
regarding higher-octane fuels. API commented that ``[g]iven the
multiple engine technology pathways available to the automakers for
achieving future fuel economy and CO2 emissions targets, the
challenge of determining future market fuel gasoline octane number
needs is complex and not yet settled. API believes that the octane
number issue should be part of a comprehensive transport policy that
addresses both vehicles and fuels as a system. API and its members are
engaged in collaborations with the automakers and other stakeholders to
better understand future fuel requirements for emerging powertrain
technologies.'' API also commented ``the future for gasoline octane
number will be driven by the stringency of regulations that set future
fuel economy and CO2 requirements, the collective responses
of the automakers to those regulations, consumer preferences regarding
vehicles and fuels, and fuel supply economics. EPA's authority to
regulate gasoline octane number is doubtful. Therefore, EPA should not
attempt to regulate gasoline octane number at this time.'' \727\
---------------------------------------------------------------------------
\727\ API, Detailed Comments, EPA-HQ-OAR-2018-0283-5458.
---------------------------------------------------------------------------
In terms of challenges associated with potential high-octane fuel
deployment, the American Fuel & Petrochemical Manufacturers (AFPM)
commented that, ``[a]side from a lack of legal authority, EPA faces
numerous technical, logistical, and legal challenges and uncertainties
in requiring the use of higher-octane fuels. Any such requirement would
need a separate rulemaking dedicated to such a purpose with an
extensive technical record in support, including test data on vehicles
designed for the higher-octane fuel and on the existing fleet with and
without higher-octane.'' \728\
---------------------------------------------------------------------------
\728\ AFPM, Detailed Comments, EPA-HQ-OAR-2018-0283-5698.
---------------------------------------------------------------------------
AFPM also commented that it does not support the potential
regulatory requirement for the production or use of higher octane
gasoline as a compliance option. AFPM commented that EPA lacks the
authority to require the use of higher octane fuels under CAA Sec.
211(c)(1)(A). AFPM further commented ``[t]he only vehicles legally
permitted to use more than 15 percent ethanol blends are flex-fuel
vehicles, which are currently certified to utilize both E10 and E85.
Without an alternative certification for an auto
[[Page 24389]]
manufacturer to build an E30 certified vehicle, which would require
extensive testing and certification procedures as well as sufficient
market availability of the certification fuel, it would be
inappropriate for the Administration to consider such vehicles as a
viable option in the 2022-2026 compliance period.''
Gasoline retailers also commented regarding higher-octane fuels.
NACS and SIGMA commented that they support examining the use of such
fuels as a potential path towards future emissions reductions and that
it will be important that the agencies appropriately consider and
address a variety of related issues, including:
1. How to allow and handle the expanded sales of higher-octane
fuels, which may include fuels that currently face barriers to sale,
such as E15;
2. Streamlining the registration and regulation of higher-level
blends of ethanol;
3. Addressing misfueling liability concerns of retailers;
4. Streamlining federal labeling requirements and ensuring federal
preemption of state requirements; and
5. Addressing any other regulatory and legislative challenges
associated with the use of higher-octane fuels.\729\
---------------------------------------------------------------------------
\729\ Joint submission on behalf of NACS and SIGMA, Detailed
Comments, EPA-HQ-OAR-2018-0283-5824.
---------------------------------------------------------------------------
NATSO commented that ``the Agencies should under no circumstances
consider `requiring that mid-level [ethanol] blends be made available
at service stations' '' and went on to say that ``retailers would need
to be assured that they will not be held responsible for customers that
misfuel . . . Federal dispenser labeling requirements would have to be
streamlined and state requirements would have to be preempted. . . Auto
manufacturers would have to warrant all new higher-octane vehicles up
to at least E15 depending upon vehicles' capabilities, and would have
to affirmatively state which cars in the existing fleet can run on E15
and ensure that the cars are warrantied or retroactively warrantied as
such.'' \730\
---------------------------------------------------------------------------
\730\ NATSO, Detailed Comment, EPA-HQ-OAR-2018-0283-5484.
---------------------------------------------------------------------------
UCS commented that ``[a]n orderly transition to high-octane fuel
would take several years to complete. It will take time for the
necessary regulations to be finalized, for vehicles optimized for high-
octane gasoline to come to market and to build out the fuel
distribution infrastructure to make this fuel broadly available. And
even once high-octane gasoline is in use, it will take more time for
automakers to phase-in new models optimized for high-octane fuel and to
fully replace the legacy E10 fleet. Another factor to consider is that
the rising share of high-octane gasoline will be buffered by falling
sales of gasoline, given increasing fuel efficiency, such that the
overall demand for ethanol will change more slowly. The agencies'
expectation is that high-octane gasoline will not significantly enter
commerce before 2026, and subsequently will only gradually gain market
share through 2040. There is no realistic prospect of completing this
process before 2025 or 2026, the timeframe of this rulemaking. The
appropriate context for this discussion within vehicle rules is the
next round of fuel economy and emission standards. Even then, an
expeditious rulemaking process will be required to achieve adequate
regulatory clarity to facilitate rapid adoption post-2026.'' UCS also
commented ``[we] strongly oppose granting fuel economy credits based on
the technical potential of vehicles to operate on high-octane fuel
before there is clear evidence that high-octane fuel is in use and the
potential fuel economy benefits are being realized on the road.'' \731\
---------------------------------------------------------------------------
\731\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
The agencies have reviewed the submissions received in response to
their solicitation of comments concerning fuel octane levels and
recognize the potential that higher-octane fuels, coupled with advanced
engine technologies, can provide for improvements to fuel economy and
tailpipe CO2 emissions. The agencies agree with commenters
that establishing a higher minimum octane for gasoline is a complex
undertaking that would require consideration of a wide array of
difficult issues. In light of the complexity of the constellation of
issues, the fact that EPA did not propose new octane requirements, and
that EPA's authority to set fuel requirements resides in CAA section
211(c)(1), the agencies recognize that the present rulemaking is not
the appropriate vehicle to set octane levels. If EPA pursues future
rulemaking action on this topic, it would consider these comments in
that context and in consideration of the appropriate statutory
provisions. The agencies note that the current vehicle certification
process provides a path to certify a vehicle requiring the use of high-
octane fuel, which allows the impact of such fuels to be captured over
the required certification test cycles for CO2 emissions and
fuel economy.
EPA also is declining to adopt new incentives for flex-fueled
vehicles (FFVs) (vehicles designed to operate on gasoline or E85 or a
mixture), as some commenters suggested. FFV incentives were not
identified by EPA in its request for comments in the proposed rule and
are outside the scope of this rulemaking.
The analyses conducted for this rulemaking assumed the use of Tier
3 fuels, where applicable, which are considered directly
representative, or a reasonable proxy for, fuels available for
consumers to purchase. As explained in the previous paragraph, agency
actions related to test fuels, consumer available fuels, or flexible-
fuel incentives are out of scope of this rulemaking. However, to the
extent that the agencies consider any additional rulemaking actions
related to fuel octane requirements and/or availability, the agencies
note that further analysis to set CAFE and CO2 standards
would also reflect any potential, related impacts of those potential
changes.
b) Engine Maps
Engine paths include numerous engine technologies that
manufacturers can use to improve fuel economy and reduce CO2
emissions. Some engine technologies can be incorporated into existing
engine design architectures with minor or moderate changes to the
engine, but many engine technologies require an entirely new engine
architecture or a major refresh. For this final rule analysis, twenty-
three unique engine technologies are available for adoption, and are
evaluated uniquely across the ten separate vehicle types (technology
classes).
For the NPRM and final rule analysis, the impact of engine
technologies on fuel consumption, torque, and other metrics was
characterized using GT-POWER(copyright) modeling conducted by IAV
Automotive Engineering, Inc. (IAV). IAV is one of the world's leading
automotive industry engineering service partners and has extensive
experience in testing and modeling engines and combustion. GT-POWER is
a commercially available engine modeling tool with detailed cylinder
and combustion modeling capabilities.\732\ GT-POWER is used to simulate
engine behavior and provides data on engine metrics, including power,
torque, airflow, volumetric efficiency, fuel consumption, turbocharger
performance, and other parameters. The primary outputs of IAV's use of
GT-POWER for this
[[Page 24390]]
analysis are the development of engine maps that provide operating
characteristics of engines equipped with specific technologies.
---------------------------------------------------------------------------
\732\ More information regarding GT Power Modeling is available
at https://www.gtisoft.com/gt-suite-applications/propulsion-systems/gt-power-engine-simulation-software.
---------------------------------------------------------------------------
When an engine is running, at any given point in time, the
operation can be characterized by the engine's crankshaft rotational
speed (typically in revolutions per minute, or RPM) and engine output
(torque) level. Engines can operate at a range of engine speed and
torque levels. Engine maps provide a visual representation of various
engine performance characteristics at each engine speed and torque
combination across the operating range of the engine. A common example
of a performance characteristic is BSFC.\733\ Other characteristics
include engine emissions, engine efficiency, and engine power.
---------------------------------------------------------------------------
\733\ The amount of fuel needed to achieve a specific power, or
how efficiently an engine uses fuel to produce work.
---------------------------------------------------------------------------
Engine maps have the appearance of topographical maps, typically
with engine speed on the horizontal axis and engine torque on the
vertical axis. A third engine characteristic, BSFC, is displayed as
contours, defining the operating regions for that BSFC with each
contour showing all operating points at a specified BSFC value. Once
created, the data they contain is referenced for engine fuel
consumption at a given engine speed and torque operating point.
For the NPRM and final rule analysis, the agencies relied on IAV to
develop engine maps representing each of the engine technologies. IAV
used benchmark production engine test data, component test data, and
manufacturers and suppliers' technical publications to develop a one-
dimensional GT-POWER engine model for the baseline engine technology
configuration. Technologies were incrementally added to the baseline
model to assess their impact on fuel consumption. The following is a
representative example of how IAV created the engine maps used in this
analysis.
First, IAV defined the characteristics of Eng01 (a base VVT engine)
and optimized it for all the combustion parameters while minimizing
fuel consumption and maintaining performance. The result of this was a
fuel map as a function of BMEP and engine RPM. IAV then took the same
Eng01 and adopted characteristics of SGDI technology to the base
engine. The new engine (Eng18, VVT and SGDI) was then optimized for all
combustion parameters while minimizing fuel consumption and maintaining
performance. The result was an engine fuel map for Eng18, as a function
of BMEP and engine speed. The engine map is directly comparable to the
engine map for Eng01 and the difference in those engine maps
specifically identifies the effectiveness impact of VVT and SGDI
technologies. This process was repeated for all of the IAV engine maps
that used Eng01 (VVT) as the baseline engine. This methodology ensured
the engine maps represent the maximum improvement in BSFC for each
engine configuration change, while considering real world design
constraints.
IAV used its global engine database that includes benchmarking
data, engine test data, single cylinder test data, prior modeling
studies, and technical publications and information presented at
conferences to populate the assumptions and inputs used for engine map
modeling, and to validate the ultimate results.\734\ Argonne used the
engine maps resulting from this analysis as inputs for the Autonomie
full vehicle modeling and simulation.
---------------------------------------------------------------------------
\734\ Friedrich, I., Pucher, H., and Offer, T., ``Automatic
Model Calibration for Engine-Process Simulation with Heat-Release
Prediction,'' SAE Technical Paper 2006-01-0655, 2006, https://doi.org/10.4271/2006-01-0655. Rezaei, R., Eckert, P., Seebode, J.,
and Behnk, K., ``Zero-Dimensional Modeling of Combustion and Heat
Release Rate in DI Diesel Engines,'' SAE Int. J. Engines 5(3):874-
885, 2012, https://doi.org/10.4271/2012-01-1065. Multistage
Supercharging for Downsizing with Reduced Compression Ratio (2015).
MTZ Rene Berndt, Rene Pohlke, Christopher Severin and Matthias
Diezemann IAV GmbH. Symbiosis of Energy Recovery and Downsizing
(2014). September 2014 MTZ Publication Heiko Neukirchner, Torsten
Semper, Daniel Luederitz and Oliver Dingel IAV GmbH.
---------------------------------------------------------------------------
As described in the NPRM and PRIA, the agencies developed engine
maps for technologies that are in production today or that are expected
to be available in the rulemaking timeframe. The agencies recognize
that engines with the same combination of technologies produced by
different manufacturers will have differences in BSFC and other
performance measures, due to differences in the design of engine
hardware (e.g., intake runners and head ports, valves, combustion
chambers, piston profile, compression ratios, exhaust runners and
ports, turbochargers, etc.), control software, and emission
calibration. Therefore, the engine maps are intended to represent the
levels of performance that can be achieved on average across the
industry in the rulemaking timeframe.
Accordingly, the agencies noted that it was expected that the
engine maps developed for this analysis will differ from engine maps
for manufacturers' specific engines. For a given engine configuration,
some production engines may be less efficient and some may be more
efficient than the engine maps presented in the analysis. However, the
agencies intended and expected that the incremental changes in
performance modeled for this analysis, due to changes in technologies
or technology combinations, will be similar to the incremental changes
in performance observed in manufacturers' engines for the same changes
in technologies or technology combinations. Most importantly, using a
single engine model as a reference provides a common base for
comparison of all incremental changes resulting from technology
changes, and anchors incremental technology effectiveness values to a
common reference. The effectiveness values from the internal simulation
results were validated against detailed engine maps produced from
engine benchmarking programs, as well as published information from
industry and academia, ensuring reasonable representation of simulated
engine technologies.\735\
---------------------------------------------------------------------------
\735\ Bottcher, L., Grigoriadis, P. ``ANL--BSFC map prediction
Engines 22-26.'' IAV (April 30, 2019). 20190430_ANL_Eng 22-26
Updated_Docket.pdf.
---------------------------------------------------------------------------
As discussed in the NPRM, the agencies updated the list of engine
technologies, before and after the Draft TAR, based on stakeholder
comments and consultations with CARB, Argonne, and IAV. The technology
list was built on the technologies that were considered in the 2012
final rule, and included technologies that are being implemented or
that are under development and feasible for production in the
rulemaking timeframe. The agencies noted that some advanced engines
were included in the simulation that were, and often still are, not yet
in production, and the engine maps for those engines were either based
on CBI or theoretical data. The agencies also stated in the NPRM that
the final rule analysis may include updated engine maps for existing
modeled engines, or entirely new maps added to the analysis if either
action could improve the quality of the fleet-wide analysis.
While there are a large number of possible combinations of engine
technologies, the agencies categorized the IAV engine maps used in the
NPRM full vehicle simulations into six categories. The categories were
based on engine architecture and include: Dual overhead camshaft (DOHC)
engines, single overhead camshaft (SOHC) engines, turbocharged engines,
hybrid Atkinson cycle engines,\736\ non-hybrid
[[Page 24391]]
Atkinson mode engines, and diesel engines. Another unique technology
that was available for adoption for the NPRM analysis was the advanced
cylinder deactivation (ADEAC) for the SOHC and DOHC engines, however
this technology was modeled using a fixed effectiveness value rather
than an engine map, because the agencies did not have sufficient data
to be used as input to the engine map or full vehicle simulation
modeling. In addition, the agencies provided potential engine maps and
additional specifications for several other technologies that could be
considered for the final rule analysis. These included a basic high
compression ratio Atkinson mode engine, a Miller cycle engine, and an
engine with an electric assist.
---------------------------------------------------------------------------
\736\ These types of Atkinson cycle engines are mainly for
hybrid applications like Toyota Prius or Ford C-Max.
---------------------------------------------------------------------------
The full list of engine maps used in the NPRM is presented in Table
VI-39 below.
BILLING CODE 4910-59-P
[[Page 24392]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.143
BILLING CODE 4910-59-C
The full list of engine maps used in this final rule analysis is
presented in Table VI-40.
[[Page 24393]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.144
[[Page 24394]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.145
[[Page 24395]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.146
BILLING CODE 4910-59-C
Comments on engine maps varied, with industry commenters generally
supporting the maps used in the NPRM analysis and CARB and
environmental advocate commenters generally objecting to the maps. The
Alliance argued that previously-modeled fuel efficiency improvements
for downsized, turbocharged engine technologies were ``highly
optimistic,'' and stated that the updated engine maps used for the NPRM
analysis were an improvement.
ICCT argued that the IAV engine maps used for the NPRM analysis
were out of date, and better engine maps benchmarked by EPA staff were
available and should have been used instead.\737\ UCS similarly stated
that Argonne work used for previous CAFE technical documents had relied
on outdated engine maps, and that the new IAV engine maps used in this
rulemaking were developed for a different purpose and had not been
benchmarked against the latest engines either on the road or in
development.\738\ ICCT questioned whether the agencies had validated
engines 13 and 14 with physical testing and/or simulation modeling to
the level of quality of EPA's simulation modeling.\739\ ICCT further
asserted that EPA's benchmarked engine maps had been ``knowingly
disregarded'' for the NPRM analysis, and stated that the NPRM analysis
was therefore arbitrary.\740\ ICCT commented that the agencies must
conduct and disclose a systematic investigation and comparison of
engine benchmarking, engine modeling, and transmission modeling
completed by EPA, Ricardo, and Argonne for model year 2014-2018
vehicles. ICCT recommended that the agencies rely on engine maps used
for past EPA ALPHA modeling while the agencies conduct such an
investigation.
---------------------------------------------------------------------------
\737\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-49.
\738\ Union of Concerned Scientists, Technical Appendix, Docket
No. NHTSA-2018-0067-12039, at 4.
\739\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-46.
\740\ ICCT, Docket No. NHTSA-2018-0067-11741, at I-49.
---------------------------------------------------------------------------
The agencies believe it is most important for engine map data to
provide accurate BSFC information for known technologies and technology
levels. The agencies disagree with statements that IAV engine maps are
outdated. The majority of the engine maps were developed specifically
to support the midterm review and encompass engine technologies that
are present in the analysis fleet and technologies that could be
applied in the rulemaking timeframe. In many cases those engine
technologies are mainstream today and will continue to be during the
rulemaking timeframe. For example, the engines on some MY 2017 vehicles
in the analysis fleet have technologies that were initially introduced
ten, or more, years ago. Having engine maps representative of those
technologies is important for the analysis. The most basic engine
technology levels also provide a useful baseline for the incremental
improvements for other engine technologies. The timeframe for the
testing or modeling is unimportant, because time by itself doesn't
impact engine map data. A given engine or model will produce the same
BSFC map regardless of when testing or modeling is conducted.
Simplistic discounting of engine maps based on temporal considerations
alone could result in discarding useful technical information. Also,
narrow use of temporal considerations would also result in the
discarding of several engine maps from Ricardo that were used for the
EPA Draft TAR and Proposed Determination analyses.\741\ Therefore, with
the engine maps used representing current technologies regardless of
development date, the agencies do not agree with commenter assertions.
---------------------------------------------------------------------------
\741\ Ricardo, Inc. ``Computer Simulation of Light-Duty Vehicle
Technologies for Greenhouse Gas Emission Reduction in the 2020-2025
Timeframe.'' Ricardo (December 2011). https://nepis.epa.gov/Exe/ZyPDF.cgi/P100D57R.PDF?Dockey=P100D57R.PDF. Last accessed Jan 14,
2020.
---------------------------------------------------------------------------
The same commenters also appear to misunderstand how the agencies'
effectiveness data, including engine maps, were used in the NPRM
analysis (and in past rulemakings). The analysis never applies absolute
BSFC levels from the engine maps to any vehicle model or configuration
for the rulemaking analysis. The absolute fuel economy values from the
full vehicle Autonomie simulations are used only to determine
incremental effectiveness for switching from one technology to another
technology. The incremental effectiveness is applied to the absolute
fuel economy of vehicles in the analysis fleet, which are based on CAFE
compliance data. For subsequent technology changes, incremental
effectiveness is applied to the absolute fuel economy level of the
previous technology configuration. Therefore, for a technically sound
analysis, it is most important that the differences in BSFC among the
engine maps be accurate, and not the absolute values of the individual
engine maps. However, achieving this can be challenging.
A technically sound approach is to use a single or very small
number of baseline engine configurations with well-defined BSFC maps,
and then, in a very systematic and controlled process, add specific
well-defined technologies and create a BSFC map for each unique
[[Page 24396]]
technology combination. This could theoretically be done through engine
or vehicle testing, but testing would need to be conducted on a single
engine, and each configuration would require physical parts and
associated engine calibrations to assess the impact of each technology
configuration, which is impractical for the rulemaking analysis because
of the extensive design, prototype part fabrication, development, and
laboratory resources that are required to evaluate each unique
configuration. Modeling is an approach used by industry to assess an
array of technologies with more limited testing. Modeling offers the
opportunity to isolate the effects of individual technologies by using
a single or small number of baseline engine configurations and
incrementally adding technologies to those baseline configurations.
This provides a consistent reference point for the BSFC maps for each
technology and for combinations of technologies which enables the
differences in effectiveness among technologies to be carefully
identified and quantified. The agencies selected this approach for the
NPRM and final rule. Engine maps were created by IAV using this
technically sound and rigorous methodology. Both absolute engine maps
and the incremental differences in engine maps were presented in the
PRIA.
Using a mix of engine maps from engine modeling and from
benchmarking data provides no common reference for measuring impacts of
adding specific technological improvements. In addition, as discussed
in further detail in Section VI.C.1.e), manufacturers often implement
multiple fuel-saving technologies simultaneously when redesigning a
vehicle and it is not possible to isolate the effect of individual
technologies by using laboratory measurements of a single production
engine or vehicle with a combination of technologies. Because so many
vehicle and engine changes are involved, it is not possible to
attribute effectiveness improvements accurately for benchmarked engines
to specific technology changes. This leads to overcounting or
undercounting technology effectiveness.
Further, while two or more different manufacturers may produce
engines with the same high level technologies (such as a DOHC engine
with VVT and SGDI), each manufacturer's engine will have unique
component designs that cause its version of the engine to have a unique
engine map. For example, engines with the same high level technologies
have unique intake manifold and exhaust manifold runners, cylinder head
ports and combustion chamber geometry that impact charge motion,
combustion and efficiency, as well as unique valve control, compression
ratios, engine friction, cooling systems, and fuel injector spray
characteristics, among other factors. The agencies developed and used a
single engine map to represent each technology and each combination of
engine technologies.
Therefore, it should not be expected that any of the agencies'
engine maps would necessarily align with a specific manufacturer's
engine, unless of course the engine map was developed from that
specific engine. The agencies do not agree that comparing an engine map
used for the rulemaking analysis to a single specific benchmarked
engine has technical relevance, beyond serving as a general
corroboration for the engine map. When a vehicle is benchmarked, the
resulting data is dictated by the unique combination of technologies
and design constraints for the whole vehicle system. For these reasons,
the agencies do not agree with ICCT that Eng13 and Eng14 should be
validated by conducting full vehicle modeling and comparing the results
with a single benchmarked vehicle. The engine maps used in this
analysis are precisely controlled for specific incremental technology
adoption and not for comparisons of absolute performance of a specific
vehicle's engine.
Differences are also explained by the NPRM and final rule analyses
using large-scale full vehicle Autonomie simulations to estimate
effectiveness instead of rough LPM approximations based on limited
ALPHA simulation work.\742\ These issues are discussed in more detail
in Section VI.B.3.
---------------------------------------------------------------------------
\742\ 2016 EPA Proposed Determination TSD at p.2-276 to 2-279
---------------------------------------------------------------------------
Accordingly, the agencies declined directly to use the Ricardo and
other EPA engine maps created from engine benchmarking as inputs for
this rulemaking because, among other reasons discussed below, they did
not afford the opportunity to evaluate the effectiveness improvements
for specific, individual technologies. For example, the 2018 Toyota
Camry 2.5L engine that EPA benchmarked had a broad array of observable
technologies, and several more that were not observable.\743\ However,
there was no baseline from which to isolate or compare any of the
individual technology improvements. For example, Toyota commented on
this benchmarking, stating:
---------------------------------------------------------------------------
\743\ EPA Test Data. 2018 Toyota Camry 2.5L A25A-FKS Engine Tier
3 Fuel. Available at https://www.epa.gov/sites/production/files/2019-04/2018-toyota-2.5l-a25a-fks-engine-tier3-fuel-test-data-package-dated-04-08-19.zip. Last accessed Nov. 20, 2019.
Past Toyota comments on Atkinson-cycle benefits have addressed
only those derived from variable valve timing (VVT) with late intake
valve closing (LIVC) that enables a 13:1 compression ratio. The
total 18.6 percent improvement of the 2018 Camry 2.5L over the
previous generation also includes benefits from cEGR and internal
engine design changes such as to the block, cylinder head, pistons,
valvetrain, as well as drivetrain and body/chassis
enhancements.\744\
---------------------------------------------------------------------------
\744\ NHTSA-2018-0067-12431. Supplemental Comments--Toyota Motor
North America, at p. 1-2.
Toyota's comments emphasize that the efficiency improvements in
this engine were driven by several additional technological
improvements, and not merely the cEGR, Atkinson cycle engine and higher
compression ratio design that was assumed for the EPA Draft TAR and
Proposed Determination analyses.\745\
---------------------------------------------------------------------------
\745\ EPA PD TSD at 2-229.
---------------------------------------------------------------------------
The agencies do agree component, engine, and vehicle test data are
very important for validating systems models, such as Autonomie, and
for validating model inputs, such as engine maps. Accordingly, the
agencies did fully consider engine maps used in prior rulemakings,
along with a broad array of other data as part of the process for
evaluating the IAV engine maps used for the NPRM and the final rule
analysis simulation work. Engine maps from Ricardo, EPA benchmarking,
NHTSA-sponsored benchmarking,\746\ information from technical papers
and conferences,\747\ extensive data and
[[Page 24397]]
expertise from the Argonne AMTL vehicle testing group and Energy
modeling group, \748\ and the 2015 NAS report,\749\ were all sources
used to confirm that incremental technology effectiveness estimates
were appropriate. The engine maps developed by IAV provided reliable
and reasonable estimates for the incremental impacts of engine
technologies. The use of this approach explains some of the
effectiveness differences between the NPRM and final rule analyses, and
the EPA Draft TAR and Proposed Determination analyses.
---------------------------------------------------------------------------
\746\ NHTSA Benchmarking, ``Laboratory Testing of a 2017 Ford F-
150 3.5 V6 EcoBoost with a 10 speed transmission.'' DOT HS 812 520.
\747\ Maruyama, F., Kojima, M., and Kanda, T., ``Development of
New CVT for Compact Car,'' SAE Technical Paper 2015-01-1091, 2015,
doi:10.4271/2015-01-1091. Shelby, M., Leone, T., Byrd, K., and Wong,
F., ``Fuel Economy Potential of Variable Compression Ratio for Light
Duty Vehicles,'' SAE Int. J. Engines 10(3):2017, doi:10.4271/2017-
01-0639. Eisazadeh-Far, K. and Younkins, M., ``Fuel Economy Gains
through Dynamic-Skip-Fire in Spark Ignition Engines,'' SAE Technical
Paper 2016-01-0672, 2016, doi:10.4271/2016-01-0672. Wade, R.,
Murphy, S., Cross, P., and Hansen, C., ``A Variable Displacement
Supercharger Performance Evaluation,'' SAE Technical Paper 2017-01-
0640, 2017, doi:10.4271/2017-01-0640. Hakariya, M., Toda, T., and
Sakai, M., ``The New Toyota Inline 4-Cylinder 2.5L Gasoline
Engine,'' SAE Technical Paper 2017-01-1021, 2017, doi:10.4271/2017-
01-1021. Ogino, K., Yakabe, Y., and Chujo, K., ``Development of the
New V6 3.5L Gasoline Direct Injection Engine,'' SAE Technical Paper
2017-01-1022, 2017, doi:10.4271/2017-01-1022. Shibata, M., Kawamata,
M., Komatsu, H., Maeyama, K. et al., ``New 1.0L I3 Turbocharged
Gasoline Direct Injection Engine,'' SAE Technical Paper 2017-01-
1029, 2017, doi:10.4271/2017-01-1029. Conway, G., Robertson, D.,
Chadwell, C., McDonald, J. et al., ``Evaluation of Emerging
Technologies on a 1.6 L Turbocharged GDI Engine,'' SAE Technical
Paper 2018-01-1423, 2018, doi:10.4271/2018-01-1423.
\748\ ANL Energy Group. https://www.anl.gov/es; ANL AMTL group.
https://www.anl.gov/es/advanced-mobility-technology-laboratory.
\749\ National Research Council. 2015. Cost, Effectiveness, and
Deployment of Fuel Economy Technologies for Light-Duty Vehicles.
Washington, DC--The National Academies Press, at pp. 294-305.
https://doi.org/10.17226/21744.
---------------------------------------------------------------------------
In considering ICCT's comment about using IAV engine maps or EPA's
engine maps, as an exercise, the agencies compared two IAV engine maps
to the EPA's benchmarked Toyota 2.5L naturally aspirated engine and
Honda's 1.5L turbocharged downsized engine.750 751 The IAV
engines were modeled and simulated in a midsize non-performance vehicle
with an automatic transmission and the same road load technologies,
MR0, ROLL0 and AERO0, to isolate for the benefits associated with the
specific engine maps.\752\ Eng 12, a 1.6L, 4 cylinder, turbocharged,
SGDI, DOHC, dual cam VVT, VVL engine was selected as the closest engine
configuration to the Honda 1.5L. Eng 22b, a 2.5L, 4 cylinder, VVT
Atkinson cycle engine, was selected as the closest engine configuration
to the Toyota 2.5L. As discussed before, both the Toyota 2.5L naturally
aspirated engine and Honda's 1.5L engine have incorporated a number of
fuel saving technologies including improved accessories and engine
friction reduction. In order to assure an ``apples-to-apples''
comparison, both IACC and EFR technologies were applied to the IAV
engine maps. IACC technology provides an additional 3.6% incremental
improvement and EFR provides an additional 1.4% incremental improvement
beyond the IAV engine maps for midsize non-performance vehicles.\753\
---------------------------------------------------------------------------
\750\ Toyota 2.5L TNGA Prototype Engine From 2016 SAE Paper--
ALPHA Map Package. Version 2017-12. Ann Arbor, MI: US EPA National
Vehicle and Fuel Emissions Laboratory, National Center for Advanced
Technology, 2017.
\751\ Honda 1.5L Turbo Prototype Engine From 2016 SAE Paper--
ALPHA Map Package. Version 2017-12. Ann Arbor, MI: US EPA National
Vehicle and Fuel Emissions Laboratory, National Center for Advanced
Technology, 2017.
\752\ See ANL--All Assumptions_Summary_FRM_06172019_FINAL and
ANL--Summary of Main Component Performance
Assumptions_FRM_06172019_FINAL for midsize class characteristics.
\753\ The NPRM and this final rule analysis allowed the adoption
of IACC technologies in the CAFE model that provided an additional
3.6% incremental improvement for the midsize car vehicle class. As
discussed in [Section VI.C Other Technologies], these benefits are
not shown in the IAV engine simulated results, so they were added
manually for this comparison.
---------------------------------------------------------------------------
The comparison shows effectiveness of the IAV engine maps and
effectiveness values for the final rule analysis are in line with the
Honda 1.5L and the Toyota 2.5L benchmarked engines. Figure VI-15 below
shows the effectiveness improvements for the EPA benchmarked engines
and the corresponding IAV engine maps incremental to a baseline
vehicle. Accordingly, the agencies believe that the methodology used in
this analysis, and the engine maps and incremental effectiveness values
used, are in line with benchmarking data and are reasonable for the
rulemaking analysis. The agencies believe the approach used in this
rulemaking analysis appropriately allows the agencies to account for a
wide array of engine technologies that could be adopted during the
rulemaking timeframe. Declining to use manufacturer-specific engines
allows the agencies to ensure that all effectiveness and cost
improvements due to the incremental addition of fuel economy improving
technologies are appropriately accounted for.
[[Page 24398]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.147
Next, Roush Industries (``Roush''), writing on behalf of the
California Air Resources Board, commented that the NPRM-modeled engines
vary in cylinder size, which would significantly alter combustion, heat
transfer, knock tolerance, and other important operating
parameters.\754\ Roush stated that a more accurate simulation, which
would improve incremental fuel economy improvement, should maintain a
consistent cylinder displacement (500cc) and vary the number of
cylinders or expected fuel consumption maps.\755\
---------------------------------------------------------------------------
\754\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 12.
\755\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 12.
---------------------------------------------------------------------------
The agencies believe that holding cylinder volume constant is the
appropriate approach to research seeking to identify the impacts of
technological changes on BSFC, torque, power, and other
characteristics, when holding cylinder volume constant. However, as
explained in Section VI.B.3.a)(2) Maintaining Vehicle Attributes and
Section VI.B.3.a)(6) Performance Neutrality, CAFE and CO2
rulemaking analyses attempt to maintain vehicle attributes, including
performance, and hold all of the attributes constant when showing
pathways that improve fuel economy. Therefore, the agencies' analyses
require engine maps that attempt to hold performance constant--not
necessarily cylinder size. Since certain fuel economy improving
technologies would increase performance if cylinder size is held
constant, such as when adding turbocharging technology, the agencies
appropriately include changes in displacement and cylinder volume for
technologies that have a significant impact on engine torque and power,
such as turbocharging. For a number of fuel economy improving
technologies that had smaller impacts on engine torque and power, the
engine maps were created with cylinder volume held constant. Table VI-
39 identifies the engine displacement information for each of the
engine maps. For example, the same engine displacement (2.0 L) and
cylinder displacement (500 cc) was used for creating engine maps for
naturally aspirated engines Eng01, Eng02, Eng03, Eng04, Eng05a, Eng5b,
Eng06a, Eng07a, and Eng08a, whereas engine displacement (1.6 L) and
cylinder displacement (400 cc) is used for creating the engine map for
turbocharged engine Eng12 in order to maintain performance. The
agencies have concluded that the approach used for the NPRM and the
final rule analysis is the most technically sound approach given the
data needs and assessments required for CAFE and CO2
rulemaking.
Roush also commented as follows:
[S]everal of the base engine maps used in the 2018 PRIA analysis
exhibit maximum thermal efficiency (lowest fuel consumption) at
2000-3000 rpm and at maximum load, which is unrealistic for normal
passenger vehicle engines. Such maps will over predict fuel economy
for extremely down-sized applications (very small engine in a heavy
vehicle). This is because there is no fuel economy penalty for
running the engine at a high loads point where, in reality, BSFC is
high due to retarding spark timing to prevent knocking and fuel
enrichment to reduce exhaust temperatures to protect exhaust valves
and turbocharger components.\756\
---------------------------------------------------------------------------
\756\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 11.
For example, Roush stated that Eng12 is predicted to have its
highest efficiency at very high load and high engine speeds with no
degradation in brake specific fuel consumption (BSFC) at engine speeds
between 2,000 rpm and 4,500 rpm all the way up to peak load, which is
unrealistic because
[[Page 24399]]
turbocharged engines at high loads require retarded spark timing to
prevent knock and fuel enrichment to prevent overheating of the
turbocharger and related components.\757\ Roush stated that these
factors would increase fuel consumption and reduce efficiency under
real-world conditions.\758\ Roush also stated that another effect of
the Eng12 fuel consumption curve would be to predict unreasonably good
fuel consumption at very high power levels for downsized turbocharged
engines. Roush stated this could bias technology pathways in over-
predicting fuel economy benefits for small engines installed in heavier
vehicles, causing an overly optimistic predicted performance of the
vehicle with regard to drivability, acceleration, and fuel consumption,
which would create unrealistic real-world pathways to compliance.\759\
---------------------------------------------------------------------------
\757\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 18.
\758\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 19.
\759\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 23.
---------------------------------------------------------------------------
As discussed in the Argonne model documentation for the final rule
analysis, the simulations used to determine incremental effectiveness
for the NPRM and final rule analyses were conducted using 2-cycle test
procedures, because they are the test procedures used for CAFE and
CO2 compliance.\760\ Therefore, the engines maps are
intended to represent BSFC accurately under those test conditions and
do not need to capture BSFC under every operating condition. During 2-
cycle test conditions, engines do not operate for extended periods at
the speed and high load conditions noted by Roush. A few vehicle and
engine combinations may operate at those speed and load points only
briefly during the 2-cycle CAFE and CO2 tests. Engines are
capable of operating for short periods of time under higher exhaust
temperature conditions and manufacturers commonly delay fuel enrichment
until it is needed to protect engine components (in particular exhaust
valves and exhaust manifolds) from excessive temperatures that can
impact engine durability. Fuel enrichment can be delayed because it
takes a period of time at higher temperature for components to heat up
and reach a temperature that would impact durability. Because these
high speed and load conditions occur for a relatively short time during
the CAFE and CO2 test cycles, and then return to lower speed
and/or load conditions with lower exhaust temperature, engines operate
for the entire CAFE and CO2 test cycles without triggering
fuel enrichment. The fuel enrichment delay also enables vehicles to
comply with criteria emission regulations and improves real world fuel
economy. Therefore, the engine maps used for the NPRM and final rule
analysis fully represent how engines operate during CAFE and
CO2 test cycles, and properly do not include fuel enrichment
at all 2-cycle operating conditions. Also, a trained knock model was
used to develop the engine maps, and the spark timing reflects
appropriate levels for engine operation during the delay in fuel
enrichment.
---------------------------------------------------------------------------
\760\ A Detailed Vehicle Simulation Process To Support CAFE and
CO2 Standards for the MY 2021-2026 Final Rule Analysis.
---------------------------------------------------------------------------
Next, regarding developing the NPRM engine maps to account for Tier
3 test fuel, the Alliance and Ford stated that the engine maps using
Tier 3 test fuel represented an improvement over prior analyses. The
Alliance stated that previous EPA modeling had incorrectly used Tier 2
premium octane fuel to predict the benefits of engine technologies,
which overstated fuel economy gains that would be achievable when using
regular-grade octane Tier 3 fuel. Ford provided similar comments, and
also noted that regular grade octane fuel will be required for
compliance after the 2020 model year.\761\
---------------------------------------------------------------------------
\761\ Ford Motors, Attachment, Docket No. EPA-HQ-OAR-2018-0283-
5691, at 7.
---------------------------------------------------------------------------
In contrast, ICCT and UCS both commented that the agencies had
incorrectly updated the IAV engine maps developed with Tier 2 test fuel
to account for Tier 3 fuel.\762\ ICCT stated that the update reduced
the effectiveness of the turbo technologies and suggested that the fuel
update adjustment should not have been done at all, stating
manufacturers that label vehicles as ``premium fuel recommended'' are
required to show no emissions changes over all test cycles when using
premium octane fuel and therefore reducing effectiveness for fuel
differences, as the agencies did with the IAV engine maps, is
unrealistic and inappropriate.
---------------------------------------------------------------------------
\762\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-82; Union of Concerned
Scientists, Technical Appendix, Docket No. NHTSA-2018-0067-12039, at
p. 15.
---------------------------------------------------------------------------
UCS also commented more specifically on the impact of the
adjustment from Tier 2 to Tier 3 fuel related to the knock threshold
for advanced engines, noting that manufacturers consider different
approaches to different fuels, and not all of those approaches
necessitate reductions in efficiency, as the agencies' assumption
suggests. UCS stated that charge cooling can reduce knock in direct
injection engines, resulting in an ``effective octane'' difference of a
six point increase for E10, thus potentially compensating for the
difference in octane between Tier 2 (E0 93 AKI) and Tier 3 (E10 87 AKI)
fuels. UCS argued that excluding this consideration led the agencies to
restrict advanced engines like HCR2 and reduce the effectiveness of
turbocharged engines with CEGR. UCS suggested that there would be a
reduction in the costs between the baseline and proposed standards if
the analysis allowed the application of HCR2 engines and corrected the
effectiveness of turbocharged CEGR engines.
Both ICCT and UCS also stated that the adjustment ignored a 2018
EPA study showing that, while fuel consumption increases with the
switch from Tier 2 to Tier 3 test fuel, emissions are reduced, meaning
that the agencies' adjustment is wrong ``for some technologies because
[CO2]-per-mile emissions can be lower with the switch to
higher octane ethanol blends.'' UCS also stated that the adjustment
factor applied is wrong for two reasons, first because converting
solely with energy density would assume a 3.7 percent increase in fuel
consumption compared to the observed 2.7 percent increase, and second
because the adjustment goes in the wrong direction when applied to
CO2 emissions, which show a reduction of 1.4 percent on the
test cycle. UCS stated that the Autonomie model accordingly overstates
CO2 emissions on Tier 3 fuel by 4.2 percent. UCS argued that
the adjustment to account for Tier 3 test fuel therefore double counts
any penalty in fuel economy and ignores CO2 tailpipe
reductions, which would result in an improvement on the test cycle.
Because the CAFE test procedure already has an adjustment in place to
correct for fuel properties relative to 1975 test fuel, but carbon-
related exhaust emissions do not, UCS stated that the fuel adjustment
could lead to drastically conservative fuel economy and CO2
curves.
ICCT stated that the agencies could fix this issue by relying on
EPA's engine maps, where EPA had accounted for cost and effectiveness
of technology used to protect operation on regular octane fuel by
increasing costs and reducing effectiveness.
Some of these comments can be addressed with a simple
clarification: The NPRM contained text that was
[[Page 24400]]
inconsistent regarding how the analysis accounted for the engine maps
(which were based on Tier 3 fuel). The separate model documentation
correctly described that, for the NPRM analysis, the agencies developed
fuel maps for Tier 3 fuel and did not adjust the final Autonomie
outputs.\763\ The NPRM text, however, incorrectly stated that ``(a)n
adjustment factor was applied to the Autonomie simulation results to
adjust them to reflect Tier 2 certification fuel. Argonne adjusted the
vehicle fuel economy results to present certification fuel by using the
ratio of the lower heating values to the rest and certification
fuels.'' In fact, no adjustments were made to the NPRM Autonomie
simulation outputs, as the modeled engine maps were appropriately
modeled using Tier 3 fuel.
---------------------------------------------------------------------------
\763\ NHTSA-2018-0067-0007 at 177-178 and 191.
---------------------------------------------------------------------------
As discussed in detail in VI.C.1.a) Fuel Octane, engine
specifications used to create the engine maps for the NPRM and the
final rule were developed using Tier 3 fuel. Tier 3 fuel was used to
ensure the engines were capable of operating on real world regular
octane (87 pump octane = (R+M/2)). This capability is in line with what
manufacturers must do to ensure engines have acceptable noise,
vibration, harshness, drivability and performance levels, and will not
fail prematurely when operated on regular octane fuel. If the agencies
developed engine maps based on Tier 2 fuel alone, the engine maps would
reflect the engines' ability to have higher compression ratios and to
operate with greater levels of spark advance than could be implemented
by manufacturers, who must take into account operation on regular
octane fuels used by a majority of U.S. consumers.\764\ Not considering
regular octane fuel operation by manufacturers would lead to engine
durability, and engine noise, vibration, harshness, and drivability
issues. Manufacturers have told the agencies that even for vehicles
designed to operate on high octane fuel, the engines and controls must
be designed to operate on every fuel octane level available in the U.S.
to avoid these issues.\765\ Thus, developing engine maps based on Tier
2 fuel alone would incorrectly overstate the BSFC improvements
achievable in the real world.
---------------------------------------------------------------------------
\764\ Tamm, D.C., Devenish, G.N. Finelt, D.N. Kalt, L.K.
``Analysis of Gasoline Octane Costs'' Baiker and O'brien, Inc.
Prepared for EIA. October 18, 2018. https://www.eia.gov/analysis/octanestudy/pdf/phase1.pdf at 11-13.
\765\ Ford Motor Company. NHTSA-2016-0068-0048 at 3. Auto
Alliance comments for 2016 draft TAR. Attachment 7 Limitations of
Ricardo Fuel Economy Analysis of Downsizing. NHTSA-2016-0068-0070.
---------------------------------------------------------------------------
Based on these comments and considerations, the agencies determined
the engine maps developed for the NPRM appropriately account for fuel
octane, and better approximate BSFC achieved by the majority of engines
used in the U.S. vehicle fleet. The agencies believe ICCT's and other
commenters' assertions that the engine maps should reflect Tier 2 fuel
and not be updated for Tier 3 fuel would ignore these important
considerations, and would provide engine maps that could not be
achieved by engines in the real world. The agencies determined that
engine maps developed for the Draft TAR and EPA Proposed Determination
that were based on Tier 2 fuel should not be used for the NPRM and
final rule analyses for these reasons.
EPA is addressing the impact of Tier 3 fuel on fuel economy and
CO2 emissions compliance test results as part of a separate
rulemaking. The separate rulemaking may establish an adjustment to
account for the impacts of the change in test fuel. Those impacts are
beyond the scope of this rulemaking. The analysis for this rule uses
fuel economy and CO2 emissions of the vehicles in the MY
2017 analysis fleet as the reference for absolute fuel economy and
CO2 emissions. The analysis starts with absolute compliance
data from MY 2017 and adopts technologies incrementally to determine
future compliance. Because MY 2017 absolute compliance values are based
on Tier 2 fuel, and standards are based on the use of Tier 2 fuel,
there is no need to make any adjustments for the differences in energy
content and carbon content of Tier 2 and Tier 3 fuel.\766\
---------------------------------------------------------------------------
\766\ During the 1980s, the U.S. Environmental Protection Agency
(EPA) incorporated the R factor into fuel economy calculations in
order to address concerns about the impacts of test fuel property
variations on corporate average fuel economy (CAFE) compliance,
which is determined using the Federal Test Procedure (FTP) and
Highway Fuel Economy Test (HFET) cycles. The R factor is defined as
the ratio of the percent change in fuel economy to the percent
change in volumetric heating value for tests conducted using two
differing fuels.
---------------------------------------------------------------------------
The agencies considered ICCT's statement that manufacturers that
label vehicles as ``premium fuel recommended'' are required to show no
emissions changes over all test cycles when using regular octane fuel,
and therefore reducing effectiveness for fuel differences as the
agencies did with the IAV engine maps is unrealistic and inappropriate.
The agencies believe these conclusions are technically incorrect. The
existence of an EPA compliance regulation does not impact the laws of
nature, which govern issues associated with the impact of fuel octane
on the ability to improve engine BSFC and on engine durability, noise,
vibration, harshness, and drivability. It is widely recognized and
accepted that higher octane fuels allow engines to be designed with
higher compression ratios, faster combustion rates, and more optimal
spark advance, which improve BSFC. Section VI.C.1.a) discusses comments
advocating for increasing the minimum fuel octane specification to
enable these improvements. The engine maps developed by IAV and used
for the Draft TAR and NPRM were consistent with these trends and showed
that BSFC is better with Tier 2 (higher octane) fuel than Tier 3 (lower
octane) fuel.\767\ ICCT did not provide any data supporting the concept
that there is no shift in BSFC, fuel economy, or CO2
emissions when engines are optimized with different octane fuels, or
between Tier 2 and Tier 3 fuel. It is appropriate to note that the EPA
regulation does provide a tolerance which in practice allows a small
level of shift in emissions.\768\
---------------------------------------------------------------------------
\767\ See BSFC difference between engines modeled with Tier 3
fuel versus high octane fuel by IAV in PRIA 6.3.2.2.20.9 at 288 to
PRIA 6.3.2.20.11 at 292.
\768\ 40 CFR 1066.210 (b) Accuracy and Precision.
---------------------------------------------------------------------------
Regarding comments that certain combinations of technologies can
enable BSFC improvements while controlling spark knock, the agencies in
fact considered a very broad array of engine technology combinations
for the analysis, including several added technologies as discussed
further below. The agencies believe the rigorous methodology used to
develop the engine maps resulted in engine maps representing the
maximum improvement in BSFC for each engine configuration, while also
addressing real world constraints. Engine maps for the new technologies
were presented in PRIA Chapter 6.3.2.2.16.4. The PRIA also discussed
that IAV maps were developed considering a very comprehensive list of
combustion operating parameters as part of the IAV GT-Power engine
modeling. IAV's GT-Power engine modeling included sub-models to account
for heat release through a predictive combustion model, knock
characteristic through a kinetic fit knock model, physics-based heat
flow model physics based friction model, and IAV's proprietary
Optimization Tool Box.\769\ These independent models were
[[Page 24401]]
run concurrently to make sure engine design requirements were met for
each engine configuration that was modeled.
---------------------------------------------------------------------------
\769\ IAV's Optimization Tool Box is a module of IAV Engine. IAV
Engine, as the basic platform for designing engine mechanics,
provides a large number of tools that have proven their worth across
the globe in several decades of automotive development work at IAV.
The modules help designers, computation engineers and simulation
specialists in designing mechanical engine components--for example,
in laying out valvetrains and timing gears as well as crankshafts.
---------------------------------------------------------------------------
Finally, in response to the agencies' request for comment on
including the additional engine maps presented in the NPRM as potential
technological pathways, several commenters stated that the agencies
should include those technologies, in addition to other emerging engine
technologies.\770\ After considering these comments, the agencies added
several engine technologies and technology combinations to the final
rule analysis. The additions included a basic high compression ratio
Atkinson mode engine (HCR0), a variable compression ratio engine (VCR),
a variable turbo geometry engine (VTG), and a variable turbo geometry
with electric assist engine (VTGe). The agencies also added advanced
cylinder deactivation technology (TURBOAD) to Eng12 (TURBOD) in the
Autonomie modeling for the final rule analysis. Like with ADEAC, the
agencies did not have IAV engine maps for TURBOAD, so the agencies took
the effectiveness values as predicted by full vehicle simulations of a
TURBOD and added 1.5 percent or 3 percent respectively for I-4 engines
and V-6 or V-8 engines, as explained in more detail further below. The
agencies also included more iterations of existing technologies, like
diesel engines with cylinder deactivation, diesel engines paired with
manual transmissions, and diesel engines paired with 12-volt start stop
technology, in addition to more combinations of hybrid technologies
that are discussed further in Section VI.C.3, below.
---------------------------------------------------------------------------
\770\ ICCT Docket # NHTSA-2018-0067-11741 at I-19--I-22; CARB
Docket # NHTSA-2018-0067-11873 at 107-108.
---------------------------------------------------------------------------
The following sections list and describe the comprehensive set of
engine technologies and combinations of engine technologies that have
been included in the analysis. The agencies also discuss the additional
engine technologies added for the final rule, and reasons for excluding
a small number of technologies proffered by commenters. The agencies
believe the wide array of engine technologies included in the final
rule analysis and the methodology used to develop the engine maps to
measure the effectiveness of those technologies reasonably represents
the scope of technologies that should be considered during the
rulemaking timeframe.
c) Engine Modeling in the CAFE Model
(1) Basic Engines
The NPRM described that there are a number of engine technologies
that manufacturers can use to improve fuel economy and CO2
emissions. Some engine technologies can be incorporated into existing
engines with minor or moderate changes to the engines, but many engine
technologies require an entirely new engine architecture. The terms
``basic engine technologies'' and ``advanced engine technologies'' are
used only to define how the CAFE model applies a specific engine
technology and handles incremental costs and effectiveness
improvements. ``Basic engine technologies'' refer to technologies that,
in many cases, can be adapted to an existing engine with minor or
moderate changes to the engine, compared to ``advanced engine
technologies'' that generally require significant changes or an
entirely new engine architecture.
In the CAFE model, basic engine technologies may be applied in
combination with other basic engine technologies; advanced engine
technologies (defined by an engine map) stand alone as an exclusive
engine technology. The words ``basic'' and ``advanced'' are not meant
to confer any information about the level of sophistication of the
technology. Also, many advanced engine technology definitions include
some basic engine technologies, but these basic technologies are
already accounted for in the costs and effectiveness values of the
advance engine. The ``basic engine technologies'' need not be (and are
not) applied in addition to the ``advanced engine technologies'' in the
CAFE model.
(a) DOHC
In the NPRM analysis, the agencies characterized dual overhead cam
(DOHC) engine technology as ``basic.'' DOHC engine configurations have
two camshafts per cylinder head, one operating the intake valves and
one operating the exhaust valves. Four basic engine technologies--
variable valve timing (VVT), variable valve lift (VVL), stoichiometric
gasoline direction injection (SGDI), and basic cylinder deactivation
(DEAC)--were considered for DOHC engines. Implementing these
technologies involves changes to the cylinder head of the engine, but
the engine block, crankshaft, pistons, and connecting rods require few,
if any, changes.
Variable valve timing (VVT) is a family of valve-train designs that
dynamically adjusts the timing of the intake valves, exhaust valves, or
both, in relation to piston position. VVT can reduce pumping losses,
provide increased engine torque and horsepower over a broad engine
operating range, and allow unique operating modes, such as Atkinson
cycle operation, to further enhance efficiency. VVT is nearly
universally used in the MY 2017 fleet.\771\ In the NPRM analysis, the
VVT technology modeled by IAV was based on dual (independent) cam
phasing. This was a more advanced VVT technology that allowed
controlling of valve overlap, which can be used to control internal EGR
to minimize fuel consumption at low engine loads.\772\ VVT enables
control of many aspects of air flow, exhaust scavenging, and combustion
relative to fixed valve timing engines. Engine parameters such as
volumetric efficiency, effective compression ratio, and internal
exhaust gas recirculation (iEGR) can all be enabled and accurately
controlled by a VVT system.
---------------------------------------------------------------------------
\771\ 98.1 percent of MY2017 vehicles are equipped with VVT. EPA
Report. The 2018 EPA Automotive Trends Report. https://nepis.epa.gov/Exe/ZyPDF.cgi/P100W5C2.PDF?Dockey=P100W5C2.PDF at
Table 4.1 Production Share by Engine technology.
\772\ 2015 NAS at p. 32.
---------------------------------------------------------------------------
Variable valve lift (VVL) dynamically adjusts the distance a valve
travels from the valve seat optimizing airflow over a broad range of
engine operating conditions. The technology can increase effectiveness
by reducing pumping losses and may improve efficiency by affecting in-
cylinder charge (fuel and air mixture), motion, and combustion. VVL is
less common in the 2017 fleet than VVT. Some manufacturers have
implemented a limited, discrete approach to VVL where just two valve
lift profiles are available versus a full-range, continuously variable
implementation.
Stoichiometric gasoline direct injection (SGDI) sprays fuel at high
pressure directly into the combustion chamber, which provides cooling
of the in-cylinder charge via in-cylinder fuel vaporization to improve
spark knock tolerance and enable an increase in compression ratio and/
or more optimal spark timing for improved efficiency. SGDI appears in
about half of basic engines produced in MY 2017, and the technology is
used in many advanced engines as well.\773\
---------------------------------------------------------------------------
\773\ 49.7 percent of MY2017 vehicles are equipped with SGDI.
EPA Report. The 2018 EPA Automotive Trends Report. https://nepis.epa.gov/Exe/ZyPDF.cgi/P100W5C2.PDF?Dockey=P100W5C2.PDF at
Table 4.1 Production Share by Engine technology.
---------------------------------------------------------------------------
Basic cylinder deactivation (DEAC) disables intake and exhaust
valves and
[[Page 24402]]
turns off fuel injection for the deactivated cylinders during light-
load operation. The engine runs temporarily as though it were a smaller
engine, which reduces pumping losses and improves efficiency. In the MY
2017 fleet, manufacturers used DEAC on V6, V8, V10, and V12 engines in
OHV, SOHC, and DOHC engine configurations. With some engine
configurations in some operating conditions, DEAC creates noise-
vibration-and-harshness (NVH) challenges. NVH challenges are
significant for V6 and I4 DEAC configurations, and limit the operating
range where DEAC can operate. For I4 engine configurations with smaller
displacements, there are fewer operating conditions where engine load
is low enough to use DEAC, which limits effectiveness. No manufacturers
produced I4 DEAC engines in MY 2017. Typically, the smaller the engine
displacement, the less opportunity DEAC provides to improve fuel
consumption.
The agencies provided engine fuel maps for each of the eight DOHC
engines (Eng01, Eng02, Eng03, Eng04, Eng18, Eng19, Eng20, and Eng21)
used for the NPRM analysis. Each of these engines incrementally added
technology to Eng01, a basic VVT engine, while holding all other
factors constant like ambient temperature, ambient pressure, and fuel
type.
For the NPRM analysis, the agencies estimated the effectiveness of
DEAC using full vehicle modeling and simulation. In the NPRM PRIA
6.2.1.2, the agencies discussed how Autonomie uses a specific control
logic for cylinder deactivation for naturally aspirated engines that
takes into consideration for noise, vibration, and harshness.\774\ For
the final rule analysis, the agencies took steps to use full vehicle
modeling and simulation to apply DEAC to both naturally aspirated and
turbocharged engines. The same control logic was applied to the
turbocharged engine cylinder deactivation (TURBOD) for the final rule
analysis.
---------------------------------------------------------------------------
\774\ NHTSA-2018-0067-1972. ``Preliminary Regulatory Impact
Analysis (PRIA) The Safer Affordable Fuel-Efficient (SAFE) Vehicles
Rule for Model Year 2021-2026 Passenger Cars and Light Trucks,'' at
191.
---------------------------------------------------------------------------
The agencies used the same assumptions for advanced cylinder
deactivation (ADEAC) in the final rule analysis. In the NPRM the
agencies stated engine maps were not available at the time of the
analysis, and said that ADEAC was estimated to improve a basic engine
with VVL, VVT, SGDO, and DEAC by three percent (for 4 cylinder engines)
and six percent (for engines with more than 4 cylinders).\775\ The new
technology combination for turbocharged advanced cylinder deactivation
(TURBOAD) uses a similar approach for determining effectiveness. The
agencies have applied a one-and-a-half percent effectiveness
improvement estimate for 4-cylinder or smaller engines and a three
percent effectiveness estimate for 6-cylinder or larger engines
relative to TURBOD.
---------------------------------------------------------------------------
\775\ 83 FR 430039 (Aug. 24, 2018).
---------------------------------------------------------------------------
For the final rule analysis the basic engine path for DOHCs are
shown in Figure VI-16 and the high-level engine specifications are
shown in Table VI-41. The baseline basic DOHC engine, Eng01, was the
starting point and other engine technologies were incrementally adopted
to determine effectiveness. Adoption of DEAC technology for
turbocharged engines is discussed in Section VI.C.1.e)(2). Similarly,
ADEAC technology is discussed in Section VI.C.1.e)(4).
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.148
[[Page 24403]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.149
BILLING CODE 4910-59-C
(b) SOHC
Similar to DOHC engines, SOHC engines were characterized as
``Basic'' engine technologies in the NPRM analysis. They are
characterized by having a single camshaft in the cylinder head
operating both the intake and exhaust valves. Four basic engine
technologies, VVT, VVL, SGDI, and DEAC were considered for SOHC
engines. Implementing these technologies involves changes to the
cylinder head of the engine, but the engine block, crankshaft, pistons,
and connecting rods require few, if any, changes.
The agencies provided engine fuel maps for each of these types of
SOHC engines and requested comments. Engine maps 5b, 6a, 7a, and 8a
were modeled SOHC engines. The SOHC engine models used engine 5a, which
was based on Eng01 as a reference, by removing one camshaft. Eng5a was
included for the Draft TAR, but not included for the NPRM analysis due
to high BSFC from higher friction that was inherited from the DOHC
engine design. A level 0.1 bar of friction reduction over the entire
operating range for engine maps 5b, 6a, 7a, and 8a was applied to
represent improvements over existing engine designs. The addition of
friction reduction to these engines was a result of consideration of
deliberative interagency comments received during the Draft TAR review
process noting higher fuel consumption on the baseline SOHC engine 5a
relative to other modern SOHC engines.
Meszler on behalf of NRDC commented that ``[a]lthough variable
valve timing (VVT) technology is identified as an available refresh
technology, the NPRM CAFE model (unlike the version used for the 2016
TAR analysis) actually assumes that all baseline vehicles include VVT
technology. As a result, the approximately 9 percent of model year 2016
sales that do not actually include VVT are not credited with any
efficiency benefit for adoption of the technology . . . . '' \776\
---------------------------------------------------------------------------
\776\ Meszler, at 32.
---------------------------------------------------------------------------
We agree with this comment, and for the final rule analysis updated
the CAFE model to add a non-VVT level engine in the 2017 analysis fleet
and to allow those vehicles to adopt VVT technologies at a refresh or
redesign. However, the agencies did not have engine maps for the non-
VVT engines, so the agencies applied a fixed-value effectiveness
estimate from similar VVT engine maps to represent the effectiveness
for non-VVT engines. The agencies used the effectiveness of a similar
configuration technology package of another engine to represent non-VVT
engines. Non-VVT SOHC engines may add any combination of VVL with SGDI
and DEAC. The agencies believe that the estimated effectiveness used
for VVT engines was appropriate because the effectiveness offset is in
line with 2015 NAS estimates for VVT engines with respect to VVL
engines.777 778
---------------------------------------------------------------------------
\777\ Baseline effectiveness references for SOHC;VVT; SGDI;
AT5;CONV;ROLL0;MR0;AERO0, SOHC;VVT; DEAC; AT5;CONV;ROLL0;MR0;AERO0,
SOHC;VVT;VVL; DEAC; AT5;CONV;ROLL0;MR0;AERO0, and SOHC;VVT;
SGDI;DEAC; AT5;CONV;ROLL0;MR0;AERO0 were used to represent SOHC;VVL;
SGDI; AT5;CONV;ROLL0;MR0;AERO0, SOHC;VVL;DEAC;
AT5;CONV;ROLL0;MR0;AERO0, and SOHC;VVL; SGDI;DEAC;
AT5;CONV;ROLL0;MR0;AERO0 baseline combinations. These combinations
represented only 2% of the models and 3.1% sales by volume in the MY
2017 baseline fleet.
\778\ 2015 NAS Table 2.7 and Table 2.8 at 32-33.
---------------------------------------------------------------------------
The basic engine path for SOHC engines used in this final rule is
shown in Figure VI-17 and the specifications are shown in Table VI-42.
Note, that Eng5a is only a reference used to build the rest of the SOHC
engines.
BILLING CODE 4910-59-P
[[Page 24404]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.150
[GRAPHIC] [TIFF OMITTED] TR30AP20.151
BILLING CODE 4910-59-C
(2) Turbocharged Downsized Engines
Engine maps 12, 13, and 14 modeled turbocharged downsized engines.
Turbocharged downsized engines are characterized by technology that can
create greater-than-atmospheric pressure in the engine intake manifold
when higher output is needed. The raised pressure results in an
increased volume of airflow into the cylinder supporting combustion,
increasing the specific power of the engine. An increased specific
power means the engine can generate more power per unit of volume,
which allows engine volume to be reduced while maintaining performance,
thereby increasing fuel efficiency. IAV Eng12 was the base engine for
all simulated turbocharged engines and was validated using engine
dynamometer test data.\779\
---------------------------------------------------------------------------
\779\ Bottcher, L. Grigoriads, P. ``ANL--BSFC map prediction
Engines 22-26'' April, 30, 2019. IAV_20190430_Eng 22-26
Updated_Docket.pdf.
---------------------------------------------------------------------------
One notable change that the agencies made for the NPRM analysis
based on stakeholder comments to the Draft TAR was to update the turbo
family engine maps to assume operation on regular octane fuel (Tier 3,
or 87 AKI), instead of premium fuel (Tier 2, or 93 AKI), to assure the
maps accounted for real world constraints that impact durability and
drivability, and noise, vibration, and harshness. Using regular octane
fuel is consistent with the fuel octane that manufacturers specify be
used in the majority of vehicles (manufacturers generally only specify
premium fuel is required for higher performance models, although that
is not always the case), and enables the modeling to account for
important design and calibration issues associated with regular octane
fuel. The agencies noted in the NPRM that using the updated engine maps
addressed over-estimation of potential fuel economy improvements and
ensured that the analysis reflected real-world constraints faced by
manufacturers to assure engine durability and acceptable drivability.
Importantly, assuming no change in fuel octane required to
[[Page 24405]]
operate a vehicle ensures that the agencies are modeling technology
pathways that can improve fuel economy while maintaining vehicle
performance, capability, and other attributes.
Compared with the NHTSA analysis in the Draft TAR, the turbocharged
and downsized engine maps adjusted at high torque and low speed
operation, and at high speed operation to account for knock limitations
when using regular octane fuel. The knock model used to develop the
turbocharged engines was trained on production and development engines
tested at IAV to quantify the effects of different octane fuels.\780\
Below the knock threshold, there is no change to the fuel consumption
maps. The agencies noted that with the fuel octane change there are
generally two major effects in the regions where the engine is knock-
limited: First, spark timing is retarded causing a reduction in
combustion efficiency and hence an increase in BSFC, and second, an
increase in combustion and exhaust temperatures requiring fuel
enrichment to cool those temperatures for engine component protection
and resulting in increased BSFC.781 782
---------------------------------------------------------------------------
\780\ Knock models are based on Gamma Technology's kinetic fit
model per the technical paper titled, ``A combustion model for IC
engine combustion simulations with multi-component fuels,'' by
YoungChul Ra, Rolf D. Reitz--Engine Research Center, University of
Wisconsin-Madison.
\781\ Fuel enrichment is extra fuel is injected at the intake
manifold port or directly into the cylinder. Fuel vaporization and
the fuel's thermal mass reduces combustion and exhaust temperatures.
Changes to the air/fuel ratio also impact combustion speed which
impacts the knock limit.
\782\ Singh, E. and Dibble, R., ``Effectiveness of Fuel
Enrichment on Knock Suppression in a Gasoline Spark-Ignited
Engine,'' SAE Technical Paper 2018-01-1665, 2018, https://doi.org/10.4271/2018-01-1665.
---------------------------------------------------------------------------
The agencies also noted that for Eng14, the turbocharged downsized
engine with cooled exhaust gas recirculation (cEGR), cEGR was added at
the higher speeds where further reduction in combustion temperature was
required. The higher specific heat capacity of cEGR reduced the need
for fuel enrichment by lowering combustion temperatures and limiting
the amount of spark retardation necessary to manage spark knock. With
increasing load, cEGR is also used to lower combustion temperatures to
reduce NOx emissions. The agencies explained that because IAV's models
are not trained for emissions, cEGR was only considered for areas that
are knock-limited and/or to reduce combustion temperatures. Because
cEGR has the impact of slowing down burn rates, the amount of cEGR that
could be utilized was balanced to maintain efficient combustion.
Combustion stability was also evaluated to assure cEGR rates did not
cause excessive cycle-to-cycle combustion variations, which adversely
impact drivability.\783\
---------------------------------------------------------------------------
\783\ Heywood. B. J, Internal Combustion Engine Fundamentals, at
413-37, McGraw-Hill (1988).
---------------------------------------------------------------------------
Some commenters criticized these downsized turbocharged IAV maps,
referencing deliberative EPA comments docketed pursuant to the Clean
Air Act procedural requirements at 42 U.S.C. 7607, which stated that
the assumptions for Eng12's fuel octane, heating value, and carbon
content were not representative of certification fuel and did not
appear to be consistently used for the various engine maps, concluding
that the resultant engine maps were not representative of
CO2 performance of turbocharged engines over the
certification cycle. ICCT stated it appeared these concerns had not
been addressed for the NPRM, and that ``this problem essentially
affect[ed] all engines on the turbocharged engine pathway.'' \784\
---------------------------------------------------------------------------
\784\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-46.
---------------------------------------------------------------------------
The agencies disagree with ICCT's comments relating both to whether
fuel specifications were used consistently and whether the fuel
specifications for fuel octane, heating value and carbon content were
representative of the same fuel. First, the EPA deliberative comments
were resolved in the deliberative process through the clarification
that a single fuel specification was used to develop all of the engines
and engine maps. Therefore, the engine maps are internally consistent.
The fuel specification was presented in the NPRM section PRIA Chapter
6.3.2.2.17. Second, the agencies considered future fuel and emissions
standards by using regular octane fuel for this analysis. The
assumptions for the fuel used in this analysis align with the EPA's
Tier 3 standards that went into effect January 1, 2017.\785\ For the
reasons discussed further above, the agencies believe it is important
to use Tier 3 fuel for engine maps used for rulemaking analysis.
---------------------------------------------------------------------------
\785\ Final Rule for Control of Air Pollution from Motor
Vehicles: Tier 3 Motor Vehicle Emission and Fuel Standards. https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-air-pollution-motor-vehicles-tier-3. Last accessed September
26, 2019. Docket EPA-HQ-OAR-2011-0135.
---------------------------------------------------------------------------
Roush claimed that the turbocharged engine maps used in the
analysis were responsible for an overly-conservative estimate of
underlying combustion engine efficiencies, arguing that many production
engines available today use the same technology packages identified in
the PRIA but with significantly higher efficiencies.\786\ Roush noted
that the base turbocharged engine map used in the PRIA, Eng12, is
assumed to have variable valve lift (VVL), but with a turbocharged
engine the benefit of VVL over dual variable valve timing (VVT) is
limited.\787\ Roush argued that almost all vehicle manufacturers use
lower-cost dual VVT systems in their turbocharged engines, and that the
agencies' base turbocharged engine assumption is unrealistic with a
correspondingly high cost.\788\
---------------------------------------------------------------------------
\786\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 16.
\787\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 17.
\788\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 17.
---------------------------------------------------------------------------
Roush contrasted its critique of Eng12 with an EPA ALPHA run of a
2016 Honda Civic 1.5L turbocharged engine (L15B7) with continuously
variable intake and exhaust camshaft phasing (CVVT), which is less
expensive than the CVVL, arguing that it showed greater efficiency over
more of the engine map at a lower cost than Eng12. Roush further argued
that since the L15B7 engine is the first generation of the new Honda
turbocharged engine, ``even further fuel consumption improvement is
highly likely in the period through MY2025.'' \789\
---------------------------------------------------------------------------
\789\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 18.
---------------------------------------------------------------------------
As the agencies explained further above, from a technical
perspective there is no reason why the 2016 Honda Civic 1.5 L Turbo
should have an engine map that is the same as Eng12, Eng13, or Eng14.
The turbocharged engine technologies represented by Eng12, Eng13 and
Eng14 are not representative of any specific engine from any one
manufacturer. Honda's 1.5L turbocharged engine incorporates a unique
combination of technologies including electric wastegate, sodium-filled
exhaust valves, light weight internal components, friction reduction
technologies, 2-stage oil pump, low viscosity oil (0W-20), and a unique
exhaust system.\790\
---------------------------------------------------------------------------
\790\ Honda Press Release. ``2016 Honda Civic Sedan Press Kit--
Powertrain'' October 18, 2015. https://hondanews.com/en-US/releases/2016-honda-civic-sedan-press-kit-overview?page=178. Last accessed
Feb. 12, 2020.
---------------------------------------------------------------------------
While there are an enormous number of different technology
combinations that manufacturers could apply on their
[[Page 24406]]
engines, the agencies' analysis must select a reasonable number of
configurations--in fact, the agencies analyze thousands of unique make/
model/powertrain combinations and apply them to over one hundred
thousand unique technology combinations for each of ten classes for
this rulemaking. See Section VI.B.3.a)(6) and Section VI.B.3 for more
details. For turbocharged engines, the agencies selected eight
combinations which represent a wide range of technologies, combinations
of technologies, and effectiveness improvements for the rulemaking
analysis, as listed in Table VI-40. Three of the combinations were
added based on commenter's recommendations. While it is possible to
identify other combinations, such as the unique technologies Honda
chose for its 1.5L Turbo engine, agencies do not believe it would be
appropriate to select all of the technologies on one specific
manufacturer's engine for the rulemaking analysis. Doing so would,
appropriately, raise questions about the availability of proprietary
designs and controls to other manufacturers, among other
considerations.
The agencies also believe that the engine maps for Eng12, Eng13 and
Eng14 show reasonable differences in BSFC maps that characterize the
impact of each of these technology combinations, and differences
relative to naturally aspirated engines. As discussed further above,
incremental differences in BSFC are used for the rulemaking analysis.
Roush's comments center on the comparison of absolute effectiveness
values for a specific production vehicle, and do not address
incremental effectiveness among a range of technologies, nor the
appropriate baseline reference for the Honda 1.5L Turbo for technology
content and for effectiveness. The ALPHA simulation for the 2016 Honda
Civic 1.5L turbocharged engine provides absolute test data and has no
baseline for assessing incremental effectiveness. Because there is no
baseline, there is no basis for identifying which specific technologies
have changed, nor any basis for determining the incremental
effectiveness of each individual technology.
Regarding Roush's comment that that further fuel consumption
improvement for the Honda L15B7 is highly likely in the period through
MY 2025, Roush provided no information or data on what specific
technologies would further improve the fuel consumption of that engine.
With no defined new technology to consider, there is no basis for
estimating the costs, nor for estimating the effectiveness of Roush's
assertion. Without further information, the agencies can only point to
the additional engine technologies considered for this final rule,
discussed further below.
ICCT also stated that IAV's handling of cooled EGR (cEGR) in the
engine maps was inappropriate, as IAV analyzed cEGR as a knock-
abatement technology instead of a fuel efficiency technology. ICCT
stated that this is reason that the NPRM analysis showed no benefit to
cEGR, and if the agencies had used EPA's properly modeled cEGR
effectiveness based on validated data, the effectiveness of cEGR would
have been more realistic.
Similarly, Roush commented that cEGR application in the modeled
turbocharged engines is excluded in engine operating modes that highly
influence vehicle fuel economy. Roush contrasted Eng13, a turbocharged
engine with VVT, direct injection, and cEGR, with the Mazda 2.5L
SkyActiv Turbo engine available in the 2016 Mazda CX-9, which also
employs cEGR.
The agencies believe Eng14 was created and modeled using a sound
technical methodology, using constraints that the industry uses to
ensure the engines would meet durability and customer acceptability
criteria. IAV turbocharged engines adopted VVT and VVL to maximize
volumetric efficiency and improve the combustion process. Engines with
VVT control intake and exhaust valve timing to recycle burned exhaust
gas into the combustion chamber. The recycling of exhaust gases using
VVT is commonly called internal EGR. Cooled EGR (cEGR) is a second
method for diluting the incoming air that takes exhaust gases, passes
them through a cooler to reduce their temperature, and then mixes them
with incoming air in the intake manifold. Diluting the incoming air
with inert exhaust gas reduces pumping losses, thereby improving BSFC.
The dilution also reduces combustion rates, temperatures, and
pressures, which mitigates spark knock and reduces the need for fuel
enrichment at higher loads to control exhaust temperature for component
durability (typically, exhaust valves and exhaust manifold). Not only
does this exhaust gas displace some incoming air, but it also heats the
incoming air and lowers its density. Both interactions lower the
volumetric efficiency of the engine.\791\ Cooled EGR is a more
effective way of reducing combustion temperature in higher load and
higher speed engines like turbocharged engines.
---------------------------------------------------------------------------
\791\ Volumetric efficiency (VE) in internal combustion engine
engineering is defined as the ratio of the mass density of the air-
fuel mixture drawn into the cylinder at atmospheric pressure (during
the intake stroke) to the mass density of the same volume of air in
the intake manifold. Ideally, you want this to be high as possible
to maximize thermal efficiency during the power stroke (combustion
phase).
---------------------------------------------------------------------------
As mentioned above, IAV developed engine specifications, including
the rate of internal EGR and cEGR, using variation in combustion
criteria used by industry to ensure the engines would meet durability
and customer acceptability criteria. In addition to reducing pumping
losses, EGR slows the combustion rate and causes combustion to be less
consistent cycle-to-cycle as the concentration increases. Industry and
researchers use a measurement known as coefficient of variation of
indicated mean effective pressure (COV of IMEP) to evaluate combustion
stability. Industry commonly recognizes values greater than 3.0 percent
as unacceptable because above those levels, the combustion instability
creates a noticeable and objectionable drivability problem for vehicle
occupants, referred to as ``surge.'' Surge is perceived as the vehicle
accelerating and decelerating erratically, instead of running smoothly.
IAV set EGR rates at each of the engine operating conditions at the
highest level that did not exceed 3.0 percent COV of IMEP. Therefore,
the IAV engine maps did maximize efficiency within real-world
constraints, similar to how manufacturers develop their engines. At the
lower speed and load conditions of the 2-cycle tests, the COV of IMEP
threshold was reached using internal EGR alone, so additional cEGR was
not applied. At higher load conditions, such as the US06 cycle, cEGR
was applied.
ICCT's statement that the engine maps were only developed
considering knock-abatement is inaccurate. In the PRIA Chapter
6.3.2.2.11, the agencies discussed the application of internal EGR in
combination with cEGR for Eng14. VVT technology, with which Eng14 is
equipped, maximizes EGR usage first in areas where the engine primarily
operates, such as low load and low speed area like city cycle and
highway cycle tests used in CAFE compliance testing. Cooled EGR is
applied at higher speed and higher load conditions, such as the US06
test cycle.
Using EPA's modeled cEGR would have resulted in infeasible engine
maps because they were developed assuming the exclusive use of high
octane Tier 2 fuel, and using a COV of IMEP threshold of 5 percent,
which is beyond the level that is deemed acceptable to consumers in the
real world.\792\ The use of these
[[Page 24407]]
criteria results in engine maps with BSFC levels that cannot be
achieved by manufacturers that must ensure their engines are durable
and are acceptable to customers with fuels that are used and available.
The reference engine for EPA's cEGR concept was a 2010 Ricardo
prototype V6 engine that used 98 RON fuel (93AKI or premium fuel) to
determine effectiveness.\793\ The problems associated with using high
octane Tier 2 to develop engine maps are discussed in detail in Section
VI.C.1.a). The issues associated with excessive cEGR rates and COV of
IMEP, are discussed immediately above. In addition, the cEGR engine
maps that EPA used were never evaluated with regular octane Tier 3 fuel
to assess the further degradation in BSFC and COV of IMEP that would
occur where spark advance would need to be decreased to address spark
knock, as decreasing spark advance directionally makes both BSFC and
COV of IMEP worse.\794\ Also, because some models are still under
development, ALPHA effectiveness estimates in the Draft TAR and derived
for the Proposed Determination do not provide the best available basis
for assessing effectiveness impacts.\795\ Therefore, the assumptions
used for the EPA Draft TAR and Proposed Determination engine maps
overstate feasible improvements and therefore do not provide meaningful
comparisons to the engine maps used for the NPRM and final rule
analyses.
---------------------------------------------------------------------------
\792\ EPA Proposed Determination TSD at 2-295.
\793\ 2016 EPA Technical Support Document at p. 2-312 in section
2.3.4.1.9 Table 2.69. EPA-420-R-16-021, November 2016. Available at
https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100Q3L4.pdf.
\794\ 2016 EPA Technical Support Document at p. 2-312 in section
2.3.4.1.9. EPA-420-R-16-021, November 2016. Available at https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100Q3L4.pdf.
\795\ Advanced Light-Duty Powertrain and Hybrid Analysis (ALPHA)
Tool. Available at https://www.epa.gov/regulations-emissions-vehicles-and-engines/advanced-light-duty-powertrain-and-hybrid-analysis-alpha#v1.0. Version 2.2. Incomplete Models in
ALPHA2.2_TechWalkExamples\Ford Tech Walk\publish_Escape_AWD_matrix.
---------------------------------------------------------------------------
Finally, with regards to Roush's comparison of Eng13 to the 2016
Mazda SkyActiv-G 2.5L Turbo, the agencies believe these engines use
technologies that are sufficiently different so as to render a
comparison not useful, even for a very rough validation of Eng13. Most
fundamentally, as discussed in PRIA Chapter 6.3.2.2.11 and 6.3.2.2.13,
the Mazda 2.5L Turbo is a Miller cycle engine, whereas Eng13 is an Otto
cycle engine. Also, the Mazda 2.5L Turbo has cEGR, whereas Eng13 does
not.\796\ On a more detailed level, as described in PRIA Chapter
6.3.2.2.20.10, Eng13 has a BSFC of 238 g/kwh, whereas Roush refers to
an engine having a BSFC of 250 g/kwh.\797\ The agencies therefore
believe comparing the 2016 Mazda SkyActiv-G 2.5L Turbo to Eng13 is not
a useful or relevant comparison. In the PRIA, the agencies included an
engine map for a Miller cycle engine and requested comments on whether
it should be included in the final rule analysis. Based on the
comments, as discussed further below, the agencies added a Miller cycle
engine to the final rule analysis.
---------------------------------------------------------------------------
\796\ NHTSA Benchmarking, ``Laboratory Testing of a 2016 Mazda
CX9 2.5 I4 with a 6 Speed Transmission.'' DOT HS 812 519.
\797\ NHTSA-2018-0067-11984 at p. 20 of 37 Figure 8.
---------------------------------------------------------------------------
(3) Non-HEV Atkinson Mode Engines
Manufacturers use a variety of designs and technologies to obtain
an engine's highest thermal efficiency while maintaining drivability
and performance. While the Otto cycle has historically been used by the
vast majority of gasoline based engines, one way to improve thermal
efficiency is by using alternative combustion cycles. One such
alternative combustion cycle that can be used in place of the Otto
cycle to achieve a higher maximum thermal efficiency is the Atkinson
cycle. Atkinson cycle operation is achieved by modifying the Otto cycle
engines' crank and valvetrain mechanics to maintain compression ratio
while increasing expansion ratio.798 799 800 Specifically,
in Otto cycle operation, the exhaust valve is opened near the end of
the power stroke, allowing exhaust gases out of the cylinder. The
pressure in the cylinder is still about three to five atmospheres.\801\
Currently, there are two common approaches to achieving Atkinson Cycle
operation: Either the exhaust valve timing or the intake valve timing
are modified. In the first instance, the exhaust valve is not opened
until enough expansion has occurred for the cylinder pressure to be
equivalent to atmospheric pressure. The energy that typically is lost
when the exhaust valve opens in Otto cycle is captured in the Atkinson
cycle, leading to higher thermal efficiency. Modifying the intake valve
timing, the most common way to achieve Atkinson cycle operation,
involves allowing the intake valve to stay open during some portion of
compression stroke. As a result, some of the fresh charge is driven
back into the intake manifold by the raising piston so the cylinder is
never completely filled with air, allowing optimized capture of
combustion-created pressure.
---------------------------------------------------------------------------
\798\ Otto cycle is a four-stroke cycle that has four piston
movements over two engine revolutions for each cycle. First stroke:
Intake or induction; seconds stroke: Compression; third stroke:
Expansion or power stroke; and finally, fourth stroke: Exhaust.
\799\ Compression ratio is the ratio of the maximum to minimum
volume in the cylinder of an internal combustion engine.
\800\ Expansion ratio is the ratio of maximum to minimum volume
in the cylinder of an IC engine when the valves are closed (i.e.,
the piston is traveling from top to bottom to produce work).
\801\ Pulkrabek. W.W. ``Engineering Fundamentals of the Internal
Combustion Engine.'' 2nd edition. Pearson Prentice Hall, at p. 118.
---------------------------------------------------------------------------
While Atkinson cycle engines have higher theoretical thermal
efficiency compared to Otto cycle engines, the Atkinson cycle engine
delivers that higher efficiency at the cost of power density.\802\ The
reduced power density is because of lower operation pressures in the
cylinder than in a typical Otto cycle engine. Accordingly, Atkinson
cycle engines have been ideal for hybrid vehicles because their
electric motor can make up for lost power density.
---------------------------------------------------------------------------
\802\ Power density is the engine power per unit of displacement
(= [Engine Power]/[Engine Displacement]).
---------------------------------------------------------------------------
As vehicle technologies have become more sophisticated,
descriptions of Atkinson cycle engines and Atkinson mode engine
technologies have been used interchangeably, and often incorrectly, in
association with high compression ratio (HCR) engines by the agencies
and stakeholders. Although they both achieve an overall higher thermal
efficiency than Otto cycle-only engines, they differ in execution
depending on engine load. For the following discussion, Atkinson
technologies considered in the analysis can be categorized into three
groups: (1) Atkinson engines, (2) Atkinson-mode engines, and (3)
Atkinson-enabled engines, which are variable valve timing engines with
late intake closing that enables the Atkinson cycle mode. As discussed
earlier, because power density is traded for efficiency, there is a
limit to where Atkinson technology can be applied. While any vehicle
could, theoretically, adopt an Atkinson-mode engine or an engine that
enables operating in Atkinson cycle mode, the difference in vehicle
application (high-performance versus standard-performance vehicles,
towing requirements, trucks) leads to different effectiveness levels.
The range of effectiveness appeared to create confusion among
stakeholders regarding how the technology is applied to vehicles for
compliance modeling and simulation.
Atkinson engines are engines that operate full-time in the Atkinson
cycle. As mentioned above, the most common method of operation used by
Atkinson engines currently is late intake closing.
[[Page 24408]]
This approach allows backflow from the combustion chamber into the
intake manifold, reducing the dynamic compression ratio, but providing
a higher expansion ratio. This improves thermal efficiency but reduces
power density. As a result of limited engine operation, these engines
tend to have lower specific power.\803\ The lower specific power tends
to relegate these engines to hybrid vehicles applications, as coupling
the engines to electric motors can compensate for the lower specific
power. The Toyota Prius is an example of a vehicle that uses an
Atkinson engine. Typically, vehicles that use an Atkinson cycle engine
incorporate various fuel-efficient technologies like aerodynamic
improvements, advanced continuously variable transmissions, mass
reduction, and many other technologies to minimize engine load and
attain high thermal efficiency.\804\ The 2017 Toyota Prius achieved a
peak thermal efficiency of 40 percent.\805\
---------------------------------------------------------------------------
\803\ Specific power is the maximum power produced per
displacement typically in units of hp/L or kw/l.
\804\ Toyota. ``Under the Hood of the All-new Toyota Prius.''
Oct. 13, 2015. Available at https://global.toyota/en/detail/9827044.
Last accessed Nov. 22, 2019.
\805\ Matsuo, S., Ikeda, E., Ito, Y., and Nishiura, H., ``The
New Toyota Inline 4 Cylinder 1.8L ESTEC 2ZR-FXE Gasoline Engine for
Hybrid Car,'' SAE Technical Paper 2016-01-0684, 2016, https://doi.org/10.4271/2016-01-0684.
---------------------------------------------------------------------------
Atkinson-mode engines are engines that use both the Otto cycle and
Atkinson cycle during operation, switching between the modes of
operation based on engine loads. During high loads the engine will
operate in the power-dense Otto cycle mode, while at low loads the
engine will operate in the higher-efficiency Atkinson cycle mode. The
magnitude of efficiency improvement experienced by a vehicle using this
technology is directly related to how much of the vehicle's operation
time is spent in Atkinson mode. This means vehicles that typically
operate at a high load, like a truck towing a trailer, will spend more
time in the Otto mode and less time in the Atkinson cycle mode, and
will achieve a lower overall efficiency improvement over a traditional
Atkinson engine that operates full-time in the Atkinson cycle. As a
result, manufacturers will try to use this type of engine in
conjunction with other technologies that reduce engine load, which
allows the engine to operate more frequently in Atkinson cycle mode.
For example, manufacturers could reduce parasitic losses by
incorporating more efficient accessory technologies, or reducing
overall vehicle mass and aerodynamic drag. These technologies are
enablers for Atkinson-mode engines. When these types of technologies
are adopted, it reduces the parasitic losses and, in turn, reduces the
time the engine is in high load region. An example of an Atkinson-mode
engine is the MY 2017 Mazda 3.
The last type of Atkinson-type engine, the Atkinson-enabled engine,
can be characterized by primarily running the Otto cycle, but can
achieve Atkinson-mode using variable valve timing (VVT) technology.
Some engines use changes in VVT on the intake side to enable Atkinson
cycle operation in low load, low speed operation, like city driving.
These types of engines are typically used in applications that
generally require higher specific power such that it would be
infeasible to use Atkinson-mode engines or Atkinson engines. These
vehicles tend to have higher load demands due to towing requirements,
payload requirements, greater aerodynamic drag from larger frontal
areas, greater tire rolling resistance from larger tires and higher
driveline losses from four-wheel drive or all-wheel drive (e.g., SUVs
and pickup trucks). These higher load demands tend to push these
engines more frequently to the less efficient region of the engine map
and limit the amount of Atkinson operation. An example of the Atkinson-
enabled engine is the Toyota MY 2017 Tacoma 3.5L 6-cylinder engine.
EPA developed two engine maps representing non-hybrid Atkinson
engines to support the 2016 Draft TAR, Proposed Determination, and
first Final Determination.\806\ Referred to as ATK and ATK2, the
engines represented a current non-hybrid Atkinson cycle engine based on
the 2.0L 2014 Mazda SkyActiv-G (ATK) engine, and a future Atkinson
engine concept based on the Mazda engines, but adding cooled EGR,
cylinder deactivation, and an increased compression ratio (14:1)
developed for full vehicle modeling and simulation (ATK2). For the 2016
Draft TAR, the agencies adopted EPA's high compression ratio (HCR)
engine maps as Eng24 and Eng25, which corresponded to HCR1 and HCR2 in
the CAFE modeling.
---------------------------------------------------------------------------
\806\ 2016 LD Draft Technical Assessment Report (TAR), Vehicle
Greenhouse Gas Emission Standards and Corporate Average Fuel Economy
Standards for Model Years 2022-2025; at p. 5-282. Proposed
Determination on the Appropriateness of the Model Year 2022-2025
Light-Duty Vehicle Greenhouse Gas Emissions Standards under the
Midterm Evaluation; pp. 22 & A-7. Final Determination on the
Appropriateness of the Model Year 2022-2025 Light-Duty Vehicle
Greenhouse Gas Emissions Standards under the Midterm Evaluation,
Response to Comments; pp. 29 & 52.
---------------------------------------------------------------------------
The Alliance had provided significant comments on the 2016 Draft
TAR regarding the engine maps for HCR engines.\807\ The Alliance
detailed concerns regarding the feasibility and effectiveness of Eng24
(HCR1) and Eng25 (HCR2). Many of the comments on the 2016 Draft TAR
noted that the modeling projected an implausible rapid fleet
penetration for these technologies, and overestimated effectiveness.
Commenters stated the overestimation was due largely to modeling with
use of high-octane fuel and the addition of other technologies like
cEGR and cylinder deactivation (DEAC) using theoretical assumptions
that exceed the bounds of operation of components. In contrast, other
commenters had stated that EPA's work on the future Atkinson concept
``has shown this pathway to be a promising alternative way to match the
levels of improvement from a 27-bar BMEP turbocharged engine,'' and
that ``it is prudent to assume that the robust body of evidence EPA is
putting together based on benchmarking and modeling data is a
reasonable assessment of the technology's potential.'' \808\
---------------------------------------------------------------------------
\807\ Alliance of Automobile Manufacturers, Alliance of
Automobile Manufacturers Comments on Draft Technical Assessment
Report: Midterm Evaluation of Light-Duty Greenhouse Gas Emission
Standards and Corporate Average Fuel Economy Standards for Model
Years 2022-2025 (EPA-420-D-16-900, July 2016), at 45 (Sept. 26,
2016), Docket ID EPA-HQ-OAR-2015-0827-4089 and NHTSA-2016-0068-0072.
\808\ Union of Concerned Scientists Comments Concerning the
Draft Technical Assessment Report for the Mid-term Evaluation of
Model Year 2022-2025 Light-duty Vehicle Greenhouse Gas Emissions and
Fuel Economy Standards, at 10-11.
---------------------------------------------------------------------------
For the NPRM analysis, the agencies included EPA's engine maps. The
agencies allowed HCR1 to be applied only for a few manufacturers that
indicated they would pursue this technology pathway versus alternative
pathways, such as downsized turbocharged engines. The agencies were
also careful to maintain vehicle performance and utility attributes
when considering the application of Atkinson-type technologies. Current
Atkinson capable engines have incorporated other technologies to reduce
load in order to maximize time in Atkinson operation and to offset the
power loss partially. This includes improved accessories, addition of
friction reduction technologies, and other technologies that reduce
engine load. Although modern improvements to engines have allowed
Atkinson operation to occur more often (because of lower engine loads)
for passenger cars, larger vehicles capable of carrying more cargo and
occupants, and towing larger and heavier trailers, have more limited
potential Atkinson operation. Those
[[Page 24409]]
adoption features are discussed further in Section VI.C.1.e) Adoption
Features, below.
As stated in the NPRM, the agencies excluded the HCR2 concept
engine from the central analysis for several reasons. First, the
concept was not subjected to validation to assess its technical
feasibility. The concept was only modeled with high octane Tier 2 fuel.
The HCR2's capability to operate on regular octane Tier 3 fuel was
assessed using non-cycle specific operation, necessitating adjustments
to the final results to account for Tier 3 fuel properties from Tier 2
operation, instead of simply operating the engine on Tier 3 to generate
effectiveness estimates.\809\ As discussed further above and in Section
VI.C.1.a), fuel octane affects engine durability, performance,
drivability, and noise, vibration and harshness. Assumptions about
compression ratio, EGR rates, and use of cylinder deactivation were not
adequately validated. PRIA Chapter 6.3.2.2.20.18 discussed many
questions about HCR2 technology's practicability as specified,
especially in high load, low engine speed operating conditions. There
also has been no observable physical demonstration of the technology
assumptions. Many manufacturer engine experts questioned its technical
feasibility and commercial practicability during the model years
covered by the rulemaking. Stakeholders like the Alliance had
previously asked for the engine to be removed from the rulemaking
analyses until the performance could be validated with engine
hardware.\810\ For these reasons, the agencies considered the HCR2
engine too speculative to include in the NPRM central analysis.
However, the agencies did provide a sensitivity analysis that included
the HCR2 engine.
---------------------------------------------------------------------------
\809\ EPA PD TSD at 2-210.
\810\ NHTSA-2016-0068-0070 at 45.
---------------------------------------------------------------------------
Comments on HCR1 and HCR2 varied, with commenters split on issues
like whether HCR2 was speculative or real, whether there was technology
in the fleet that could adequately be represented by HCR2, and the
effectiveness of HCR2 in the analysis.
The Alliance commented in support of the decision to exclude HCR2
from the analysis, citing previous comments to the Draft TAR and
proposed determination ``detailing concerns of feasibility and
effectiveness of the non-hybrid Atkinson engine technology packages,
including cooled exhaust gas recirculation (``CEGR'') and cylinder
deactivation.'' \811\ Specifically, the Alliance's comments ``noted
that the modeling projected an implausibly rapid fleet penetration of
this complex engine technology and overestimated its effectiveness, due
largely to modeling with high-octane fuel and the theoretical addition
of CEGR plus cylinder deactivation.'' The Alliance concluded that ``the
inexplicably high benefits ascribed to this theoretical combination of
technologies has not been validated by physical testing.'' Ford
commented that previous assessments had ``over-estimated both the
effectiveness and near-term penetration of advanced Atkinson technology
powertrains,'' stating that ``[t]he effectiveness of the `futured'
Atkinson package (HCR2) that includes cooled exhaust gas recirculation
(CEGR) and cylinder deactivation (DEAC) is excessively high, primarily
due to overly-optimistic efficiencies in the base engine map,
insufficient accounting of CEGR and DEAC integration losses, and no
accounting of the impact of 91RON Tier 3 test fuel. Given the
speculative and optimistic modeling of this technology combination,
Ford supports limiting the use of HCR2 technology to reference only, as
described in the Proposed Rule.'' \812\ Separately, in support of its
overarching comments that the NPRM modeling better reflected reality
over prior regulatory assessments, Toyota commented that the
effectiveness estimates for Atkinson cycle engine technology in the
NPRM may still have been overstated.\813\
---------------------------------------------------------------------------
\811\ NHTSA-2018-0067-12073.
\812\ NHTSA-2018-0067-11928.
\813\ NHTSA-2018-0067-12150.
---------------------------------------------------------------------------
In contrast, CARB, ICCT, Meszler Engineering Services, UCS, and
other stakeholders commented in different respects, with the broad
themes being: (1) That the change in approach towards HCR engines from
the Draft TAR and Proposed Determination to the NPRM was not justified,
was inadequately justified, or was based on justification from the
industry and not the agencies' own independent judgment; (2) that HCR2
as defined by EPA does exist and therefore should be used in the
analysis; and (3) that even if HCR2 technology does not exist exactly
as EPA defined it, other technologies in the fleet provide the same
level of efficiency improvement as HCR2 and therefore it should be used
in the analysis. Many of these commenters stated that if HCR2 had been
allowed in the compliance analysis, as shown in the NPRM sensitivity
analysis allowing HCR2 to be applied, compliance costs would have been
reduced dramatically, ``on par with NHTSA and EPA estimates in the
TAR.'' 814 815
---------------------------------------------------------------------------
\814\ NHTSA-2018-0067-11741.
\815\ NRDC, Attachement2_CAFE Model Tech Issues.pdf. Docket No.
NHTSA-2018-0067-11723, at 7-13. ICCT, Full Comments Summary. Docket
No. NHTSA-2018-0067-117411, at I-2.
---------------------------------------------------------------------------
Specifically, ICCT, CARB, and UCS took issue with the agencies'
description of HCR2 technology as speculative, stating that description
contrasted with how EPA described the technology in prior documents.
ICCT commented that ``in the Draft TAR and Final Determination, EPA
observed the real-world advances toward production vehicles using HCR2
technology, and determined that that technology could be adopted by
automakers during the compliance period.'' \816\ ICCT stated that in
the NPRM, ``without rational explanation, the agencies now describe
this technology as `speculative' and have omitted the technology from
their primary compliance scenarios altogether.'' CARB similarly
commented that ``[t]he fact that the Agencies, especially EPA, make [a
statement that HCR2 is entirely speculative] is genuinely impossible to
credit.'' \817\ In support, all three commenters referenced EPA's
hardware testing of a European Mazda engine,\818\ with ICCT stating
that HCR2 was dismissed as entirely speculative ``despite the careful
benchmarking of improved HCR engines by EPA,'' while CARB and UCS
similarly cited this hardware testing to rebut the Alliance's assertion
that the effectiveness values for HCR2 was ``seriously overestimated.''
---------------------------------------------------------------------------
\816\ NHTSA-2018-0067-11741.
\817\ NHTSA-2018-0067-11873.
\818\ Schenk, C. and Dekraker, P., ``Potential Fuel Economy
Improvements from the Implementation of cEGR and CDA on an Atkinson
Cycle Engine,'' SAE Technical Paper 2017-01-1016, 2017, doi:10.4271/
2017-01-1016.
---------------------------------------------------------------------------
ICCT also took issue with the NPRM statements that ``many engine
experts questioned [HCR2's] technical feasibility and near-term
commercial practicability,'' \819\ and that ``[s]takeholders asked for
the engine to be removed from compliance simulations until the
performance could be validated with engine hardware,'' with references
to comments from Fiat-Chrysler (stating ``Remove ATK2 from OMEGA model
until the performance is validated'' and ``ATK2--High Compression
engines coupled with Cylinder Deactivation and Cooled EGR are unlikely
to deliver modeled results, meet customer needs, or be ready for
commercial application.''),\820\ and comments from the Alliance of
Automobile Manufacturers, stating that
[[Page 24410]]
``[There] is no current example of combined Atkinson, plus cooled EGR,
plus cylinder deactivation technology in the present fleet to verify
EPA's modeled benefits and . . . EPA could not provide physical test
results replicating its modeled benefits of these combined
technologies.'' \821\ ICCT stated that the agencies did not identify
any such comments or evidence from engine experts, or agency analysis
of them. ICCT stated that ``it is clear that NHTSA is deferring to
stakeholders, and that EPA has been forced to defer to NHTSA.''
---------------------------------------------------------------------------
\819\ 83 FR 43038.
\820\ Id. (citing NHTSA-2016-0068-0082).
\821\ Id. (citing EPA-HQ-OAR-2015-0827-6156).
---------------------------------------------------------------------------
ICCT also cited interagency review documents where EPA stated
``[t]here are Atkinson engine vehicles on the road today (2018 [Toyota]
Camry and Corolla with cooled EGR and the 2019 Mazda CX5 and Mazda6
with cylinder deac) that use high geometric compression ratio Atkinson
cycle technology that is improved from the first generation, MY2012
vintage ``HCR1'' technology. While it is true that no production
vehicle has both cooled EGR and cylinder deac, as the EPA ``HCR2''
engine did, nonetheless, these existing engines demonstrate better
efficiency than estimated by EPA. Therefore, it would be appropriate to
continue to use EPA's cooled EGR + deac engine map to represent
``HCR2'' engines.'' \822\
---------------------------------------------------------------------------
\822\ NHTSA-2018-0067-11741, Attachment3_ICCT 15page summary and
full comments appendix, at I-10 (citing Docket Entry: E.O. 12866
Review Materials for The Safer Affordable Fuel-Efficient (SAFE)
Vehicles Rule for Model Years 2021-2026 Passenger Cars and Light
Trucks NPRM, Docket ID EPA-HQ-OAR-2018-0283-0453 (hereinafter
``EO12866 Review Materials''), File:
``EO_12866_Review_EPA_comments_on_the_NPRM_sent_to_OMB,_June_29,_2018
'' at 82, https://www.regulations.gov/document?D=EPA-HQ-OAR-2018-0283-0453).
---------------------------------------------------------------------------
More specifically regarding the technical specifications of the
HCR2 engine, ICCT and others stated that EPA had already addressed
concerns brought by the Alliance \823\ on (1) the base engine fuel
consumption maps used as the foundation of the HCR2 engine map; \824\
(2) practical limitations for cEGR to limit engine knock; \825\ (3) the
reliance on the availability of cylinder deactivation at unrealistic
speed and load operating points; (4) the impact of 91 RON market and
certification test fuels; and (5) the ability to implement HCR2
technology in existing vehicle architectures.\826\
---------------------------------------------------------------------------
\823\ EPA-HQ-OAR-2015-0827-4089; EPA-HQ-OAR-2015-0827-6156.
\824\ NHTSA-2018-0067-11741 (``EPA showed how its ``difference''
engine maps validly represented performance of the ATK2 [HCR2]
packages including on different fuels (pp. 301-02); and that the
difference maps submitted in the industry comment ``provided no
information to compare vintage or application of the actual engine
or engines tested, and did not state whether or not testing was
conducted,'' lacking any information on ``test and/or analytical
methods, assumptions, fuel properties, environment test conditions,
how the engine was controlled or how control was modeled, the number
of data points gathered to generate the AAM `difference map' to
assure that identical testing and a sufficient fit of data was
performed'' (p. 301). In addition, EPA showed that concerns about
knock due to use of cooled exhaust gas recirculation had been
considered and resolved by ignition improvements (p. 302).'').
\825\ NHTSA-2018-0067-12039 (``The agencies appear to have
relied upon the differences between anti-knock properties of Tier 2
and Tier 3 fuels, mistakenly focusing solely on octane while
ignoring ethanol content. . . . this fails to acknowledge the anti-
knock benefit of charge cooling related to ethanol, which more than
compensates for the change in octane. HCR2 therefore should not be
omitted out of concerns around knock.'').
\826\ NHTSA-2018-0067-11741. ICCT stated that EPA had previously
concluded that existing engine architectures were ``well adapted for
[HCR] technology, and well adapted for the emerging next level HCR2
package of technologies, since the foundational technologies of
gasoline direct injection, increased valve phasing authority, higher
compression ratios, and cooled exhaust gas recirculation are already
in widespread use.'' ICCT also commented that ``EPA correctly
observed that there was sufficient lead time to adopt the HCR2
technology before MY2022 and that it could be incorporated without
requiring major vehicle redesigns.''
---------------------------------------------------------------------------
CARB, UCS, and ICCT all stated, in different terms, that even if
HCR2 technology does not exist exactly as EPA defined it, other
technologies that exist in the fleet provide the same level of
efficiency improvement as HCR2, specifically referencing the MY 2018
Toyota Camry engine and various Mazda engines, and claiming that HCR2
should therefore be used in the analysis. Specifically, CARB stated
that these engines ``are already achieving similar efficiency as the
modeled HCR2 package even though they don't have the full complement of
technologies (i.e., CEGR and DEAC) used in the HCR2 package.'' \827\
CARB stated that these engines' ``existence as production engines today
certainly speaks to the feasibility of this technology for modeling
that goes out to 2030MY.'' \828\ Similarly, UCS stated that while the
2018 Toyota Camry engine ``does not have all of the features of the
HCR2 package constructed by EPA, it achieves similar levels of
performance, thus rendering the agencies' rationale for excluding HCR2
moot--this is a production vehicle using Tier 3 fuel which achieves
performance equivalent to HCR2.'' \829\ Similarly, ICCT cited their own
analysis of the 2018 Toyota Camry for the propositions that the package
of technologies on the Camry exceeds the efficiency gains projected by
EPA's OMEGA model, meaning that EPA's projections for the HCR2 engine
might understate its effectiveness, and the early problems with low-end
torque losses associated with Atkinson cycle engines have been
completely solved.\830\ ICCT stated that ``[t]his evaluation of a real
world vehicle that comes close to meeting all of the elements of an
HCR2 engine makes it clear that HCR2 engines are far from a speculative
technology.''
---------------------------------------------------------------------------
\827\ NHTSA-2018-0067-11873.
\828\ NHTSA-2018-0067-11873.
\829\ NHTSA-2018-0067-12039.
\830\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
ICCT and CARB also took issue with the agencies' justification for
not using the HCR2 engine map as a simulation proxy for other new
engine technology, specifically the statement that:
It is important to conduct a thorough evaluation of the actual
new production engines to measure the brake specific fuel
consumption and to characterize the improvements attributable to
friction and thermal efficiency before drawing conclusions. Using
vehicle level data may misrepresent or conflate complex interactions
between a high thermal efficiency engine, engine friction reduction,
accessory load improvements, transmission technologies, mass
reduction, aerodynamics, rolling resistance, and other vehicle
technologies.\831\
---------------------------------------------------------------------------
\831\ 83 FR 43038.
Both commenters also took issue with the agencies' statement that
existing technologies in the NPRM version of the CAFE model could work
together appropriately to represent an HCR1 engine with additional
efficiency improvements.\832\
---------------------------------------------------------------------------
\832\ 83 FR 43038.
---------------------------------------------------------------------------
ICCT stated that the complexity associated with the package of
improvements in the Camry engine was common to all of the technology
packages included in either OMEGA or CAFE modeling, and was neither a
new issue nor an issue that precludes making reasonable engineering
judgments. ICCT stated that the agencies projected efficiency estimates
for other technology packages without engine maps from a production
engine, citing the agencies' approach to modeling ADEAC technology, and
concluded that the purpose of full vehicle simulation modeling is to
project the efficiency impact when several different parts of the
vehicle are simultaneously upgraded. ICCT stated that ``[i]f reasonable
estimates could be made for ADEAC without fully validated engine maps,
there is no reason to exclude other technologies on these grounds,
especially considering the deep expertise by the agencies and their
state-of-the-art technology simulation capabilities with the ALPHA
modeling.'' Similarly, HDS noted that in contrast to the agencies'
exclusion of HCR2 due to
[[Page 24411]]
unresolved issues associated with knock mitigation and cylinder
deactivation, ``the 2018 analysis included Advanced Cylinder De-
activation (ADEAC) which has recently come to market readiness.'' \833\
---------------------------------------------------------------------------
\833\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
Merriam-Webster's dictionary defines speculative as ``involving,
based on, or constituting intellectual speculation,'' and also,
``theoretical rather than demonstrable.'' \834\ To be clear, most
engines maps used in this analysis--IAV engine maps included--are
theoretical, although they are built based on benchmarked engine data,
and additional fuel-economy-improving technologies are added through
modeling and simulation. But that does not mean that these engines are
speculative. Although the IAV engine maps are not meant to model any
manufacturer's particular engine, many, if not all, technology
combinations have been implemented in real-world engines.
---------------------------------------------------------------------------
\834\ Definition of ``speculative,'' https://www.merriam-webster.com/dictionary/speculative.
---------------------------------------------------------------------------
The agencies qualified the HCR2 engine as speculative because ``no
production engine as outlined in the EPA SAE paper has ever been
commercially produced or even produced as a prototype in a lab setting.
Furthermore, the engine map has not been validated with hardware and
bench data, even on a prototype level (as no such engine exists to test
to validate the engine map).'' \835\ It is important to distinguish
theoretical engines maps with technology combinations that have been
proven through real-world testing and operation, from the HCR2 engine
map, that was created using a combination of validated individual
component models, but the resulting engine system model and generated
engine map were not fully validated against actual hardware.
---------------------------------------------------------------------------
\835\ 83 FR 43038.
---------------------------------------------------------------------------
The Alliance and individual automakers have repeatedly provided
comments on agency actions with their assessment of the feasibility of
the HCR2 engine, including comments ICCT referenced, stating the EPA
had addressed concerns brought by the Alliance in the Proposed
Determination Technical Support Document.\836\ The agencies agree with
ICCT that EPA provided responses to comments about HCR2 assumptions and
engine maps in the Technical Support Document, the Proposed
Determination, and the 2017 Final Determination. However, the agencies
considered the matter further after receiving extensive comments on
HCR2 for the NPRM. The agencies have concluded responses did not
directly and fully address the technical concerns raised by the
Alliance. Further, new data and information has become available since
the Proposed and Final Determination that is directly relevant to the
use of EPA's engine maps in this analysis.
---------------------------------------------------------------------------
\836\ Also important to note regarding ICCT's comment, the
Alliance comment cited in the NPRM came from a section of the
Alliance's comments titled, ``EPA's Response to Alliance Comments
Regarding Atkinson Cycle Engine Technology Benefits is Inadequate,''
which seems to suggest that EPA did not address concerns brought by
the Alliance in the Proposed Determination Technical Support
Document.
---------------------------------------------------------------------------
First, it is important to provide background information about
ICCT's comments referencing previous discussions from the TAR, Proposed
Determination and Final Determination. For the 2016 Draft TAR, EPA
initially created the ATK1 and ATK2 engine maps based on the MY 2014
Mazda 2.0L SKYACTIV-G engine. The EPA benchmarked the Mazda engine,
then modeled increasing the efficiency of the Mazda engine map by
simulating the application of additional technologies using GT-Power
models. The Alliance and FCA commented on the 2016 Draft TAR suggesting
the EPA's development of the ATK1 and ATK2 engine maps were flawed
because the maps were developed based on optimistic baseline engine
characterization of the Mazda engine. The Alliance provided evidence of
the flaws in EPA's characterization by comparing EPA's published base
engine data, developed using Tier 2 certification gasoline, to engine
data benchmarked by USCAR. USCAR benchmarked their own Mazda Skyactiv
engine map using a 91 RON fuel. The comparison resulted in the creation
of a ``difference map'' that showed where the two data sets diverged.
The ``difference map'' implied there were areas of significant
divergence, calling into question the data upon which the ATK1 and ATK2
models are based. The EPA responded stating ``[the Alliance] did not
provide data or other information to substantiate its claim that EPA's
engine dynamometer fuel consumption measurements using a MY2014 Mazda
OEM production 2.0L SKYACTIV-G, upon which the ATK2 packages from the
TAR analysis are based, were in any way unrepresentative of this
engine's actual performance.'' \837\ ICCT cited in their NPRM comments
that the EPA's discussion of these ``difference maps'' supported their
statement that ``[i]n fact, in the Technical Support Document for EPA's
Proposed and 2017 Final Determination, EPA addressed all these concerns
brought forth by the Alliance [regarding HCR2] (including the costs and
effectiveness impacts of using regular octane fuel instead of premium
fuel).''
---------------------------------------------------------------------------
\837\ EPA PD TSD at 2-299.
---------------------------------------------------------------------------
It is understandable why ICCT may have thought this discussion
addressed concerns raised about the HCR2 map; however, review of the
Alliance's original Draft TAR comments makes it clear the Alliance's
initial comments addressed the benchmarking of the MY 2014 Mazda 13:1
SKYACTIV-G engine itself. The Alliance's original comments, expressed
concern over the modeled effectiveness of the advanced Atkinson
technology packages because of the baseline engine data used. The
Alliance suggested the effectiveness is likely overestimated due to
multiple flaws in the benchmarking and modeling approaches taken by
EPA. Only the benchmarking is addressed by EPA's response to the
``difference maps,'' not the concerns about modeling approach.
The Alliance's concerns about modeling included the accuracy of the
base engine fuel consumption maps (to the extent the baseline engine
maps were overly optimistic, the modeled ATK maps were optimistic),
limitations for cEGR to mitigate engine knock, limitations of cylinder
deactivation, and the impact of fuels.\838\ After further review, the
agencies determined the Alliance's concerns were not fully addressed,
resulting in a closer review of the ATK model development process.
---------------------------------------------------------------------------
\838\ EPA-HQ-OAR-2015-0827-4089.
---------------------------------------------------------------------------
Review of the engine model development showed the engine map was
generated assuming the use of high octane fuel, and the follow-up
engine dynamometer validation testing also used high octane fuel.\839\
The characterization of the baseline Mazda Skyactiv engine showed 1-3
percent increase in thermal efficiency across a large portion of the
engine map when operated on Tier 2 fuel versus lower octane
fuel.840 841 The increase in engine
[[Page 24412]]
thermal efficiency, caused by the higher octane fuel, is anticipated to
be amplified when applying ATK technologies. ATK technologies increase
efficiency by increasing the pressure in cylinder during combustion;
however, at the same time the increased pressure increases risk of
knock. For more discussion on engine knock, see Section VI.C.1.a).
Ultimately, it is expected that the ATK1 and ATK2 engines would show a
larger improvement in thermal efficiency as a result of being developed
assuming a high-octane fuel versus the 1-3 percent improvement observed
on the baseline Mazda Skyactiv engine.
---------------------------------------------------------------------------
\839\ Ellies, B., Schenk, C., and Dekraker, P., ``Benchmarking
and Hardware-in-the-Loop Operation of a 2014 MAZDA SkyActiv 2.0L
13:1 Compression Ratio Engine,'' SAE Technical Paper 2016-01-1007,
2016, doi:10.4271/2016-01-1007.
\840\ The engine was first run on LEVIII-compliant certification
fuel which has a 7 psi vapor pressure and 88aki. This fuel is
similar to Tier 3 fuel with exception of the vapor pressure which is
required to be 9 psi to meet Tier 3 certification. It was then
tested on Tier 2 certification fuel (93aki) to assess effects of
higher octane fuel on engine operation and efficiency.
\841\ Ellies, B., Schenk, C., and Dekraker, P., ``Benchmarking
and Hardware-in-the-Loop Operation of a 2014 MAZDA SkyActiv 2.0L
13:1 Compression Ratio Engine,'' SAE Technical Paper 2016-01-1007,
2016, doi:10.4271/2016-01-1007.
Schenk, C. and Dekraker, P., ``Potential Fuel Economy
Improvements from the Implementation of cEGR and CDA on an Atkinson
Cycle Engine,'' SAE Technical Paper 2017-01-1016, 2017, doi:10.4271/
2017-01-1016.
---------------------------------------------------------------------------
A further limitation was revealed during the agencies review of the
ATK model development. The limitation was in how COV of IMEP, an
important indicator of combustion stability, was not accounted for
directly in the model. The 0-D/1-D models used for investigating cEGR
effectiveness could not adequately simulate changes to COV of IMEP. To
compensate for the lack of an appropriate model, limits on cEGR were
based on literature values for unrelated engine technologies.\842\ As a
result, there was no direct evaluation of combustion stability while
evaluating the feasibility of the engine concept.
---------------------------------------------------------------------------
\842\ Schenk, C. and Dekraker, P., ``Potential Fuel Economy
Improvements from the Implementation of cEGR and CDA on an Atkinson
Cycle Engine,'' SAE Technical Paper 2017-01-1016, 2017, doi:10.4271/
2017-01-1016.
---------------------------------------------------------------------------
In contrast, for the NPRM and final rule analysis, IAV engines were
optimized using Tier 3 fuel, to balance performance and fuel
consumption. The majority of baseline vehicles are specified to operate
on 87 AKI fuel, therefore lower octane fuel was used to maintain
baseline functionality. The IAV engine maps were all derived from a
consistent baseline engine and were also optimized using a validated
kinetic knock model, and using a COV of IMEP threshold of 3 percent.
These differences in model construction caused an inconsistency
that resulted in unrealistic improvements in fuel economy and
CO2 emissions for the HCR engine technologies, whereas the
IAV engine maps reflect more realistic accounting for the improvements.
The use of high octane fuel and lack of combustion stability modeling
are complimentary issues that have compounded effects when combined.
For example, the use of high octane fuel allows more advanced spark
timing which both increases efficiency and improves combustion
stability, allowing higher cEGR rates before reaching acceptable limits
for drivability. The compound effect is greater than the simply adding
together individual effects, causing a potentially further unrealistic
increase in effectiveness. At a minimum, it is uncertain how using Tier
3 fuel in the HCR2 engine would impact the BSFC of the engine, as there
was no direct evaluation of the feasibility of the engine concept's
ability to operate on regular octane fuel. The cost for the
effectiveness of the HCR2 technology also is inconsistent with the cost
of the effectiveness improvement values for the technologies in the
2015 NAS report.\843\ In considering all of this information, the
agencies, believe the HCR2 engine map overstates the capabilities of
the technology and decided not to use that engine map for the final
rule analysis.
---------------------------------------------------------------------------
\843\ 2015 NAS at p. 90 and 91.
---------------------------------------------------------------------------
However, the agencies believe the HCR1 engine map does reflect
improvements that are representative of the technology in the
rulemaking timeframe. For the final rule, to reflect better the
incremental effectiveness for a low-cost version of HCR technology, the
agencies added the HCR0 engine for the analysis. The specification of
this engine was provided in the NPRM PRIA as Eng22b. Using this engine
improves the estimated incremental effectiveness because the
incremental engine changes from were directly specified for the
modeling. HCR0 is the first engine in the HCR path that a manufacturer
could adopt. Accordingly, the non-HEV Atkinson engine maps used for the
NPRM and final rule central analysis fit into the three defined
categories as follows: (1) Eng26 is an HEV Atkinson Cycle engine; (2)
in the NPRM analysis, Atkinson-mode engines were characterized by Eng24
(HCR1), and for the final rule analysis, Atkinson-mode engines are
characterized by Eng22b (HCR0) and Eng24 (HCR1); and (3) Atkinson-
enabled engines are characterized by the different VVT engine
technologies identified earlier in basic engine discussions and shown
on Table VI-41 and Table VI-42.
Regarding the ability of manufacturers to adapt the engine
architecture to practical use, the agencies see merit in observations
from both manufacturers and other groups. ICCT is correct in their
observation that some production engines have integrated combinations
of the technologies, including SGDI, VVT and cEGR. Furthermore, the
agencies agree with ICCT that an engine could be built integrating all
the technologies represented in the HCR2 engine model. However, the
agencies also agree with the Alliance's comments to the 2016 Draft TAR
that applying all the technologies to an engine that only has some of
the technologies would require a significant redesign of the powertrain
package. The redesign would need to accommodate the new hardware
integration, controls and emissions calibration, OBD development and
other major efforts. As discussed further in Section VI.C.1.e), the
agencies believe these considerations impact how quickly and widely the
technology could be implemented in the rulemaking timeframe.
The agencies also disagree with commenters that the HCR2 engine map
should be used as a proxy for other vehicles in the fleet that achieve
high thermal efficiency. None of the existing vehicles that commenters
cited, like the 2019 Toyota Camry and Corolla with cEGR or the 2019
Mazda CX5 and Mazda 6 with cylinder deactivation, include the same
combination of technologies as the HCR2 engine. Unlike other engine
technologies in the NPRM and the final rule analysis, no engines in the
market or in prototype stages exist that have the combined technology
specifications of the HCR2. Accordingly, there is no production vehicle
that demonstrates the combination of technologies as applied in the
HCR2 engine that (1) is feasible, and (2) can achieve the same
effectiveness as the modeled HCR2 engine. The NPRM highlighted concerns
about using the HCR2 engine map as a proxy for new engine technologies
that achieve high thermal efficiency, specifically that:
It is important to conduct a thorough evaluation of the actual
new production engines to measure the brake specific fuel
consumption and to characterize the improvements attributable to
friction and thermal efficiency before drawing conclusions. Using
vehicle level data may misrepresent or conflate complex interactions
between a high thermal efficiency engine, engine friction reduction,
accessory load improvements, transmission technologies, mass
reduction, aerodynamics, rolling resistance, and other vehicle
technologies.\844\
---------------------------------------------------------------------------
\844\ 83 FR 43038.
The agencies continue to believe this is true, and Toyota's
comments that the Camry improvements were due to more than just the
engine improvements, as discussed further below, provide further
support to this conclusion.
Several commenters cited EPA's SAE paper discussing the use of the
HCR2 engine model and comparing it to the benchmarking of a 2018 Toyota
Camry
[[Page 24413]]
2.5L engine.845 846 The commenters cited the HCR2 engine's
similarities to the Toyota Camry engine as a reason to employ the
technology model broadly across the entire vehicle fleet, including
applying it to pickup trucks such as the Toyota Tacoma. In the paper,
EPA benchmarked a 2018 Toyota Camry 2.5L Atkinson cycle engine equipped
with cEGR. EPA created a full vehicle model (the exemplar vehicle)
based on the benchmarked data for use in the ALPHA modeling tool. The
full vehicle simulation was used to compare the HCR2 engine to the
Camry's 2.5L engine, and showed some similarities. The paper implied
that it is possible to adopt more technologies to the MY 2018 Camry,
like cylinder deactivation, to meet future standards.
---------------------------------------------------------------------------
\845\ Kargul, J., Stuhldreher, M., Barba, D., Schenk, C. et al.,
``Benchmarking a 2018 Toyota Camry 2.5-Liter Atkinson Cycle Engine
with Cooled-EGR,'' SAE Int. J. Adv. & Curr. Prac. in Mobility
1(2):601-638, 2019, https://doi.org/10.4271/2019-01-0249.
\846\ Duleep, K.G., ``Review of the Technology Costs and
Effectiveness Utilizing in the Proposed SAFE Rule,'' Final Report,
H-D Systems, October 2018, at p. 37.
---------------------------------------------------------------------------
This paper, and the comments relying on it--specifically that it
shows that additional technologies can be added to the MY 2018 Camry
engine to meet future standards--were the subject of considerable
debate in the rulemaking docket. Toyota provided supplemental comments
regarding issues Toyota had with the modeling and simulation. These
included a detailed discussion on why HCR2 is not a reasonable model of
the 2018 Toyota Camry engine. Toyota identified other technologies that
contributed to the overall thermal efficiency of the 2018 Camry
compared to previous generation.\847\ Toyota stated that the 2018
Toyota Camry employed numerous technologies like SGDI, cEGR, optimized
intake system, optimized exhaust system, optimized piston design,
laser-cladded valve seats, VVT, engine friction reduction, variable oil
pump, and electric coolant pump, that all contributed to the engine's
improved efficiency over the previous version.\848\
---------------------------------------------------------------------------
\847\ NHTSA-2018-0067-12431. Supplemental Comments of Toyota
Motor North America, Inc. (7/15/19) at 1-2; NHTSA-2018-0067-12376.
Supplemental Comments of Toyota Motor North America, Inc. (3/25/19)
at 1.
\848\ Hakariya, M., Toda, T., and Sakai, M., ``The New Toyota
Inline 4-Cylinder 2.5L Gasoline Engine,'' SAE Technical Paper 2017-
01-1021, 2017, available at https://doi.org/10.4271/2017-01-1021.
---------------------------------------------------------------------------
In addition, Toyota stated:
[T]he 2018 Exemplar Vehicle that is based on the baseline 2018
Toyota Camry was equipped with engine start stop that doesn't exist
on the production vehicle. Cylinder deactivation was added to the
2025 exemplar vehicle as a protentional enhancement. We acknowledged
that adding cylinder deactivation to the Atkinson-cycle engines is
technically possible and would provide some fuel economy benefits.
However, the primary function of cylinder deactivation is to reduce
engine pumping losses which the Atkinson cycle and EGR already
accomplish. The diminishing return on the cylinder deactivation,
Atkinson cycle and EGR are further exaggerated by smaller 4-cylinder
engines.
This assessment aligns with the 2015 NAS committee report that
estimated a 0.7 percent fuel consumption improvement for adoption of
cylinder deactivation for DOHC and SOHC V6 and V8 engines.\849\ The
agencies agree with Toyota and the NAS assessment that applying
cylinder deactivation in small cylinder count engines is subject to
diminishing returns.
---------------------------------------------------------------------------
\849\ 2015 NAS at p. 34.
---------------------------------------------------------------------------
The agencies agree with Toyota that the presence of the advanced
technologies, in addition to the HCR technology, contributed to the
performance of the Camry. The analysis already provides benefits for
the other advanced technologies individually, and risks, if not
ensures, double counting these benefits if the HCR2 model is used (as
discussed above and in VI.B). Likely double counting of technology
effectiveness further supported the agencies' choice not to use the
HCR2 model for the final rule analysis.
The agencies disagree that the approach taken to modeling ADEAC
technology should similarly apply to modeling the HCR2 engine, or that
because ADEAC just recently entered the market and was employed in the
modeling, HCR2 should be as well. As discussed further below, the
effectiveness estimates for ADEAC were based on extensive discussions
with suppliers and manufacturers that provided CBI data, and technical
publications.\850\ The effectiveness estimates provided for ADEAC
represented the effects of applying a single technology, and not a
combined estimate for several technologies applied at once. Moreover,
as commenters noted, ADEAC had recently ``come to market readiness,''
\851\ compared to the HCR2 technology which cannot be found, as
modeled, in the market, or even in prototype form. As discussed
throughout this document, the preferred approach for the NPRM and final
rule was to isolate the effectiveness improvement attributable to
specific technologies and apply those through full vehicle simulations
to capture technology synergies and dis-synergies appropriately.
---------------------------------------------------------------------------
\850\ Eisazadeh-Far, K. and Younkins, M., ``Fuel Economy Gains
through Dynamic-Skip-Fire in Spark Ignition Engines,'' SAE Technical
Paper 2016-01-0672, 2016, doi:10.4271/2016-01-0672.
\851\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
The agencies also disagree with ICCT's comment that the agencies
were simply deferring to stakeholders, or that EPA was simply deferring
to NHTSA regarding the feasibility of the HCR2 engine. It is reasonable
to assume that the automobile manufacturers that belong to the Alliance
employ some engine experts that are qualified to speak on the
feasibility of an engine. Not just one or two manufacturers objected to
the HCR2 engine; the Alliance commented on behalf of its members in
support of the exclusion of the engine from the analysis,\852\ and this
exclusion was further supported by comments from individual automakers
as well. Toyota, the automaker cited by several commenters as closest
to implementing HCR2 technology stated in supplemental comments that
(1) the HCR2 is not representative of its engine technology; \853\ and
(2) Toyota believes there are diminishing returns for implementing the
HCR2 technologies.\854\ The agencies received no comments from
stakeholders that manufacture engines in support of the HCR2
technology's feasibility and potential future adoption.
---------------------------------------------------------------------------
\852\ NHTSA-2018-0067-12073, at 139.
\853\ Comment from Toyota NHTSA-2018-0067-12376 (``While the
agencies' definitions for the different levels of Atkinson
technology seem to have evolved, the 2018 Camry is clearly not
equipped with HCR2 technology.'').
\854\ Comment from Toyota NHTSA-2018-0067-12376 (``advanced
cylinder deactivation has not yet been established when packaged
with an Atkinson-cycle engine. Both technologies play similar roles
in reducing engine pumping losses which can led to diminishing
returns when combined.'').
---------------------------------------------------------------------------
For HCR technology, the agencies carefully considered comments to
the NPRM and the available data, and concluded it is appropriate to
include HCR0 and HCR1 engine models for the final rule analysis. The
engine maps for those technologies provide the best estimates for the
effectiveness of HCR technology relative to the engine maps for the
other engine technologies used for the analysis. The agencies have
reconsidered issues associated with the HCR2 engine models and maps.
The agencies find that significant technical questions and issues
remain and the engine maps very likely overstate the feasible amount of
effectiveness that could be achieved by the represented technologies.
Therefore, HCR2 technology is not included for the final rule analysis.
(4) HEV Atkinson Cycle Engines
Three types of Atkinson technology were discussed in the previous
section.
[[Page 24414]]
HEV Atkinson cycle engines fall in the first category, operating solely
or primarily in Atkinson mode, supported by an electric drive.
Engine map 26 (Eng26) is the model of the HEV/PHEV Atkinson cycle
engine used for the NPRM and final rule analysis. The engine was based
on Argonne's Advanced Mobility Technology Laboratory (AMTL) 2010 Toyota
Prius test data and published literature.\855\ Argonne's AMTL is
continuously involved in research and testing of advanced technologies,
especially in areas of electrification, and has a large existing
database of test data from advanced technology vehicles.\856\ As a
result of Argonne's continued research, a 2017 Toyota Prius was
characterized for an independent project. Argonne updated the HEV
Atkinson cycle engine using the new Prius data to reflect the 41
percent thermal efficiency of the new 2017 system.\857\ The
electrification technology groups that used Eng26 include powersplit
hybrid vehicles (SHEVPS) and plug-in powersplit hybrid vehicles
(PHEV20/50).
---------------------------------------------------------------------------
\855\ ``2010 Toyota Prius.'' http://www.anl.gov/energy-systems/group/downloadable-dynamometer-database/hybrid-electric-vehicles/2010-toyota-prius. Last accessed April, 2018.
\856\ ANL AMTL Downloadable Dynamometer Database (D3). https://www.anl.gov/es/downloadable-dynamometer-database. Last accessed Dec.
05, 2019.
\857\ Carney, D. ``Toyota unveils more new gasoline ICEs with
40% thermal efficiency.'' SAE. April 4, 2018. https://www.sae.org/news/2018/04/toyota-unveils-more-new-gasoline-ices-with-40-thermal-efficiency. Last accessed Dec. 5, 2019.
---------------------------------------------------------------------------
(5) Advanced Cylinder Deactivation Technologies
Advanced cylinder deactivation (ADEAC) systems, also known as
rolling or dynamic cylinder deactivation systems, allow a further
degree of cylinder deactivation than the base DEAC. ADEAC allows the
engine to vary the percentage of cylinders deactivated and the sequence
in which cylinders are deactivated, essentially providing
``displacement on demand'' for low load operations.
ADEAC systems may be integrated into the valvetrains with moderate
modifications on OHV engines. However, while the ADEAC operating
concept remains the same on DOHC engines, the valvetrain hardware
configuration is very different, and application on DOHC engines is
projected to be more costly per cylinder due to the valvetrain
differences.
The agencies discussed assumptions and effectiveness for the ADEAC
package in the NPRM preamble.\858\ The initial review of this
technology was based on a technical publication that used a MY 2010
engine design that had incorporated a SOHC VVT basic engine.\859\ Other
preproduction 8-cylinder OHV prototype vehicles with ADEAC were briefly
evaluated for this analysis, but no production versions of the
technology have been studied.\860\ For ADEAC fuel consumption
effectiveness values, no engine map was available at the time of the
NPRM analysis. Accordingly, the agencies took the effectiveness values
as predicted by full vehicle simulations of a DEAC engine with SGDI,
VVL, and VVT, and added 3 percent or 6 percent respectively for I-4
engines and V-6 or V-8 engines, and cross-referenced CBI data to
quality check this approach.
---------------------------------------------------------------------------
\858\ 83 FR 43038-39.
\859\ Wilcutts, M., Switkes, J., Shost, M., and Tripathi, A.,
``Design and Benefits of Dynamic Skip Fire Strategies for Cylinder
Deactivated Engines,'' SAE Int. J. Engines 6(1):278-288, 2013,
available at https://doi.org/10.4271/2013-01-0359. Eisazadeh-Far, K.
and Younkins, M., ``Fuel Economy Gains through Dynamic-Skip-Fire in
Spark Ignition Engines,'' SAE Technical Paper 2016-01-0672, 2016,
available at https://doi.org/10.4271/2016-01-0672.
\860\ EPA, 2018. ``Benchmarking and Characterization of a Full
Continuous Cylinder Deactivation System.'' Presented at the SAE
World Congress, April 10-12, 2018. Retrieved from https://www.regulations.gov/document?D=EPA-HQOAR-2018-0283-0029.
---------------------------------------------------------------------------
The agencies noted two potential approaches to including advanced
cylinder deactivation in the full-scale Argonne simulation modeling
analysis for the final rule. First, the agencies proposed using IAV
Eng25a, which was developed to capture the maximum benefits of advanced
cylinder deactivation with several constraints that could include
emissions, cold start, NVH, and durability. Second, the agencies
proposed using a technique developed by Argonne in coordination with
NHTSA to split the overall engine data into individual cylinder data
and compute overall torque and the fuel consumption rate by accounting
for whether each cylinder is active or inactive. The agencies sought
comment on using either approach in the final rule analysis to capture
best the benefits of advanced cylinder deactivation.
CARB, ICCT, Meszler Engineering Services, HDS, and UCS provided a
mixed set of comments on numerous aspects of ADEAC in the NPRM
analysis.\861\ Broadly, HDS commented on a need to describe ADEAC
technology better: ``The 2018 analysis also utilized Advanced Cylinder
Deactivation in its analysis but the package components were not
completely explained in the PRIA.'' \862\ Other stakeholders provided
comments on ADEAC adoption features, effectiveness, and cost, which are
discussed below.
---------------------------------------------------------------------------
\861\ ICCT Docket # NHTSA-2018-0067-11741 at I-12, Duleep Docket
# NHTSA-2018-0067-11873 at 108, Meszler Docket # NHTSA-2018-0067-
11723 at p. 26.
\862\ Duleep, K.G., ``Review of the Technology Costs and
Effectiveness Utilizing in the Proposed SAFE Rule,'' Final Report,
H-D Systems, October 2018, at p. 17.
---------------------------------------------------------------------------
The agencies discussed assumptions and effectiveness for the ADEAC
package in the NPRM preamble.\863\ The initial review of this
technology was based on a technical publication that used a MY 2010
engine design incorporating SOHC and VVT.\864\ After determining the
MY2010 engine design was not representative of the analysis fleet, the
agencies used effectiveness values based on CBI data. The MY2017
baseline fleet reflects technology updates such as SGDI and DEAC that
could adopt ADEAC incrementally in the final rule analysis. The cost
and effectiveness for ADEAC reflects the baseline engine. The 2015 NAS
Committee estimated an 0.7 percent fuel consumption improvement for
adoption of cylinder deactivation for V6s and V8s
engines.865 866
---------------------------------------------------------------------------
\863\ 83 FR 43038-39.
\864\ Wilcutts, M., Switkes, J., Shost, M., and Tripathi, A.,
``Design and Benefits of Dynamic Skip Fire Strategies for Cylinder
Deactivated Engines,'' SAE Int. J. Engines 6(1):278-288, 2013,
available at https://doi.org/10.4271/2013-01-0359. Eisazadeh-Far, K.
and Younkins, M., ``Fuel Economy Gains through Dynamic-Skip-Fire in
Spark Ignition Engines,'' SAE Technical Paper 2016-01-0672, 2016,
available at https://doi.org/10.4271/2016-01-0672.
\865\ Applied after VVT and VVL.
\866\ Applied before VVT and VVL.
---------------------------------------------------------------------------
The agencies requested comments on alternative methods to estimate
ADEAC effectiveness but received no comments regarding either approach
mentioned in the NPRM. For the final rule analysis, the agencies used
effectiveness values as predicted by full vehicle simulations of a DEAC
engine with SGDI, VVL, and VVT, and added 3 percent or 6 percent
respectively for I-4 engines and V-6 or V-8 engines for the naturally
aspirated engines. Effectiveness for turbocharged engines used 1.5
percent and 3 percent values, as predicted by full vehicle simulation
of a TURBOD engine for I4 and V6/V8, respectively. Without sufficient
data to simulate ADEAC, both the IAV and Argonne methodologies
described in the NPRM provided questionable estimates for ADEAC. These
errors would have propagated across other technology combinations in
the analysis. The estimates used for ADEAC and TURBOD for the final
rule analysis are also in line with EPA
[[Page 24415]]
estimates discussed in their SAE technical publications.\867\
---------------------------------------------------------------------------
\867\ Kargul, J., Stuhldreher, M., Barba, D., Schenk, C. et al.,
``Benchmarking a 2018 Toyota Camry 2.5-Liter Atkinson Cycle Engine
with Cooled-EGR,'' SAE Int. J. Adv. & Curr. Prac. in Mobility
1(2):601-638, 2019, https://doi.org/10.4271/2019-01-0249 at pp. 19-
21.
---------------------------------------------------------------------------
For the final rule analysis, the agencies used the same
effectiveness values for ADEAC applied to naturally aspirated engines
as in the NPRM, and incorporated estimated effectiveness values for
TURBOAD to represent ADEAC on downsized turbocharged engines.
(6) Miller Cycle Engines
In the proposed rule, the agencies provided two engine maps
representative of Miller cycle and Eboost engines with 48V battery
systems. The Miller cycle engine (Eng23b) and Miller cycle engine with
Eboost (Eng23c) specifications were provided in the PRIA but were not
used in the NPRM analysis,\868\ although the agencies sought comment on
the specifications used for the modeling.
---------------------------------------------------------------------------
\868\ NPRM PRIA at p. 307-09.
---------------------------------------------------------------------------
Roush on behalf of CARB, ICCT, Meszler Engineering on behalf NRDC,
HDS, and UCS, commented that the agencies did not consider the
combination of turbocharging and Miller cycle.\869\ Specifically, Roush
argued that the agencies' omission of an engine that utilizes a
combination of turbocharging and Miller cycle was unreasonable because
it is already in production, specifically on the VW 2.0L EA888 Gen3B--
DI. Roush stated this omission would limit the effectiveness for
turbocharged engines and cause the adoption of more expensive
solutions, thereby overstating the cost to achieve target fuel economy
levels. Similarly, Roush pointed to the omission of an engine that uses
a variable geometry turbocharger as an error in the agencies' vehicle
modeling; Roush pointed to VW's EA211 TSI Evo engine available in
Europe in 2017 as an example of an engine in production that enables
cost-effective Miller cycle applications.
---------------------------------------------------------------------------
\869\ NHTSA-2018-0067-11985. HD systems at p, 34; ICCT at p.
102; NRDC Attachment 2 at p.16.
---------------------------------------------------------------------------
In response to these comments, the agencies added and used both
Miller cycle-type engines and Miller cycle engines with electric assist
for the final rule analysis. Discussed earlier in this section, the
agencies developed engine maps for additional combinations of
technologies for the final rule, including engine maps that became
available after the NPRM analysis was completed but before the NPRM was
published. For the final rule analysis, the agencies have included a
Miller cycle engine, Eng23b (VTG), as another available engine
technology. The specification of this engine was discussed in PRIA
Chapter 6.3.2.2.20.20.2.2 and the costs are based on the 2015 NAS
estimates for this technology.
(7) Variable Compression Ratio Engines
Variable compression ratio (VCR) engines work by changing the
length of the piston stroke of the engine to operate at a more optimal
compression ratio and improve thermal efficiency over the full range of
engine operating conditions. Engines using VCR technology are currently
in production, but appear to be targeted primarily towards limited
production, high performance and very high BMEP (27-30 bar)
applications.
A few manufacturers and suppliers provided information about VCR
technologies, and several design concepts were reviewed that could
achieve a similar functional outcome. In addition to design concept
differences, intellectual property ownership complicates the ability of
the agencies to define a VCR hardware system that could be widely
adopted across the industry.
For the NPRM analysis, the agencies provided specifications of a
VCR engine (Eng26a) in the PRIA for review and comment.\870\ However
the VCR engine was not used in the NPRM analysis.
---------------------------------------------------------------------------
\870\ NPRM PRIA at pp. 304-06.
---------------------------------------------------------------------------
The Alliance commented in support of the exclusion of variable
compression ratio engines from the analysis, stating that the
technology is still in early development, and too speculative to be
included at this time. The Alliance also stated that the technology is
unlikely to attain significant penetration in the MY 2026 timeframe due
to intellectual property protection associated with early
implementations and its likely application primarily to high-
performance vehicles. The Alliance also cited the technology's price as
a potential barrier to adoption.\871\ Similarly, Ford commented that:
---------------------------------------------------------------------------
\871\ NHTSA-2018-0067-12073 (``At least one source also
indicates a steep price to this technology--``at least $3,000 more
to produce than a standard 16-valve double-overhead-camshaft four-
cylinder.'').
[VCR technology] is likely to be adopted only for premium/
limited-market vehicles in the near future. We also agree that
intellectual property protections on early implementations will
further inhibit significant fleet penetration. Incorporation of VCR
requires a new or highly modified engine architecture, necessitating
major investment from both the engineering and manufacturing
standpoints. Sharing/commonality across engine families would be
greatly limited.'' 872 873
---------------------------------------------------------------------------
\872\ NHTSA-2018-0067-11928.
\873\ NHTSA-2018-0067-11928 at p. 9.
Similarly, other automakers commented on a confidential basis that
several main hurdles prevented them from employing VCR engines,
including the complexity of VCR engines and the associated cost of
those complex parts.
UCS commented that the agencies did not consider VCR engine
technologies in the NPRM analysis.\874\ They stated that the technology
was not modeled, nor was it incorporated into the CAFE model. UCS
argued that Nissan's VC-Turbo engine is part of a strategy to improve
fuel efficiency for Nissan's luxury vehicles by 30-35 percent over
previous models, which would be enough to exceed the vehicle's
regulatory targets without any credits. UCS concluded that given VCR
technology is being put into production in a high-volume vehicle, there
is no reason for the agencies to exclude its adoption.
---------------------------------------------------------------------------
\874\ NHTSA-2018-0067-12039 at p. 6.
---------------------------------------------------------------------------
The agencies agreed with comments to include VCR engine
technologies in the final rule analysis and on further technical
consideration, the agencies have added a VCR engine to the engine
technologies list manufacturers could adopt. However, the agencies
limited the adoption of the VCR engine technology to Nissan only. VCR
engines are complex, costly by design, and synergetic with mainstream
technologies like downsize turbocharging, making it unlikely that a
manufacturer that has already started down an incongruent technology
path would adopt VCR technology.
(8) Diesel Engines
Diesel engines have several characteristics that result in superior
fuel efficiency over traditional gasoline engines, including reduced
pumping losses due to lack of (or greatly reduced) throttling, high
pressure direct injection of fuel, a combustion cycle that operates at
a higher compression ratio, and a very lean air/fuel mixture relative
to an equivalent-performance gasoline engine.\875\ However, diesel
technologies requires additional enablers, such as a NOX
adsorption catalyst system or a urea/ammonia selective catalytic
reduction system, for control of NOX emissions.
---------------------------------------------------------------------------
\875\ Diesel cycle is also a four-stroke cycle like the Otto
Cycle, except in the Intake stroke no fuel is injected and fuel is
injected late in the compression stroke at higher pressure and
temperature.
---------------------------------------------------------------------------
For the NPRM, the agencies modeled one diesel engine, represented
by
[[Page 24416]]
Eng17,\876\ which was termed ``ADSL'' in the CAFE modeling. DSLI, a
more advanced diesel engine, was modeled using a 4.5 percent fixed
effectiveness improvements over ADSL.
---------------------------------------------------------------------------
\876\ Docket ID NHTSA-2018-0067-1972. NPRM PRIA at p. 295.
---------------------------------------------------------------------------
CARB commented that diesel technologies are essentially locked out
of being selected in the CAFE model because of the high cost.\877\ They
state that diesel technology is only selected in rare instances.
---------------------------------------------------------------------------
\877\ Docket ID NHTSA-2018-0067-11873. CARB at 108.
---------------------------------------------------------------------------
The agencies agree that diesel technology is rarely selected. The
technologies required to meet diesel emissions standards are costlier
compared to gasoline technologies, particularly in the rulemaking
timeframe. For example, the 2015 NAS report determined that in the
current market, ``vehicles with diesel engines are priced an average of
more than $4,000 more than comparably equipped gasoline vehicles.''
\878\ Furthermore, the NAS report stated that the ``Carbon Penalty''
makes it harder for manufactures to meet CO2 standards
because of the higher carbon density in the diesel fuel compared to
gasoline that results in higher CO2 per gallon.\879\ In
addition, the market for diesel vehicles has stagnated at around 1
percent for many years after it peaked at 5.9 percent in 1981,
according to the EPA Trends Report.\880\ The agencies believe that the
modeled cost of diesel engines appropriately prevents their widespread
adoption in the analysis.
---------------------------------------------------------------------------
\878\ 2015 NAS at 123-24.
\879\ 2015 NAS Findings 3.3 and 3.4 at p. 120.
\880\ EPA, ``The 2018 EPA Automotive Trends Report.'' March
2019. EPA-420-R-19-002. https://nepis.epa.gov/Exe/ZyPDF.cgi/P100W5C2.PDF?Dockey=P100W5C2.PDF at pp. 5 & 6. Last accessed
December 16, 2019.
---------------------------------------------------------------------------
UCS commented that the agencies restricted cylinder deactivation
technologies to only naturally aspirated gasoline engines.\881\ In
response to this and other comments, the agencies have allowed diesel
engines to adopt ADEAC for this final rule analysis. These engines were
designated as DSLIAD to represent diesel engines with ADEAC, and were
modeled using a 7.5 percent fixed effectiveness improvement on top of
DSLI. This effectiveness improvement of ADEAC on diesel engines is
based on the review of technical publications discussed earlier in
Section VI.C.1.c)(5).
---------------------------------------------------------------------------
\881\ Docket ID NHTSA-2018-0067-12039, at p. 3.
---------------------------------------------------------------------------
(9) Alternative Fuel Engines
CNG engines use compressed natural gas as a fuel source. The fuel
storage and supply systems for these engines differ tremendously from
gasoline, diesel, and flex fuel vehicles. CNG engines were a baseline-
only technology and were not applied to any vehicle that was not
already CNG-based in NHTSA's analysis, per EPCA/EISA's restrictions on
considering dedicated alternative fueled vehicles to set fuel economy
standards.882 883 However, for the EPA program the agencies
allowed any vehicle to adopt CNG engines. The NPRM MY 2016 analysis
fleet did not include any dedicated CNG vehicles to simulate in the
CAFE Model.
---------------------------------------------------------------------------
\882\ NHTSA's provisions for dedicated alternative fuel vehicles
in 49 U.S.C. 32905(a) state that the fuel economy of any dedicated
automobile manufactured after 1992 shall be measured based on the
fuel content of the alternative fuel used to operate the automobile.
A gallon of liquid alternative fuel used to operate a dedicated
automobile is deemed to contain 0.15 gallon of fuel. Under EPCA, for
dedicated alternative fuel vehicles, there are no limits or phase-
out for this special fuel economy calculation, unlike for duel-
fueled vehicles, as discussed below.
\883\ EPA's provisions for dedicated alternative fuel vehicles
that are able to run on compressed natural gas (CNG) currently are
eligible for an advanced technology multiplier credit for MYs 2017-
2021.
---------------------------------------------------------------------------
In addition, for the NPRM and this final rule analysis, NHTSA
modified the CAFE model to include the specific provisions related to
AFVs under the CO2 standards. In particular, the CAFE model
now carries a full representation of the production multipliers related
to electric vehicles, fuel cell vehicles, plug-in hybrids, and CNG
vehicles, all of which vary by year through MY 2021.
(10) Emerging Gasoline Engine Technologies
Manufacturers, suppliers, and researchers continue to create a
diverse set of fuel economy technologies, some of which are still in
the early stages of the development and commercialization process. Due
to uncertainties in the cost and capabilities of emerging technologies,
some new and pre-production technologies are not a part of the CAFE
model simulation. As discussed throughout this section and in VI.B.3,
the agencies declined to include technologies in the analysis where the
agencies did not believe those technologies would be feasible in the
rulemaking timeframe, or the agencies did not have appropriate data
upon which to generate an estimate of how effective the technology is
that could be applied across the ten vehicle classes. Evaluating and
benchmarking promising fuel economy technologies as they enter
production-intent stages of development continues to be a priority as
commercial development matures.
UCS and ICCT commented that the agencies should consider novel
engine designs.\884\ Specifically, ICCT stated that the agencies should
consider a more advanced HCR technology called HCCI (similar to Mazda's
Skyactiv-X) by estimating efficiency and cost to EPA's process that
assigned effectiveness estimates using LPM. They stated that ``the
agencies developed estimates for ADEAC in the NPRM and the associated
modeling even without conclusive and independently verifiable
effectiveness.''
---------------------------------------------------------------------------
\884\ ICCT, Full Comments Summary. Docket No. NHTSA-2018-0067-
117411, at I-17 to I-19.
UCS, Comment. Docket No. NHTSA-2018-0067-12039, at pp. 6 & 7.
---------------------------------------------------------------------------
In response to comments, a number of technologies were added for
the final rule analysis, and adoption features were refined
accordingly, as discussed further in Section VI.C.1.e). New engine
technologies and combinations include Atkinson engine technology
allowed with P2 HEV, new high compression ratio engine (HCR0), variable
compression ratio engine, variable geometry turbo engine, variable
geometry turbo with electric assist engine, diesel with advanced
cylinder deactivation engine, turbo with cylinder deactivation engine,
diesel with manual transmission, diesel with start-stop, and PHEV-turbo
with 20 mile range, and PHEV-turbo with 50 mile range.
The agencies also disagree with ICCT's comment that because ADEAC
was developed without ``conclusive and independently verifiable
effectiveness'' estimates, and as such the agencies should allow HCCI
technology as well. First, conclusive estimates for ADEAC effectiveness
were based on CBI data from both manufacturers and suppliers, technical
publications, and engineering judgement. The references can be reviewed
in the previous Section VI.C.1.c)(5) Advanced Cylinder Deactivation
Technologies. In addition, the agencies benchmarked the first prototype
vehicle equipped with skip-fire, and discussed potential application of
it for other engines. A similar level of data has not been made
available for HCCI engine technologies.
The agencies also believe that the technology associated with Mazda
SkyActiv-X has been mischaracterized by ICCT and other commenters, and
declined to include a specific representation of the SkyActiv-X family
of technologies in the analysis for two reasons. The engine known as
Skyactiv-X is characterized by Mazda as a unique spark plug controlled
compression ignition (SPCCI) technology, 2-liter displacement, 4-
cylinder engine with mechanical compression ratio of 16.3:1 operating
on 95 RON fuel (91 AKI) with
[[Page 24417]]
a mild hybrid system.\885\ The NPRM and this final rule analysis may
not have the exact technology combination associated with this vehicle,
but the analysis does include technologies that are representative of
them, that could enable the benefits employed by the Mazda engine. A
mild hybrid system is available for adoption in both the NPRM and this
final rule analysis.
---------------------------------------------------------------------------
\885\ Mazda Press Release. ``Revolutionary Mazda Skyactiv-x
engine details confirmed sales start.'' May 6, 2019. https://www.mazda-press.com/eu/news/2019/revolutionary-mazda-skyactiv-x-engine-details-confirmed-as-sales-start/. Last accessed Dec, 11,
2019.
---------------------------------------------------------------------------
Also, the effectiveness associated with this engine was from
European test cycles and cannot be compared for U.S. application.
European compliance tests are significantly different than those in the
U.S., especially when it comes to fuel type and test cycles. Any
effectiveness data provided for this engine or any non-U.S. engine
cannot be used for U.S. vehicle application without an adjustment for
fuel and emissions. For example, the higher-octane fuel used in Europe
enables engines to operate at higher compression ratios across wider
areas of engine operation.
The agencies further believe that with the technology additions for
the final rule discussed in previous sections, the analysis reasonably
represents the suite of engine technologies that could be available in
the rulemaking time frame. Manufacturers, suppliers, and researchers
continue to create a diverse set of fuel economy technologies. However,
due to the uncertainties in the cost, manufacturing, and intellectual
property concerns like those identified by commenters, the agencies did
not consider prototype technologies in the final rule analysis.
(11) Engine Lubrication and Friction Reduction Technologies
Manufacturers have already widely adopted both lubrication and
friction reduction technologies. Previous agency analysis considered
these improvements in combination as Improved Low Friction Lubricants
and Engine Friction Reduction (LUBEFR). The NPRM analysis included
advanced engine maps that already assume application of low-friction
lubricants and engine friction reduction technologies, and therefore
additional levels of friction reduction were not considered. Low-
friction lubricants including low viscosity and advanced low-friction
lubricant oils are now available, and widely used. Manufacturers may
make engine changes and conduct durability testing to accommodate the
lubricants. The level of low-friction lubricants exceeded 85 percent
penetration in the MY 2016 fleet.\886\ Reduction of engine friction can
be achieved through low-tension piston rings, roller cam followers,
improved material coatings, more optimal thermal management, piston
surface treatments, and other improvements in the design of engine
components and subsystems that improve efficient engine operation.
---------------------------------------------------------------------------
\886\ NPRM CAFE Model Market Data file.
---------------------------------------------------------------------------
Meszler Engineering on behalf of NRDC commented that ``the NPRM
CAFE model no longer considers advanced lubricants and evolutionary
friction reduction (LUBEFR) to be adoptable. As a result, no fuel
efficiency improvement credits are available. Engine friction reduction
is an ongoing evolutionary process that should generate benefits on the
order of 5 percent or so increase in fuel economy over a multiyear
forecast period, with costs totaling approximately $100. Moreover, the
technology is a benefit of ongoing industry research and evolutionary
engine improvements so that it is easily `adoptable' and deployed
throughout the fleet. Accordingly, NHTSA should revise the NPRM CAFE
model to reinstate the ability to adopt evolutionary friction reduction
technology.'' \887\
---------------------------------------------------------------------------
\887\ Meszler Engineering. Docket ID NHTSA-2018-0067-11723, at
p. 32.
---------------------------------------------------------------------------
The agencies disagree with Meszler that a five percent fuel economy
improvement attributable to lubricants and evolutionary friction
reduction is continuously feasible. The MY 2017 baseline vehicles have
incorporated many technologies like low viscosity engine oil,
integrated exhaust manifold for faster oil warmup, and internal
component friction reduction.\888\ \889\ \890\ The LUB and EFR
technologies are a legacy of the existing rulemaking work going back to
the 2010 CAFE and CO2 rule for MY 2012 to MY 2016.\891\ The
agencies believe that many of these technologies have been incorporated
in many of the engines in the baseline fleet, and therefore the engine
maps used for the NPRM and final rule analysis incorporated them as
well. Furthermore, manufactures have raised concerns over issues with
further decreasing oil viscosity; specifically, manufacturers have
articulated concerns that damage caused by low speed pre-ignition
(LSPI) \892\ can damage an engine.\893\ \894\ \895\
---------------------------------------------------------------------------
\888\ Wards Auto. ``Infiniti's Brilliantly Downsized V-6 Turbo
Shines.'' July 11, 2017. Available at https://www.wardsauto.com/print/engines/infiniti-s-brilliantly-downsized-v-6-turbo-shines.
Last accessed Dec. 11, 2019. Nissan Motor Corp. ``Mirror Bore
Coating.'' Available at https://www.nissan-global.com/EN/TECHNOLOGY/OVERVIEW/mirror_bore_coating.html. Last accessed Dec 11, 2019.
\889\ Toyota's 2AR-FE I4 and 2GR-FE V6 use 0-W20.
\890\ Audi Media Center. ``Efficiency and driving pleasure:
innovative V engines at Audi.'' Available at https://www.audi-mediacenter.com/en/techday-on-combustion-engine-technology-8738/efficiency-and-driving-pleasure-innovative-v-engines-at-audi-8748.
Last accessed Dec.11, 2019.
\891\ 75 FR 25373.
\892\ LSPI is an abnormal combustion event in which the fuel-air
mixture ignites before intended, causing excessive pressures inside
the engine's cylinders. In mild cases, this can cause engine noise,
but when severe enough, LSPI can cause engine damage. There are
several factors that contribute to LSPI, of which lubricating oil
has been observed to be one.
\893\ Motor Magazine. ``Will ILSAC GF-6 Ever Be Approved?'' Nov,
20, 2018. Available at http://newsletter.motor.com/2018/20181120/!ID_Infineum_ILSAC_GF-6.html. Last accessed Dec 11, 2019.
\894\ Chevron. ``Low Speed Pre-ignition.'' Available at https://www.oronite.com/about/news/low-speed-pre-ignition.aspx. Last
accessed Dec. 11, 2019.
\895\ Elliott, I., Sztenderowicz, M., Sinha, K., Takeuchi, Y. et
al., ``Understanding Low Speed Pre-Ignition Phenomena across Turbo-
Charged GDI Engines and Impact on Future Engine Oil Design.'' SAE
Technical Paper 2015-01-2028, 2015, available at https://doi.org/10.4271/2015-01-2028.
---------------------------------------------------------------------------
In response to the comment that engine friction reduction
technology is evolutionary technology, the agencies introduced one
level of friction reduction (EFR) for the final rule analysis. The
agencies estimated a 1.4 percent effectiveness for this type of
technology based on the 2015 NAS report assessment of further
improvements in lubrication and friction.\896\
---------------------------------------------------------------------------
\896\ 2015 NAS at pp. 28 & 29.
---------------------------------------------------------------------------
d) How the Agencies Assign Engine Technologies to the Baseline Fleet
[[Page 24418]]
Manufacturers have made significant improvements in fuel economy
and CO2 emissions reductions since the MY 2012 rulemaking
analysis.\897\ \898\ The agencies expended substantial effort to update
the analysis fleet from the MY 2016 representative fleet used for the
NPRM to a MY 2017 analysis fleet used for this final rulemaking to
capture the technologies manufacturers have used to increase their
fleet's fuel economy and CO2 emissions performance. Detailed
discussion of the model year 2017 fleet development and application can
be found in VI.B.1. The agencies extensively updated the new MY 2017
fleet engine technologies using available manufacturer final model year
CAFE compliance submissions to the agencies, as well as manufacturer
press release specifications, agency-sponsored vehicle benchmarking
studies, review of available technical publications, and through
manufacturer CBI.\899\
---------------------------------------------------------------------------
\897\ EPA. ``2018 EPA Automotive Trends Report'' 12 pp, 421 K,
EPA-420-S-19-001, March 2019. https://www.epa.gov/automotive-trends/download-automotive-trends-report#Full%20Report last accessed Feb.
12, 2020
\898\ FOTW #1108, Nov 18, 2019: Fuel Economy Guide Shows the
Number of Conventional Gasoline Vehicle Models Achieving 45 miles
per gallon or Greater is Increasing. DOE VTO. Available at https://www.energy.gov/eere/vehicles/articles/fotw-1108-november-18-2019-fuel-economy-guide-shows-number-conventional. Last accessed Nov 18,
2019.
\899\ NPRM CAFE Market Data file.
---------------------------------------------------------------------------
The data for each manufacturer was used to determine which
platforms shared engines and to establish the leader-follower
relationships between vehicles. Within each manufacturer's fleet,
engines were assigned unique identification designations based on
configuration, and technologies applied, along with other
characteristics. The data were also used to identify the most similar
engine among the IAV engine maps, as discussed in Section VI.C.1.
Just like the real-world vehicle variants, the CAFE model considers
differences between each vehicle like base performance and higher
performance levels. For example, the 2017 Ford F150 has many variants
with different types of engines like the 2.7L turbocharged V6, 3.3L
naturally-aspirated V6, 3.5L turbocharged V6, and 5L naturally-
aspirated V8. In contrast to the LPM, the CAFE model rosters each
variant level and powertrain application individually. This variation
is accounted for as engine technologies are assigned in the analysis
fleet.
As a result of new information available since publication of the
NPRM and comments received to the NPRM, the agencies included
additional engine technologies in the compliance analysis, expanding
the total number of engine technologies available from 16 to 23. This
expansion is a direct result of comments received to the NPRM and
further enables the agencies' capabilities to accurately and,
realistically, characterize the technologies present on an engine found
in the analysis fleet. This collection of technologies represents the
best available information the agencies have, at the time of this
action, regarding both currently available engine technologies and
engine technologies that could be feasible for application to the U.S.
fleet during the rulemaking timeframe. The agencies believe this effort
has yielded the most technology-rich and accurate analysis fleet
utilized by the CAFE model to date.
In some cases, however, it was necessary for the agencies to
substitute an engine map that closely represented an engine technology
that were effectively the same, or, based on engineering judgement,
were the best available proxy at the time of the analysis. For example,
many manufacturers offer their own proprietary VVT engine technologies
and so the agencies assigned the same engine map for all of these VVT
in the baseline fleet. The CAFE model uses compliance CAFE and
CO2 values for baseline vehicles and so it's not as relevant
to have exact technology assignment type as it more important to
provide the advanced vehicle have adopted to date. For further
discussion of this see section VI.A.3 Fuel-Savings Technologies. This
substitution was necessary, in some cases, where an ``exact-match''
engine map was not available for application to a specific vehicle and/
or vehicle specific engine application. The agencies leveraged a series
of engine operating characteristic maps developed by industry suppliers
and, in some cases, the agencies themselves, to assign the closest
baseline engine map for the analysis.
As discussed in Section VI.C.1.b), these engine maps provide
operational characteristics such as horsepower, torque, or efficiency
at a specified point in an engine's operational range. These
operational maps are developed based on a given set of engine
characteristics and technologies applied to that engine. Engine maps
are closely held by vehicle manufacturers and are typically considered
intellectual property. As such, vehicle manufacturers are not typically
willing provide the operational maps to the agencies, where it would
ultimately be in the purview of competitors. In some instances,
manufacturer engine maps are published in media such as technical
papers or conference presentation materials. However, these publicly
available engine maps are, in nearly all instances, void of critical
information that would enable their use for meaningful simulation and
modeling.
Therefore, the agencies are generally limited to the catalog of
engine maps they have developed through contracts and, where possible,
in-house which, in turn, yields the need for sound, engineering
judgement-based substitution of an engine map as a proxy for an engine
application in the marketplace. Unfortunately, this is necessary as the
agencies are unable to fund the development of engines maps for every
possible engine and technology combination available for sale. However,
it is important to note the agencies do have a substantial catalog of
engine maps to leverage and continue to fund the development of new
maps as new technologies enter the marketplace. Additional information
on the agencies' catalog of engine maps used for this this final
rulemaking can be found in Section VI.C.1.b).
Some engine technologies are designated in the CAFE Model as
``baseline only'' technologies, meaning these are characteristics such
as engine configuration, architecture, or a technology that is
considered inherent to the fleet for the given model year, an example
for the MY 2017 fleet used in this analysis is variable-valve-timing
(VVT). Beyond the aforementioned configurations and technology, engine
technologies that can be applied to a future engine and, eventually, to
a vehicle in the compliance modeling are only available at a vehicle
redesign. As such, a vehicle will only adopt a new engine according to
the application schedule defined as a CAFE model input.
e) Engine Adoption Features
Engine adoption features are defined through mechanisms like
technology path logic or the application of selection logic, refresh
and redesign cycles, and phase-in capacity limits. Most of the
technology adoption features from the NPRM have been carried over for
the final rule analysis. However, the final rule analysis also included
adoption features for the new technologies incorporated in the final
rule analysis. For a detailed discussion of CAFE model path logic for
the final rule analysis, including technology supersession logic and
technology mutual exclusivity logic, please see Section IV.
Figure VI-18 and Figure VI-19 below show the engine technology
paths used for the NPRM and this final rule analysis, respectively. The
engine
[[Page 24419]]
technology paths have increased to incorporate new advanced
technologies manufacturers could adopt into their fleet.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.152
[GRAPHIC] [TIFF OMITTED] TR30AP20.153
BILLING CODE 4910-59-C
Similar to the 2012 final rule for MYs 2017-2025, this final rule
analysis also considered real-world limits when the defining the rate
at which technologies can be deployed.\900\ During the rulemaking
timeframe, manufacturers are expected to go through the normal
automotive business cycle of redesigning and upgrading their light-duty
vehicle products. This allows manufacturers the time needed to
incorporate fuel economy improving and CO2 reducing
technologies into their normal business cycle. This is important
because it has the potential to avoid the much higher costs that could
occur if manufacturers need to add or change technology at times other
than their scheduled vehicle redesigns. This time period also provides
manufacturers the opportunity to plan for compliance using a multi-year
time frame, again consistent with normal business practice.
---------------------------------------------------------------------------
\900\ 77 FR 62712.
---------------------------------------------------------------------------
Section II.G.3.a of the NPRM provided substantial discussion of how
an ``application schedule'' is used by the CAFE model to determine when
manufacturers are assumed to be able to apply a given technology to a
vehicle. The NPRM application schedule for engine technologies is
reproduced in Table VI-43, which shows that all of the
[[Page 24420]]
engine technologies may only be applied (for the first time) during
redesign.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.154
For this final rulemaking action, a similar schedule is employed,
and has been updated with information gathered since the NPRM and
through comments provided to the agencies.
Table VI-44 presents the engine technology application schedule
used for the final rule CAFE modeling.
[[Page 24421]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.155
Fuel economy improving and CO2 reducing technologies for
vehicle applications vary widely in function, cost, effectiveness, and
availability. Some of these attributes, like cost and availability,
vary from year to year. New technologies often take several years to
become available across the entire market. The agencies use phase-in
caps to manage the maximum rate that the CAFE model can apply new
technologies. Phase-in caps are intended to function as a proxy for a
number of real-world limitations in deploying new technologies in the
auto industry. These limitations can include but are not limited to,
engineering resources at the OEM or supplier level, restrictions on
intellectual property that limit deployment, and/or limitations in
material or component supply as a market for a new technology develops.
Without phase-in caps, the model may apply technologies at rates that
are not representative of what the industry is actually capable of
producing, which would suggest that more stringent standards might be
feasible than actually would be. Table VI-45 and Table VI-46 below
shows the phase-in caps between the NPRM and this final rule analysis,
respectively.
Most engine technologies are available at a rate of 100 percent in
MY2017 for the final rule analysis. Some advanced technologies that
have been recently introduced for one or two vehicle models are phased
in at lower rates. Technologies such as ADEAC and TURBOD are phase in
at rates that represent manufacturers' adoption capability and
typically have complementary effectiveness compared to other advanced
technologies. These lower phase-in caps also represent intellectual
property and functional performance concerns.
BILLING CODE 4910-59-P
[[Page 24422]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.156
[[Page 24423]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.157
BILLING CODE 4910-59-C
Comments received on engine adoption features were mixed, with
manufacturers generally supporting the NPRM methodology, and CARB and
NGOs opposing it. Several manufacturers commented, both in their public
comments or on a CBI basis, that many of the emerging engine
technologies had the potential to improve vehicle fuel economy, but
were technically complex and addressed many of the same issues as other
existing engine technologies.
We agree with manufacturers that broadly, there are technologies
that, in theory, present large potential effectiveness improvements
like VCR, ADEAC, and others. However, the agencies believe it is
important to assure realistic adoption of these technologies into the
fleet in the rulemaking time frame, so that the rulemaking analysis
accurately represents the costs and benefits of different regulatory
alternatives considered. If the agencies were to select stringency
based on an assumption that an emerging technology would see widespread
adoption, and then it does not, the benefits of that stringency level
would not be realized. The agencies have taken steps in the NPRM and
this final rule analysis to consider the manufacturability and
feasibility of these technologies for different vehicle types and
manufacturers. Discussed earlier, the analysis considers these and
other concerns by accounting for product cadence, and by implementing
phase-in caps and skips, and by designating technology phase-in and
phase-out years. Similar to the 2012 final rule, this final rule
analysis employed these strategies to reflect better the real-world
considerations faced by manufacturers.
EDF commented, referencing EPA's statutory command prescribed in
Section 202(a) of the Clean Air Act that:
EPA's task is thus to identify the major steps necessary for
`development and application of the requisite technology,' and then
the respective standard `shall take effect.' These individual
decisions are highly consequential: As noted above, without changing
anything else about the agencies' analysis, allowing HCR2 would
reduce augural compliance costs by $619--or about 30% of the total
difference between the augural and rollback scenarios. The
proposal's rejection of these technologies nowhere justifies how the
(unfounded and cursorily justified) concerns accord with the
agency's limited discretion under Section 202(a)(2) and duty to
`press for the development and application of improved technology
rather than be limited by that which exists today.' If the agency is
to predict more than the results of merely assembling pre-existing
components, it must have some leeway to deduce results that are not
represented by present data.\901\
---------------------------------------------------------------------------
\901\ NHTSA-2018-0067-12108 at 104.
CARB also commented that the CAFE Model prevents manufacturers
``from
[[Page 24424]]
switching between a turbocharged and HCR pathways under the premise
that manufacturers either would not develop both or would be committed
irreversibly to one path or the other. This assumption is not based in
reality and is not reflective of actual industry practice--
manufacturers who have pursued turbocharging have also already pursued
HCR engines for other vehicles in their line-up. For example, General
Motors (GM) utilizes downsized turbocharging in some vehicles, such as
the newly designed 2019MY Silverado pick-up and the Malibu sedan which
has two different turbocharged engine options. GM also has a third
offering in the Malibu sedan which is an HCR naturally aspirated 1.8L
equipped with cooled exhaust gas recirculation (CEGR) mated to a hybrid
electric system.'' \902\
---------------------------------------------------------------------------
\902\ NHTSA-2018-0067-11873 at 109.
---------------------------------------------------------------------------
CARB's observation was true for the NPRM analysis, however for the
final rule analysis the agencies allowed manufacturers to adopt engine
technologies from alternate tree paths, when incorporating
electrification technology, see Section VI.C.3.c). The agencies still
believe that if manufacturers have invested in one type of engine
technology for their vehicles that they would not transition to another
technology except in the case of a major vehicle powertrain redesign,
such as the inclusion of an HEV system. Additional discussion on this
issue is presented in Section VI.B.1.
The following sections discuss adoption features specific to
individual engine technologies, including comments received and updates
(or not) for the final rule analysis.
(1) Basic Engines
Most vehicles in the MY 2017 analysis fleet that are DOHC or SOHC/
OHV spark ignited engines and are not downsized turbocharged engines
have any two combinations of VVT, VVL, SGDI or DEAC.\903\ For the NPRM,
only engines with 6-cylinders or more could adopt DEAC and ADEAC.
---------------------------------------------------------------------------
\903\ EPA. ``2018 EPA Automotive Trends Report'' 12 pp, 421 K,
EPA-420-S-19-001, March 2019. https://www.epa.gov/automotive-trends/download-automotive-trends-report#Full%20Report (last accessed Feb.
12, 2020) p. 72.
---------------------------------------------------------------------------
HDS on behalf of CARB commented that in the NPRM analysis VVL,
which is cost ineffective compared to other conventional technologies,
was always included in an adopted technology package.\904\ HDS further
stated that the ``effectiveness of VVL is even smaller when the
technology is combined with turbocharged downsized engines.''
Accordingly, HDS stated that removing VVL from the base pathway would
save $314 but reduce fuel economy by only 1.4 percent, according to the
LPM.
---------------------------------------------------------------------------
\904\ NHTSA-2018-0067-11985 at p.34.
---------------------------------------------------------------------------
The agencies did not agree with HDS' assessment of the NPRM
analysis. The agencies do not agree VVL was forced to be adopted in the
analysis fleet and do not agree with how technology effectiveness
values compare to LPM estimates. As discussed earlier in the
effectiveness and modeling section, each engine technology was modeled
independently and the CAFE model was allowed to adopt the most cost
effective technology. Therefore, it is inaccurate to state, a
technology is less effective, especially when comparing LPM.
Particularly because VVL technologies reduce pumping losses in engines,
so it is realistic that other technologies, that also reduce pumping
losses, have synergetic effect. This is specifically true for
turbocharged engines.
ICCT commented that DEAC technology should be available for every
engine, and should not be limited to 6-cylinder and higher cylinder
count engines. ICCT and CARB also commented that DEAC should be allowed
on turbocharged engines. ICCT also commented that ADEAC should be
widely available as it can be a viable technology application for
various other powertrain technology combinations.\905\ Furthermore,
CARB commented ``automakers will combine technologies like
turbocharging, HCR and DEAC as well as more technologies when they have
cost-effectiveness synergies.'' \906\
---------------------------------------------------------------------------
\905\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-13.
\906\ CARB at p. 6.
---------------------------------------------------------------------------
The agencies agree with ICCT that DEAC and ADEAC could be applied
to additional engine types, including turbocharged engines. However,
the agencies disagree with ICCT that ADEAC should be widely applied to
all powertrain technology combinations in this analysis. The agencies
have updated the final rule analysis to allow DEAC and ADEAC for
various engine cylinder counts and for turbocharged engines.
For the final rule analysis, both DEAC and ADEAC technologies can
be adopted by any naturally aspirated engine. Similarly, any
turbocharged engine can also adopt cylinder deactivation technology, as
characterized by TURBOD and TURBOAD in the CAFE model. In this final
rule analysis, the agencies distinguished cylinder deactivation
technologies between naturally aspirated and forced air induction
systems.
For the final rule analysis, the agencies allow any combination of
VVT, VVL, SGDI and DEAC to be adopted for any engine displacement and
cylinder count. Figure VI-18 below shows the basic engine paths a
vehicle could traverse for the final rule analysis. Similar to the
NPRM, the agencies have not changed the adoption features of the
technologies shown in Figure VI-18, with one exception. Vehicles that
are SOHC or DOHC configuration that do not have VVT in the baseline can
now adopt it.
Finally, the agencies disagree with ICCT and CARB that these DEAC,
ADEAC, TURBOD, and TURBOAD should apply beyond these configurations.
DEAC's fundamental benefits are driven by reducing pumping losses and
by enabling the engine to operate in a more thermal efficient region of
the engine fuel map. Conventional spark-ignited engines control airflow
into the cylinders via a throttle operated by the driver to provide the
level of power that is delivered.\907\ In an 8-cylinder engine, when
driving in light load conditions such as highway driving, there are
lower engine power requirements. In a throttle controlled system,
engine pumping losses increase as air flow decreases. A way to reduce
pumping loss in an engine is by increasing the airflow into the
cylinders. By deactivating a set of cylinders, the same power output
can be delivered by a ``smaller'' engine. Many technologies modeled for
this analysis work to reduce pumping losses, but through other
mechanisms like VVT, VVL, downsized engines with turbochargers, high
compression Atkinson mode cycle, and Miller Cycle.\908\ Transmissions
with a higher number of gears also provide the opportunity to reduce
pumping work of the engine.\909\
---------------------------------------------------------------------------
\907\ A throttle is the mechanism by which fluid flow is managed
by constriction or obstruction. An engine's power can be increased
or decreased by the restriction of inlet gases, but usually
decreased.
\908\ 2015 NAS at p. 23.
\909\ 2015 NAS at p.173.
---------------------------------------------------------------------------
As discussed earlier, DEAC can reduce pumping losses, so when
combined with other technologies that also reduce pumping losses, like
downsized turbocharged engines, the benefits for cylinder deactivation
are lower than for naturally aspirated engines because downsized
turbocharged engines already have lower pumping losses due to having a
downsized engine.\910\
---------------------------------------------------------------------------
\910\ 2015 NAS at p. 34.
---------------------------------------------------------------------------
[[Page 24425]]
(2) Turbocharged Downsized Engines
About 23 percent of vehicles in the MY 2017 baseline fleet had
turbocharged engines. For the final rule analysis, the agencies allowed
any basic engine to adopt turbo engine technology (TURBO1, TURBO2 and
CEGR1) from the Turbo path similar to the NPRM analysis. This includes
any combination of VVT, VVL, SGDI and DEAC for both SOHC and DOHC
configurations. Vehicles that have turbocharged engines in the baseline
fleet will stay on the turbo engine path to prevent unrealistic engine
technology change in a short timeframe considered in the rulemaking
analysis. Turbo path is a mutually exclusive technology in that it
cannot be adopted for HCR, diesel, ADEAC, CNG and powersplit PHEVs.
(3) Non-HEV Atkinson Mode Engines
The NPRM analysis allowed limited application of HCR engines (HCR1
and HCR2) to vehicles in the MY 2016 baseline fleet.\911\ As discussed
above, applying HCR1 or HCR2 technologies to a vehicle resulted in
overstated effectiveness values relative to the baseline VVT
engine,\912\ because of differences in how those maps were developed
compared to the IAV engine maps used for the majority of the technology
analysis. In an attempt to avoid unrealistic results in the NPRM,
adoption of HCR1 (Eng24) technology was limited to only manufacturers
that demonstrated existing use of high compression ratio technology.
HCR was disallowed for other manufacturers that demonstrated an intent
to develop other advanced technologies incompatible with HCR
technology. In addition, the agencies disallowed HCR engines from being
applied to vehicles with greater performance requirements, like 6- and
8-cylinder vehicles, because the higher load requirements from these
vehicles would force the engine to exit the Atkinson mode, where
maximum efficiency is achieved.
---------------------------------------------------------------------------
\911\ 83 FR 43037.
\912\ 83 FR 43029 Figure II-1--Simulated Technology
Effectiveness Value.
---------------------------------------------------------------------------
The Alliance commented in agreement with the application
restrictions for HCR1 in the NPRM, listing the following
justifications: ``Packaging and emission constraints associated with
intricate exhaust manifolds needed to mitigate high load/low
revolutions per minute knock; Inherent performance limitations of
Atkinson cycle engines; and Extensive capital and resources required
for manufacturers to shift to HCR from other established technology
pathways (e.g., downsized turbocharging).'' \913\ Ford similarly
commented in support of ``the more restrained application of HCR1 in
the Proposed Rule, an approach that recognizes the investment,
packaging, performance and emissions factors that will limit
penetration of this technology.'' \914\
---------------------------------------------------------------------------
\913\ NHTSA-2018-0067-12073.
\914\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
In contrast, CARB stated that the constraint on HCR1 engines was
inappropriate and did not reflect reality,\915\ and stated that the
agencies failed to supply any detailed rationale as to why HCR
applications were so constrained in the CAFE Model. Specifically, CARB
took issue with the justification that HCR1 is limited in the CAFE
model because it is ``not suitable for MY 2016 baseline vehicle models
that have 8-cylinder engines and in many cases 6-cylinder engines.''
\916\ CARB stated that ``the HCR1 technology is declared not suitable
on 207 of the 288 engines cumulatively used by all of industry
including over 50 percent of the 4 cylinder engines and nearly 90
percent of the 6 cylinder engines instead of only being restricted from
8 cylinder and `in many cases 6 cylinder engines.' '' CARB also stated
that the implied rationale for not allowing HCR1 to be applied to 6-
and 8-cylinder engines because trucks or larger vehicles could not
utilize it is unreasonable, as the Toyota Tacoma used a 3.5L V6 HCR
Atkinson-like engine since MY 2016. CARB stated that the Toyota Tacoma
was properly assigned a HCR1 engine in the MY 2016 analysis fleet file,
but the engine was disallowed from other Toyota V6 engines utilized in
vehicles like the Sienna minivan and 4Runner SUV. CARB commented that
``[i]f the intended rationale is that HCR engines will have
insufficient low end torque to satisfy truck-like towing demands, it
would be inappropriate to restrict the engine from minivan and SUV
applications which have a lower tow rating and lower expected towing
demands.'' Finally, CARB stated that the HCR1 package restrictions were
inappropriate, as there was no mechanism in the CAFE model to represent
appropriately the MY 2019 Dodge Ram 1500 5.7L V8 that uses ``a higher
compression ratio than earlier versions and using its VVT system to
reduce pumping losses via delayed, or late, intake valve closing--
resulting in an HCR-like engine with an over-expanded or Atkinson
cycle.''
---------------------------------------------------------------------------
\915\ NHTSA-2018-0067-11873.
\916\ 83 FR 43038.
---------------------------------------------------------------------------
Similarly, Meszler Engineering Services, commenting on behalf of
NRDC, commented that HCR1 appears as a baseline technology on vehicles
representing about 4 percent of the baseline non-hybrid vehicle market,
and is subsequently applied to only 23 percent of the market. Meszler
stated that the ``relative cost effectiveness of the technology is
perhaps best illustrated by the fact that the market penetration of HCR
technology on non-hybrid vehicles under the augural standard is modeled
to be 27 percent of 2032 sales, exactly equal to the baseline
penetration of 4 percent and the allowable adoption fraction of 23
percent. In other words, the technology was adopted by every vehicle
that was not explicitly prohibited (by NHTSA) from doing so.'' EDF
commented that ``NHTSA has further imposed artificial and unreasonable
constrains on the use of certain technologies that does not match how
automakers are applying them in vehicles today,'' stating that HCR1
represented a technology that had been in the marketplace for many
years and had been applied by several manufacturers, ``[y]et, even for
MY 2030 vehicles and beyond, NHTSA only allows the use of HCR1 by about
30 percent of the U.S. fleet.'' \917\
---------------------------------------------------------------------------
\917\ NHTSA-2018-0067-12108.
---------------------------------------------------------------------------
In considering the comments, the agencies agree with commenters
that the HCR1 engine application was overly limited for the NPRM
analysis. As a result, the agencies have expanded the availability of
HCR1 technology for the final rule analysis. The refined adoption
features for HCR1 are discussed below. The new adoption features do
maintain considerations for performance neutrality. Comments about how
the characterization of engine technologies in the analysis fleet
impacted HCR technology adoption in subsequent model years are
addressed in Section VI.C.1.d) Baseline Fleet Engine Tech.
Regarding HCR2, the Alliance commented in support of ``the decision
to exclude the speculative HCR2 technology from the analysis.'' \918\
The Alliance continued, ``[a]s previously documented in Alliance
comments, the inexplicably high benefits ascribed to this theoretical
combination of technologies has not been validated by physical
testing.'' Similarly, Ford stated that ``[t]he effectiveness of the
`futured' Atkinson package (HCR2) that includes cooled exhaust gas
recirculation (CEGR) and cylinder deactivation (DEAC) is excessively
high, primarily due to overly-optimistic efficiencies in the base
engine map, insufficient accounting of CEGR and DEAC integration
losses, and no accounting of the impact of 91RON
[[Page 24426]]
Tier 3 test fuel. Given the speculative and optimistic modeling of this
technology combination, Ford supports limiting the use of HCR2
technology to reference only, as described in the Proposed Rule.''
\919\
---------------------------------------------------------------------------
\918\ NHTSA-2018-0067-12073.
\919\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
In contrast, several commenters disagreed with the agencies'
decision to limit the adoption of HCR2 engines, stating that the
technology was clearly applicable during the rulemaking timeframe, as
the technology was already being applied by manufacturers, and that the
technology was cost-effective, as shown by the agencies' own modeling.
ICCT commented that ``[i]t is clear that the agencies have
artificially excluded a known technology that is applicable in the
timeframe of the rulemaking.'' \920\ ICCT commented that ``[d]espite
the facts that (as discussed above) the agencies have cost and
effectiveness data for this technology, many automakers are already
deploying the HCR1 technology, and the 2018 Camry has already put most
of the HCR2 technologies into production, the agencies did not allow
any application of HCR2 by 2025.'' \921\ ICCT concluded that the ``only
explanations . . . for the agencies' system of omissions and
constraints are that the agencies have biased the analysis against
including all the viable technologies by inserting their own artificial
constraints (either for lack of research, lack of analytical effort, or
not fully utilizing all the agencies' best analytical tools and data)
or that the auto industry is providing information that erroneously
suggests their innovation is far less than what is demonstrated both
above and in the agencies' own previous analyses.'' ICCT stated that
``[t]he great lengths the agencies have gone to artificially impose
`skip' constraints for HCR in the CAFE modeling system demonstrates
that the agencies have exerted an explicable and apparently deliberate
bias towards forcing most of the automaker compliance technology toward
higher cost, non-HCR turbocharging paths.'' \922\
---------------------------------------------------------------------------
\920\ NHTSA-2018-0067-11741.
\921\ NHTSA-2018-0067-11741.
\922\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
Several commenters also stated that HCR should not have been
restricted because it is clearly a cost-effective technology, citing
the sensitivity runs conducted that allowed unrestricted HCR
application in the analysis. For example, ICCT commented that allowing
HCR2 application across the fleet reduced total per-vehicle cost of
compliance with the augural standards by $690, which ``shows that the
agencies intentionally excluded a highly cost-effective technology (by
their own analysis) in the rulemaking analysis.'' \923\ Similarly, EDF
performed software modifications of the CAFE model, including allowing
the use of both HCR1 and HCR2 technology for all manufacturers by MY
2028. The analysis performed by EDF using their modified version of the
CAFE model, showed reductions in the per-vehicle compliance cost
projections by nearly $600.\924\
---------------------------------------------------------------------------
\923\ NHTSA-2018-0067-11741.
\924\ NHTSA-2018-0067-12108.
---------------------------------------------------------------------------
ICCT concluded that ``[t]he only reasonable and technically valid
assumption is that HCR be allowed for application to all vehicle
models' engine redesigns through all the model years of the compliance
modeling analysis.'' \925\ ICCT stated that ``[f]or the agencies to
constrain HCR technology for use by other automakers, they have a
responsibility to demonstrate why each of the other automakers cannot
adopt this known technology in their fleet.''
---------------------------------------------------------------------------
\925\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
The agencies agree with commenters' observations about the results
of the sensitivity runs performed as part of the NPRM analysis.
However, the agencies also believe the adoption features for HCR1 and
HCR2 were appropriate for the NPRM analysis. Had the agencies not
applied adoption features in that way, the agencies would have shown
unrealistic pathways for compliance for manufacturers that would have
understated costs and overstated benefits of potential CAFE and
CO2 standards.
The agencies disagree with commenters' statements that HCR has been
widely available in the automotive market and that the HCR technology
accordingly should not be limited in the CAFE model. For reasons
discussed in the NPRM and explained in more detail in Section
VI.C.1.c)(3), depending on vehicle type and use, Atkinson cycle
operation may be enabled for low and moderate engine demand conditions,
whereas Otto cycle operation may be needed for higher load conditions
to meet performance needs, such as to move more passengers, cargo, or
for towing. In addition, there may be issues on some platforms to
package the larger exhaust manifolds needed to enable Atkinson
operation, particularly with V6 and V8 engines. Manufacturers have
applied Atkinson technologies in unique ways to meet the needs and
capabilities of their vehicles to operate using the Atkinson and Otto
cycles. The agencies agree with comments from stakeholders, including
Toyota, who observed HCR technology is not suitable for all vehicle
configurations, and may not meet performance requirements for high-load
applications. As discussed earlier, the agencies believe the variation
of technologies can be categorized into three different forms of
Atkinson engine technologies for this analysis: (1) Atkinson engines,
(2) Atkinson-mode engines, and (3) Atkinson-enabled engines using
variable valve timing with late intake closing. Manufacturers typically
apply one of these technologies and tune that technology for specific
applications. Some commenters have consistently conflated the
technologies and asserted the capabilities of all three types of
Atkinson technologies can be represented by a single engine model. The
agencies do not agree with stakeholder assertions that a single HCR
engine map should be applied to every technology class or vehicle
platform.
To reflect better the incremental effectiveness for a low-cost
version of HCR technology, the agencies added the HCR0 engine for the
analysis. The specification of this engine was provided in the NPRM
PRIA as Eng22b. Using this engine improves the estimated incremental
effectiveness because the incremental engine changes were directly
specified for the modeling and are relative to the other engine
technologies in the analysis.\926\ HCR0 is the first engine in the HCR
path that a manufacturer could adopt. HCR0 represents technology that
could incrementally be adopted to the VVT engine, increasing
compression ratio and adding Atkinson cycle capability. The use of the
HCR0 technology, applied in the final rule analysis, allowed the
agencies to update HCR adoption features. Once a basic engine adopts
HCR technology (i.e., HCR0 and HCR1 for the central analysis, or HCR2
for a sensitivity case) the vehicle will not switch to a different
engine technology path. For example, if a vehicle had adopted HCR or is
equipped with HCR technology it is not allowed to adopt turbocharged
engine technologies. The HCR0 technology appropriately captures the
benefits of applying transitional Atkinson technologies to conventional
basic engine technologies. The agencies note that VVT technology valve
control has late intake valve closing under some operating conditions
to take some advantage of Atkinson cycle-like operation; however, that
operation is not as extensive as HCR technology and is not coupled with
a higher
[[Page 24427]]
compression ratio as is the case for HCR technologies.
---------------------------------------------------------------------------
\926\ PRIA 6.3.2.2.21.20.2.1 IAV Engine 22b--High Compression
Atkinson Cycle Engine at p. 307.
---------------------------------------------------------------------------
The agencies also allowed all 4-cylinder engines on the basic
engine path to adopt HCR technology similar to turbocharged
technologies. This allowed any small and midsize vehicles, including
small and midsize SUVs, that had any combinations of basic engine path
technologies to move to the HCR path. However, there are two exceptions
to this feature, including: (1) When the vehicle is a pickup including
both standard and performance class; and (2) when the base engine is
shared with a pickup including both standard and performance class. The
agencies discussed earlier in the non-HEV Atkinson section why HCR
technology cannot be applied to all vehicle applications.
Finally, engines with advanced engine technology already in the
baseline vehicle such as turbocharged engines are not allowed to adopt
HCR technology. The agencies continue to believe this constraint is
reasonable given the extensive capital resources and stranded capital
that would be involved if a manufacturer who focused on and invested
heavily in non-HCR advanced technologies were to abandon those
technologies abruptly and switch to HCR technologies.\927\ For example,
Ford has incorporated turbocharged engines across 75 percent to 80
percent of their fleet in MY2017, and these engines are shared across
multiple technology classes.\928\ The abovementioned modeling,
limitation for this analysis assumes that manufacturers will not change
advanced engine technology applied to a platform due to the high cost
and lead time required for research and development, and for the
development and implementation of new manufacturing plants and
equipment to implement an entirely new powertrain in the rule making
time frame. For further discussion see Section VI.B.1.
---------------------------------------------------------------------------
\927\ 83 FR 43038.
\928\ The 2018 EPA Automotive Trends Report figure 4.23. at
p.68.
---------------------------------------------------------------------------
In response to ICCT's comment that agencies must discuss the
reasoning for allowing and disallowing HCR technology for each
individual manufacturer, these updated adoption features now allow more
manufacturers to adopt HCR engine technology. The agencies no longer
apply adoption features based on manufacturer, but now base them on
individual platforms. The agencies believe a manufacturer that has
already invested in advanced engine technologies for a specific
platform would face very high costs and incur significant stranded
capital to switch that platform to another advanced technology. And
doing so would not be reasonable given the small incremental fuel
economy improvement that would be gained, for example, for switching
from advanced turbocharging to HCR technologies. Specifically,
manufacturers that have invested in turbocharging technology for
certain platforms, like Honda, Ford, and the German manufacturers,
would incur unreasonable costs to switch to another advanced technology
path. However, manufacturers that use turbo technology on one platform
are not precluded from implementing HCR technology on another of its
platforms. HCR adoption is still limited for all manufacturers based on
vehicle performance requirements discussed earlier.
(4) Advanced Cylinder Deactivation Technology
In the NPRM, any basic engine technology could adopt ADEAC.
Commenters stated that the agencies restricted ADEAC technologies in
the NPRM analysis to naturally aspirated engines.
ICCT provided a broad comment regarding the treatment of advanced
technologies, including ADEAC, and criticized how the NPRM ``removed
many technologies that are viable and being actively deployed by the
auto industry.'' ICCT specifically criticized ``cases where viable
technology combinations are disallowed'' such as ``turbocharging and
cylinder deactivation (DEAC).'' \929\
---------------------------------------------------------------------------
\929\ NHTSA-2018-0067-11741 at p.6.
---------------------------------------------------------------------------
UCS also commented on how ADEAC technology was applied in the NPRM,
stating ``While the agencies have acknowledged the existence of dynamic
cylinder deactivation, they have not appropriately included it as an
available technology, dramatically limiting its availability.'' UCS
specifically disagreed with adoption features of the ADEC, noting the
technology ``is restricted to naturally aspirated, low-compression
ratio engines--it cannot be combined with turbocharged engines, high
compression ratio engines, or variable compression ratio engines due to
pathway exclusivity in the Volpe model.'' \930\ CARB and Meszler
mirrored these concerns.\931\
---------------------------------------------------------------------------
\930\ NHTSA-2018-0067-12039 at p.4
\931\ NHTSA-2018-0067-12039 at p.4.
---------------------------------------------------------------------------
The agencies agreed with commenters and in response have allowed
both naturally aspirated engines and turbocharged engines to adopt
ADEAC in the final rule analysis. The new Advanced Turbocharging path
includes TURBOD and TURBOAD, while naturally aspirated engines use the
same ADEAC engine designation. There is some potential for this type of
technology to improve fuel economy and reduce CO2 emissions,
however, the technology provides diminishing returns if it is included
with engine downsizing or other technologies that already reduce
pumping losses. Accordingly, once a vehicle has adopted ADEAC, TURBOD,
or TURBOAD, the agencies did not allow further adoption of other engine
technologies that reduce pumping losses such as VCR and VTG.
(5) Miller Cycle Engines
Miller cycle engine technologies (VTG and VTGe) are new for this
final rule analysis, and VTG engines could be applied to any basic and
turbocharged engine. Discussed earlier, the VTGe technology is enabled
by the use of a 48V system that presents an improvement from
traditional turbocharged engines, and accordingly VTGe could only be
applied with a mild hybrid system.
(6) Variable Compression Ratio Engines
In the NPRM analysis, variable compression ratio (VCR) technology
was not available for adoption, but the engine map and specifications
were provided for review. For this final rule analysis, VCR engines are
included in the analysis and can be applied to basic and turbocharged
engines, however the technology is limited to Nissan. VCR technology
requires a complete redesign of the engine, and in MY2020, only two of
Nissan's models had incorporated this technology. In addition, the
technology showed lower fuel savings than expected.\932\ The agencies
do not believe any other manufacturers will invest to develop and
market this technology in their fleet in the rulemaking time frame.
---------------------------------------------------------------------------
\932\ VanderWerp, D. ``Why Nissan's Holy-Grail VC-T Engine
Doesn't Achieve Better Fuel Economy,'' C/D Nov 1, 2018. Available at
https://www.caranddriver.com/features/a24434937/nissan-new-vc-t-engine-fuel-economy/. Last accessed Dec. 19, 2019.
---------------------------------------------------------------------------
(7) Diesel Engines
Diesel engine adoption and features have been carried from the NPRM
analysis for this final rule analysis for ADSL and DSLI. Any basic
engine technologies (VVT, VVL, SGDI, and DEAC) can adopt ADSL and DSLI
engine technologies. New for the final rule analysis is the adoption of
advanced cylinder deactivation for diesel engines (DSLIAD). Any basic
engine and diesel engine can adopt this technology in the final rule
analysis;
[[Page 24428]]
however, the agencies have applied a phase in cap and year for this
technology at 34 percent and MY 2023, respectively. In the agencies'
engineering judgement, the agencies have concluded that this is a
rather complex and costly technology to adopt and think that it could
take significant investment to develop. For more than a decade, diesel
engine technologies have been used in less than one percent of the
total light-duty fleet production,\933\ and the investment for this
cylinder deactivation technologies may not be justifiable.
---------------------------------------------------------------------------
\933\ The 2018 EPA Automotive Trends Report Table 4.1 at p. 72.
---------------------------------------------------------------------------
(8) Alternative Fuel Engines
Adoption features for alternative fueled compressed natural gas
(CNG) engines have been carried over from the NPRM for this final rule
analysis. Because CNG is considered an alternative fuel under EPCA/
EISA, it cannot be adopted during the rulemaking timeframe for NHTSA's
standard setting analysis. The EPA analysis was modeled separately in
the CAFE model without such constraints.
(9) Engine Lubrication and Friction Reduction
Finally, new for this analysis is the addition of EFR. The agencies
allow EFR to apply to any engine technology except for DSLI and DSLIAD.
DSLI and DSLIAD inherently have incorporated engine friction
technologies from ADSL. In addition, friction reduction technologies
that apply to gasoline engines cannot necessarily be applied to diesel
engines due to the higher temperature and pressure operation in diesel
engines.
f) Engine Effectiveness Modeling and Effectiveness Values
Figure VI-20 below shows the effectivness estimates from all the
vehicle types for the NPRM analysis using Autonomie full vehicle
modeling and simulation.
[GRAPHIC] [TIFF OMITTED] TR30AP20.158
Roush commented that they had observed wide variations in estimated
incremental effectiveness associated with individual technology
packages between the 2016 Draft TAR and NPRM analysis.\934\
---------------------------------------------------------------------------
\934\ NHTSA-2018-0067-11984. Roush at p. 16.
---------------------------------------------------------------------------
The agencies agree that to predict potential incremental
improvements in fuel efficiency accurately, it is extremely important
to understand the nature of the improvements being sought by each
increment (improved thermodynamics, reduced friction, reduced vehicle
weight, etc.). The technology modeling
[[Page 24429]]
and large scale simulation used for the proposal and updated for the
final rule does exactly that. In fact, the NPRM and final rule use
these methods more expansively than any previous CAFE and
CO2 rulemaking, including the 2016 Draft TAR and 2016 EPA
Proposed Determination.
One commenter stated the effectiveness for ADEAC was overestimated
for the NPRM, and that data from compliance shows much lower
effectiveness. The agencies disagree with this comment, as it is
invalid to compare effectiveness of full vehicle compliance data
directly to the incremental effectiveness modeled for ADEAC. For
reasons discussed in Section VI.B.3 data from full vehicle benchmarking
cannot be used as a comparison for specific technology effectiveness.
The effectiveness estimated for this technology is in line with test
data, CBI, and engineering analysis.\935\
---------------------------------------------------------------------------
\935\ Boha, Stani. ``Benchmarking and Characterization of a Full
Continuous Cylinder Deactivation System.'' EPA. April 10-12, 2018
SAEA World Congress. https://www.epa.gov/sites/production/files/2018-10/documents/deact-sae-world-congress-bohac-2018-04.pdf last
access Feb 12, 2020.
---------------------------------------------------------------------------
Engine effectiveness estimates remained the same for most
technologies from the NPRM analysis, with the exception of some
technologies that had characteristics updated, and the new added engine
technologies. For the final rule analysis, the agencies used the same
effectiveness values for ADEAC applied to naturally aspirated engines
as in the NPRM, and incorporated estimated effectiveness values for
TURBOAD to represent ADEAC on downsized turbocharged engines.
Other technology-specific comments and the agencies' responses are
provided within the discussion of each technology throughout this
section, as those comments tended to be predicated on issues
surrounding the engine maps used to model technologies or technology-
specific adoption features. For the final rule analysis, the technical
merits of the substantive comments and any accompanying publications
and information were carefully considered and discussed in the
subsections where appropriate.
Figure VI-21 below shows the effectivness estimates from compact
car and midsize car vehicle types for the final rule analysis using
Autonomie full vehicle modeling and simulation.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.159
g) Engine Costs
Discussed in the PRIA, the agencies spent millions of dollars
sponsoring research to determine direct manufacturing costs (DMCs) for
fuel saving technologies since the 2012 rule.\936\ Because a major
objective of the studies was to consider costs in the rulemaking
timeframe, the agencies believed that these costs were appropriate to
use for the NPRM and final rule analysis. Table VI-47 below shows the
DMC used for IC engine technologies for the NPRM analysis.
---------------------------------------------------------------------------
\936\ FEV prepared several cost analysis studies for EPA on
subjects ranging from advanced 8-speed transmissions to belt
alternator starter, or Start/Stop systems. NHTSA also contracted
with Electricore, EDAG, and Southwest Research on teardown studies
evaluating mass reduction and transmissions. The 2015 NAS report on
fuel economy technologies for light-duty vehicles also evaluated the
agencies' technology costs developed based on these teardown
studies, and the technology costs used in this proposal were updated
accordingly. These studies are discussed in detail in Chapter 6 of
the RIA accompanying the NPRM proposal.
---------------------------------------------------------------------------
[[Page 24430]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.160
[[Page 24431]]
CARB commented that costs associated with IC engines were not
excluded from the final costs of BEV vehicles.\937\ CARB continued,
stating that ``the final costs of BEV vehicles are higher due to the
inclusion of the base absolute costs, to which the assigned BEV
incremental cost would be added.''
---------------------------------------------------------------------------
\937\ NHTSA-2018-0067-11873 at p.122.
---------------------------------------------------------------------------
The agencies agree with CARB that inclusion of IC engine costs in
the BEV cost was an error in the analysis. In response to this comment,
the agencies have developed absolute costs for baseline engines for the
CAFE model in order to account for appropriate cost of removing engines
from BEVs. In the final rule analysis, once a vehicle adopts BEV
technology, the costs associated with powertrain systems are removed.
Due to the extensive variations in engine technologies in real world
production, the agencies relied on discrete publication costs and
historical studies to assign costs for base engines.938 939
For this final rule analysis, the agencies have included these costs
for base engines shown in Table VI-48.
---------------------------------------------------------------------------
\938\ FEV P311732-02 Oct13, 2015 at p. 259.
\939\ UBS Limited. ``UBS Evidence Lab Electric Car Teardown--
Disruption ahead?'' May 18, 2017.
[GRAPHIC] [TIFF OMITTED] TR30AP20.161
BILLING CODE 4910-59-C
Commenters compared engine cost data from the NPRM to other
sources, in many cases to support their comments that the technology
costs used in the NPRM were too high. ICCT commented that the agencies
did not consider the latest reports on technology cost data, and
specifically referenced an ICCT-sponsored FEV cost study for the
European EU6b regulations in MY 2025,\940\ as well as prior EPA cost
estimates for several engine technologies including SGDI, cEGR, HCR,
and others, to point out differences in cost.\941\ ICCT also commented
on the difficulty they had in locating the cost data used in the NPRM,
stating that ``because the agencies present cost data in so many
different ways in dozens of different places in the NPRM, impact
assessment, and supporting data files, the precise agencies' costs are
obscured and not transparent.'' ICCT stated that ``[w]ithout a clear
explanation of the methodology, it is unclear precisely how price
increases are determined, as well as the relationship between
technology costs, fines, and price increases.'' Despite this claim,
ICCT was able to provide several pages comparing engine technology
costs.
---------------------------------------------------------------------------
\940\ FEV. '' 2025 Passenger Car and Light Commercial Vehicle
Powertrain Technology Analysis'' September 2015. https://theicct.org/sites/default/files/publications/PV-LCV-Powertrain-Tech-Analysis_FEV-ICCT_2015.pdf.
\941\ NHTSA-2018-0067-11741 at p. I-68.
---------------------------------------------------------------------------
In the NPRM PRIA Chapter 6.3.2.2.20.22, the agencies provided DMCs
for all engine technologies in 2016 dollars without inclusion of RPE
and learning for review. In the same chapter, the agencies also
provided absolute costs that incorporated costs in 2016 dollars, RPE
and learning data as used by the CAFE model to assess cost
effectiveness for future MY vehicles. Where appropriate, the agencies
discussed in the individual technology sections where costs were
updated for this final rule analysis with the latest data. This also
includes cost data for new technologies available in the CAFE model for
the final rule analysis.
Some engine costs were carried over from prior rulemakings, but may
have looked different because they were updated to current dollars
(2016 for the NPRM and 2018 for the final rule), and for engine
architecture and cylinder count. In addition, costs were updated based
on appropriate vehicle class. This was important to consider to
maintain performance neutrality, as technology effectiveness associated
with one engine technology type for a vehicle class cannot be used for
the same engine technology for higher performance vehicle class. This
affected total costs. For further discussion on the cost-effectiveness
metric used in the CAFE model, see discussions in the Section VI.A
Overview of the CAFE model and VI.B.3 Technology Effectiveness Values.
The agencies do not believe that the FEV report referenced by ICCT
is applicable for this analysis for a few reasons. First, the primary
focus of the FEV study ``is the European Market according to the EU6b
regulation as well as the consideration of emissions under both the
NEDC and WLTP test procedures.'' This final rule analysis specifically
considered the U.S. automotive market during the rulemaking timeframe
based on U.S.-specific regulatory test cycles. Accordingly, the costs
reflect incremental technology effectiveness for achieving improvements
as measured through U.S. regulatory test methods. The agencies had
discussed these test cycles and methods further in Section VI.B.3
Technology Effectiveness Values.
Second, FEV did not conduct original teardown studies for this
report, as indicated by project tasks, but rather used engineering
judgement and external studies in assessing incremental costs.\942\ The
FEV report did not provide sources for each individual cost and it is
unclear how costs in many scenarios were developed since no teardowns
were used. Note that for this final rule analysis, the agencies have
used previously conducted FEV cost teardown studies and the referenced
2015 NAS costs that referenced FEV teardowns. The agencies are not
concluding that FEV is an unreliable source. The agencies preferred to
specifically identify incremental costs of adding technology to account
appropriately for the costs of those technologies in the analysis.
---------------------------------------------------------------------------
\942\ FEV EU Costs Tasks: ``Definition of reference hardware or
description made by experience of development and design engineers
as well as additional research as base for cost analysis (no
purchase of hardware)''.
---------------------------------------------------------------------------
Finally, the cost for different vehicle classes identified by the
FEV study does not line up with the vehicle classes discussed in the
NPRM and this final rule analysis. FEV stated specifically, ``the
configuration of the vehicles has not been optimized for the US market
and may not be representative of this market.''\943\ The agencies have
discussed the importance of aligning the CAFE vehicle models with the
U.S. market earlier in Section VI.B.3
[[Page 24432]]
Technology Effectiveness Values and Section VI.C.1.d) Baseline Fleet.
All of these factors make it difficult to compare directly the
agencies' estimates and estimates presented in the FEV report cited by
ICCT in their comments.
---------------------------------------------------------------------------
\943\ Id. at p.141.
---------------------------------------------------------------------------
HDS provided a variety of costs and effectiveness comparisons
between the NPRM and previous 2012 final rule and the 2016 Draft
TAR.\944\ Specifically, HDS stated that the data presented in the 2016
TAR indicated a $60 per CO2/mile reduction for most
conventional engine technologies.
---------------------------------------------------------------------------
\944\ Duleep, K.G., ``Review of the Technology Costs and
Effectiveness Utilizing in the Proposed SAFE Rule,'' Final Report,
H-D Systems, October 2018, at p. 18-19.
---------------------------------------------------------------------------
Although the comparison was technically sound, there are
significant differences between the Draft TAR and NPRM analyses that
clearly account for the differences in engine cost. First, the NPRM
analysis used the MY 2016 fleet as a starting point to model
manufacturers' potential responses to CAFE and CO2
standards, whereas the 2012 final rule and Draft TAR used older
baseline fleets. Vehicles in the MY 2016 fleet already included more
advanced technologies than their predecessors in prior MY fleets, which
would make it more expensive for vehicles that have already adopted
advanced technologies to adopt more advanced technology. Second, the
agencies refined the engine modeling from previous analysis to the NPRM
to account for engine configurations and cylinder count more precisely.
For the final rule analysis, the same approach was taken to account
appropriately for costs for different type engine designs and
configurations.
Aside from these updates, engine costs were carried over from the
NPRM analysis, except for newly added technologies, where costs were
obtained from various sources such as NAS studies, technical
publications, and CBI data. Finally, the cost estimates have been
updated to account for dollar year (updated from 2016 dollars to 2018
dollars), and learning rate.
(1) Basic Engines
DMCs used for the final rule analysis for basic engine technologies
were the same as NPRM costs. Table VI-49 below shows the basic engine
DMC used for this final rule analysis.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.162
[GRAPHIC] [TIFF OMITTED] TR30AP20.163
[[Page 24433]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.164
BILLING CODE 4910-59-C
(2) Turbocharged Downsized Engines
DMCs used for the final rule analysis for the turbocharged engine
technologies were the same as NPRM costs. When these technologies are
applied to V6 and V8 non-turbocharged engines, the incremental I4 and
V6 turbocharged costs are applied, respectively. Table VI-52 below
shows the DMC used for turbochared technologies for FRM analysis in
2018 dollars.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.165
[GRAPHIC] [TIFF OMITTED] TR30AP20.166
[[Page 24434]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.167
BILLING CODE 4910-59-C
(3) Non-HEV Atkinson and Atkinson Engines
DMCs used for the final rule analysis for HCR0 and HCR1 were based
on HCR1 and HCR2 from NPRM, respectively. Discussed in Section
VI.C.1.c).(3), the agencies aligned the cost of HCR technologies to
align with 2015 NAS effectiveness and costs.
Stakeholders commented on the costs of HCR technology compared to
previous analysis. ICCT compared the NPRM costs to EPA's Proposed
Determination costs, stating that ``[t]his is a clear case where the
agencies appear to have not used the best available data from EPA which
has extensively analyzed this technology and its associated cost, nor
have the agencies justified how they have increased the associated
costs, apparently by a factor of three.'' Similarly, Roush Industries
commenting on behalf of CARB stated that the costs for implementing HCR
technology were 5-6 times the 2016 Draft TAR estimated costs, which are
``extremely high'' and ``will significantly overstate the incremental
cost and bias technology pathways.''\945\ HDS also commented that the
costs for HCR technology were higher than the costs from the 2016 Draft
TAR, and speculated that was due to ``the bulky exhaust system used in
the Mazda ATK1 engine, which apart from being expensive also requires
the vehicle to be modified to accommodate the exhaust system.''\946\
HDS cited the 2018 Camry as an example of a vehicle that does not use
the same exhaust system, but stated the sources of the new cost data
were not documented in the PRIA. ICCT stated that ``[t]he agencies
should reinstate the better justified and more deeply analyzed original
Proposed Determination HCR cost numbers from EPA for this rulemaking.''
---------------------------------------------------------------------------
\945\ NHTSA-2018-0067-11984.
\946\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
The NPRM analysis and the final rule analysis used the same DMCs
established by the 2015 NAS report for the Atkinson cycle technologies.
However, because there are many various engine configurations in the
market, the agencies do not use the same fixed costs that were set for
each type of vehicle described in the 2015 NAS report, such as pickup
and sedan. The agencies have expanded costs by taking into account the
type of technology in the baseline, like SGDI, and the configuration of
the engine, such as SOHC versus DOHC. In addition, the cost used in the
NPRM also included updated dollar year, learning rate, and RPE.
Although EPA also used costs from the 2015 NAS report for the Proposed
Determination analysis, they used a different approach to account for
components.\947\ For the final rule analysis the agencies continued to
use the same DMC for HCR technologies. Table VI-55 below shows HCR DMCs
used for the final rule analysis in 2018 dollars.
---------------------------------------------------------------------------
\947\ EPA PD TSD at 2-307 to 2-308 ``Note that the NAS costs
include the costs of gasoline direct injection (shown as ``DI'' in
the NAS report row header). EPA has removed those costs (using the
NAS reported values) since EPA accounts for those costs separately
rather than including them in the Atkinson-2 costs. Note also that
EPA always includes costs for direct injection, along with variable
valve timing and other costs, when building an Atkinson-2 package.''
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.168
[[Page 24435]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.169
[GRAPHIC] [TIFF OMITTED] TR30AP20.170
[[Page 24436]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.171
[[Page 24437]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.172
[[Page 24438]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.173
[[Page 24439]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.174
[[Page 24440]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.175
BILLING CODE 4910-59-C
(4) Advanced Cylinder Deactivation Technologies
DMCs used for the final rule analysis for the advanced cylinder
deactivation technologies were the same as NPRM costs.
Roush commented that in the NPRM analysis, the agencies did not
properly
[[Page 24441]]
consider the ``very cost-effective benefits of skip-fire technology,''
referred to in the analysis as ADEAC. Roush stated that ``due to
extremely high estimated cost ($1,250.00 in MY2016), the benefits of
this technology will likely not be chosen in any reasonable technology
pathway. If included, the predicted cost for that pathway will be
overestimated by $750-$1,000.''\948\ Similarly, Meszler commented on
the cost for the ADEAC system stating ``advanced cylinder deactivation
paths are assumed (by NHTSA) to be expensive, and are selected only in
rare instances.'' \949\ ICCT also stated ``The agencies estimated a
greatly exaggerated cost of advanced cylinder deactivation for that
level of the technology.'' \950\
---------------------------------------------------------------------------
\948\ Roush at p.13.
\949\ Meszler Comments, Attachment 2, NHTSA Docket No. NHTSA-
2018-0067-11723.
\950\ ICCT comments, NHTSA-2018-0067-11741, Page I-71.
---------------------------------------------------------------------------
The agencies do not agree with the commenter's statement that the
analysis did not consider ADEAC as a cost effective technology or that
the agencies overestimated costs for the technology. The agencies
considered the most up to date information and data for the NPRM and
final rule analysis.\951\ The agencies rely on the CAFE model to
determine technology cost effectiveness, and if the technology was cost
effective for a manufacturer to adopt, then the model would apply it to
a manufacturer's vehicle. The adoption of ADEAC was applied to vehicles
with corresponding technology combinations to reflect appropriate cost
and effectiveness, as discussed in the paragraph above. The purpose of
ADEAC is to reduce pumping losses, but if the engine has been
downsized, or has already incorporated technologies that also reduce
pumping loss, then it is likely the ADEAC has reached a point of
diminishing return. As far as the agencies are aware, Roush did not
provide alternative DMCs for ADEAC technology. Table VI-58 below shows
the examples of advanced cylinder deactivation DMC used for both
naturally aspirated and turbocharged engines for the final rule
analysis in 2018$.
---------------------------------------------------------------------------
\951\ Boha, Stani. ``Benchmarking and Characterization of a Full
Continuous Cylinder Deactivation System.'' EPA. April 10-12, 2018
SAEA World Congress. https://www.epa.gov/sites/production/files/2018-10/documents/deact-sae-world-congress-bohac-2018-04.pdf. (last
accessed Feb 12, 2020).
CARB. ``Tula Technology's Dynamic Skip Fire.'' September 28,
2016. CARB_2016 Tula ppt skipfire_NHTSA-2018-0067-11985.pdf
---------------------------------------------------------------------------
(5) Miller Cycle Engines
The agencies estimated costs for Miller cycle engines with VTG from
2016 ICCT-sponsored FEV technology cost assessment report. The agencies
considered costs from 2015 NAS study that referenced a NESCCAF 2004
report,952 953 but believed that the reference material from
the ICCT report had more updated cost estimates for this technology
that represented what was discussed in the NPRM and modeled in the
final rule analysis.
---------------------------------------------------------------------------
\952\ ``Reducing Greenhouse Gas Emissions from Light-Duty Motor
Vehicles.'' NESCCAF. September 23, 2004 Report. Available at https://www.nesccaf.org/documents/rpt040923ghglightduty.pdf/. Last accessed
Dec. 22, 2019.
\953\ ``VGT gasoline turbo, charge air cooler, piston upgrade,
piston cooling, steel crankshaft, cooling system upsize, plumbing,
rings, pressure sensor & bearing upgrade. Excludes any needed
increase in transmission torque capacity or modifications to
aftertreatment system.'' NESCCAF Report comment (2004).
---------------------------------------------------------------------------
NAS estimated the incremental cost for VTG as $525 in 2010$, but
this cost assumes many of the traditional turbocharged components and
adds VVT, VVL and SGDI. In addition, VTG (Eng23b) and VTGe (Eng23c)
engines both have similar modeled BMEP levels and a cooled EGR system
to CEGR1 (Eng14), implying that the components such as cooling systems
and piping will have similar costs.
The NAS template to calculating the final DMCs for the Miller cycle
engines for the different engine configuration is the $525 (2010$) plus
cost of cEGR1 minus cost of VVT, VVL, and SGDI. The agencies estimated
the cost for electrically-assisted variable supercharger VTGe (Eng23c)
engines based on the 2015 NAS study that uses a cost of $1050 (2010$)
plus the cost of the mild hybrid battery. For the final rule analysis,
the total costs for these technologies are shown below.
BILLING CODE 4910-59-P
[[Page 24442]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.176
[[Page 24443]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.177
[[Page 24444]]
BILLING CODE 4910-59-C
(6) Variable Compression Ratio Engines
DMCs used for the final rule analysis for the VCR engines were
based on the 2015 NAS report.\954\ The 2015 NAS reported cost for VCR
in MY2025 used a naturally aspirated engine; however, for this final
rule analysis the agencies have added cEGR and other engine
technologies to the engine. Total costs were updated to reflect 2018
dollars and MY2017 learning rate which is based on the NPRM ADEAC
learning rate. Table VI-67 below shows examples of VCR DMCs used for
this this final rule analysis in 2018 dollars.
---------------------------------------------------------------------------
\954\ 2015 NAS at p. 93.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24445]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.178
[[Page 24446]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.179
[[Page 24447]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.180
BILLING CODE 4910-59-C
(7) Diesel Engines
DMCs used for the final rule analysis for diesel engine
technologies were the same as the NPRM analysis. For DSLIAD
technologies, the agencies have added the incremental cost of ADEAC to
DSLI.
[[Page 24448]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.181
[GRAPHIC] [TIFF OMITTED] TR30AP20.182
(8) Alternative Fuel Engines
DMCs used for the final rule analysis for CNG engine technologies
were the same as the NPRM analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.183
[GRAPHIC] [TIFF OMITTED] TR30AP20.184
[[Page 24449]]
(9) Engine Lubrication and Friction Reduction Technologies
EFR costs used for the final rule analysis are based on the 2015
NAS assessment for low friction lubrication and engine friction
reduction level 2 (LUB2_EFR2). The 2015 NAS report provided estimates
of $51 (I4 DOHC), and $72 (V6 SOHC and DOHC) for midsize cars, in 2015
dollars, relative to level 1 engine friction reduction (EFR1), which
costs about $12 per cylinder. For this analysis, EFR technologies DMCs
are estimated to be $14.05 per cylinder in 2016 dollars. Total costs
were updated to reflect 2018 dollars and MY 2017 learning rate. Table
VI-74 shows the EFR DMC used for the final rule analysis in 2018
dollars.
[[Page 24450]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.185
[[Page 24451]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.186
[[Page 24452]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.187
BILLING CODE 4910-59-C
[[Page 24453]]
2. Transmission Paths
Transmissions transmit torque from the engine to the wheels.
Transmissions primarily use two mechanisms to improve fuel efficiency:
(1) A higher gear count, as more gears allow the engine to operate
longer at higher efficiency speed-load points; and (2) improvements in
friction or shifting efficiency (e.g., improved gears, bearings, seals,
and other components), which reduce parasitic losses.
There are two major categories of transmission types modeled in the
analysis: Automatic and manual. Automatic transmissions automatically
select and shift between transmission gears for the driver during
vehicle operation. The automatic transmission category is further
subdivided into four subcategories: Traditional automatic
transmissions, dual clutch transmissions, continuously variable
transmissions, and direct drive transmissions. Manual transmissions
require direct control by the driver to select and shift between gears
during vehicle operation.
Conventional planetary gear automatic transmissions (AT) are the
most popular transmission.\955\ ATs typically contain three or four
planetary gear sets that provide the various gear ratios. Gear ratios
are selected by activating solenoids which engage or release multiple
clutches and brakes as needed. ATs with gear counts ranging from five
speeds to ten speeds were considered in the NPRM and final rule
analysis.\956\
---------------------------------------------------------------------------
\955\ ``The 2018 EPA Automotive Trends Report,'' https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends, Accessed Aug 23, 2019.
\956\ Specifically, the agencies considered five-speed automatic
transmissions (AT5), six-speed automatic transmissions (AT6), seven-
speed automatic transmission (AT7), eight-speed automatic
transmissions (AT8), nine-speed automatic transmissions (AT9), and
ten-speed automatic transmissions (AT10).
---------------------------------------------------------------------------
ATs are packaged with torque converters, which provide a fluid
coupling between the engine and the driveline, and provide a
significant increase in launch torque. When transmitting torque through
this fluid coupling, energy is lost due to the churning fluid. These
losses can be eliminated by engaging the torque convertor clutch to
directly connect the engine and transmission (``lockup'').
Conventional continuously variable transmissions (CVT) consist of
two cone-shaped pulleys, connected with a belt or chain. Moving the
pulley halves allows the belt to ride inward or outward radially on
each pulley, effectively changing the speed ratio between the pulleys.
This ratio change is smooth and continuous, unlike the step changes of
other transmission varieties. CVTs were not initially chosen in the
fleet modeling for the 2012 rulemaking analysis for MYs 2017 and later
because of the predicted low effectiveness associated with CVTs (due to
the high internal losses and narrow ratio spans of CVTs in the fleet at
that time).\957\ However, improvements in CVTs in the current fleet
have increased their effectiveness, leading to increased adoption rates
in the fleet. In its 2015 report, the NAS recommended CVTs be added to
the list of considered technologies. The agencies included CVT
technology for the NPRM and this final rule analyses.
---------------------------------------------------------------------------
\957\ Morihiro, S., ``Fuel Economy Improvement by
Transmission,'' presented at the CTI Symposium 8th International
2014 Automotive Transmissions, HEV and EV Drives.
---------------------------------------------------------------------------
Dual clutch transmissions (DCT), like automatic transmissions,
automate shift and launch functions. DCTs use separate clutches for
even-numbered and odd-numbered gears, allowing the next gear needed to
be pre-selected, resulting in faster shifting. The use of multiple
clutches in place of a torque converter result in lower parasitic
losses than ATs. However, DCTs are seeing limited penetration in the
fleet, and because of the low penetration rate, only two DCTs were
considered in the analysis.
Direct drive (DD) transmissions are a direct connection between the
wheels and a drive motor. In a DD transmission, the ratio between wheel
speed and motor speed remains constant. A DD transmission is only used
in battery electric vehicles, and in the NPRM the agencies provided the
specification for comments.\958\
---------------------------------------------------------------------------
\958\ NHTSA-2018-0067-0003. ANL Autonomie Summary of Main
Component Assumptions. Aug 21, 2018. NHTSA-2018-0067-0007. Islam, E.
S, Moawad, A., Kim, N, Rousseau, A. ``A Detailed Vehicle Simulation
Process To Support CAFE Standards 04262018--Report'' ANL Autonomie
Documentation. Aug 21, 2018.Aug 21, 2018 NHTSA-2018-0067-0004. ANL
Autonomie Data Dictionary. Aug 21, 2018.
---------------------------------------------------------------------------
Manual transmissions (MT) are transmissions that require direct
control by the driver to operate the clutch and shift between gears.
Manual transmissions have seen a significant reduction in application
by automakers over recent years. As a result of the reduced market
presence, only three variants are used in the analysis.
a) Transmission Modeling in the CAFE Model
The NPRM analysis modeled pathways for applying improved technology
for each of the transmission categories and subcategories, except for
the direct drive, which was only available in the battery electric
vehicles. The MT and DCT pathways only included increasing gear counts
(e.g., 5-speed manual transmission, 6-speed manual transmission, and 7-
speed manual transmission) as improved technologies.
The traditional ATs and CVTs included both increased gear counts
and high efficiency gearbox (HEG) technology improvements as options.
HEG improvements for transmissions represent incremental advancement in
technology that improves efficiency, such as: Reduced friction seals,
bearings and clutches, super finishing of gearbox parts, and improved
lubrication. All these advancements are aimed at reducing frictional
and other parasitic loads in transmissions to improve efficiency. Three
levels of HEG improvements are considered in this analysis, based on
2015 NAS recommendations and based on CBI data.\959\ HEG efficiency
improvements were applied to ATs and CVTs, as those transmissions
inherently have higher friction and parasitic loads related to
hydraulic control systems and greater component complexity, compared to
MTs and DCTs.
---------------------------------------------------------------------------
\959\ 2015 NAS Report, at 191.
---------------------------------------------------------------------------
In total, 18 unique transmission technology combinations were
simulated, using explicit input values for gear ratios, gear
efficiencies, gear spans, shift logic, and transmission
architecture.\960\ \961\ Table VI-77 shows a list of the multi-gear
transmissions used for the NPRM.\962\
---------------------------------------------------------------------------
\960\ See PRIA Chapter 6.3.
\961\ Ehsan, I.S., Moawad, A., Kim, N., & Rousseau, A., ``A
Detailed Vehicle Simulation Process To Support CAFE Standards.''
ANL/ESD-18/6. Energy Systems Division, Argonne National Laboratory.
2018.
\962\ The NPRM and final rule also included a direct drive
transmission (single ratio) for BEVs.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24454]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.188
The technologies that made up the four transmission/level paths
defined by the modeling system for the NPRM analysis are shown in
Figure VI-22. Each vehicle model in the analysis fleet is assigned an
initial transmission type and level that most closely matches its
configuration and characteristics. The baseline-level technologies
(AT5, MT5 and CVT) appear in gray boxes and are only used to represent
the initial configuration of a vehicle's transmission in the analysis
fleet. Because there are only a few manual transmissions with less than
five forward gears in the analysis fleet, for simplicity, all manual
transmissions with five forward gears or fewer were designated MT5 for
the analysis. Similarly, all automatic transmissions with five forward
gears or fewer have been assigned the AT5 technology. For the NPRM
analysis, the agencies included a 7-speed automatic and a 9-speed
automatic to account for effectiveness of those transmissions in the
analysis fleet. These two transmissions were not available for adoption
but were available as initial configurations, and appear in gray boxes
in Figure VI-22.
[[Page 24455]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.189
BILLING CODE 4910-59-C
The model generally may apply any of the more efficient
transmission technologies that are contained within the pathway of the
baseline vehicle initial transmission configuration. The model
prohibits manual transmissions from becoming automatic transmissions.
Automatic transmissions may become CVT level 2 after progressing though
the 6-speed automatic, as shown in Figure VI-22. While the structure of
the model could allow automatic transmissions to consider applying a
DCT, the market data file was used to preclude the application of DCTs
to automatic transmission vehicles, as discussed
[[Page 24456]]
further in Section VI.C.2.c) Transmission Adoption Features, below.
The model does not attempt to simulate ``reversion'' to less
advanced transmission technologies, such as replacing a 6-speed AT with
a DCT and then replacing that DCT with a 10-speed AT. The agencies
invited comment on whether the model should be modified to simulate
``reversion'' and, if so, how this possible behavior might be
practicably simulated. Richard Rykowski, supporting comments from the
Environmental Defense Fund (EDF), broadly discussed the concept of
reversion in the CAFE model, and included an example relating to the
transmission technology paths.\963\ Mr. Rykowski stated that it is
``possible that the model could add a 10-speed transmission to a
vehicle with a very basic engine'' and then as the simulation
progressed and ``the manufacturer required greater fuel or
CO2 emission control, the Volpe Model might move to a TURBO1
or HCR engine'' and the vehicle would no longer need the 10-speed
transmission to meet standards, and a 6-speed or 8-speed transmission
might be more cost effective.
---------------------------------------------------------------------------
\963\ Comments from Environmental Defense Fund, Attachment B,
NPRM Docket No. NHTSA-2018-0067-12108, at 70.
---------------------------------------------------------------------------
The scenario discussed by Mr. Rykowski is very unlikely. The CAFE
model cost optimization algorithm considers both current and future
standard requirements when selecting current MY technologies. The
algorithm will look multiple years into the future and compare multiple
potential technology paths going forward for the most cost-effective
path. For a more detailed discussion on the cost optimization algorithm
see Section VI.A.4, Compliance Simulation.
Regarding the types of transmission technologies modeled, Meszler
Engineering Services provided a comment criticizing the limited number
of manual transmission model options and the limited technology paths
available to vehicles with manual transmissions.\964\ The agencies do
not agree with Meszler Engineering Service's assessment. The manual
transmission path includes three model options and allows for the
vehicles to receive electrification in the form of SS12V and BISG
technologies. The agencies believe the technology paths dedicated to
manual transmission was appropriate for vehicles that typically
represent manufacturers' specialty performance cars, such as the Subaru
STI or BMW M-series, that comprise an overall fleet share of less than
2 percent.
---------------------------------------------------------------------------
\964\ Comments from Meszler Engineering Services,
Attachment2_CAFE Model Tech Issues, Docket No. NHTSA-2018-0067-
11723, at 33.
---------------------------------------------------------------------------
Commenters also discussed potential missing transmission
technologies in the NPRM analysis. ICCT stated that the agencies failed
to consider transmission warm-up technologies, which are available in
3.7 million new vehicles in the MY 2016 fleet, that are being deployed
due to regulatory test-cycle benefits and off-cycle credits.\965\ In
addition, the Fiat Chrysler Automobiles (FCA) also expressed concern
over the lack of inclusion of thermal bypass devices in the modeling of
transmission technologies.\966\
---------------------------------------------------------------------------
\965\ Comments from ICCT, NPRM Docket No. NHTSA-2018-0067-11741
full comments, at I-28.
\966\ Comments from Fiat Chrysler Automobiles, Attachment 1,
NPRM Docket No. NHTSA-2018-0067-11943, at 97.
---------------------------------------------------------------------------
The agencies agree with parts of ICCT's and the FCAs comments and
disagree with other parts. The agencies do agree with ICCT and the Auto
Alliance that the analysis should consider the off-cycle benefits of
transmission warm-up technology. For the final rule analysis, the
agencies applied off-cycle technologies in the CAFE model. For the
final rule analysis, the agencies applied off-cycle technologies at the
maximum menu regulatory value of 10 g/mile for all manufacturers by MY
2023. The modeled adoption included benefits of transmission warm-up as
a menu item. The modeling of off-cycle technologies is further
discussed in Section VI.C.8. The agencies disagree with ICCT and the
Auto Alliance comments that transmission warm-up technologies were not
included in the NPRM on-cycle analysis. For the NPRM, and for the final
rule, the HEG level 2 technology package includes rapid transmission
oil warm-up technology.\967\ The inclusion of the HEG2 technology
package in AT and CVT models accounts for impacts of this technology to
performance on the standard test-cycle.
---------------------------------------------------------------------------
\967\ 2015 NAS Report, at 191.
---------------------------------------------------------------------------
For the final rule analysis the transmission model paths are shown
in Figure VI-23. For the final rule analysis, the baseline-only
technologies (MT5, AT5, AT7L2, AT9L2, and CVT) are grayed and are only
used to signify initial vehicle transmission configurations. For
simplicity, all manual transmissions with five forward gears or fewer
are assigned the MT5 technology in the analysis fleet. Similarly, all
automatic transmissions with five forward gears or fewer are assigned
the AT5 technology.
BILLING CODE 4910-59-P
[[Page 24457]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.190
Since the Manual Transmission path terminates with MT7, the system
assumes that all manual transmissions with seven or more gears are
mapped to the MT7 technology. Moreover, all dual-clutch (DCT) or auto-
manual (AMT) transmissions with five or six forward gears are mapped to
the DCT6 technology, and all DCTs or AMTs with seven or more forward
gears are mapped to DCT8.
For the final rule analysis, the naming convention for the
transmission technology models was updated to identify better the
technologies represented in each transmission. Although the
technologies in each transmission configuration were described in the
NPRM, there appears to have been confusion among some commenters about
the technology content of some transmission configurations. Some
commenters compared the NPRM AT10 to the NPRM AT8, and commented on
unexpected differences in effectiveness relative to the differences in
transmission gear count.\968\ For the given example, the NPRM AT8
represented a baseline 8-speed automatic transmission, with level 1 HEG
technology applied, and the NPRM AT10 represented a 10-speed automatic
transmission with level 2 HEG technology applied. A direct comparison
of gear count would occur by comparing the NPRM AT8L2 to the NPRM AT10.
The updated naming convention identifies the transmission technology
type, gear count and HEG technology level. Table VI-78 shows the final
rule names for transmission models compared to the names used for the
NPRM analysis.
---------------------------------------------------------------------------
\968\ Comments from CARB, Attachment 2018-10-26 FINAL CARB
Detailed Comments on SAFE, NPRM Docket No. NHTSA-2018-0067 at 110-
13.
---------------------------------------------------------------------------
[[Page 24458]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.191
BILLING CODE 4910-59-C
b) Transmission Analysis Fleet Assignments
The agencies discussed in the NPRM the process for developing the
2016 analysis fleet, including how the agencies weighed using
confidential business information versus publicly-releasable sources,
the use of compliance data, and decision to use a 2016 analysis fleet
over other alternatives.\969\ As discussed above, this final rule
analysis used the 2017 vehicle fleet as the analysis fleet input, and
the agencies followed largely the same process for assigning initial
transmission assignments as in the NPRM.
---------------------------------------------------------------------------
\969\ 83 FR 43003.
---------------------------------------------------------------------------
For the 2017 analysis fleet, transmission data was gathered from
the manufacturer final model year CAFE compliance submissions to the
agencies as well as manufacturer press releases. The data for each
manufacturer was used to determine which platforms shared transmissions
and to establish the leader-follower relationships between vehicles.
Within each manufacturer fleet, transmissions were assigned unique
identification designations based on technology type, drive type, gear
count, and technology version. The data were also used to identify the
most similar transmission among the Autonomie transmission models, as
discussed further below.
The transmission characteristics of vehicles in the analysis fleet
show manufacturers use transmissions that are the same or similar on
multiple vehicle models. Manufacturers have told the agencies they do
this to control component complexity and associated costs for
development, manufacturing, assembly, and service. Both the NPRM and
final rule analyses account for this sharing. To identify common
transmissions, the agencies considered the transmission type (manual,
automatic, dual-clutch, continuously variable), number of gears, and
vehicle architecture (front-wheel-drive, rear-wheel-drive, all-wheel-
drive based on a front-wheel-drive platform, or all-wheel-drive based
on a rear-wheel-drive platform). If multiple vehicle models shared
these attributes, the transmissions were treated as single group for
the analysis. Vehicles in the analysis fleet with the same transmission
configuration adopted transmission technology together.
For ATs and CVTs, the identification of the most similar Autonomie
transmission model required additional steps beyond just assigning gear
count for ATs, or just assigning the CVT model. A review of the age of
the transmission design, relative performance versus previous designs,
and technologies incorporated was conducted, and the information
obtained was used to assign a HEG level. Engineering judgment was used
to compare the technologies and performance improvements reported
versus descriptions of HEG technology discussed in the NAS report.\970\
---------------------------------------------------------------------------
\970\ 2015 NAS Report, at 191.
---------------------------------------------------------------------------
In addition, no automatic transmissions in the 2017 analysis fleet
were determined to be initially at a HEG Level 3. However, all 7-speed
automatic transmissions, all 9-speed automatic transmissions, all 10-
speed automatic transmissions and some 8-speed automatic transmissions
were found to be advanced transmissions operating at a Level 2 HEG
equivalence. All other transmissions were assigned at the minimum
level.
c) Transmission Adoption Features
The agencies included several transmission adoption features in the
NPRM that have been carried over for the final rule analysis. For a
detailed discussion of path logic applied in the final rule analysis,
including technology supersession logic and technology mutual
exclusivity logic, please see FRM CAFE Model Documentation Section
S4.5, Technology Constraints
[[Page 24459]]
(Supersession and Mutual Exclusivity).\971\
---------------------------------------------------------------------------
\971\ Available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
(1) Automatic Transmissions
Automatic transmission technology adoption is defined by path logic
and technology availability. The transmission path precludes adoption
of other transmission types once a platform progresses past an AT6.
This restriction is used to avoid the significant level of stranded
capital that could result from adopting a completely different
transmission type shortly after adopting an advanced transmission,
which would occur if a different transmission type was adopted after
AT6 in the rulemaking timeframe. Stranded capital is discussed in more
detail in Section VI.B.4.c), Stranded Capital Costs. In addition, any
automatic transmissions that use HEG3 technology cannot be phased in
until the 2020 model year. The technology phase-in year is based on the
estimated availability of HEG3 technology from the NAS (2015) report
and confidential data obtained from OEM's and suppliers. Finally, all
P2HEVs are paired with an AT8 transmission, which is also discussed
further in Section VI.C.3.c).
One commenter expressed concern that all P2HEVs were paired with an
AT8 transmission, and argued that the full slate of transmission
technology should be available for adoption with that powertrain
technology.\972\ The commenter correctly observed a limit of
transmission technologies for use only with the P2HEV technology
option; all other HEV based technology options did not have this
limitation.
---------------------------------------------------------------------------
\972\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723 at 32.
---------------------------------------------------------------------------
The agencies disagree that a greater variety of transmission
technologies are necessary to model the P2HEV technology reasonably.
The P2HEV demonstrated limited response to transmission technologies
beyond the AT8L2, and access to those technologies were limited to
reflect the diminishing returns anticipated for higher gear counts used
in conjunction with the P2 system, and trends in industry.\973\
Adopting P2HEV to a conventional vehicle provides a significant fuel
consumption improvement, agnostic of transmission type, based on the
agencies' full vehicle simulation results.
---------------------------------------------------------------------------
\973\ Greimel, H. ``ZF CEO--We're not chasing 10-speeds,''
Automotive News, November 23, 2014, http://www.autonews.com/article/20141123/OEM10/311249990/zf-ceo:-were-not-chasing-10-speeds.
---------------------------------------------------------------------------
(2) Continuously Variable Transmissions
Application of CVTs in the NPRM and final rule analysis was not
allowed for high torque vehicle applications. The launch, acceleration,
and ratio variation characteristics of powertrains with CVTs may be
significantly different than ATs leading to potential consumer
acceptance issues and/or complaints. Several manufacturers have told
the agencies that they employ strategies that mimic AT shifting under
some conditions to address these issues. Some manufacturers have also
encountered significant engineering challenges in employing CVTs for
use in high torque or high load applications.
In addition, the CVT adoption was limited by technology path logic.
CVTs cannot be adopted by vehicles that do not start with a CVT or by
vehicles beyond the AT6 in the baseline fleet which have a greater
number of gear ratios and therefore increased ability to operate the
engine at a highly efficient speed and load. Once on the CVT path the
platform is only allowed to apply improved CVT technologies. This
restriction is used to avoid the significant level of stranded capital
that could result from adopting a completely different transmission
type shortly after adopting an advanced transmission, which would occur
if a different transmission type was adopted in the rulemaking
timeframe. Stranded capital is discussed in more detail in Section
VI.B.4.c), Stranded Capital Costs.
The Alliance commented that the analysis ``appropriately restricts
the application of CVT technology on larger vehicles.'' \974\ The
agencies concurred with the Alliance's observations and thus the
limitations on CVT application were continued in the final rule
analysis.
---------------------------------------------------------------------------
\974\ Comments from Auto Alliance, Attachment 1, NHTSA-2018-
0067-12073, at 142.
---------------------------------------------------------------------------
(3) Dual Clutch Transmission
For DCTs, while the structure of the model could allow automatic
transmissions to consider applying a DCT, the market data file was used
to preclude the application of DCTs to vehicles that had already
adopted an automatic transmission with six or more gears (e.g., AT6
through AT10). The model allows baseline vehicles that have DCTs to
apply an improved DCT (if opportunities to do so exist), and allows
vehicles with an AT5 to consider DCTs. This was done to ensure vehicle
functionality is maintained as technologies are applied, and accounts
for consumer acceptance issues related to the drivability and launch
performance tradeoffs. These issues with DCTs resulted in a low
relative adoption rate over the last decade.\975\ It also is broadly
consistent with manufacturers' technology choices.
---------------------------------------------------------------------------
\975\ ``The 2018 EPA Automotive Trends Report,'' Page 60, figure
4.18, https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends, Accessed Aug 23, 2019.
---------------------------------------------------------------------------
(4) Manual Transmissions
Manual transmission technology adoption in the CAFE model remained
unchanged from the NPRM and is only limited by the technology path
limits discussed above. Manual transmissions cannot be adopted by
vehicles that do not start with a manual transmission in the analysis
fleet. Vehicles with manual transmissions cannot receive an alternate
transmission technology, and may only progress to more advanced manual
transmissions. These restrictions are in recognition of the low
customer demand for manual transmissions.\976\
---------------------------------------------------------------------------
\976\ ``The 2018 EPA Automotive Trends Report,'' https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends, Accessed Aug 23, 2019.
---------------------------------------------------------------------------
d) Transmission Effectiveness Modeling and Resulting Effectiveness
Values
For the NPRM and final rule analysis, full vehicle simulation was
used to understand how transmissions work within the full vehicle
system to improve fuel economy, and how changes to the transmission
subsystem influence the performance of the full vehicle system.
The Autonomie tool models transmissions as a sequence of mechanical
torque gains. The torque and speed are multiplied and divided,
respectively, by the current ratio for the selected operating
condition. Furthermore, torque losses corresponding to the torque/speed
operating point are subtracted from the torque input. Torque losses are
defined based on a three-dimensional efficiency lookup table that has
as inputs: Input shaft rotational speed, input shaft torque, and
operating condition.\977\
---------------------------------------------------------------------------
\977\ Detailed discussion of transmission modeling can be found
in the ANL Model Documentation at Chapter 4 and Chapter 5.
---------------------------------------------------------------------------
The general transmission models are populated with characteristics
data to model specific transmissions. Characteristics data are
typically provided in the form of tabulated data for transmission gear
ratios, maps for transmission efficiency, and maps for torque converter
performance, as applicable. The quantity of data needed
[[Page 24460]]
depends on the transmission technology being modeled. The
characteristics data for these models was collected from peer-reviewed
sources, transmission and vehicle testing programs, results from
simulating current and future transmission configurations, and
confidential data obtained from OEMs and suppliers.\978\
---------------------------------------------------------------------------
\978\ Downloadable Dynamometer Database.: https://www.anl.gov/energy-systems/group/downloadable-dynamometer-database, Kim, N.,
Rousseau, N., Lohse-Bush, H., ``Advanced Automatic Transmission
Model Validation Using Dynamometer Test Data,'' SAE 2014-01-1778,
SAE World Congress, Detroit, April 2014. Kim, N., Lohse-Bush, H.,
Rousseau, A., ``Development of a model of the dual clutch
transmission in Autonomie and validation with dynamometer test
data,'' International Journal of Automotive Technologies, March
2014, Volume 15, Issue 2, pp 263-271.
---------------------------------------------------------------------------
The level of HEG improvement applied to a given transmission was
modeled by improvements made to the efficiency map of the transmission.
As an example, the 8-speed automatic transmission models show how a
model can be incrementally improved with the addition of the HEG
enhancement. The AT8 is the model of a baseline transmission developed
from a transmission characterization report.\979\ The AT8L2 has the
same gear ratios as the AT8, however the gear efficiency map has been
improved to represent application of the HEG level 2 technologies. The
AT8L3 models the application of HEG level 3 technologies using the same
principle, further improving the gear efficiency map over the AT8L2
improvements.
---------------------------------------------------------------------------
\979\ See PRIA Section 6.3.3.2
---------------------------------------------------------------------------
The NPRM and final rule analysis, using the Autonomie tool,
comprehensively simulated each of the 18 transmission technologies.
Each transmission was modeled with explicit gear ratios, gear
efficiencies, gear spans, adaptive shift logic, and transmission
architecture individually for each of the ten vehicle types. The NPRM
and final rule analysis clearly showed the specific contributions to
effectiveness provided by each transmission technology combination and
the associated cost. This provided greater transparency for public
review and comment.
The implementation of the full vehicle simulation approach used in
the NPRM analysis, and carried forward to the final rule analysis,
clearly defines the contribution of individual transmission
technologies and separates those contributions from other technologies.
This modeling approach comports with the National Academy of Science
2015 recommendation to use full vehicle modeling supported by
application of collected improvements at the sub-model level.\980\ The
approach allows the isolation of technology effects in the analysis
which contributes to an accurate cost assessment.
---------------------------------------------------------------------------
\980\ 2015 NAS Report, at 292.
---------------------------------------------------------------------------
This approach was supported by the Auto Alliance, who commented in
support of the agencies' explicit and transparent modeling of the cost
and effectiveness for each of the transmission technologies. The
Alliance contrasted the NPRM approach with the transmission modeling
methodology used in the Proposed Determination--which they strongly
objected to--which had lumped together fundamentally different
transmission technologies into bundles with identical cost and
efficiencies, ``making it impossible to fully comprehend the
rationale'' for the Proposed Determination's high effectiveness
estimates.\981\
---------------------------------------------------------------------------
\981\ Comments from Alliance of Automobile Manufacturers, NHTSA-
2018-0067-12073, at 142.
---------------------------------------------------------------------------
However, other stakeholders were not supportive of the modeling
approach used in the NPRM. The Union of Concerned Scientists (UCS)
thought a level of abstraction was necessary to account for
unpredictability in the market, such as the failure of the dual-clutch
transmission to reach widespread use as anticipated in the agencies
2012 analysis for MYs 2017 and later. UCS thought that keeping the
transmission technology generalized would avoid the pitfalls of
potentially picking the wrong technology leader, but would still
predict the general trend of behavior, stating that ``[i]ncidentally,
this is an example of why we supported EPA's move to a more generic
representation of transmissions in its OMEGA modeling.'' \982\
---------------------------------------------------------------------------
\982\ Comments from Union of Concerned Scientists, NHTSA-2018-
0067-12039, at 20-21.
---------------------------------------------------------------------------
The agencies disagree with UCS's suggestion to generalize the
transmission technology groupings for the analysis. By grouping the
technologies into overly broad, generic categories, the analysis loses
accuracy on the costs and the effectiveness for specific systems. The
OMEGA model used general transmission categories, asked for by UCS's
comments, as part of the CO2 analysis in the Draft TAR and
in the Proposed Determination, and the assumptions and limitations were
acknowledged at the time.983 984 One assumption used by the
OMEGA model approach was ``[t]he incremental effectiveness and cost for
all automated transmissions are based on data from conventional
automatics.'' \985\ In response, the Alliance observed that the
transmission groups used ``do not recognize unique efficiencies of
different transmission technologies.'' \986\ At the time EPA stated
``the potential effectiveness gains between TRX levels, while arising
from different technology packages within each transmission type, will
be very similar among the transmission types.'' \987\ However, as shown
in Table VI-81 and Table VI-82, there are nontrivial differences in the
costs of different transmission technologies.
---------------------------------------------------------------------------
\983\ ``Midterm Evaluation of Light duty Vehicle Greenhouse Gas
Emission Standards and Corporate Average Fuel Economy Standards for
Model Years 2022-2025,'' Paragraph 5.3.4.2.1, EPA-420-D-16-900, July
2016.
\984\ ``Proposed Determination on the Appropriateness of the
Model Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions
Standards under the Midterm Evaluation, Technical Support
Document,'' Pages 2-328--2-329, EPA-420-R-16-021, November 2016.
\985\ ``Proposed Determination on the Appropriateness of the
Model Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions
Standards under the Midterm Evaluation, Technical Support
Document,'' Pages 2-327, EPA-420-R-16-021, November 2016.
\986\ ``Proposed Determination on the Appropriateness of the
Model Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions
Standards under the Midterm Evaluation, Technical Support
Document,'' Pages 2-329, EPA-420-R-16-021, November 2016.
\987\ ``Proposed Determination on the Appropriateness of the
Model Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions
Standards under the Midterm Evaluation, Technical Support
Document,'' Pages 2-329, EPA-420-R-16-021, November 2016.
---------------------------------------------------------------------------
The approach used in the NPRM analysis and this final rule analysis
is an evolution of the approach used for the Proposed Determination
model, and avoids the issue described above. The NPRM and final rule
analyses reduce the span of transmission technology groupings, with the
intent to provide an increase in fidelity and precision for cost and
performance, as was requested by stakeholders such as the Auto
Alliance, while including tools to mitigate market effects, which
addresses other concerns such as those expressed by UCS. In the
analysis for the final rule the transmissions are grouped by technology
type (AT, DCT, CVT, etc.) and gear count (5,6,7, etc.). The level of
HEG technology applied as a separate factor further subdivided the
transmission groups. Defining technology adoption features addresses
the potential for market forces, such as those that affected the sales
of DCTs, and supports the narrower technology groupings. Technology
adoption features are defined through market research, historic and
current fleet composition analysis, and dialogue with manufacturers.
Commenters also provided general comments regarding the values of
effectiveness for advanced transmissions used for the NPRM
[[Page 24461]]
analysis versus values used for the Draft TAR. For example, CARB noted
a ``2 percent-3 percent lower efficiency assumed for advanced 8- and 9-
speed transmissions relative to the data EPA itself previously
developed with back to back testing on FCA vehicles,'' \988\ with
similar concerns expressed by other commenters.\989\ Meszler
Engineering Services wondered ``why the AT10 technology was being so
widely adopted when its associated benefits appeared negligible for a
particular vehicle'' and noted ``[t]he wide ranging effectiveness
estimates were unexpected.'' \990\ Senator Tom Carper also noted ``the
most advanced eight speed transmission technology are assigned
unrealistically low fuel efficiency effectiveness values for some
vehicle types.'' \991\
---------------------------------------------------------------------------
\988\ Comments from CARB, Attachment 2018-10-26 FINAL CARB
Detailed Comments on SAFE, NPRM Docket No. NHTSA-2018-0067-11873, at
110-113.
\989\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 5; Comments from CARB, Attachment HDS
Final Report, NPRM Docket No. NHTSA-2018-0067-11985, at 26, 47.
\990\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723, at 5-6.
\991\ Comments from Senator Tom Carper, Attachment 1, NPRM
Docket No. NHTSA-2018-0067-11910, at 4.
---------------------------------------------------------------------------
The Auto Alliance also provided comments with regards to the larger
variation of effectiveness values that were of concern to commenters
such as Meszler Engineering Services and Senator Tom Carper. The Auto
Alliance acknowledged that the use of full vehicle simulation, with
more details, results in greater diversity of results. The comment
stated, ``Over an entire fleet, a more reasonable expectation is that
there will be some vehicles with higher fuel economy than expected for
a given technology set and some vehicles with a lower fuel economy than
expected for a given technology set. As discussed above, these
differences arise for a variety of reasons, and cannot simply be
attributed to ``less than optimal technology integration.'' \992\
---------------------------------------------------------------------------
\992\ Comments from Alliance of Automobile Manufacturers,
Attachment 1, NPRM Docket No NHTSA-2018-0067-12385, at 9.
---------------------------------------------------------------------------
The Auto Alliance also specifically commented on the FCA vehicle
study used to support CARB's comment and used to generate the TAR
analysis values. The Auto Alliance pointed out that the vehicles used
in the study had other technology differences, however the study still
``proceeds to compare the fuel economy of these variants to assert
support for its own estimate of transmission effectiveness. This
comparison neglects that the 2.4L engines in these variants are not the
same and that the variant with the nine-speed transmission was a
redesigned vehicle.'' The Alliance concluded, therefore, that ``the
Chrysler 200 comparison provided by H-D Systems does not compare a
transmission change in isolation from other changes that impact fuel
economy and likely overestimates the benefits associated with the
transmission change.'' The Auto Alliance summarized the analysis of the
study by noting that ``[s]uch differences also impact fuel economy,
confounding an analysis which purports to compare the fuel economy
benefits associated directly with the transmission.'' \993\
---------------------------------------------------------------------------
\993\ Comments from Alliance of Automobile Manufacturers,
Attachment 1, NPRM Docket No NHTSA-2018-0067-12385, at 27-28.
---------------------------------------------------------------------------
The agencies agree with the Auto Alliance assessment of the 8- and
9-speed FCA vehicles, and have based analysis inputs on alternate
information sources.\994\ However, the observations by commenters of a
wider range of values for the NPRM effectiveness when compared to the
Draft TAR compliance analyses are a direct result of the improvements
in modeling approach. As discussed above the NPRM compliance analysis
increased the number of transmission technology paths considered by
further subdividing the technology groupings. The change resulted in a
wider range of effectiveness, as the specific transmission technologies
are paired across all the configurations of vehicle technologies. In
addition to this greater range, there were also specific effectiveness
issues identified for some of the transmission technologies, which are
addressed in the sections below.
---------------------------------------------------------------------------
\994\ See Data discussed in PRIA Section 6.3.3.2. and Kim, N.,
Rousseau, N., Lohse-Bush, H. ``Advanced Automatic Transmission Model
Validation Using Dynamometer Test Data,'' SAE 2014-01-1778, SAE
World Congress, Detroit, April 2014. Kim, N., Lohse-Bush, H.,
Rousseau, A. ``Development of a model of the dual clutch
transmission in Autonomie and validation with dynamometer test
data,'' International Journal of Automotive Technologies, March
2014, Volume 15, Issue 2, pp 263-271.
---------------------------------------------------------------------------
Commenters may also be observing, with comments like ``advanced
transmissions have low effectiveness with some vehicles types,'' an
expected effect when an advanced transmission is coupled to an advanced
engine. The National Academy of Science, in their 2015 report, noted
that ``as engines incorporate new technologies to improve fuel
consumption, including variable valve timing and lift, direct
injection, and turbocharging and downsizing, the benefits of increasing
transmission ratios or switching to a CVT diminish.'' \995\ This is not
to say that transmissions are not an important technology going
forward, but rather a recognition that advanced engines have larger
``islands'' of low fuel consumption that rely less on the transmission
to improve the overall efficiency of the vehicle. Thus, effectiveness
percentages reported for transmissions paired with unimproved engines
would be expected to be reduced when the same transmission is paired
with a more advanced engine.
---------------------------------------------------------------------------
\995\ 2015 NAS Report, at 175.
---------------------------------------------------------------------------
Commenters also expressed concern for the transmission gear set and
final drive values used for the NPRM analysis, or, more specifically,
that the gear ratios were held constant across applications. Roush
commented that ``all transmissions with a given number of ratios (8-
speed, 10-speed) maintain the same individual step ratios'' and that
this would lead to ``powertrain inefficiencies and under-predict
potential fuel economy benefits.'' \996\ CARB, quoting a report from
its contractor, noted that ``the final drive ratio was kept constant as
powertrains were changed and that transmission gear ratios were not
optimized,'' and suggested that manufacturers forgoing improvements
from gear ratio or final drive ratio changes is unrealistic and results
in an underestimation of the benefits from advanced transmissions.\997\
---------------------------------------------------------------------------
\996\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 14-15.
\997\ Comments from CARB, Attachment 1, NPRM Docket No. NHTSA-
2018-0067-11873, at 110.
---------------------------------------------------------------------------
However, the Auto Alliance stated that ``[m]anufacturers share
major technologies such as transmissions and engines across multiple
vehicle models and platforms.'' The Auto Alliance also supported the
agencies' approach of not including final drive ratio changes,
particularly when only minor system changes are incurred. The Auto
Alliance continued further stating that ``[i]n the case of passenger
cars, the final drive ratio is frequently the same across multiple
models that use the same transmission.'' \998\
---------------------------------------------------------------------------
\998\ Comments from Auto Alliance, Attachment 1, NPRM Docket No.
NHTSA-2018-0067-12073, at 142.
---------------------------------------------------------------------------
The agencies disagree with Roush, Duleep, and CARB's assessment. It
is an observable practice in industry to use a common gear set across
multiple platforms and applications. The most recent example is the GM
10L90, a 10-speed automatic transmission that used the same gear set in
both pick-up truck
[[Page 24462]]
and passenger car applications.\999\ Optimization of performance is
achieved through shift control logic rather than customized hardware
for each vehicle line. The use of a single gear set for each
transmission technology also supports the overall analysis approach.
The level of technology performance modeled must reasonably represent a
typical level of performance representative of the industry range of
performance. If the systems were over-optimized for the agencies'
modeling, such as applying a unique gear set for each individual
vehicle configuration, the analysis would likely over-predict the
reasonably achievable fuel economy improvement for the technology.
Over-prediction would be exaggerated when applied under real-world
large-scale manufacturing constraints necessary to achieve the
estimated costs for the transmission technologies. Accordingly, the
agencies used the NPRM approach for the final rule analysis.
---------------------------------------------------------------------------
\999\ ``GM Global Propulsion Systems--USA Information Guide
Model Year 2018'' (PDF). General Motors Powertrain. Retrieved 26
September 2019. https://www.gmpowertrain.com/assets/docs/2018R_F3F_Information_Guide_031918.pdf.
---------------------------------------------------------------------------
In response to comments related to the effectiveness of micro-HEV
systems, which are discussed in Section VI.C.3.d)(2)(a), and comments
related to the effectiveness of diesel engines, which are discussed in
Section VI.C.1.c)(8), the agencies took a close look at NPRM
effectiveness results. Two issues were identified related to the
interaction between Autonomie transmission models and other Autonomie
powertrain technology models. First, a logic issue was found in a
transmission control subroutine and, second, there was an issue with a
sub-model input. While these items were caused by issues in the
transmission model sub-systems, the effects manifested in the
effectiveness of the micro-HEV systems and the diesel engine systems.
Autonomie uses a gearbox transient sub-model to control the simulated
state of powertrain components during a transmission event, such as
shifting or vehicle starting and stopping. The simulated powertrain
component states include conditions such as clutch engagement, or
engine operation mode. A detailed discussion of the Autonomie control
model can be found FRM Argonne Model Documentation file at Section 4.4.
Different versions of the sub-model are used for micro-HEV technologies
(12VSS and ISG) than for conventional drivetrains, mild-HEV or Strong-
HEV systems.
An issue was found in the control logic used in the micro-HEV
version related to the sequence of powertrain component modes during
shifting events for automatic transmissions, regenerative braking
events for automatic transmissions, and stop start events for manual
transmissions. While these issues reduced the effectiveness of the
micro-HEV technology in the Argonne modeling results, they had very
minimal effect on the overall NPRM Analysis. The control logic issue
was resolved for the final rule analysis. There also was an issue with
the gearbox transient sub-model used for micro HEVs that impacted
calculation of the CVT best efficiency operating ratio targets under
low torque conditions. This resulted in some negative effectiveness
values for certain CVT technology combinations, but had very minimal
effect on the overall NPRM results. This software item was also
resolved for the final rule analysis.
As discussed in the Autonomie model documentation, FRM Argonne
Model Documentation file at Section 4, the full vehicle model is
created from a network of subsystem models. The subsystems all interact
through data connections transferring outputs from one subsystem model
to the inputs of another. An issue was identified with the definition
of the connection between the gearbox transient sub-model for DCT's
with diesel engines, which impacted the values provided to the diesel
control model. This caused reduced effectiveness values for the diesel
engines with DCTs in the Argonne modeling results, however it had very
minimal effect on the overall NPRM analysis. The data connection issue
was resolved for the final rule analysis.
Lastly, the agencies received several comments on transmission
shifting logic, which are addressed in the following section.
(1) Shift Logic
Transmission shifting logic has a significant impact on vehicle
energy consumption and was modeled in Autonomie to maximize the
powertrain efficiency while maintaining acceptable drive quality. The
logic used in the Autonomie full vehicle modeling relied on two
components: (1) The shifting controller, which provides the logic to
select appropriate gears during simulation; and (2) the shifting
initializer, an algorithm that defines shifting maps (i.e., values of
the parameters of the shifting controller) specific to the selected set
of modeled vehicle characteristics and modeled powertrain
components.\1000\
---------------------------------------------------------------------------
\1000\ See FRM ANL Model Documentation file at Paragraph 4.4.5.
---------------------------------------------------------------------------
(a) Shifting Controller
The shift controller is the logic that governs shifting behavior
during simulated operation. The shift controller performance was
informed by inputs from the model. The inputs included: Specific engine
or transmission used, and instantaneous conditions in the simulation.
Instantaneous conditions included values such as vehicle speed, driver
demand and a shifting map unique to the full vehicle
configuration.\1001\ The shift controller logic was consistently
applied for all vehicles simulated.
---------------------------------------------------------------------------
\1001\ See FRM ANL Model Documentation file at Paragraph 4.4.5.
---------------------------------------------------------------------------
Although no comments were received specifically on shift control
logic, the agencies tracked several effectiveness concerns identified
by commenters back to how the agencies modeled some transmissions
paired with turbocharged engines. Meszler Engineering Services
discussed an unexpected range of effectiveness observed for
transmissions when coupled to different engine technologies, and
concluded that ``[m]oreover, the variation across technology
combinations is markedly different.'' \1002\ Senator Carper's comments
mirrored Meszler's, noting that ``the more expensive version of an
engine technology (TURBO2), which would be expected to be more fuel-
efficient, was instead assigned a negative fuel-efficiency value for
some types of vehicles.'' \1003\ The Senator also observed the same
phenomenon for cooled exhaust gas recirculation (CEGR I), which ``was
assigned a fuel-efficiency effectiveness of at or near zero.''
Similarly, UCS noted that ``many simulations of improved transmissions
and turbocharged engines show little incremental improvement over less
complex technologies.'' \1004\
---------------------------------------------------------------------------
\1002\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723, at 5-6.
\1003\ Comments from Senator Tom Carper, Attachment 1, NPRM
Docket No. NHTSA-2018-0067-11910, at 4.
\1004\ Comments from UCS, Attachment 1, NPRM Docket No. NHTSA-
2018-0067-12039, at 32.
---------------------------------------------------------------------------
In response to the comments, the agencies conducted an in-depth
review of these technology combinations. The agencies determined the
minimum lugging speed for turbocharged engines, which controls the
minimum engine speed allowed before down-shifting, caused the observed
behavior. The issue was isolated to some combinations of advanced
transmissions and
[[Page 24463]]
turbocharged engines. For the final rule analysis, a modification was
made to the shift controller logic of transmissions coupled to
turbocharged engines. Specifically, the minimum lugging speed allowed
for turbocharged engines was increased in the shift controller. An
increase in lugging speed increases the minimum speed at which the
shift controller will allow the engine to operate before down-shifting,
resulting in increased operation in better efficiency regions of the
engine map.\1005\ The updated lugging speeds are based on Argonne
benchmarking data of the 2017 F150.\1006\ The updated values are shown
in Table VI-79, the lugging speeds for naturally aspirated engines are
shown as reference and remain unchanged from the NPRM.
---------------------------------------------------------------------------
\1005\ See FRM ANL Model Documentation at Paragraph 4.4.5.1, for
more details on lugging speed.
\1006\ NHTSA Benchmarking, ``Laboratory Testing of a 2017 Ford
F-150 3.5 V6 EcoBoost with a 10-speed transmission.'' DOT HS 812
520.
[GRAPHIC] [TIFF OMITTED] TR30AP20.192
(b) Shift Initializer
As defined above, the shifting initializer is an algorithm that
defines shifting maps (i.e., values of the parameters of the shifting
controller) specific to the selected set of modeled vehicle
characteristics and modeled powertrain components.
Commenters stated that the model did not customize shifting maps
for each transmission application. Roush Industries commented, ``[t]he
2018 PRIA analysis assumes that all transmissions with a given number
of ratios maintain the same individual step ratios and shift maps.''
\1007\ Roush also commented that the effectiveness of transmissions
were understated due to inaccurate transmission maps or ``the lack of
vehicle system optimization and calibration.'' \1008\ UCS stated that
the ``transmission shift strategy does not deploy gear-skipping or
other more modern control strategies.'' \1009\ HDS provided similar
comments to Roush, observing that the Autonomie models ``do not
optimize engine efficiency after most changes in tractive load because
the model employs fixed shift points, gear ratios, and axle ratios.''
\1010\ Finally, CARB expressed that ``[f]or the Autonomie modeling, a
fixed final drive ratio was utilized and, presumably, a fixed shift
logic based on the selected transmission.'' \1011\
---------------------------------------------------------------------------
\1007\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 14-15.
\1008\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 5.
\1009\ Comments from UCS, Attachment 1, NPRM Docket No. NHTSA-
2018-0067-12039, at 23.
\1010\ Comments from K. Gopal Duleep, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-12395, at 4-5.
\1011\ Comments from CARB, Attachment 2018-10-26 FINAL CARB
Detailed Comments on SAFE, NPRM Docket No. NHTSA-2018-0067-11873, at
185.
---------------------------------------------------------------------------
The commenters seem to conflate the practice in the analysis of
using the same gear sets across vehicle configuration with using the
same shift maps. As commenters stated, they assumed the same maps were
applied across vehicle models. However, the shift initializer routine
was run for every unique Autonomie full vehicle model configuration and
generated customized shifting maps. The algorithms' optimization was
designed to balance minimization of energy consumption and vehicle
performance.\1012\ This balance was necessary to achieve the best fuel
efficiency while maintaining customer acceptability by meeting
performance neutrality requirements, as discussed in Performance
Neutrality, Section VI.B.3.a)(6).
---------------------------------------------------------------------------
\1012\ See FRM ANL Model Documentation at Paragraph 4.4.5.2.
---------------------------------------------------------------------------
While discussing shift logic, commenters also expressed concern
about the capturing of fuel efficiency losses associated with shifting
events. Roush stated, ``[t]he 2018 PRIA transmission modeling does not
accurately capture the losses and FE penalty associated with a shift
event.'' \1013\ The agencies disagree with this statement. While losses
associated with a shifting event are not modeled as a single factor,
the mechanisms that cause the loss are appropriately incorporated in
the Autonomie transmission models.
---------------------------------------------------------------------------
\1013\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 14-15.
---------------------------------------------------------------------------
[[Page 24464]]
The automatic transmission models have an associated torque converter
model.\1014\ The torque converter model is designed to simulate the
inertial and torque loads imposed on an engine because of shift events.
Other clutch-based transmission models, MTs and DCTs, apply a general
loss of efficiency across transmission efficiency maps to account for
losses due to shift events.
---------------------------------------------------------------------------
\1014\ See FRM ANL Model Documentation at Paragraph 4.5 and
Paragraph 5.4.
---------------------------------------------------------------------------
(2) Transmission Effectiveness Values
The NPRM technology effectiveness modeling results showed that the
effectiveness of a technology often varies with the type of vehicle and
the other technologies that are on the vehicle. Figure VI-24 shows the
range of effectiveness for each transmission technology across the
range of vehicle types and technology combinations in the NPRM
analysis. The data reflect the change in effectiveness for applying
each transmission technology by itself while all other technologies are
held unchanged. The effectiveness improvement range is over a 5-speed
automatic transmission.
[GRAPHIC] [TIFF OMITTED] TR30AP20.193
(a) Automatic Transmissions
Regarding AT effectiveness values, commenters pointed out the
unusually high level of effectiveness displayed by the AT6L2
transmission. ICCT and UCS both specifically expressed concern with the
effectiveness of the AT6L2 compared to other advanced
transmissions.1015 1016 The performance of the AT6L2 was
central to ICCT's analysis of the NPRM inputs, which highlighted the
AT6L2 models' performance, showing the cost versus effectiveness of the
AT6L2 outperformed more advanced transmission options.\1017\
---------------------------------------------------------------------------
\1015\ Comments from International Council on Clean
Transportation, Attachment 3, NPRM Docket No. NHTSA-2018-0067-11741,
at I-26, I-64 (`` ``However, the impact of adding level 2
transmission efficiency technologies varies wildly and produces
absurd results. A 6-speed AT6L2 Is modeled as much more efficient
(12.0% improvement) than a comparable 8-speed AT8L2 (9.1%) and even
slightly more efficient than a comparable 10-speed AT10L2
(11.5%).'')%).''.
\1016\ Comments from Union of Concerned Scientists, Attachment
1, NPRM Docket No. NHTSA-2018-0067-12039, at 32. (``[I]n the NPRM
analysis, 0 percent of vehicles had an AT6L2 transmission while 52.4
percent adopted AT10L2 transmissions, even though the latter
supplies virtually identical modeled efficiency.'').
\1017\ Comments from International Council on Clean
Transportation, Attachment 3, NPRM Docket No. NHTSA-2018-0067-11741,
at I-64--I-65.
---------------------------------------------------------------------------
Evaluation of the AT6L2 transmission model in response to these
comments revealed an overestimated efficiency map was developed for the
NPRM model. The high level of efficiency assigned to the transmission
surpassed benchmarked advanced transmissions.\1018\ To address the
issue, the agencies replaced the effectiveness values of the AT6L2
model for the final rule analysis with AT7L2 effectiveness values.
---------------------------------------------------------------------------
\1018\ See PRIA Section 6.3.3.2. Sources of Transmission
Effectiveness Data.
---------------------------------------------------------------------------
The updated estimate of effectiveness is supported by values shown
in the NAS 2015 analysis.\1019\ The study estimated the difference in
effectiveness between a 6-speed automatic transmission and a 7-speed
automatic transmission of approximately the same technology level to be
0.8 percent. The difference is reduced further when application of high
efficiency gear box technology ranges of effectiveness is applied.
Because the 7-speed automatic transmission and the advanced 6-speed
automatic transmission technologies are parallel on the technology
tree, the agencies felt using the same effectiveness value was
reasonable and appropriate.
---------------------------------------------------------------------------
\1019\ 2015 NAS Report, at page 189.
---------------------------------------------------------------------------
Commenters also pointed out a lack of skip-shift logic used in the
NPRM analysis, and an increase in the shift busyness observed for the
high gear count transmissions. Roush commented on the NPRM analysis
``not incorporating the concept of `Skip shifting' which is important
for reducing shift busyness and increasing FE especially in vehicles
equipped with transmission with a large number of ratios (8-10).''
\1020\ Both CARB and UCS repeated similar concerns.\1021\
---------------------------------------------------------------------------
\1020\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984 at 14-15.
\1021\ Comments from CARB, Attachment 2018-10-26 FINAL CARB
Detailed Comments on SAFE, NPRM Docket No. NHTSA-2018-0067-11873, at
110-113 (``Rogers found that the modeling did not consider `skip-
shifting' where a transmission can upshift or downshift in a non-
sequential manner''). Comments from UCS, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-12039, at 23 ``including that ANL's transmission
shift strategy does not deploy gear-skipping'').''.
---------------------------------------------------------------------------
After consideration of the comments and re-evaluation of the NPRM
results, the agencies concurred with the commenters. The lack of skip-
shift logic and increased shift busyness can result in lower overall
efficiency and decreased consumer acceptance. For the final rule
analysis, a skip-shift logic was applied to the 10 speed automatic
transmissions. The logic was based on the baseline 2017 Ford F150 10-
speed transmission benchmarking performed by Argonne.\1022\ The
introduction of the skip-shift logic impacted effectiveness and reduced
the number of shifts by 23 percent for the 10-speed automatic
transmission over the UDDS cycle.\1023\
---------------------------------------------------------------------------
\1022\ NHTSA Benchmarking, ``Laboratory Testing of a 2017 Ford
F-150 3.5 V6 EcoBoost with a 10-speed transmission.'' DOT HS 812
520.
\1023\ See FRM ANL Model Documentation file at Paragraph
4.4.5.5. This update reduced the number of shift events from 231 to
178.
---------------------------------------------------------------------------
In the NPRM analysis, transmission gear spans increased as the
number of
[[Page 24465]]
gears increased.\1024\ However, to address further the comments related
to optimization, the gear span of the AT10L3 was increased over the
AT10L2, based on gear span data for the Honda 2018 10-speed
transmission.\1025\ The AT10L3 span was increased to 10.10 in the final
rule analysis from 7.34 in the NPRM analysis. However, the efficiency
map for the AT10L3 remained the same for the final rule analysis.\1026\
---------------------------------------------------------------------------
\1024\ See FRM ANL Model Documentation file at 5.3.2.1.
\1025\ Sugino, S., SAE Internation Presentation., ``ALL-NEW
HONDA 10-SPEED FWD TRANSMISSION.'' November 2017. ``2018 Honda
Odyssey Press Kit--Overview.'' internet: Honda News, https://hondanews.com/en-US/releases/2018-honda-odyssey-press-kit-overview.
Last accessed October 8, 2019.
\1026\ See FRM ANL Model Documentation file at 5.3.4.1.
---------------------------------------------------------------------------
Finally, in the agencies' review of NPRM model inputs, a weight
discrepancy for the AT10 transmissions was identified. The weight
assigned to the AT10 transmission in the NPRM analysis was too high.
The weights were corrected for the final rule analysis. The AT10
transmission weights were reduced by 20-45 kg, depending upon vehicle
type.\1027\
---------------------------------------------------------------------------
\1027\ See FRIA VI.C.2.d.2.
---------------------------------------------------------------------------
The AT effectiveness values used for the final rule analysis can be
seen in Figure VI-25. For automatic transmission technologies, the
effectiveness improvement range is relative to a 5-speed automatic
transmission. The new effectiveness values are a result of the
aforementioned changes implemented to address comments. To summarize,
the changes included an adjustment to the modeled effectiveness of the
AT6L2, the use of skip-shift logic on the 10-speed transmissions, and
the increase of the AT10L2 gear span.
Figure VI-25 shows the automatic transmission's effectiveness
increases progressively in a logical order and behaves in an expected
manner. Gains in effectiveness can be observed increasing as gear count
increases, and as HEG levels increase. The effects of diminishing
returns can be observed as gear count reaches higher levels, and
effectiveness effects for increased gear count are reduced. This agrees
with observed data reported by the NAS and industry
stakeholders.1028 1029
---------------------------------------------------------------------------
\1028\ 2015 NAS Report, at 175.
\1029\ Greimel, H., ``ZF CEO--We're not chasing 10-speeds,''
Automotive News, November 23, 2014, http://www.autonews.com/article/20141123/OEM10/311249990/zf-ceo:-were-not-chasing-10-speeds.
---------------------------------------------------------------------------
(b) Continuously Variable Transmissions
For CVTs, the agencies also identified a discrepancy with the NPRM
CVT weights. The weight assigned to the CVT class during the NPRM
analysis was incorrect. Corrected values were assigned for the final
rule analysis. The CVT weights were reduced by 9-10 kg based on vehicle
type.\1030\
---------------------------------------------------------------------------
\1030\ See FRIA VI.C.2.d.2.
---------------------------------------------------------------------------
The CVT effectiveness values used for the final rule analysis can
be seen in Figure VI-26, shown as an effectiveness improvement over a
5-speed automatic transmission. The effectiveness values were not
changed significantly from the values used in the NPRM analysis.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.194
(c) Dual Clutch Transmissions
The DCT effectiveness values used for the final rule analysis can
be seen in Figure VI-27, shown as an effectiveness improvement over a
5-speed automatic transmission. The effectiveness values were not
changed significantly from the values used in the NPRM analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.195
[[Page 24466]]
(d) Manual Transmission
The MT effectiveness values used for the final rule analysis can be
seen in Figure VI-28, shown as an effectiveness improvement over a 5-
speed manual transmission. The effectiveness values were not changed
significantly from the values used in the NPRM analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.196
e) Transmission Costs
For the NPRM, the transmission technology costs used as inputs for
the CAFE model were retail price equivalent costs with learning curves
applied. For a complete discussion on how the retail price equivalent
and learning effects were applied to direct manufacturing costs see
Section VI.B.4.b), Indirect Costs, and Section VI.B.4.d), Cost
Learning. The direct manufacturing costs for the transmission
technologies used in the NPRM were derived from technical sources and
manufacturer's CBI.\1031\
---------------------------------------------------------------------------
\1031\ See PRIA Section 6.3.7.3.
---------------------------------------------------------------------------
Table VI-80 below shows the relative costs of the transmissions
used in the NPRM analysis including learning and retail price
equivalent.
[GRAPHIC] [TIFF OMITTED] TR30AP20.197
BILLING CODE 4910-59-C
(1) Automatic Transmissions
Several comments were received on technology costs, or cost
effectiveness. Meszler Engineering Services noted that ``AT10L2 (level
2 ten-speed automatic) transmission technology is another example of an
end-of-path technology with very poor cost effectiveness relative to
other transmission options.'' \1032\ A cost analysis by ICCT also
showed relative costs of transmission technologies may not be in line
with the modeled effectiveness.\1033\
---------------------------------------------------------------------------
\1032\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723, at 33.
\1033\ Comments from International Council on Clean
Transportation, Attachment 3, NPRM Docket No. NHTSA-2018-0067-11741,
at I-64.
---------------------------------------------------------------------------
The agencies conducted a review of transmission costs in response
to the comments. For the final rule analysis, adjustments were made to
costs of the AT6L2, AT7L2, AT9L2, AT10L2, and the AT10L3. The costs
were adjusted based on reviewing the recommended relative costs
discussed in the NAS 2015 report. Table VI-81 shows the cost for the
automatic transmissions in the final rule analysis.
The direct manufacturing cost (DMC) estimate for the AT6 is drawn
from Table 5.7 of the NAS report. The DMC estimate for the AT6L2 is
based on the cost of the AT6 with HEG level 2 technology costs applied.
This cost change is applied in accordance with the effectiveness
adjustment made for the AT6L2.
A DMC estimate for the AT7 was drawn from Table 5.9 of the NAS
report and was based on the cost of a system already equipped with HEG
technology. The DMC estimate was given in 2007 dollars and relative to
an AT5/AT4. The new DMC replaces the DMC from the NPRM, which did not
account for the HEG technology.
The DMC for the AT9 technology was drawn from Table 8A.2a of the
NAS (2015) report and per the NPRM description of the technology made
relative to the AT8L2. The AT9 is assumed to have at least the level 2
HEG technology applied. The NPRM analysis assumed the AT9 cost was only
relative
[[Page 24467]]
to the AT8 and did not account for the cost of the HEG technology.
The DMC for the AT10 technologies was drawn from Table 8A.2a of the
NAS report and per the NPRM description of the technology made relative
to the AT8L2. The AT10L2 is assumed to have at least the level 2 HEG
technology applied. The AT10L3 has the HEG3 technology applied. The
NPRM analysis assumed the AT10 costs were only relative to the AT8 and
did not account for the cost of the HEG technology.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.198
(2) Continuously Variable Transmissions
No adjustments were made to the NPRM costs of the CVT technologies
for the final rule analysis. Table VI-82 shows the cost for the CVTs in
the final rule analysis.
[[Page 24468]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.199
(3) Dual Clutch Transmissions
The agencies received one comment on cost learning over time for
DCT technologies. Roush Industries ``believes that the [actual]
learning factors for such systems are significantly better than those
estimated by either the 2018 PRIA or the 2016 Draft TAR.'' Roush stated
that ``eight-speed DCTs (DCT8) are currently in production (MY2018),
with quantities increasing significantly,''\1034\ but provided no
specific supporting data.
---------------------------------------------------------------------------
\1034\ Comments from Roush Industries, Attachment 1, NPRM Docket
No. NHTSA-2018-0067-11984, at 14-15.
---------------------------------------------------------------------------
The current learning curve for the DCT technologies was established
based on recommendations from the NAS 2015 report and on CBI data
collected from manufacturers and suppliers. Since Roush did not supply
any data to support its comment, the agencies decided it was reasonable
to make no change to the DCT learning curve for the final rule
analysis. Table VI-83 shows the cost for the DCTs in the final rule
analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.200
(4) Manual Transmissions
No adjustments were made to the NPRM costs of the manual
transmission technologies for the final rule analysis. Table VI-84
shows the cost for the MTs in the final rule analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.201
[[Page 24469]]
BILLING CODE 4910-59-C
3. Electric Paths
The electric paths include a large set of technologies that share
the common element of using electrical power for certain vehicle
functions that were traditionally powered mechanically by engine power.
Electrification technologies thus can range from electrification of
specific accessories (for example, electric power steering to reduce
engine loads by eliminating parasitic loss) to electrification of the
entire powertrain (as in the case of a battery electric vehicle).
Electrified vehicles are considered, for this analysis, to mean
vehicles with a fully or partly electrified powertrain. These include
several electrified vehicle categories, including: Battery electric
vehicles (BEVs), which have an all-electric powertrain and use only
batteries for propulsion energy; plug-in hybrid electric vehicles
(PHEVs), which have a primarily electric powertrain and use a
combination of batteries and an engine for propulsion energy; and
hybrid electric vehicles (HEVs), which use electrical components and a
battery to manage power flows and assist the engine for improved
efficiency and/or performance. HEVs are further divided into strong
hybrids (including P2 and power-split hybrids) that provide strong
electrical assist and in many cases, can support a limited amount of
all-electric propulsion, and mild hybrids (such as belt integrated
starter generator (BISG) hybrids, crankshaft integrated starter
generator (CISG) hybrids, and 48V mild hybrids) that typically provide
only engine on/off with minimum electrical assist.
Fuel cell electric vehicles (FCEVs) are also another form of
electrified vehicle having a fully electric powertrain, and are
distinguished by the use of a fuel cell system rather than grid power
as the primary energy source.
The factors that influence the cost and effectiveness of
electrification technologies are their components. These include:
Energy storage components such as battery packs; propulsion components
such as electric motors; and power electronics components, such as
inverters and controllers, that process and route electric power
between the energy storage and propulsion components. For the purpose
of this analysis, these components are divided into battery components
and non-battery components.
Battery components strongly influence the cost of electrified
vehicles.\1035\ Because developments in battery technology may apply to
more than one category of electrified vehicles, they are discussed
collectively in Section VI.C.3.e). That section details battery-related
topics that directly affect the specification and costing of batteries
for all types of electrified vehicles considered in this analysis.
---------------------------------------------------------------------------
\1035\ Battery costs are not necessarily a strong influence on
fuel Cell Electric Vehicles, where the cost of the fuel cell
technology has a larger influence.
---------------------------------------------------------------------------
Non-battery components also have an influence on both the cost and
effectiveness of electrified vehicles. The selection and configuration
of non-battery technologies distinguish the different architecture
among electrified vehicles. Non-battery components largely consist of
propulsion components and power electronics.
Propulsion components typically include one or more electric
machines (an umbrella term that includes what are commonly known as
motors, generators, and motor/generators). Depending on how they are
employed in the design of a vehicle, electric machines commonly act as
motors to provide propulsion, and/or act as generators to enable
regenerative braking and conversion of mechanical energy to electrical
energy for storage in the battery.
``Power electronics'' refers to the various components that control
or route power between the battery system and the propulsion
components, and includes components such as: Motor controllers, which
issue complex commands to control torque and speed of the propulsion
components precisely; inverters and rectifiers, which convert and
manage DC and AC power flows between the battery and the propulsion
components; onboard battery chargers, for charging the BEV or PHEV
battery from AC line power; and DC-to-DC converters that are sometimes
needed to allow DC components of different voltages to work together.
Onboard chargers are charging devices permanently installed in
electrified vehicles to allow charging from grid electrical power.
Onboard chargers travel with the vehicle and are distinct from
stationary charging equipment. Level 1 charging refers to charging
powered by a standard household 110-120V AC power outlet. Level 2
charging refers to charging at 220-240V AC power.
The agencies included a more extensive overview of charging
technology and the state of charging infrastructure in the NPRM and
PRIA, however, this was purely qualitative because charging was not
accounted for in any respect in the NPRM analysis. The Alliance
commented that ``[w]hile the costs of installing chargers and charger
convenience were not taken into account within the Volpe model . . .
these factors will continue to have an impact on the overall
penetration of electrification technologies that the market will be
willing to accept.'' \1036\ In contrast, the National Coalition for
Advanced Transportation (NCAT) commented that the qualitative
discussion overstated the risks and understated the benefits of
electric vehicle charging.\1037\ Specifically, NCAT took issue with the
characterization of potential risks of charging to the electric grid,
stating that ``the PRIA's focus on worst case hypotheticals does not
reflect the current capabilities of the grid, nor the dynamic nature of
EV charging to mitigate any potential negative impacts. In both in the
short-term and long-term, the impact of EVs with respect to the
electric grid would have a net-positive impact to society, including
the EV owners and utility customers broadly.'' NCAT also commented that
``[w]hile substantial investments in EV infrastructure have and will be
made, the costs and benefits to consumers must be put into the
appropriate context.'' NCAT cited two studies for the proposition that
the average lifetime distribution electric vehicle infrastructure
impact is about $80-$90 per electric vehicle sold, with the adoption of
time of use rates and assuming a diversity of charging rates. NCAT also
cited the California Public Utilities Commission 2016-2017 Electric
Vehicle Load Research Report in support of their statement that the
additional service and distribution system upgrades due to additional
plug-in electric vehicle load is minimal, as ``of the approximately
275,000 [electric] vehicles estimated to be on the road as of October
2017 in the service areas of California's three investor-owned
utilities, only 460, or 0.16 percent required a service line or
distribution system upgrade solely to support the plug-in electric
vehicle load at their residential charging location.''\1038\
---------------------------------------------------------------------------
\1036\ NHTSA-2018-0067-12073.
\1037\ NHTSA-2018-0067-11969.
\1038\ Citing Joint IOU Electric Vehicle Load Research Report
(December 29, 2017), pp. 1-2, 12, available at http://www.cpuc.ca.gov/zev/ (2016-2017 Load Research Report).
---------------------------------------------------------------------------
The agencies agree that adding electric vehicle infrastructure will
require additional costs, and information about what that cost is and
how it can or should be accounted for
[[Page 24470]]
in the analysis is helpful for commenters to submit in order to put
those considerations in the appropriate context. For this final rule,
the agencies did not incorporate any costs related to electric vehicle
charging infrastructure in the technology compliance analysis because
those costs are separate from the costs that manufacturers and
consumers would directly incur from a manufacturer transitioning part
of their fleet to plug-in electric vehicles and consumers paying for
those vehicles, even though local electric ratepayers will in all
likelihood pay higher rates to upgrade local power grids to accommodate
any widespread adoption of electrified vehicles. Accordingly, this
means that the actual costs associated with electrified vehicles have
been underestimated for the final rule analysis. The agencies did
refine the estimates for the value of refueling time for electric
vehicles, and that topic is discussed in Section VI.D.1.b)(11)(b). The
agencies will continue to explore whether and how charging
infrastructure should be incorporated into the analysis for future
actions.
The following sections discuss vehicle electrification issues that
were accounted for in the analysis, including the agencies'
characterizations of electric vehicle technology, additional electric
vehicle configurations added for the final rule analysis per
commenters' requests, and the sources and methods used to develop
battery and non-battery components, which were also refined for this
final rule.
a) Electrification Modeling in the CAFE Model
A set of technologies was chosen to represent the spectrum of
electrification methods observed in the baseline fleet and that the
agencies believed could be applied to vehicles in the rulemaking
timeframe. Each technology was placed in a specific electrification
pathway, grouping and defining the progression of related technologies.
In the NPRM analysis, a total of eleven electrification technologies
were contained in four electrification pathways. In consideration of
comments received, the electrification technologies and associated
pathways were modified for the final rule analysis, resulting in a
total of eighteen variants of electrification technologies. Each of
these NPRM and final rule technologies, and the electrification
pathways they belong to, are detailed below. Operational modes of
electrified vehicles are further described in the Argonne Model
Documentation for the final rule.
(1) Electrification Technologies
(a) Electric Improvements
The electrification of power steering (EPS) and other accessories
(IACC) have the potential of reducing fuel consumption by facilitating
power-saving control strategies that avoid parasitic loss of engine
power. These accessories traditionally are directly coupled to and
driven by the conventional combustion engine; any time the engine is
running some energy is continuously consumed by each accessory, even
when it is not needed. By decoupling these accessories from the engine
and instead driving them ``on-demand'' with electric motors, a more
energy-efficient control strategy can be employed to reduce fuel
consumption. EPS and IACC are discussed in detail in Section VI.C.7,
Other Vehicle Technologies.
(b) Micro Hybrid
12-volt stop-start (SS12V), sometimes referred to as start-stop,
idle-stop or 12-volt micro hybrid, is the most basic hybrid system that
facilitates idle-stop capability. In this system, the integrated
starter generator is coupled to the internal combustion (IC) engine.
When the vehicle comes to an idle-stop the IC engine completely shuts
off and, with the help of 12-volt battery, the engine cranks and starts
again in response to throttle to move the vehicle, or release of the
brake pedal. The 12-volt battery used for the start-stop system is an
improved unit capable of higher power, increased life cycle, and
capable of minimizing voltage drop on restart. This technology is
beneficial to reduce fuel consumption and emissions when the vehicle
frequently stops, such as in city driving conditions or in stop and go
traffic, and can be applied to all vehicle technology classes.
(c) Mild Hybrids
The belt integrated starter generator (BISG) and crank integrated
starter generator (CISG), sometimes referred to as mild hybrid systems,
provide idle-stop capability and use a higher voltage battery with
increased energy capacity over typical automotive batteries. The higher
voltage allows the use of a smaller, more powerful and efficient
electric motor/generator, which replaces the standard alternator. In
BISG systems, the motor/generator is coupled to the engine via belt
(similar to a standard alternator), while the CISG integrates it to the
crankshaft between the engine and transmission; both of these systems
allow the engine to be automatically turned off as soon as the vehicle
comes to a full stop. In addition, these motor/generators can recover
braking energy while the vehicle slows down (regenerative braking) and
in turn can propel the vehicle at the beginning of launch, allowing the
engine to be restarted later. Some limited electric assist is also
provided during acceleration to improve engine efficiency. The CISG
system has a higher efficiency, but also higher cost than the BISG.
The agencies received limited high-level comments on CISG systems,
with CARB stating that CISG systems are generally considered more
capable and more efficient relative to BISG systems because they do not
have the same belt-related constraints including maximum torque
limitations, load restrictions on the front crank to avoid uneven
crankshaft bearing wear, and mechanical energy transfer losses.\1039\
CARB also noted that the decision to implement a CISG system is
typically made early in the design process because doing so often
requires an engine block casting change. CARB stated that the current
high costs and larger dimensions, compared to BISGs, will likely delay
major market penetration of CISG systems until beyond the MY 2025
timeframe.
---------------------------------------------------------------------------
\1039\ Roush Industries on behalf of California Air Resources
Board, Rogers_Final_Final_NPRM_10.26.2018, Docket No. NHTSA-2018-
0067-11984, at 15.
---------------------------------------------------------------------------
For the final rule analysis, the agencies did not include CISG
systems. The effectiveness of CISG systems were similar to the BISG,
and the high cost of the CISG caused it to be applied infrequently.
Other packaging and integration issues make it difficult for most
vehicles to adopt CISG technology. Typically, a manufacturer would have
to modify the flywheel housing to allow the installation of an electric
motor, which must also fit where the system is mounted between the
transmission and the engine block. Space in that part of the vehicle
also comes at a premium because other components such as exhaust
systems and piping systems must also be housed in the same area. In the
final rule analysis, all vehicles previously considered to possess CISG
technology were instead assigned a BISG system.
(d) Strong Hybrids
A hybrid vehicle is a vehicle that combines two or more sources of
propulsion energy, where one uses a consumable fuel (like gasoline),
and one is rechargeable (during operation, or by another energy
source). Hybrids reduce fuel consumption through three major
mechanisms, including (1) potential engine downsizing, (2) optimizing
the performance of the engine to operate at
[[Page 24471]]
the most efficient operating point and under some conditions storing
excess energy such as by charging the battery, and (3) capturing energy
during braking and some decelerations that might otherwise be lost to
the braking system and using the stored energy to provide launch
assist, coasting, and propulsion during stop and go traffic conditions.
The effectiveness of the hybrid systems depends on how the above
factors are balanced, taking into account complementary equipment and
vehicle application. For some performance vehicles, the hybrid
technologies are used for performance improvement without any engine
downsizing.
The NPRM analysis evaluated the following strong hybrid vehicles:
Hybrids with ``P2'' parallel drivetrain architecture (SHEVP2),\1040\
and hybrids with power-split architecture (SHEVPS). The parallel hybrid
drivetrain, although enhanced by the electric portion, remains
fundamentally similar to a conventional powertrain. In contrast, the
power-split hybrid drivetrain is novel and considerably different than
a conventional powertrain. Although these hybrid architectures are
quite different, both types provide start-stop or idle-stop
functionality, regenerative braking capability, and vehicle launch
assist. A SHEVPS has a higher potential for fuel economy improvement
than a SHEVP2, although its cost is also higher.
---------------------------------------------------------------------------
\1040\ Depending on the location of electric machine (motor with
or without inverter), the parallel hybrid technologies are
classified as P0-motor located at the primary side of the engine,
P1-motor located at the flywheel side of the engine, P2-motor
located between engine and transmission, P3-motor located at the
transmission output, and P4-motor located on the axle.
---------------------------------------------------------------------------
Power-split hybrid (SHEVPS) is a hybrid electric drive system that
replaces the traditional transmission with a single planetary gear set
(the power-split device) and a motor/generator. This motor/generator
uses the engine either to charge the battery or to supply additional
power to the drive motor. A second, more powerful motor/generator is
permanently connected to the vehicle's final drive and always turns
with the wheels. The planetary gear splits engine power between the
first motor/generator and the drive motor either to charge the battery
or to supply power to the wheels. During vehicle launch, or when the
battery state of charge (SOC) is high, the engine, which is not as
efficient as the electric drive, is turned off and the electric machine
propels the vehicle. During normal driving, the engine output is used
both to propel the vehicle and to generate electricity. The electricity
generated can be stored in the battery and/or used to drive the
electric machine. During heavy acceleration, both the engine and
electric machine (by consuming battery energy) work together to propel
the vehicle. When braking, the electric machine acts as a generator to
convert the kinetic energy of the vehicle into electricity to charge
the battery.
The Autonomie simulations assumed all SHEVPS' used an Atkinson
cycle engine (Eng26). Therefore, all vehicles equipped with SHEVPS
technology in the CAFE model simulations were assumed to have Atkinson
cycle engines. This Atkinson cycle engine with high compression ratio
is optimized for efficiency, rather than performance. Accordingly,
SHEVPS technology as modeled in this analysis was not suitable for
large vehicles that must handle high loads.\1041\ Further discussion of
Atkinson engines and their capabilities is discussed in Section VI.C.1
Engine Paths.
---------------------------------------------------------------------------
\1041\ Kapadia, J., Kok, D., Jennings, M., Kuang, M. et al.,
``Powersplit or Parallel--Selecting the Right Hybrid Architecture,''
SAE Int. J. Alt. Power. 6(1):68-76, 2017, https://doi.org/10.4271/2017-01-1154.
---------------------------------------------------------------------------
P2 parallel hybrids (SHEVP2) are a type of hybrid vehicle that uses
a transmission-integrated electric motor placed between the engine and
a gearbox or CVT, with a clutch that allows decoupling of the motor/
transmission from the engine. Although similar to the configuration of
the CISG system discussed previously, a P2 hybrid would typically be
equipped with a larger electric machine and battery in comparison to
the CISG. Disengaging the clutch allows all-electric operation and more
efficient brake-energy recovery. Engaging the clutch allows efficient
coupling of the engine and electric motor and, when combined with a
transmission, reduces gear-train losses relative to power-split or 2-
mode hybrid systems. P2 hybrid systems typically rely on the internal
combustion engine to deliver high, sustained power levels. Only low and
medium power demands are allowed for electric-only mode.
In the NPRM CAFE modeling, the SHEVP2 system represented a hybrid
system paired with an existing engine on a given vehicle, while the
SHEVPS removed and replaced the previous engine with an Atkinson cycle
engine. The agencies explained that while many vehicles may use HCR1
engines as part of a hybrid powertrain, HCR1 engines may not be
suitable for some vehicles, such as high performance vehicles or
vehicles designed to carry or tow large loads (this is further
discussed in Section VI.C.1, Engine Paths). Many manufacturers may
prefer turbocharged engines (with high specific power output) for P2
hybrid systems, in order to maintain performance. Accordingly, in the
NPRM analysis, to satisfy power demands, many SHEVP2 systems were
paired with non-HCR powertrains.
ICCT and Meszler Engineering Services commented that as a result of
NPRM CAFE model constraints, low-cost, HCR engines were too
infrequently paired with SHEVP2 technology. These commenters claimed
that frequent pairing of SHEVP2 with downsized turbocharged engines
resulted in higher cost and lower effectiveness for these strong
hybrids.1042 1043
---------------------------------------------------------------------------
\1042\ Meszler Engineering Services, Attachment 2, Docket No.
NHTSA-2018-0067-11723, at 15.
\1043\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-25.
---------------------------------------------------------------------------
In consideration of these comments, the final rule analysis
includes additional strong hybrids (P2HCR0, P2HCR1, and P2HCR2\1044\)
that use HCR engines in a P2 parallel hybrid system. The SHEVP2
technology allows the engine type to be inherited from the outgoing
engine; this is unchanged from the NPRM and provides a good solution
for vehicles that need to undergo hybridization but require other
engine technologies (such as turbocharging) to meet performance
requirements. In addition, this final rule analysis allows any
conventional engine technology to go to P2HCR strong hybrid technology
within the set performance requirements. This is further discussed in
the Section VI.C.3.c), Electrification Adoption Features.
---------------------------------------------------------------------------
\1044\ P2HCR2 was included in simulations used for sensitivity
studies, but was excluded in the central analysis simulations for
reasons surrounding the HCR2 engine, as discussed in Section VI.C.1.
---------------------------------------------------------------------------
(e) Plug-In Hybrids
Plug-in hybrid electric vehicles (PHEV) are hybrid electric
vehicles with the means to charge their battery packs from an outside
source of electricity (usually the electric grid). These vehicles have
larger battery packs with more energy storage and a greater capability
to be discharged than other non-plug-in hybrid electric vehicles. PHEVs
also generally use a control system that allows the battery pack to be
substantially depleted under electric-only or blended mechanical/
electric operation and batteries that can be cycled in charge-
sustaining operation at a lower state of charge than is typical of
other hybrid electric vehicles. These vehicles generally have a greater
all-electric range than the typical SHEVs discussed above. In the NPRM
analysis,
[[Page 24472]]
PHEVs with two all-electric ranges--a 30 mile and a 50 mile all-
electric range (AER)--were included as technologies that vehicles could
adopt. The PHEV30 represented a ``blended-type'' plug-in hybrid, which
can operate in all-electric (engine off) mode only at light loads and
low speeds, and must blend electric machine and engine power together
to propel the vehicle at medium or high loads and speeds. The PHEV50
represented an extended range electric vehicle (EREV), which is capable
of travelling in all-electric mode even at higher speeds and loads.
Unlike other alternative fuel systems that require specific
infrastructure for refueling or recharging (e.g., hydrogen vehicles or
rapidly charged battery electric vehicles), PHEV batteries can be
charged using existing infrastructure, although widespread adoption may
require upgrades to electrical power distribution systems.\1045\ PHEVs
are considerably more expensive than conventional vehicles and more
expensive than SHEVPS technologies because of larger battery packs and
charging systems capable of connecting to the electric grid.
---------------------------------------------------------------------------
\1045\ See above for a discussion of electrical vehicle
infrastructure.
---------------------------------------------------------------------------
Commenters, such as CARB, stated that in the NPRM analysis the PHEV
motors were oversized and overpowered, and that model-built PHEV30s
have excessive battery pack size and electric range when compared to
actual production vehicles.\1046\ In response to such comments, the
agencies, in collaboration with Argonne, conducted further market study
to confirm CARB's observations and determined that replacing PHEV30
(with a nominal 30 mile AER) with PHEV20 (with a nominal 20 mile AER)
would more closely characterize the PHEVs actually in production.\1047\
The agencies therefore elected to replace PHEV30 with PHEV20 in the
final rule.
---------------------------------------------------------------------------
\1046\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 150, 153.
\1047\ ``ANL response on NPRM comments (PHEV sizing)-
181112.pptx,'' available in Docket No. NHTSA-2018-0067.
---------------------------------------------------------------------------
The final rule also includes four additional types of plug-in
hybrids; two additional plug-in hybrids were added to allow the use of
turbocharged engines (PHEV20T, PHEV50T), and two additional plug-in
hybrids were added to provide maximum efficiency by utilizing an
Atkinson cycle engine (PHEV20H, PHEV50H).
In practice, many PHEVs recently introduced in the marketplace use
turbocharged engines in the PHEV system, and this is particularly true
for PHEVs produced by European manufacturers and for other PHEV
performance vehicle applications. However, the NPRM Autonomie
simulations (and thus all the CAFE model simulations) assumed all PHEVs
used a naturally aspirated, Atkinson cycle engine. The agencies
determined through continued marketplace observation that PHEV vehicles
should indeed be allowed to adopt or retain turbocharged engines. Also,
BorgWarner commented that modeling of PHEVs should include turbocharged
engines, since these engines can be downsized to reduce vehicle mass
and fit into smaller engine compartments, and offer efficiency and
performance advantages especially when paired with a higher expansion
ratio.\1048\ Thus, in addition to the PHEV20 and PHEV30, the final rule
analysis included PHEV20T and PHEV50T variations which are,
respectively, 20 and 50 mile all electric range PHEVs with turbocharged
engines.
---------------------------------------------------------------------------
\1048\ BorgWarner, Attachment 2, Docket No. NHTSA-2018-0067-
11873, at 150,153.
---------------------------------------------------------------------------
This final rule also added PHEV20H and PHEV50H, although
effectively these are not used by the model simulations. These plug-in
types represent 20 and 50 mile all electric range plug-in hybrids that
use particularly efficient high-compression, Atkinson cycle engines.
These were added with the intent to provide PHEVs with a maximum level
of fuel economy at a lower cost. However, they proved to be too similar
to existing plug-in technology choices and were thus assigned identical
characteristics as the PHEV20 and PHEV50. In this final rule analysis,
PHEV20 and PHEV50 sizing were updated and so the similarities in
performance between different engines converged. For further discussion
on PHEV sizing, see Section VI.C.3.d), Electrification Effectiveness
Modeling and resulting Effectiveness values.\1049\ The PHEV20H and
PHEV50H technologies are still considered by the CAFE model but they
remain as ``placeholders'' for potential incorporation in future
analyses.
---------------------------------------------------------------------------
\1049\ This final rule analysis used Atkinson Engine for PHEVPS
electrified vehicles. The components such as electric motor and
engine power in these hybrid systems were sized in ways to meet
vehicle class performance characteristics and efficiency. And after
these vehicle components were sized, the Atkinson engines in these
vehicles were operating in similar efficiency as HCR engines as the
full vehicle modeling and simulation. As discussed in PO 06 C.1.c.1
Non-HEV Atkinson Engine Modes, power-split hybrid-based Atkinson
engines attempt to operate in the most efficient regions while using
electric motors to meet deficiencies in performance. And so, PHEV20H
and PHEV50H HCR engines compared to PHEV20 and PHEV50 Atkinsons
engines would have be sized to operate in the most efficiency
regions and the thermal efficiency between these two set of
combinations would have had similar efficiency for this analysis.
---------------------------------------------------------------------------
(f) Battery Electric Vehicles
Electric vehicles (EVs), or battery electric vehicles (BEVs) are
equipped with all-electric drive and with systems powered by energy-
optimized batteries charged primarily from grid electricity. The range
of a battery electric vehicle depends on the vehicle's class and the
battery pack size. The NPRM analysis included BEVs with a range of 200
miles.
Following the NPRM, the agencies conducted continued market
analysis of production BEVs, and observed a growing number of vehicles
with nominal ranges above 200 miles. CARB also commented that certain
BEVs modeled as BEV200 in the NPRM in fact had ``well over 200 miles of
range.'' \1050\ The agencies thus concluded that a 300-mile-range
BEV300 should be included in the final rule to represent better these
higher-range electric vehicles as well as a potential future range
alternative more comparable to IC engines. The agencies still believe
that, in the rulemaking timeframe, BEV300 will be the most cost
effective extended range BEVs that could be available for adoption.
Longer-range electric vehicles could have been modeled in the analysis,
but the compliance simulation would likely not have selected the
longer-range vehicle if lower-range vehicles were still available. This
is because the CAFE model only applies technologies until a
manufacturer meets its CAFE or CO2 standard, and the BEV200
and BEV300 vehicles operate functionally the same in helping a
manufacturer towards meeting its compliance obligations. The only
difference between these vehicles is cost. As discussed further in
Section VI.C.3.c), the agencies used phase-in caps to control expected
BEV200 and BEV300 penetration based on the current trend and future
assumption that consumers will transition towards longer-range electric
vehicles.
---------------------------------------------------------------------------
\1050\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 147.
---------------------------------------------------------------------------
(g) Fuel Cell Vehicles
Fuel cell electric vehicles (FCEVs or FCVs) utilize a full electric
drive platform but consume hydrogen fuel to generate electricity in an
onboard fuel cell. Fuel cells are electrochemical devices that directly
convert reactants (hydrogen and oxygen via air) into electricity, with
the potential of achieving more than twice the efficiency of
conventional internal combustion engines. High pressure gaseous
hydrogen storage tanks are used by most
[[Page 24473]]
automakers for FCEVs. These high-pressure tanks are similar to those
used for compressed gas storage in more than 10 million CNG vehicles
worldwide, except that they are designed to operate at a higher
pressure (350 bar or 700 bar vs. 250 bar for CNG), and to contain the
very small, and very flammable, gaseous hydrogen molecule. FCEVs are
currently produced in limited numbers and are available in limited
geographic areas.
(2) Electrification Pathways
The electrification technologies described above were applied in
the CAFE model through a number of technological pathways. Three main
electrification technology pathways were modeled: The Electric
Improvements Path, the Electrification Path, and the Hybrid/Electric
Path. These three electrification pathways are evaluated in parallel by
the CAFE model; the model can consider any of the three right away, and
does not need to go ``through'' one pathway in order to begin
evaluating another. Any superseded technology is also disabled whenever
a succeeding technology is applied to a vehicle, even if a specific
superseded technology was not previously utilized on that vehicle. As
previously explained, this requirement exists so that the modeling
system does not downgrade technologies during analysis.
The Electrics Improvements Path defined in the NPRM and final rule
is shown in Figure VI-29 below, which starts with EPS and progresses to
IACC. While these two electrified-accessory technologies are mutually
exclusive, either one can be modularly paired with any other
technology, including those in the other electrification pathways.
[GRAPHIC] [TIFF OMITTED] TR30AP20.203
The Electrification Path shown in Figure VI-29 allows a
conventional powertrain to become a micro-hybrid with SS12V, or a mild
hybrid with BISG, or CISG (which is no longer available for the final
rule analysis, as discussed previously) technologies. All three of the
Electrification Path technologies are mutually exclusive with respect
to all conventional powertrain technologies, as well as technologies
contained in the Hybrid/Electric path discussed below. The model first
evaluates SS12V, and then progresses to BISG or CISG (NPRM-only). The
conventional engine technology CONV is grayed out to indicate that the
model uses information about the previous conventional (non-
electrified) powertrain to map properly to simulation results found in
the vehicle simulation database. Although the adoption of these
technologies will classify a vehicle as a micro/mild hybrid (MHEV) and
no longer a conventional (CONV), the vehicle is allowed to retain the
engine and transmission technologies possessed before entering the
Electrification Path.
[GRAPHIC] [TIFF OMITTED] TR30AP20.204
The Hybrid/Electric Pathways are shown in Figure VI-30. Both the
NPRM and final rule Hybrid/Electric paths begin at the ``strong
hybrid'' technology types, each of which is mutually exclusive of the
others; once one is chosen, the other is eliminated from future
selection for that vehicle. The paths then progress into plug-in
hybrids and then culminate with the mutually exclusive battery electric
vehicles or fuel cell vehicles. The additional final rule technologies
described above can be found in the final rule Hybrid/Electric pathway
on the right side of Figure VI-31, in comparison to the NPRM
technologies shown on the left
[[Page 24474]]
side of the figure.\1051\ The hybrid/electric pathways contains
multiple ``roots,'' or starting points, which force a vehicle to remain
within the branches of a chosen root. For example, the final rule
hybrid/electric pathway has three roots: SHEVP2, SHEVPS, and P2HCR0. If
a vehicle uses SHEVPS, then SHEVP2 technology and the entire P2HCR0
through PHEV50H branch will be disabled from further consideration. In
other words, from one technology in the pathway, a vehicle can only
move forward along any of the indicated arrows, and never in the
reverse direction. Also, when using any technology in the Hybrid/
Electric pathway, with the exception of SHEVP2, all engine and
transmission technologies as well as the Electrification Path
technologies shown in Figure VI-31 are prohibited. SHEVP2 is an
exception because it allows engine technologies previously held by the
vehicle to be inherited into the parallel hybrid system.
---------------------------------------------------------------------------
\1051\ Note that the NPRM Hybrid/Electric Path (left side of
Figure I-3) refers to a portion of the path containing plug-in
hybrids and electric vehicles as the ``Advanced Hybrid/Electric
Path.'' For this discussion, we will simply refer to the entire
collection of these technologies, including the ``Advanced''
technologies, as the ``Hybrid/Electric Path.''
[GRAPHIC] [TIFF OMITTED] TR30AP20.205
b) Electrification Analysis Fleet Assignments
Since the 2012 rulemaking, manufacturers have implemented a number
of powertrain electrification technologies, including 48V mild hybrid,
strong HEV, PHEV, and BEV powertrains.1052 1053 For the NPRM
analysis, the agencies identified the specific electrification
technologies in each vehicle model in the MY 2016 analysis fleet, and
used those technology levels as the starting point for the regulatory
analysis. The agencies assigned electrification technology levels based
on manufacturer-submitted CAFE compliance information, vehicle
technical specifications released publicly by manufacturers, agency-
sponsored vehicle benchmarking studies, technical publications, and
manufacturer CBI.\1054\ For the final rule analysis, the agencies used
a similar process and data sources to identify the electrification
technologies in the MY 2017 analysis fleet.\1055\
---------------------------------------------------------------------------
\1052\ ``The 2018 EPA Automotive Trends Report,'' https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends, Accessed Aug 23, 2019.
\1053\ FOTW #1108, Nov 18, 2019: Fuel Economy Guide Shows the
Number of Conventional Gasoline Vehicle Models Achieving 45 miles
per gallon or Greater is Increasing. DOE VTO. Available at https://www.energy.gov/eere/vehicles/articles/fotw-1108-november-18-2019-fuel-economy-guide-shows-number-conventional. Last accessed Nov 18,
2019.
\1054\ NPRM Market Data central analysis input file.
\1055\ FRM Market Data central analysis input file.
---------------------------------------------------------------------------
The agencies received comments regarding the application of
electrification technologies in the MY 2016 analysis fleet. Commenters,
such as the California Air Resources Board, stated the agencies
mischaracterized some hybrid technologies, such as power-split and P2
hybrid architectures.\1056\ Specifically CARB was concerned about the
``misclassification of the 2016 Chevrolet Malibu Hybrid as having a P2
hybrid,'' noting the Malibu shared many of its drivetrain components
with the 2016 Chevy Volt, a vehicle classified as a power-split HEV.
---------------------------------------------------------------------------
\1056\ Comments from CARB, Attachment 2, NHTSA Docket No. NHTSA-
2018-0067-11873, at 136.
---------------------------------------------------------------------------
BorgWarner stated that the ``modeling should be inclusive of all
approaches of PHEV and HEV and not be limited only to Atkinson Cycle
engines,'' suggesting that it was appropriate for the NPRM analysis to
include turbocharged engines in combination with PHEV and HEV
technologies.\1057\
---------------------------------------------------------------------------
\1057\ Comments from BorgWarner, Attachment 1, Appendix, NHTSA
Docket No. NHTSA-2018-0067-11895, at 10.
---------------------------------------------------------------------------
The agencies agree with the underlying issue identified by both
CARB and BorgWarner's comments. In
[[Page 24475]]
both cases a limitation of modeling classification, and not a lack of
academic understanding of HEV systems, is the crux of the issue. In the
specific case of the 2016 Chevy Malibu, the electrical architecture is
a power split, however, the vehicle uses a non-Atkinson, basic direct
injection engine. These characteristics put the Malibu HEV in an
overlap with the powertrain models used to represent HEV systems in the
agencies' analysis. If the system had been classified as a PS HEV
system in the analysis fleet, the engine would have incorrectly been
modeled as an Atkinson engine, resulting in overestimation of the
baseline system's level of efficiency and technology applied. The
overestimation of the baseline fleet model would have limited the
potential for the baseline system to improve over the timeframe of the
analysis. With the system classified as the P2 HEV, the engine can be
accurately modeled while still accounting for the benefits of an HEV
system. This allowed the platform the full potential for technology and
efficiency improvement in the analysis.
The agencies considered the issues identified in comments and
reviewed the MY 2017 analysis fleet information to determine what
changes could improve the final rule analysis. The agencies determined
that expanding the number of electrification technologies would address
the CARB and BorgWarner comments, as well as the comments from others
that are discussed in Section VI.C.3.a)(1) Electrification
Technologies. The agencies increased the number of unique
electrification technologies from twelve in the NPRM to eighteen for
the final rule analysis. The expanded list enabled greater precision in
the assignment of technologies to the MY 2017 analysis fleet, and
enabled the agencies to characterize the electrification technologies
found in the fleet accurately and realistically. The expanded list also
provided more granularity for the application of technologies for the
rulemaking analysis. Table VI-85 shows the full list of electrification
technologies for the final rule analysis.
This collection of technologies represents the best available
information the agencies have, at the time of this action, regarding
both currently available electrification technologies and
electrification technologies that could be feasible for application to
the U.S. fleet during the rulemaking timeframe. The agencies believe
this effort has yielded the most accurate analysis fleet utilized for
rulemakings to date.
As discussed in the previous section and shown in Figure VI-29,
Figure VI-30, and Figure VI-31, electrification may be added to
vehicles as shown on the decision tree pathways. Further application of
electrification technologies to vehicle platforms was dependent on
electrification technology already present on vehicles in the MY 2017
analysis fleet. Electrification may also be predicated on whether a
vehicle has a dedicated platform that accommodates battery electric
capability or whether a platform is designed (``package protected'')
\1058\ to enable the addition of some form(s) of hybridization. The
agencies' assessment of each existing platform's capability to adopt
electrification technologies is identified in the CAFE model market
data input file.\1059\
---------------------------------------------------------------------------
\1058\ `Package Protected' is an automotive industry term used
to describe the purposeful design of a vehicle to include space and
weight allowances for future technology additions.
\1059\ FRM Market Data central analysis input file.
---------------------------------------------------------------------------
c) Electrification Adoption Features
In the NPRM and final rule analysis, electrification adoption
features were applied in multiple ways. First, when an electrification
technology is selected, a path logic is applied that dictates what
other technologies are either superseded or mutually exclusive to the
applied technology. For a detailed discussion of path logic for the
final rule analysis, including technology supersession logic and
technology mutual exclusivity logic, please see CAFE model
documentation section. Second, application of the more advanced
electrification technologies, such as the strong hybrids, plug-in
hybrids, and full BEVs, result in major changes to the whole
powertrain. The changes to the powertrain include substitution of
transmission and engine technologies, and accordingly these
technologies can only be applied at a vehicle redesign, as shown in
Table VI-85 below. Finally, some of electrification technologies are
restricted from application to certain vehicle classifications. These
restrictions will be discussed under the specific technology sections.
The fully-electric technologies, BEV technology and FCV technology,
qualify as alternative fuel technologies. As a result, these
technologies are not considered during portions of the agencies'
analysis. Specifically, the exclusion of dedicated alternative fuel
technology from NHTSA's analysis of potential fuel economy standards is
a result of statutory obligations prescribed under EPCA/EISA.\1060\
However, NHTSA performed two fuel economy analyses, a standard-setting
analysis that constrained the use of the technologies, and an
unconstrained analysis that did not exclude the technologies, which
provides an estimation of real-world environmental impacts used as
inputs for the Environmental Impact Statement (EIS). The unconstrained
analysis included the alternative fuel technologies, and used the
adoption features for BEVs and FCVs discussed below. Further, for
purposes of analyzing EPA's tailpipe CO2 emissions
rulemaking pursuant to the Clean Air Act, consideration of these
technologies is likewise unconstrained. For a detailed discussion of
the analysis versions and statutory obligations please refer to Section
VI.A Analytical Approach as Applied to Regulatory Alternatives,
Overview of Methods and Section VI.A.4 Compliance Simulation.
---------------------------------------------------------------------------
\1060\ 49 U.S.C. 32902(b)(1). A ``dedicated automobile'' is
defined in 49 U.S.C. 32901 as ``an automobile that only operates on
alternative fuel.''
---------------------------------------------------------------------------
The exclusion of the BEV and FCV technology from the standard-
setting analysis resulted in a comment from ICCT. ICCT stated, ``the
agencies prevented their fleet compliance model from allowing battery
electric vehicles from being applied in their analysis of the Augural
standards.'' \1061\ The agencies believe this reflects a
misunderstanding of NHTSA's statutory obligation under EPCA/EISA and
how the agencies ran the analysis. NHTSA did consider alternative
fueled vehicles in the unconstrained analysis--but as discussed further
in Section VIII, is prohibited from considering the availability of
such technologies when setting maximum feasible standards.
---------------------------------------------------------------------------
\1061\ Comments from ICCT, Attachment 3, Appendix, NPRM Docket
No. NHTSA-2018-0067-11741, at 182.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24476]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.206
BILLING CODE 4910-59-C
(1) Micro and Mild Hybrid
For the NPRM and final rule analysis, the only adoption features
for the SS12V and BISG technologies were functions of path logic. The
SS12V and BISG technologies were allowed for consideration in any
existing vehicle configuration that did not already have a more
advanced electrification technology applied. Per Table VI-85 above, the
BISG technology was considered more advanced than the SS12V technology.
Meszler Engineering commented that 48V batteries used in
conjunction with 12 volt systems (what are referred to in the analysis
as BISG systems) are one example of a ``bolt-on'' technology that can
be added to a vehicle during a product refresh without causing
[[Page 24477]]
production problems or significantly increasing costs.\1062\ Meszler
Engineering stated that 48V systems do not require reengineering of the
engine and can be added at any time during a model's lifespan, as shown
by key suppliers that are expanding production capacity to meet
customer demand for the technology.\1063\ Meszler Engineering also
pointed to examples of vehicles that utilize 48V systems, including
high-volume non-luxury vehicles like the Ram pickup truck, Jeep
Wrangler, and Ford F-150.\1064\
---------------------------------------------------------------------------
\1062\ Comments by Meszler Engineering, Attachment 4 CAF[Eacute]
Model Redesign and Refresh Rates, NHTSA Docket No. NHTSA-2018-0067-
11723, at 2-4. (citing A.K. Kumawat and A.K. Thakur, A Comprehensive
Study of Automotive 48V Technology, SSRG International Journal of
Mechanical Engineering (SSRG-IJME), Vol. 4 (5) (May 2017), available
at: https://jalopnik.com/everything-you-need-to-know-about-the-upcoming-48-volt-1790364465 (last viewed 10/23/2018)).
\1063\ Comments by Meszler Engineering, Attachment 4 CAFE Model
Redesign and Refresh Rates, NHTSA Docket No. NHTSA-2018-0067-11723,
at 2-4.
\1064\ Comments by Meszler Engineering, Attachment 4 CAFE Model
Redesign and Refresh Rates, NHTSA Docket No. NHTSA-2018-0067-11723,
at 2-4.
---------------------------------------------------------------------------
The agencies disagree with Meszler Engineering's assessment of 48V
technology as a ``bolt-on'' technology. Although BISG systems represent
a first step in vehicle electrification, and the number of components
involved is fewer than most other types of hybrid systems, a BISG
system still requires engineering and packaging of motors, cooling
systems, additional wiring harnesses from the 48V battery pack to the
motors, control systems, and other components incorporated into the
front engine compartment. Further, the addition of a BISG system
requires recalibration and validation of numerous engine performance
parameters, including emissions controls, balancing torque supply to
the transmission between the BISG system and engine, and noise-
vibration-harshness controls. In addition, the examples Meszler
Engineering provided support the agencies' designation of SS12V and
BISG systems as redesign technologies; the BISG system in the MY 2019
Ram pickup and in the MY 2018 Jeep Wrangler were introduced during a
product redesign and not during a mid-cycle product
refresh.1065 1066 Although Ford has indicated that the F-150
will include hybrid variants,\1067\ the agencies do not have
information about specific plans for a 48V system on the F-150. In
consideration of this information, the agencies maintained the redesign
schedule for mild hybrids for the final rule analysis.
---------------------------------------------------------------------------
\1065\ See, e.g., K.C. Colwell, The 2019 Ram 1500 eTorque Brings
Some Hybrid Tech, If Little Performance Gain, to Pickups, Car and
Driver (Mar. 14, 2019), available at: https://www.caranddriver.com/reviews/a22815325/2019-ram-1500-etorque-hybrid-pickup-drive/ (``Any
2019 Ram 1500--the all-new one, not the Ram Classic that is just a
continuation of the previous generation--can be equipped with a
motor/generator attached to its engine's crankshaft via a belt that
is capable of adding torque, cranking the engine in a stop/start
event, or making electricity with regenerative braking.'').
\1066\ See, e.g., Tony Quiroga, The 2018 Jeep Wrangler Hybrid
Provides Effortless Thrust, Much Improved Fuel Economy, Car and
Driver (Oct. 15, 2018), available at: https://www.caranddriver.com/reviews/a23746585/2018-jeep-wrangler-unlimited-suv-turbo-four-cylinder-hybrid/ (``Completely redesigned for 2018, the Wrangler is
even more like a Power Wheels now that it's available with an
electric motor.'').
\1067\ ``Ford to Invest more than $1.45 Billion, Add 3,000 Jobs
in SE Mich. Plants to Deliver New Pickups, SUVs, EVS, and AVS,''
Ford Media Center, 17 Dec 2019. https://media.ford.com/content/fordmedia/fna/us/en/news/2019/12/17/ford-invests-adds-jobs-southeast-michigan-plants.html.
---------------------------------------------------------------------------
(2) Strong Hybrids--SHEVP2, SHEVPS, P2HCR0, P2HCR1, P2HCR2
NPRM adoption features applied to strong hybrid technologies
included path logic, powertrain substitution, and vehicle class
restrictions. For the NPRM analysis technologies on the Hybrid/Electric
path (SHEVP2 and SHEVPS) were defined as stand-alone and mutually
exclusive. When the modeling system applies one of those technologies,
the other one is immediately disabled from future application. Once a
strong hybrid technology is applied it also supersedes lower
technologies on the electrification path, allowing future application
of technology to consider only more advanced forms of electrification.
In the NPRM when the SHEVP2 technology or the SHEVPS technology
were applied, the transmission technology was superseded. Regardless of
the transmission technology present when the technology was applied,
the transmission technology was replaced by either the AT6 or DCT6. The
specific transmission technology selected was based on choosing the
best cost versus effectiveness.
During the NPRM analysis when the SHEVP2 technology was selected
the engine technology for the platform was maintained. However, the
engine technology was locked at the current level and could not be
changed. For the SHEVPS technology the existing engine was replaced
with an Atkinson cycle engine (Eng26).
The SHEVPS was also constrained from application to particular
vehicle technology classes or vehicles with specific performance
characteristics in the NPRM. Application of the power-split
architecture was restricted from high performance vehicles and vehicles
with a high towing capability requirements.\1068\ These constraints
prevented application to the pick-up and performance pick-up class of
vehicles. The constraints also prevented application to any platform
with a base horsepower rating greater than 400 HP. Additional platforms
determined to be purpose built as performance platforms were also
restricted from receiving SHEVPS technology.
---------------------------------------------------------------------------
\1068\ Kapadia, J., Kok, D., Jennings, M., Kuang, M. et al.,
``Powersplit or Parallel--Selecting the Right Hybrid Architecture,''
SAE Int. J. Alt. Power. 6(1):68-76, 2017, https://doi.org/10.4271/2017-01-1154.
---------------------------------------------------------------------------
Comments from ICCT criticized the manner in which SHEVP2 technology
was applied to a platform. ICCT stated ``the benefits of level-2
transmission efficiency and TURBO2 over TURBO1 are removed when P2
strong hybrid systems (SHEVP2) are selected on the electrification
pathway.'' \1069\
---------------------------------------------------------------------------
\1069\ Comments from ICCT, Attachment 3, 15 page summary and
full comments appendix, NPRM Docket No. NHTSA-2018-0067-11741, at
I25.
---------------------------------------------------------------------------
Additional comments regarding the adoption features of the SHEVP2
technology were received from Meszler Engineering and ICCT. Meszler
argued that the locking of engine technologies when a manufacturer
selects the SHEVP2 technology may preclude the selection of a more
cost-effective engine technology.\1070\ This concern was echoed by
ICCT, who also felt the engine technology lock-in artificially
increased cost for effectiveness on the overall SHEVP2 technology
packages.\1071\ Both commenters specifically wanted an option for a
high compression ratio engine technology to be considered in place of
any advanced engine technology carried into the SHEVP2 technology
pathway.
---------------------------------------------------------------------------
\1070\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723, at 15-16.
\1071\ Comments from ICCT, Attachment 3, 15 page summary and
full comments appendix, NPRM Docket No. NHTSA-2018-0067-11741, at
I25-I26.
---------------------------------------------------------------------------
The agencies agreed with the need for maintaining the benefits of a
higher transmission technology, and for the final rule analysis a AT8L2
transmission technology replaced the AT6 or DCT6 transmissions for all
hybrid-electric technologies. The AT8L2 was selected as the optimal
transmission technology point for HEV systems. The transmission
technology point was selected based on observed diminishing returns for
applying advanced transmission technologies to advanced engine/
powertrains.\1072\
---------------------------------------------------------------------------
\1072\ 2015 NAS Report--The National Academy of Science, in
their 2015 report, noted that ``as engines incorporate new
technologies to improve fuel consumption, the benefits of increasing
transmission ratios or switching to a CVT diminish.''
---------------------------------------------------------------------------
[[Page 24478]]
The agencies also reconsidered engine options for SHEVP2
technology, and other strong hybrid-electric technologies. The agencies
agreed with Meszler and ICCT's observation and instituted new P2 engine
technology options, as discussed above. For the final rule analysis,
when a platform considered the SHEVP2 option, the platform also
compared maintaining the current engine technology, or selecting an HCR
technology. If the SHEVP2 system chooses to apply a HCR engine, the
system diverts to the new electrification sub-path of technologies that
includes the P2HCR0, P2HCR1, and P2HCR2.
The P2HCR path introduced in the final rule analysis had similar
constraints as the SHEVPS. Performance vehicles and vehicles with a
high towing requirement were restricted from selection of the P2HCR
technology. Restrictions that were applied used the same criteria
described for the SHEVPS.
(3) Plug-In Hybrids--PHEV20/30, PHEV50, PHEV20T, PHEV50T, PHEV20H,
PHEV50H
The plug-in hybrid options in the NPRM included PHEV30 and PHEV50
technologies. The plug-in technologies superseded the micro, mild, and
strong hybrid electrification technologies and could only be replaced
by full electric technologies. The path logic also allowed a PHEV30 to
progress to a PHEV50.
In the NPRM, when a platform progressed to the plug-in hybrid
technologies the powertrain was automatically modified. The engine
technology was replaced by a high compression ratio engine (Eng26) and
the transmission was replaced by the AT6 or DCT6 technology.
PHEV30 and PHEV50 were also constrained from application to
vehicles with the potential for high towing demands.\1073\ This
constraint was applied by restricting access to the pickup truck
vehicle technology class. Additional specific vehicle platforms were
restricted based on engineering judgment.
---------------------------------------------------------------------------
\1073\ Power split or Parallel-selecting the Right Hybrid
Architecture: SAE 2017-01-1154. = Kapadia, J., Kok, D., Jennings,
M., Kuang, M. et al., ``Powersplit or Parallel--Selecting the Right
Hybrid Architecture,'' SAE Int. J. Alt. Power. 6(1):68-76, 2017,
https://doi.org/10.4271/2017-0-1154.
---------------------------------------------------------------------------
Comments were received regarding the options for PHEV battery-
electric technology. The comments are presented and discussed in
Section VI.C.3.e) Electrification Technologies above, and resulted in
the creation of additional technology options for plug-in hybrids, as
well as a modification of available ranges. Comments were also received
regarding the engine and transmission options used in the
electrification technologies, these comments are also presented and
discussed above in Section VI.C.3.e) Electrification Technologies.
For the final rule analysis, the plug-in hybrid options included
PHEV20, PHEV50, PHEV20T, PHEV50T, PHEV20H, and PHEV50H. As with the
NPRM, the plug-in technologies superseded the micro, mild, and strong
hybrid technologies. For the final rule analysis, plug-in hybrid
technologies were also mutually exclusive, and the PHEV20 technologies
can progress to the PHEV50 technologies.
When a platform applied plug-in hybrid technologies in the final
rule analysis, the engine and transmission technologies are superseded.
For all plug-in technologies, an AT8L2 transmission is used. For the
PHEV20/50 and PHEV20/50H, the engine is replaced by an Atkinson cycle
based engine (Eng26). For the PHEV20/50T, the engine is replaced by the
TURBO1 technology engine (Eng12).
The PHEV20/30 and PHEV20/50H path also had similar constraints as
the SHEVPS in the final rule analysis. Performance vehicles and
vehicles with a high towing requirement were restricted from selection
of the PHEV20/30 and PHEV20/50H technologies. Restrictions that were
applied used the same criteria described for the SHEVPS.
(4) Battery Electric Vehicles
For the NPRM analysis, the BEV200 technology was applied as an end-
of-path technology. The BEV200 technology was the only battery electric
vehicle option. For the final rule analysis, the BEV300 was added as a
technology option beyond the BEV200, as discussed in Section
VI.C.3.a)(1)(f) Battery Electric Vehicles. BEV200 and BEV300 technology
was applied in place of all engine and transmission technologies, and
was an end of path technology.
For the final rule analysis, both the BEV 200 and BEV300 had phase-
in cap limitations applied based on an analysis of the market
availability and cost of batteries.\1074\ The BEV200 was limited to a
greater extent than the BEV300, accounting for expected limits in
market demand for the shorter-range BEV.\1075\ The phase-in capacity
numbers were determined based on the results of the analysis of the
National Energy Model System (NEMS) discussed in Section
VI.D.1.b)(1)(b) Macroeconomic assumptions used to analyze economic
consequences of the final rule.
---------------------------------------------------------------------------
\1074\ John Elkin, MIT finds that it might take a long time for
EVs to be as affordable as you want, Digital Trends (November 23,
2019), https://www.digitaltrends.com/cars/mit-study-finds-ev-market-will-stall-in-the-2020s/.
\1075\ MIT Energy Initiative. 2019. Insights into Future
Mobility. Cambridge, MA: MIT Energy Initiative. http://energy.mit.edu/insightsintofuturemobility.
---------------------------------------------------------------------------
(5) Fuel Cell Vehicle
For the NPRM analysis, FCV technology was also applied as an end of
path technology. The FCV technology was also applied as end of path
technology in the final rule analysis.
For the final rule analysis, a phase-in cap was assigned to FCV
technology. The phase-in cap was assigned based on existing market
share as well as an analysis of expected infrastructure availability
during the time frame of regulation.1076 1092
---------------------------------------------------------------------------
\1076\ ``The 2018 EPA Automotive Trends Report,'' https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends. Last accessed Aug 23, 2019.
---------------------------------------------------------------------------
d) Electrification Effectiveness Modeling and Resulting Effectiveness
Values
For this analysis, the agencies considered a range of
electrification technologies which, when modeled, resulted in varying
levels of effectiveness at reducing fuel consumption. Each technology
consists of many different complex sub-systems with unique component
efficiencies and operational modes. As discussed further below, the
systems that contribute to the effectiveness of an electrified
powertrain in the analysis include the vehicle's battery, electric
motors, power electronics, and accessory load. Procedures for modeling
each of these sub-systems are discussed below, and also in Section
VI.B.3 Technology Effectiveness Values and in the FRM Argonne Model
Documentation.
The modeled electrification technologies included micro hybrids,
mild hybrids, strong hybrids, plug-in hybrids, and full electric
vehicles. This section discusses how Autonomie was used to model these
technologies' effectiveness. The models for the micro hybrids included
a SS12V system model; mild hybrid models included BISG system models
and CISG system models; strong hybrid models included SHEVP2 system
models and SHEVPS system models; and finally, electric vehicle models
included BEV system models and FCV system models.
[[Page 24479]]
(1) Electric Motors, Power Electronics and Accessory Load
Each electrified powertrain type possesses a unique effectiveness
for reducing fuel consumption. Autonomie determines the effectiveness
of each electrified powertrain type by modeling the basic components,
or building blocks, found in each powertrain, and then combining the
components modularly to determine the overall efficiency of the entire
powertrain. The basic building blocks that comprise an electrified
powertrain in the analysis included the battery, electric motors, power
electronics, and accessory loads. Autonomie identified which components
comprise each electrified powertrain type, and how these components are
interlinked within each unique electrified powertrain architecture.
This creates a model for each electrified powertrain architecture that
simulates how efficiently energy is transferred through each system.
For example, Autonomie determines a BEV's overall efficiency by
considering the efficiencies of the battery, the electric traction
drive system (the electric machine and power electronics) and
mechanical power transmission devices. Or, for a SHEVP2, Autonomie
combines a very similar set of components to model the electric portion
of the hybrid powertrain, and then also includes the combustion engine
and related power transmission components.
For the NPRM and this final rule analysis, Autonomie employed a set
of electric motor efficiency maps, which originated from two Oak Ridge
National Laboratory (ORNL) studies: one for a traction motor and an
inverter, the other for a motor/generator and
inverter.1077 1078 Autonomie also used test data validations
from technical publications to determine the efficiency of certain
electric motors. The electric motor efficiency maps are visual
measurements of percent efficiency of power as a function of torque and
motor RPM, and were based on representative production vehicles,
especially for base and maximum speeds as well as maximum torque curve.
The maps were used to determine the efficiency characteristics of the
motors, but were scaled such that their peak efficiency value
corresponded to the latest state of the art technologies for different
electrified powertrains. The maps also included some of the losses due
to power transfer through the electric machine.\1079\ Table VI-86
details the electric machine efficiency map sources for the different
powertrain configurations used for the NPRM.
---------------------------------------------------------------------------
\1077\ See PRIA, at 374.
\1078\ Oak Ridge National Laboratory (2008). Evaluation of the
2007 Toyota Camry Hybrid Synergy Drive System. Submitted to the U.S.
Department of Energy; Oak Ridge National Laboratory (2011). Annual
Progress Report for the Power Electronics and Electric Machinery
Program.
\1079\ See Chapters 4.7 and 5.5 in the FRM ANL Model
Documentation.
[GRAPHIC] [TIFF OMITTED] TR30AP20.207
For the final rule, the agencies used the same efficiency maps as
the NPRM, except for BEVs. The agencies updated the BEV electric motor
efficiency for the final rule analysis using data from a more recent
technical publication.\1081\ The agencies also scaled the maps to have
peak efficiencies ranging from 96-98 percent depending on the
powertrain type.\1082\ Table VI-87 below shows powertrain types and the
source of data used for the final rule.
---------------------------------------------------------------------------
\1080\ Burak Ozpineci, Oak Ridge National Laboratory Annual
Progress Report for the Power Electronics and Electronic Motors
Program, ORNL/SPR-2014/532, https://info.ornl.gov/sites/publications/Files/Pubs3253422.pdf, November 2014. (Nissan Leaf data
was used for FCV powertrain type).
\1081\ Faizul Momen, Electric Motor Design of General Motors'
Chevrolet Bolt Electric Vehicle, 2016-01-1228, SAE International,
April 5, 2016.
\1082\ See. Chapter 5.5 in FRM ANL Model Documentation.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24480]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.208
Battery performance data (e.g., internal resistance, open circuit
voltage) were measured using individual cell testing on a bench using
standard test procedures, and BatPaC was used to design battery packs
of different capacities and cell counts. The battery utilization (e.g.
SOC range) were developed based on numerous vehicle test data.\1083\ In
addition, as discussed further below, for the NPRM analysis, the
agencies resized the battery pack only with the addition of incremental
mass reduction technology levels. For this final rule, the agencies
updated the modeling to consider battery resizing with the application
of all road load reduction technologies. Accordingly, a more
appropriately-sized battery pack could result in lower vehicle mass,
resulting in potentially improved effectiveness.
---------------------------------------------------------------------------
\1083\ Kim, N., & Jeong, J. (2017). Control Analysis and Model
Validation for BMW i3 Range Extender. SAE Technical Paper 2017-01-
1152. doi:10.4271/2017-01-1152. Jeong, J. K. (2019). Analysis and
Model Validation of the Toyota Prius Prime. SAE World Congress. SAE.
Namdoo Kim, A. R. (2017). Vehicle Level Control Analysis for Voltec
Powertrain. Presented at the 30th International Electric Vehicle
Symposium and Exhibition (EVS30). Stuttgart, Germany. Hanho Son, N.
K. (2015). Development of Performance Simulation for a HEV with CVT
and Validation with Dynamometer Test Data. Presented at the 28th
International Electric Vehicle Symposium (EVS28). Kintex, Korea.
---------------------------------------------------------------------------
Beyond the powertrain components, Autonomie also considered on-
board accessory devices that consume energy and affect overall vehicle
effectiveness. Some electrical power is consumed by electrical
accessories such as headlights, radiator fans, wiper motors, engine
control units (ECU), transmission control unit (TCU), cooling systems,
and safety systems, in addition to driving the motor and the wheels. In
real-world driving, the electrical accessory load on the powertrain
varies depending on the how features are used and the condition the
vehicle is operating in, such as for night driving or hot weather
driving. However, for regulatory test cycles related to fuel economy,
the electrical load is repeatable because the fuel economy and
CO2 regulations control for these factors, as discussed in
Section VI.B.3 Technology Effectiveness Values.\1084\ Accessory loads
during test cycles do vary by powertrain type and vehicle technology
class, since distinctly different powertrain components and vehicle
masses will consume different amounts of energy.
---------------------------------------------------------------------------
\1084\ NHTSA Benchmarking, ``Laboratory Testing of a 2017 Ford
F-150 3.5 V6 EcoBoost with a 10-speed transmission.'' DOT HS 812
520.
---------------------------------------------------------------------------
The baseline fleet consists of hundreds of different vehicle types
that vary in the amount of accessory electrical power that they
consume. For example, vehicles with different motor and battery sizes
will require different capacities of electric cooling pumps and fans to
manage component temperatures. Autonomie has built-in models that can
simulate these varying sub-system electrical loads. However, for the
NPRM and this final rule analysis, the agencies used a fixed (by
vehicle technology class and powertrain type), constant power draw to
represent the effect of these accessory loads on the powertrain. The
agencies intended and expected that fixed accessory load values would,
on average, have similar impacts on effectiveness as found on actual
manufacturers' systems. This process was in line with the past
analyses, such as in the Draft TAR and the EPA Proposed
Determination.\1085\ \1086\ For assumptions regarding accessory load
modeling for the rulemaking timeframe, the agencies relied on research
and development data from DOE's Vehicle Technologies Office and Argonne
Advanced Mobility Technology Laboratory, as well as input from
automotive manufacturers.\1087\ \1088\ \1089\
---------------------------------------------------------------------------
\1085\ Draft Technical Assessment Report (July 2016), Chapter 5.
\1086\ EPA Proposed Determination TSD (November 2016), at p.2-
270.
\1087\ DOE VTO Power Electronics Research and Development.
https://www.energy.gov/eere/vehicles/vehicle-technologies-office-
electric-drive-systems. Last Accessed Jan 2, 2020.
\1088\ ANL Advanced Mobility Technology Laboratory (AMTL).
https://www.anl.gov/es/advanced-mobility-technology-laboratory. Last
Accessed Jan 2, 2020.
\1089\ DOE's lab years are ten years ahead of manufacturers
potential production intent (i.e 2020 Lab Year is MY 2030).
---------------------------------------------------------------------------
[[Page 24481]]
Table VI-88 below shows the NPRM assumptions for all the vehicle
classes and powertrain types for accessory loads.\1090\ Data from AMTL
D \3\ testing were used to designate electric loads for different types
of powertrains.\1091\
---------------------------------------------------------------------------
\1090\ See NPRM ANL Assumptions Summary.
\1091\ ANL Energy Systems Division Downloadable Dynamometer
Database: https://www.anl.gov/es/downloadable-dynamometer-database.
[GRAPHIC] [TIFF OMITTED] TR30AP20.209
BILLING CODE 4910-59-C
For the final rule analysis, the agencies updated the electrical
load assumptions for many of the powertrain types and classes,\1092\
based on further consideration of comments from the Alliance on the
2016 Draft TAR and EPA Proposed Determination.\1093\ \1094\ These
assumptions are provided below, in Table VI-89.
---------------------------------------------------------------------------
\1092\ See ANL Assumptions Summary, ANL--All
Assumptions_Summary_FRM_06172019_FINAL.
\1093\ Alliance of Automobile Manufacturers Comments on Draft
TAR at p. 30. September 26, 2016.
\1094\ EPA Proposed Determination TSD (November 2016), at p.2-
270.
[GRAPHIC] [TIFF OMITTED] TR30AP20.210
[[Page 24482]]
CARB commented on NPRM non-battery component efficiency assumptions
in two respects; first by claiming that the agencies relied on outdated
data for electric machines and inverter efficiencies across all
electrification applications,\1095\ and second by claiming that the
agencies did not project any efficiency gains in those components over
time.\1096\ CARB stated that the three vehicles benchmarked in the ORNL
studies (MY 2007 Toyota Camry Hybrid, a MY 2011 Hyundai Sonata Hybrid,
and MY 2012 Nissan Leaf) were inappropriate for the agencies to use to
assess the costs and efficiencies for the same components in MY 2020-
2030 vehicles, given the rapid development in the past ten years in
automotive electrification. CARB cited the MY 2016 Chevrolet Volt and
Bolt, and the MY 2016 Toyota Prius, as examples of vehicles that had
undergone electric machine efficiency improvements from one generation
to the next; those vehicles generally employed efficiency improvements
including reduced electric motor volume and mass, reduced power
inverter volume, increased electric motor peak power density, and
reduced mechanical losses through friction reduction, among other
improvements.
---------------------------------------------------------------------------
\1095\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 127.
\1096\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 128.
---------------------------------------------------------------------------
In support of their comments that the agencies did not project any
efficiency gains in non-battery components over time, CARB faulted the
agencies for not including data from the October 2015 ORNL progress
report for electric drive technologies, stating that benchmarking data
for a MY 2014 Honda Accord Hybrid inverter and traction motor
components could have been used to compare against and update the data
from the MY 2007 Toyota Camry Hybrid and MY 2011 Hyundai Sonata Hybrid
efficiency maps benchmarked in the older ORNL report. CARB stated that
the lack of consideration of this newer data was evidence that the
agencies' data selection was biased to support weakening fuel economy
standards.
CARB also cited 2017 research from Argonne's Autonomie group as a
source of updated data that showed efficiency gains over time for
electrification technologies not considered in the agencies' analysis,
including increases in high voltage system peak efficiency, increases
in high voltage specific power, and decreases in costs.\1097\ CARB
stated that had the agencies included newer data in the analysis,
including from the same data sources from which prior data came, the
analysis would have not supported the agencies' proposal.
---------------------------------------------------------------------------
\1097\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 131. Note that comments on non-battery
component costs are addressed in Section VI.C.3.e)(2) Non-Battery
Electrification Component Costs.
---------------------------------------------------------------------------
The agencies agree that there have been improvements in non-battery
component efficiency over the past few years, however CARB's
characterization of the process used to employ the ORNL benchmarking
data in the analysis was incorrect. Autonomie used high-level electric
machine characteristics such as base and max motor speed from
production vehicles along with generic efficiency map curves for each
technology type, with peak efficiencies matching the current state of
the art technologies discussed in ORNL reports. Although the source
data for the electric machines were from older production vehicles, the
peak electric motor and controller efficiencies were updated to reflect
the latest available data. Specifically, the NPRM analysis modeled a 92
percent peak efficiency for motors and controllers.\1098\
---------------------------------------------------------------------------
\1098\ See the Non_Vehicle_Attribute tab in the NPRM ANL
Assumptions_Summary.
---------------------------------------------------------------------------
That said, the agencies also agreed that the analysis could use
updated peak electric and controller efficiencies, and updated those
for the final rule. For the final rule analysis, the agencies used 96
percent efficiency for HEVs and PHEVS, and 98 percent peak efficiency
for BEVS and FCEVs.\1099\ The agencies believe the final rule
efficiencies are appropriate for the rulemaking timeframe.
---------------------------------------------------------------------------
\1099\ See the Non_Vehicle_Attribute tab in the FRM ANL
Assumptions_Summary.
---------------------------------------------------------------------------
In addition, as discussed above, other changes for the final rule
analysis include updating the electric motor sizing as a function of
electric power to account for lower electric machine mass, updating the
BEV electric machine map to use a newer efficiency map from the Chevy
Bolt, updating baseline and reference vehicle mass assumptions to
reflect latest machine weight technology development, and updating the
electrical accessory loads for vehicle modeling to reflect data from
vehicle benchmarking. Changes and updates to the Autonomie analysis are
discussed throughout this electrification section and in the FRM
Argonne Model Documentation. In addition, for this final rule analysis,
the agencies used the latest Argonne BatPaC model to determine the
battery pack mass and manufacturing costs for electric vehicle
batteries. Updates to non-battery component efficiency were small in
comparison to the impact of using updated battery modeling for the
final rule analysis. Further discussion on battery modeling can be
found in Section VI.C.3.e)(1) Battery Pack Modeling.
(2) Modeling and Simulating Vehicles With Electrified Powertrains in
Autonomie
Data from Argonne's AMTL was used to develop the electrified
powertrain models in Autonomie. The modeled electrification components
were sized based on performance neutrality needs, as discussed further
below, and the control algorithms were based on Argonne -collected
data.\1100\ Detailed discussion about the development of the HEV
drivetrains can be found in the Autonomie modeling documentation.\1101\
The modeled powertrains are not intended to represent any specific
manufacturer's architecture, but are intended to act as surrogates
predicting representative levels of effectiveness for each
electrification technology.
---------------------------------------------------------------------------
\1100\ See FRM ANL Model Documentation.
\1101\ See NPRM ANL Model Documentation at p.92.
---------------------------------------------------------------------------
The agencies also broadly discussed in Section VI.B.3 Technology
Effectiveness Values that certain technologies' effectiveness for
reducing fuel consumption requires optimization through the appropriate
sizing of the powertrain. This analysis iteratively minimizes the size
of the powertrain components to maximize efficiency while at the same
time enabling the vehicle to meet multiple performance criteria. The
Autonomie simulations use a series of resizing algorithms which contain
``loops,'' such as an ``Acceleration Performance Loop (0-60 mph),''
which automatically adjust the size of certain powertrain components
until a criterion, for example 0-60 acceleration time, converges to a
target value. As the algorithms examine different performance or
operational criteria that must be met, no single criterion is allowed
to degrade; once a resizing algorithm completes, all criteria will be
met, and some may be exceeded as a necessary consequence of meeting
others.
Autonomie applies different powertrain sizing algorithms depending
on the type of vehicle considered because different types of vehicles
not only contain different powertrain components to be optimized, but
they must also operate in different driving modes. While the
conventional powertrain sizing algorithm must consider only the power
of the engine, the more complex algorithm for
[[Page 24483]]
electrified powertrains must simultaneously consider multiple factors,
which could include the engine power, electric machine power, battery
power and battery capacity. Also, while the resizing algorithm for all
vehicles must satisfy the same performance criteria, the algorithm for
some electric powertrains must also allow those electrified vehicles to
operate in certain driving cycles without assistance of the combustion
engine, and ensure the electric motor/generator and battery can handle
the vehicle's regenerative braking power, all-electric mode operation
and intended range of travel.
To establish the effectiveness of the technology packages,
Autonomie simulated the vehicles performing compliance test cycles, as
discussed in Section VI.B.3 Technology Effectiveness
Values.1102 1103 1104 For vehicles with conventional
powertrains and micro hybrids, Autonomie simulated the vehicles using
the 2-cycle test procedures and guidelines.\1105\ For mild HEVs, strong
HEVs, and FCVs, Autonomie simulated the 2-cycle test, with the addition
of repeating the drive cycles until the final state of charge was
approximately the same as the initial state of charge, a process
described in SAE J1711. For PHEVs and BEVs, Autonomie simulated
vehicles performing the test cycles per guidance provided in SAE
J1711.\1106\ For BEVs, Autonomie simulated vehicles performing the test
cycles per guidance provided in SAE J1634.\1107\
---------------------------------------------------------------------------
\1102\ EPA, ``How Vehicles are Tested.'' https://www.fueleconomy.gov/feg/how_tested.shtml. Last accessed Nov 14,
2019.
\1103\ See FRM ANL Model Documentation at Chapter 6: Test
Procedures and Energy Consumption Calculations.
\1104\ EPA Guidance Letter. ``EPA Test Procedures for Electric
Vehicles and Plug-in Hybrids.'' Nov. 14, 2017. https://www.fueleconomy.gov/feg/pdfs/EPA%20test%20procedure%20for%20EVs-PHEVs-11-14-2017.pdf. Last accessed Nov. 7, 2019.
\1105\ 40 CFR part 600.
\1106\ PHEV testing is broken into several phases based on SAE
J1711. Charge-Sustaining on the City cycle, Charge-Sustaining on the
HWFET cycle, Charge-Depleting on the City and HWFET cycles.
\1107\ SAE J1634. ``Battery Electric Vehicle Energy Consumption
and Range Test Procedure.'' July 12, 2017.
---------------------------------------------------------------------------
A survey of comments about the modeled effectiveness of
electrification technologies showed most comments could be sorted in
three major categories. The first, and largest category of comments,
were concerned with effectiveness values used for the technologies.
Specifically, commenters were concerned the values for the modeled
effectiveness of the technologies were too low, particularly when
compared to past analysis efforts. The second major category of
comments were concerned with the size of the electrification components
selected in the Autonomie tool, and used to simulate the system
performance. Commenters were concerned because oversized components can
lead to the system violating performance neutrality constraints and
artificially increasing the cost of the technology. The third major
category of comments were concerned not enough variety of technologies
were represented in the electrification technology models.
Specifically, commenters wanted additional engine technologies allowed
to couple with electrification technologies.
Each of the comments from the first category will be referenced and
addressed under the specific technology sections, below. However,
broadly, two factors have led to the comments raised by stakeholders.
First, as discussed throughout this document, the agencies avoided
using performance values in the analysis that can be traced to specific
implementation of a technology type. Thus, when comparing simulated
performance to any specific real world vehicle, there will be a
deviation. The modeled inputs are meant to represent the typical range
of values for a technology--reasonable and realistic values--but are
not likely to result in performance outputs that would equal any
specific existing vehicle. Second, the modeling approach implemented in
the NPRM and final rule analysis succeeds in isolating the effects of
individual technologies to a higher degree then previous analysis. Due
to the greater use of parametric modeling of full vehicle systems, the
specific effects of technologies could be isolated to a higher degree
from the amplifying or muting effects of other technologies. This
isolation of effect often results in lower predicted effectiveness
values for individual technologies than has been observed in previous
analysis, where the isolation of effect was not as precise, and often
attributed efficiency gains from a combination of technological changes
to a single technology.
For the second major group of comments, the agencies mostly agreed
with the stakeholder observations. The issues identified were
investigated by the agencies and resulted in changes to the sizing
algorithms used by the agencies for the final rule analysis. The
agencies further investigated the constraints of performance neutrality
and ensured those constraints were followed for sizing of
electrification components. Further discussion of the changes made, as
well as specific answers to comments under each technology section, can
be found in the following technology subsections and in Performance
Neutrality, Section VI.B.3.a)(6).
The third major group of comments from stakeholders were concerned
with allowing more engine technologies to be incorporated in
electrification systems. The agencies agreed with these comments and
increased the number of technology combinations available. The new
combinations are discussed in Section VI.C.3.a)(1) Electrification
Technologies, as well as under each technology section below.
(a) Micro and Mild Hybrid Vehicles
The micro and mild hybrid systems modeled in Autonomie represented
SS12V and BISG technology (and CISG technology for the NPRM). SS12V and
BISG were modeled using a similar approach because both systems have
low peak power, low energy storage, and allow stop/start engine idle
reduction. The effectiveness improvement from both technologies is
attributable to the amount of fuel saved during engine idling period on
the 2-cycle test. However, only the BISG system model allowed limited
assist to propel the vehicle and limited regenerative braking. For
further discussion of these system models, see the FRM Argonne Model
Documentation.\1108\
---------------------------------------------------------------------------
\1108\ See FRM ANL Model Documentation at chapters 4.6, 4.7 and
4.13.
---------------------------------------------------------------------------
Powertrain resizing was not employed for micro or mild hybrid
system application, in either the NPRM or this final rule analysis.
These systems have little to no impact on the vehicle performance
metrics that would be adjusted by powertrain resizing, and in turn
there would be limited or no benefit in attempting to resize upon
application of these systems. For example, the micro hybrid SS12V
system allows the engine to be turned off when the vehicle is fully
stopped to reduce idle-stop fuel consumption, but the combustion engine
size must be retained to maintain performance metrics such as
acceleration. The main focus of mild hybrid vehicles is to provide
idle-stop and capture some regenerative braking energy, and although
they also can provide some assistance to the engine during the initial
propelling of the vehicle, this is done to improve efficiency and does
not significantly improve the acceleration performance of the vehicle.
With BISG mild hybrids, the electric machine is linked to the engine
through a belt, and thus the potential power assistance is usually
limited. In the NPRM, the BISG system used an 806 Wh capacity battery
[[Page 24484]]
pack and a 10 kW motor/generator. For the final rule analysis, the 10
kW motor/generator was paired with a 403 Wh battery pack to align with
BISG systems emerging in the marketplace.
ICCT commented that the agencies unjustifiably reduced the
CO2 and fuel consumption benefits of SS12V from the Draft
TAR, including a reduction in the overall effectiveness benefit when
the SS12V system was applied in combination with other
technologies.\1109\ ICCT stated that the agencies should know the
precise effectiveness improvement for SS12V technology based on EPA
compliance data, and the agencies should report a full listing of all
the baseline 2016 vehicle models with stop-start technology, with their
test-cycle, and off-cycle improvement in g/mile and percent
effectiveness. ICCT claimed that the agencies either intentionally
ignored the full compliance benefits of SS12V technology or ``ignored
the knowledge and expertise of the EPA engineering and compliance
staff,'' and argued that not reporting the requested data would be
``hiding relevant data the agencies have readily available to more
rigorously assess existing stop-start technologies and their impact for
the rulemaking.'' ICCT also stated that the agencies did not
appropriately include the full regulatory benefit (i.e., inclusion of
the additional off-cycle ``credit'' under EPA's program or fuel
consumption improvement value under NHTSA's program) of SS12V
technologies due to their off-cycle improvements.\1126\
---------------------------------------------------------------------------
\1109\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-22.
---------------------------------------------------------------------------
HDS made a similar observation, noting that the SS12V benefit from
the NRPM was similar to the 2012 TSD projection, but lower than the
benefit quoted by stakeholders in the Draft TAR.\1110\ HDS cited the
difference in fuel economy between two vehicles that were produced with
and without a SS12V option (the 2015 Ford Fusion 1.5L TGDI and the 2015
Mazda 3 i-ELOOP) which suggested effectiveness values for SS12V of
about 3.3 percent for both vehicles. HDS also cited a Bosch
presentation that claimed newer SS12V systems could provide
effectiveness of up to 6 percent. HDS argued that this actual data and
supplier data supported a benefit of at least 3.3 percent, which they
stated was double the benefit in the NRPM analysis.
---------------------------------------------------------------------------
\1110\ H-D Systems, Attachment 1, Docket No. NHTSA-2018-0067-
11985, at 44.
---------------------------------------------------------------------------
The agencies disagree with ICCT and HDS' comments regarding the
effectiveness of the SS12V technology modeled in the NPRM analysis. The
implementation of the full vehicle simulation approach used in the
NPRM, and carried forward to the final rule analysis, clearly defines
the contribution of individual technologies and separates those
contributions from other technologies. The modeling approach also shows
when technologies have amplifying or muting interactions. In some
cases, this may appear as a reduction in performance compared to
previous analysis. The agencies modeled the SS12V system in conjunction
with all the IC engine and transmission combinations. The results of
this parametric modeling accounted for each engine and transmission
combination's unique fuel consumption rate at idle.\1111\ The range of
effectiveness for the technology in the NPRM analysis is a result of
these differences. This range of values will result in some modeled
effectiveness values being close to real-world measured values, and
some modeled values that will depart from measured values, depending on
the level of similarity between the modeled hardware configuration and
the real-world hardware configuration. This modeling approach comports
with the National Academy of Science 2015 recommendation to use full
vehicle modeling supported by application of lumped improvements at the
sub-model level.\1112\ The approach allows the isolation of technology
effects in the analysis supporting an accurate assessment.
---------------------------------------------------------------------------
\1111\ For example, when idling, a larger eight-cylinder engine
has more friction and pumping losses than a smaller four-cylinder
engine, and therefore will save more fuel when the engine is shut-
off at rest.
\1112\ National Research Council. 2015. Cost, Effectiveness, and
Deployment of Fuel Economy Technologies for Light-Duty Vehicles.
Washington, DC--The National Academies Press. https://www.nap.edu/catalog/21744/cost-effectiveness-and-deployment-of-fuel-economy-technologies-for-light-duty-vehicles, at 292.
---------------------------------------------------------------------------
For both the NPRM and final rule analysis, the agencies assigned
SS12V technology to vehicles in the analysis fleet using compliance
data, and used compliance data to assign a vehicle's baseline fuel
economy value. The market data file indicated the presence of SS12V on
a vehicle, and accordingly, the vehicles reported to include SS12V
technology in the analysis fleet were modeled with the technology. For
more discussion on how technologies were assigned to the vehicle
platforms in the analysis fleet, please see Section VI.B.1 Analysis
Fleet. The agencies accounted for the contribution of the SS12V
technology in the analysis fleet by using the reported compliance fuel
economy values as the baseline fuel economy values for vehicles that
included the technology. The analysis fleet fuel economy values were
the reported final compliance values for the given vehicle platform and
should include the benefits from all technologies on the vehicle
platform.\1113\ The agencies also captured the off-cycle credits
provided to a manufacturer for the existence of the technology in the
manufacturer's fleet. For the NPRM and final rule analysis, the
manufacturers' fleets are modeled with baseline year compliance-
reported off-cycle credits. Further, for the final rule analysis, the
agencies increased the application of off-cycle credits in the
analysis, as discussed in Section VI.B.2.a) Off-cycle and A/C
Efficiency Adjustments to CAFE and Average CO2 Levels.
---------------------------------------------------------------------------
\1113\ Sec. 32904. Calculation of average fuel economy, https://uscode.house.gov/browse/prelim@title49/subtitle6/partC/chapter329&edition=prelim.
---------------------------------------------------------------------------
Commenters similarly disagreed with the BISG effectiveness
presented in the NPRM analysis, suggesting the resulting effectiveness
improvement should be at a range of 4 percent to 6 percent.\1114\ Such
commenters claimed that it was unclear why effectiveness values were so
much lower than previous effectiveness estimates. More specifically,
comments centered on (1) arguing that the agencies' modeling of BISG
and CISG systems in Autonomie likely underestimated the resulting
effectiveness values; (2) suggesting that the values in prior documents
like the Draft TAR and the 2015 NAS report were more accurate; and (3)
comparing modeled effectiveness values to claimed values achieved by
actual on-road vehicles and mild hybrid systems.
---------------------------------------------------------------------------
\1114\ ICCT, Attachment 3, Docket No. NHTSA-2018-0067-11741;
California Air Resources Board, Attachment 2, Docket No. NHTSA-2018-
0067-11873; Roush Industries, Attachment 1, NPRM Docket No. NHTSA-
2018-0067-11984; H-D Systems, ``HDS final report,'' Docket No.
NHTSA-2018-0067-11985; Union of Concerned Scientists, Attachment 2,
Docket No. NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
CARB claimed that the agencies failed to disclose the necessary
details to conclude why mild hybrid systems were projected to have
lower efficiency values than past estimates. CARB also concluded the
lack of engine downsizing when adding a BISG/CISG system and the lack
of adjusting transmission drive ratios and shift logic were reasons why
BISG/CISG effectiveness was underpredicted.\1115\ CARB claimed not
resizing the engines resulted in a ``less than optimized system that
does not take full advantage
[[Page 24485]]
of the mild hybrid system.'' \1116\ CARB argued that the agencies'
assumption that manufacturers ``would not optimize the engine and
transmission when installing a CISG is not realistic and results in
improper pairing of advanced gasoline engines and transmissions in the
modeling and leads to underestimation of the efficiency benefits.'' As
mentioned above, CARB stated that manufacturers ``often are required to
make a[n] engine casting change to accommodate the system,'' and when
doing so, ``no manufacturer would fail to pair the system with an
optimally sized engine and configured transmission to take full
advantage of the system's capabilities.'' \1117\
---------------------------------------------------------------------------
\1115\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 163.
\1116\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 185.
\1117\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 186.
---------------------------------------------------------------------------
CARB also inquired into whether the Argonne modeling ``took full
advantage'' of the system, using Daimler's EQ Boost system, that
provides temporary boosts for acceleration and enables engine shut-off
during coasting events, as an example.\1118\ Similarly, CARB noted that
CISG systems' ability to provide low end torque makes it an ``ideal
technology to pair with an engine technology that may have poor low end
torque but improved efficiency under other conditions; examples could
include an HCR engine sized with minimal low end torque to maximize
efficiency improvements in other operating conditions or a turbocharged
downsized engine equipped with a larger turbine to reduce backpressure
but provide improved efficiency over a larger portion of the engine
map.'' \1119\ CARB stated that manufacturers are using such systems to
boost engine torque at higher operating speeds so they can keep the
engine operating in a more efficient region.
---------------------------------------------------------------------------
\1118\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 163.
\1119\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 163.
---------------------------------------------------------------------------
Commenters also cited data from suppliers that produce 48V BISG
systems, including data from TULA that showed a 11 percent fuel economy
benefit from a 48V system,\1120\ data from a Delphi 48V system
prototype installed on a Honda Civic that showed a 10 percent reduction
in CO2 emissions levels,\1121\ and data from Continental
showing a 13 percent fuel savings improvement from its BISG
system.\1122\ ICCT also cited its supplier and technology report on
hybrids that estimated the benefit of mild hybrid technology at 12.5
percent, which it characterized as ``remarkably similar'' to that
achieved by the 2019 RAM pickup truck.\1123\ HDS noted that even if the
effectiveness values from TULA are regarded as optimistic because they
are the developers of the technology, EPA's previous modeling results
of 8-9 percent effectiveness ``appear reasonable in light of what is
observed from certification data.'' \1124\ ICCT ultimately recommended
the agencies revise the effectiveness value for mild hybrid systems to
include a CO2 effectiveness value of 12.5 percent.\1125\
---------------------------------------------------------------------------
\1120\ H-D Systems, Attachment 1, Docket No. NHTSA-2018-0067-
11985, at 45.
\1121\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 160.
\1122\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 160.
\1123\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-24.
\1124\ H-D Systems, Attachment 1, Docket No. NHTSA-2018-0067-
11985, at 45.
\1125\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-25.
---------------------------------------------------------------------------
Commenters also stated that the effectiveness estimates for CISG
systems were significantly understated, \1126\ with UCS characterizing
CISG systems as showing ``virtually no benefit whatsoever for CISG over
BISG, and in many cases actually show[ing] an increase in fuel
consumption.'' \1127\ UCS stated this was a dramatic departure from
previous Autonomie results, and with ``no explanation whatsoever''
given for the decrease in technology effectiveness.
---------------------------------------------------------------------------
\1126\ Union of Concerned Scientists, Attachment 2, Docket No.
NHTSA-2018-0067-12039; Roush-Industries, Attachment 1, Docket No.
NHTSA-2018-0067-11984; California Air Resources Board, Attachment 2,
Docket No. NHTSA-2018-0067-11873.
\1127\ Union of Concerned Scientists, Attachment 2, Docket No.
NHTSA-2018-0067-12039, at 3.
---------------------------------------------------------------------------
The agencies agree with commenters that the NPRM analysis of mild
hybrid technologies could be more representative of production vehicles
and vehicles likely to be produced during the rulemaking time period.
The agencies further conclude that the NPRM analysis overestimated the
costs of such technologies. Thus, for the final rule analysis, the
agencies only considered one 48V BISG system in the mild hybrid
technology category. The 48V mild hybrid BISG system used the same 10
kW electric motor as the one used in the NPRM analysis, and the 48V
BISG battery pack was also reduced in size to 403 W-hr from 806 W-hr to
reflect more accurately the size of battery packs available in the
market. In addition, the Autonomie model increased the usable battery
capacity, increasing the duration of electric motor use by the vehicle
before starting the engine. The specifications and assumptions for the
48V BISG system are further discussed in the FRM Argonne Model
Documentation and FRM Argonne Assumptions Summary.1128 1129
The discontinued use of the CISG technology is discussed in Section
VI.C.3.a)(1)(c) Electrification Technologies, Mild Hybrids.
---------------------------------------------------------------------------
\1128\ See FRM ANL Model Documentation, at 4.6, 4.13, and 5.7.
\1129\ FRM ANL Assumptions Summary (see Model Documentation
tables in Section VI.A.7 Structure of Model Inputs and Outputs).
---------------------------------------------------------------------------
The agencies disagree with comments stating incremental
effectiveness estimated by Autonomie modeling was incorrect because the
effectiveness values deviated from past effectiveness values estimated
in the agencies' rulemakings or from real-world values measured on
specific vehicles. As discussed in previous sections, the
implementation of the full vehicle simulation approach used in the NPRM
analysis and carried forward to the final rule analysis clearly defines
the contribution of individual technologies through the application of
parametric modeling. This approach clearly separated the contributions
of each technology. The modeling approach also showed the amplifying or
muting interactions between technologies. In some cases, this may
appear as reduced performance in comparison to previous analysis. The
agencies also strongly disagree that they should use the performance
values for any specific vehicle as representative of all mild hybrid
systems.
CARB also commented that the agencies' decision to use a fixed
final drive ratio and fixed shift logic based on the selected
transmission did not allow for efficiency improvements when mated with
electrified powertrains, with specific regards to mild hybrid BISG and
CISG systems.\1130\ CARB stated that based on the information disclosed
in the NPRM, ``it appears that Argonne did not utilize the system in
these manners nor did they allow for changes in gear ratios, final
drive ratio, or transmission shift logic to optimize for efficiency
improvements when mated with different electrified powertrains.''
\1131\ Roush Industries similarly stated that the analysis under-
predicted the potential improvements of employing a BISG system because
the engine could operate at a lower RPM with the help of the torque
assist of the electric motor/generator, with a change to the final
[[Page 24486]]
drive ratio and transmission shift logic, but the analysis did not do
so.\1132\
---------------------------------------------------------------------------
\1130\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 185.
\1131\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 185.
\1132\ Roush Industries, Attachment 1, Docket No. NHTSA-2018-
0067-11984, at 16.
---------------------------------------------------------------------------
The agencies disagree with CARB and Roush Industries' claims about
the gear ratio and shift logic used for the NPRM. As discussed in
Section VI.C.2.d) Transmission Effectiveness Modeling and Resulting
Effectiveness Values, manufacturers commonly maintain the same gear
hardware across vehicle platforms and applications, relying on controls
and shift strategy to achieve optimization. Autonomie maintained gear
hardware but customized the shifting strategy for each unique vehicle
system modeled \1133\ to reflect real-world manufacturing strategies
more accurately.
---------------------------------------------------------------------------
\1133\ FRM ANL Model Documentation, at 4.4.5.
---------------------------------------------------------------------------
CARB also commented that the performance modeled by the Autonomie
tool in the NPRM analysis failed to remain neutral for over 80 percent
of the modeled systems with mild hybrids. CARB felt the over-
performance was ``indicating some portion of the system capability was
improperly modeled to improve performance rather than reduce
CO2 emissions.'' \1134\
---------------------------------------------------------------------------
\1134\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 163.
---------------------------------------------------------------------------
The agencies agree with CARB's observations about the performance
of mild hybrid combinations. The mild hybrid configuration exhibited
higher performance in comparison to non-mild hybrid configurations in
the NPRM analysis. For the final rule analysis, the agencies updated
sizing and control of the mild hybrid systems to minimize performance
changes and maintain neutrality. As discussed earlier in this chapter,
updates include using smaller battery systems, updated algorithms, and
updated component weights. For further discussion of performance
neutrality for the final rule, see the Performance Neutrality Section
VI.B.3.a)(6).
Finally, ICCT commented that the agencies should include off-cycle
and ``game-changing'' pickup truck credits in the effectiveness
estimates for hybrid pickup trucks, as ``[i]t is the responsibility of
the agencies to include all applicable credits with their technology
packages calculations and their projections, including any additional
credits that will automatically accrue.'' \1135\
---------------------------------------------------------------------------
\1135\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-25.
---------------------------------------------------------------------------
While the agencies included many compliance flexibilities in the
modeling for the final rule analysis, hybrid pickup truck credits were
not modeled. The referenced pickup truck credit is set to expire for
all pickup trucks after MY 2021, so in analyzing this comment the
agencies considered what technologies manufacturers could apply to
pickup trucks through that model year to meet the requirements
specified in the regulation. To receive credit in a model year,
manufacturers must produce a quantity of improved full size pickup
trucks--improvement characterized by including either hybrid technology
or improved emissions performance--such that the proportion of
production of such vehicles, when compared to the manufacturer's total
production of full size pickup trucks, is not less than an amount
specified in that model year. The agencies determined that, based on
manufacturers' MY 2017 pickup truck offerings characterized in the
analysis fleet and with the technology considered in this rule, no
pickup truck manufacturer could meet the criteria set by EPA to qualify
for the mild credit before the credit is set to expire. For the strong
HEV credit, the agencies considered that forcing the application of
strong HEV pickups to meet the minimum threshold of 10 percent of the
fleet in order to earn the incentive credits would significantly
increase the cost for compliance and be less cost-effective than other
technology pathways. As the analysis seeks the most cost-effective
pathway for compliance, the agencies disagree the analysis should force
the application of strong HEV technology to at least 10 percent of full
size pickup trucks. However, the agencies did allow and simulated
maximum off-cycle and A/C off-cycle FCIVs for all manufacturers in the
CAFE model for both the CAFE and CO2 programs during the
rulemaking time frame. So, while the agencies did not model pickup
truck credits specifically, the final rule analysis allowed
manufacturers to reach the maximum off-cycle credit cap during the
rulemaking timeframe.
(b) Strong Hybrid Vehicles
The power-split hybrid (SHEVPS) model in Autonomie included a
power-split device, two electric machines and an engine, and allowed
various interactions between these components. The SHEVP2 model in
Autonomie is based on the pre-transmission (P2) configuration where the
electric motor is placed between the engine and transmission for direct
flow of power to the wheels. The vehicle can be propelled either by the
combustion engine, electric motor, or both simultaneously, but the
speed/efficiency region of operation for SHEVP2s under any engine/motor
combination is ultimately dictated by the transmission gearing and
speed. Detailed discussion of SHEVPS and SHEVP2 modeling and validation
are provided in the Argonne Model Documentations.\1136\ Autonomie full
vehicle models representing strong hybrids were based on vehicle test
data from vehicle benchmarking.
---------------------------------------------------------------------------
\1136\ FRM ANL Model Documentation, at Chapters 4.13, 4.16 and
6.0.
---------------------------------------------------------------------------
As discussed previously in this section, power-split hybrids
utilize a combustion engine, two electric machines and a planetary gear
set along with a battery pack to propel the vehicle. The smaller motor/
generator (EM1) is used to control the engine speed and uses the engine
to either charge the battery or to supply additional electric power to
the second ``drive'' motor. The more powerful drive motor/generator
(EM2) is permanently connected to the vehicle's final drive and always
turns with the wheels. The SHEVPS resizing algorithm makes an initial
estimate of the size of the engine, battery, and electric motors. The
initial estimates for the combustion engine and EM2 sizes are based on
the peak power required for acceleration performance and the continuous
power required for gradeability performance. The initial estimates for
the battery and EM1 powers are based the maximum regenerative braking
power. With these initial size estimates, the algorithm computes the
vehicle mass, and simulations are run to determine if 0-60 and 50-80
mph acceleration performance is acceptable. If acceleration is not
satisfactory (too fast or too slow), the algorithm iteratively adjusts
the sizes of the engine, motors, and battery, and runs simulations
until a minimum powertrain size is found that meets all requirements.
With each iteration, the engine, battery, and motor characteristics
were also updated for gradeability performance and regeneration, if
necessary. Figure VI-32 below shows the general steps of the SHEVPS
sizing algorithm. Detailed descriptions are available in section 8.3 of
the FRM Argonne Model Documentation.
BILLING CODE 4910-59-P
[[Page 24487]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.211
A parallel hybrid (SHEVP2) uses a combustion engine and a multi-
speed transmission-integrated electric motor (EM1), as discussed
previously in this section. As is done with SHEVPS, the SHEVP2 resizing
algorithm creates a starting point by making an initial estimate of the
size of the engine, battery, and electric motor based on performance
criteria or an estimated regenerative braking power, in turn
calculating the associated vehicle mass. The algorithm then uses a
simulation loop to find a more precise value of regenerative braking
power generated in the UDDS ``city driving'' cycle, and adjusts the
electric motor size and vehicle mass accordingly. Next, the algorithm
uses simulation loops to optimize the engine, motor, and battery sizes
in relation to acceleration performance criteria. In the event that the
acceleration criteria requires downsizing the powertrain, the electric
motor size is not reduced as this would not be suitable for the
handling of regenerative braking power. If the acceleration criteria
cause the electric motor to increase in size, the algorithm then
returns to the regenerative braking loop and subsequently all other
loops until all components are optimized. Figure VI-33 below shows a
simplified sizing algorithm for SHEVP2s.
BILLING CODE 4910-59-C
In the NPRM, the acceleration optimization loops in the SHEVP2
algorithm did not resize the powertrain if the resulting acceleration
time was less than the target. This strategy was intended to avoid
reducing the engine size compared to the conventional vehicle,
mimicking an observed marketplace trend in which parallel hybrid models
tend to retain similar engine sizes as the non-hybrid models bearing
the same nameplate. However, in some cases this resulted in overly
aggressive SHEVP2 acceleration times; to further maintain performance
neutrality, the final rule sizing algorithm for standard (non-
performance) SHEVP2 vehicle powertrains was changed to allow engine
downsizing such that acceleration performance could converge toward the
target value. This algorithm update is also detailed in Section
VI.B.3.a)(6), Performance Neutrality.
CARB, ICCT, Meszler and ACEEE commented that some combinations of
advanced engines mated with strong hybrids were illogical and
inefficient.\1137\ \1138\ \1139\ \1140\ The commenters specifically
discussed combinations of SHEVP2 with TURBO2 and CEGR1 technologies
that stated the incremental effectiveness resulted in near zero to
negative value, but also clarified that not all combinations showed
inappropriate effectiveness. CARB further expanded that ``[t]hese are
not likely combinations utilized by manufacturers as they unnecessarily
add both gasoline technology and hybrid technology that negates many of
the benefits of the advanced gasoline technology. This error in the
Agencies' modeling leads to inflated technology costs on vehicles that
are converted into P2HEVs.'' \1141\
---------------------------------------------------------------------------
\1137\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 155.
\1138\ American Council for an Energy-Efficient Economy, ACEEE
SAFE NPRM comments, Docket No. NHTSA-2018-0067-12122-22, at 8.
\1139\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-25.
\1140\ Comments from Meszler Engineering Services, Attachment 2,
NPRM Docket No. NHTSA-2018-0067-11723, at 14.
\1141\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 186.
---------------------------------------------------------------------------
The agencies now conclude that the NPRM included certain engine and
strong hybrid pairings that resulted in incremental effectiveness that
exceeded a reasonable level of performance neutrality. The agencies
also agree that Autonomie should model strong hybrid technology
combinations with other engine technologies. In response to these
comments, for the final rule analysis the agencies updated the CAFE
model to allow the use of HCR engine technologies with strong hybrids,
as discussed in Section VI.C.1.c)(4) Engine Maps, HEV Atkinson Cycle
Engines, and improved full vehicle modeling of turbocharged engine
combinations. These changes were discussed in Section VI.B.3.a)(1)
Full-Vehicle Modeling, Simulation Inputs and Data Assumptions and
Section VI.C.2.d)(1)(a) Shifting Controller.
In addition, the agencies limited adoption of advanced engine
technologies with strong hybrids in cases where the electrification
technology would have little effectiveness benefit beyond the benefit
of the advanced engine system, but
[[Page 24488]]
would substantially increase costs. Specifically, the agencies did not
model strong hybrid technologies with VCR engines (eng26a) and eBoost
engines (eng23c). The agencies believe that manufacturers would not
consider these combinations because the combination of electrification
and advanced engine technologies are not as cost-effective as other
technologies.
c) Plug-In Hybrid Vehicles
The effectiveness of the PHEV systems in the analysis was dependent
on both the vehicle's battery pack size and range, in addition to the
other fuel economy-improving technologies on the vehicle (e.g.,
aerodynamic and mass reduction technologies). For the NPRM analysis,
the electrification components were sized to achieve the specified all-
electric range (AER) on the combined cycle (UDDS + HWFET) on the basis
of adjusted energy values. As mentioned above, the PHEV would provide
propulsion energy for a limited range in addition to start-stop or
idle-stop. The NPRM analysis classified PHEVs into two levels: (1)
PHEV30 indicating a vehicle with an AER of 30 miles; and (2) PHEV50
indicating a vehicle with AER of 50 miles.
The resizing algorithm for plug-in hybrid (PHEV) vehicles,
similarly as for SHEVs, considered the power needed for acceleration
performance and all-electric mode operation (compared to regenerative
braking for SHEVs); the PHEV resizing algorithms used those metrics for
an initial estimation of engine, motor(s) and battery powers, and
battery capacity. The initial mass of the vehicle was then computed,
including weight for a larger battery pack and charging
components.\1142\ However, since PHEVs offer expanded electric driving
capacity, their resizing algorithm must also yield a powertrain with
the ability to achieve certain driving cycles and range in electric
mode, in which the engine remains off all or the majority of the
operation. The analysis sized the PHEV electric motors and battery
powers to be capable of completing either the City Cycle (UDDS) or US06
(aggressive, high speed) driving cycle in electric mode, and the
battery energy storage capacity to achieve the specified all-electric
range on the 2-cycle tests on the basis of adjusted energy
values.1143 1144
---------------------------------------------------------------------------
\1142\ FRM ANL Model Documentation, at 8.3 Vehicle Powertrain
Sizing Algorithms.
\1143\ Battery sizing and definition of combined 2-cycle tests
all-electric range is discussed in detail in ANL Autonomie Model
Documentation Chapter 6 Test Procedure and Energy Consumption
Calculation.
\1144\ ANL has incorporated SAE J1711 standard into Autonomie
Modeling. J1711: Society of Automotive Engineers Recommend Practice
for Measuring Exhaust Emissions and Fuel Economy of Hybrid-Electric
Vehicles, Including Plug-In Hybrid Vehicles.
---------------------------------------------------------------------------
The final rule analysis classified PHEVs into four technology
levels, as discussed previously: (1) PHEV20 indicating a vehicle with
an AER of 20 miles and powertrain system based on SHEVPS hybrid
architecture; (2) PHEV50 indicating a vehicle with an AER of 50 miles
and powertrain system based on SHEVPS hybrid architecture; (3) PHEV20T
indicating a vehicle with an AER of 20 miles and powertrain system
based on SHEVP2 hybrid architecture; and (4) PHEV50T indicating a
vehicle with AER of 50 miles and powertrain system based on SHEVP2
hybrid architecture.\1145\ The PHEV20, PHEV20T, PHEV50, and PHVE50T
resizing algorithms were functionally equal, and differed only in the
type of electric mode driving cycle simulated in each one (UDDS for
PHEV20/20T, or US06 for PHEV50/50T). These algorithms simulated the
driving cycles in an iterative loop to determine the size of the
electric motors and the battery required to complete the cycles. In the
case of PHEV20 and PHEV20T, the power of the electric motors and
battery must be sized to propel the vehicle through the UDDS cycle in
``charge-depleting (CD) mode;'' in this mode, the electric machine
alone propels the vehicle except during high power demands, at which
point the engine may turn on and provide propulsion assistance. The
PHEV50 and PHEV50T motor(s) and battery must be sized to power the
vehicle through the US06 cycle in ``electric vehicle (EV) mode,'' where
the engine is off at all times. Then, all PHEV algorithms adjusted the
battery capacity, or vehicle range, by ensuring the battery energy
content was sufficient to complete a simulated UDDS+HWFET combined
driving cycle, based on EPA-adjusted energy consumption. Finally, the
engine, electric motor(s), and battery powers were then sized
accordingly to meet 0-60 and 50-80 mph acceleration targets. All loops
were repeated until the acceleration targets were met without needing
to resize the electric motors, at which point the resizing algorithm
finished. Figure VI-34 below shows the general steps of the PHEV sizing
algorithm. Detailed steps can be seen in section 8.3 of the FRM Argonne
Model Documentation.
---------------------------------------------------------------------------
\1145\ As discussed previously, the NPRM analysis included
PHEV30 instead of PHEV20. However, the related resizing algorithm is
applicable to either.
---------------------------------------------------------------------------
[[Page 24489]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.213
Meszler, CARB, and BorgWarner provided comments on the
effectiveness of the PHEV models. The commenters were concerned with
underperformance of the technology, sizing of the components, and the
variety of PHEV technologies available.
Meszler commented that PHEVs in the 2016 analysis fleet were
inappropriately constrained in their future fuel economy potential by
the ratio of baseline electric-only fuel economy to baseline engine-on
fuel economy; and those vehicles should be allowed to improve that
ratio over time, identically to vehicles that adopt PHEV technology
during the analysis period.\1146\
---------------------------------------------------------------------------
\1146\ Meszler Engineering Services, Attachment 2, NPRM Docket
No. NHTSA-2018-0067-11723 at 32.
---------------------------------------------------------------------------
The agencies must use the SAE J1711 method for determining the fuel
economy for the PHEV systems. The use of SAE J1711 and the underlying
duel fuel vehicle fuel economy calculations are defined by
statute.\1147\ However, it is important to note that PHEVs are not
excluded from applying greater range technologies within the PHEV
technology paths; that is, a PHEV with a lower AER can progress to
become a PHEV with a longer AER.
---------------------------------------------------------------------------
\1147\ 49 U.S.C. 32901(b)(1).
---------------------------------------------------------------------------
CARB commented that several aspects of the agencies' PHEV modeling
contributed to increased PHEV costs. CARB stated that the electric
motors were oversized, that all-electric vehicle efficiencies were low,
and that the lack of battery resizing for road load reductions other
than mass reduction resulted in battery energy capacities much higher
than production vehicles.\1148\ CARB stated the modeled battery
capacity to achieve a given range (kWh/mi) was larger than what exists
on several representative production vehicles.
---------------------------------------------------------------------------
\1148\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 149. Specific comments related to costs
are discussed in Section VI.C.3.e) Overview of Electrification
Costs, below.
---------------------------------------------------------------------------
The agencies agreed with CARB's comments that electric motors and
batteries may be oversized. As a result, the agencies reviewed the
sizing algorithms and methods used in the NPRM analysis and updated the
model for the final rule analysis. The updates resulted in smaller
motor sizes and battery pack sizes for electrified powertrains, as
discussed above. In addition, the review also resulted in a change to
the range categories used for the PHEVs in the final rule analysis; the
final rule analysis classified PHEVs into two levels: (1) PHEV20
indicating a vehicle with an AER of 20 miles; and (2) PHEV50 indicating
a vehicle with AER of 50 miles. For more discussion on the change in
classifications see Section VI.C.3.a)(1)(e) Electrification
Technologies, Plug-in Hybrids.
BorgWarner commented that ``PHEVs and HEVs are complex systems and
should be modeled in detail,'' and further provided, ``[t]herefore,
modeling should be inclusive of all approaches of PHEV and HEV and not
be limited only to Atkinson Cycle engines.'' \1149\ In response, the
agencies created additional powertrain options for PHEV technologies
for the final rule analysis. The additional PHEV technologies included
a plug-in HEV using a turbocharged engine. The additional PHEV paths
used in the final rule analysis are described in Section
VI.C.3.a)(1)(e) Electrification Technologies, Plug-in Hybrids.
---------------------------------------------------------------------------
\1149\ BorgWarner, BorgWarner NPRM public comments 10-26-2018
Final, Docket No. NHTSA-2018-0067-11895, at 10.
---------------------------------------------------------------------------
d) Battery Electric Vehicles
Battery electric vehicles (BEVs) are vehicles with all-electric
drive and with vehicle systems powered by energy-optimized batteries
charged primarily from grid electricity. The effectiveness
[[Page 24490]]
of BEV powertrains is dependent on the efficiency of the components
that transfer power from the battery to the driven wheels. These
components include the battery, electric machine, power electronics,
and mechanical gearing. For the analysis, electric machine efficiency
was based on efficiency maps derived from actual electrified vehicles,
and was scaled such that the peak efficiency value corresponded to the
latest state-of-the-art technologies. The range of the battery electric
vehicles depends on the vehicle's class and the battery pack size. For
the NPRM analysis, manufacturers could apply BEV technology with an AER
of 200 miles. As discussed previously, the final rule analysis added a
BEV 300 to reflect vehicles in the market for the MY 2017 analysis
fleet. For further detailed discussion of how BEV sub-models are
simulated in Autonomie see the FRM Argonne model documentation.\1150\
---------------------------------------------------------------------------
\1150\ FRM ANL Model Documentation, at 4.6, 4.7, 4.13, 4.14, and
5.8.
---------------------------------------------------------------------------
The resizing algorithm for BEVs is functionally the same as the
PHEV algorithm; the difference is that BEVs do not use a combustion
engine, and thus this component was not included in the BEV algorithm.
To begin, initial estimates of motor and battery powers were calculated
based on the criteria of acceleration performance, gradeability
performance, and vehicle range. Then, the algorithm successively ran
four simulation loops to fine tune the powertrain size to ensure that
all performance and operational criteria were maintained. First, the
BEV motor and battery were sized to power the vehicle through the US06
cycle. Next, the battery capacity was adjusted to ensure the energy
content is sufficient to complete a simulated UDDS+HWFET combined
driving cycle, based on EPA adjustment factors to represent sticker
values, and meet the vehicle range requirement. Finally, the electric
motor and battery powers were sized accordingly to meet 0-60 and 50-80
mph acceleration targets. If either acceleration simulation loop
resulted in a change to the electric motor size, the algorithm repeated
all simulation loops. Once the acceleration targets were met without
any resizing of the electric motors, the algorithm finished. Figure VI-
35 below shows a simplified sizing algorithm for BEVs.
[GRAPHIC] [TIFF OMITTED] TR30AP20.214
Meszler Engineering Services, commenting on behalf of NRDC, argued
that the fuel economy for a vehicle adopting BEV technology was
inappropriately dependent on the petroleum-based fuel economy of the
transforming vehicle.\1151\ Meszler reiterated that the fuel economy of
the internal combustion engine that BEV technology replaces does not
have any impact on the efficiency of the resulting BEV, and the
electric machine ``should not care'' whether it replaces a high or low
efficiency engine, and should be modeled accordingly.
---------------------------------------------------------------------------
\1151\ Meszler Engineering Services, Attachment 2, NPRM Docket
No. NHTSA-2018-0067-11723 at 33.
---------------------------------------------------------------------------
The agencies agree with Meszler that BEV effectiveness should be
independent of the vehicle powertrain it will replace in production.
This is, in fact, how the vehicle model and simulation was performed in
Autonomie. Autonomie models the capabilities of each unique full
vehicle system independently, including BEVs. As BEV technology is
adopted by vehicles, the CAFE model uses the Autonomie databases to
determine the added incremental efficiency that will
[[Page 24491]]
bring a specific vehicle up to the appropriate level. Since the CAFE
model considers a variety of vehicle types with differing powertrain
types, vehicle technology classes, performance criteria, and physical
properties (curb weight, etc.), each with a different overall
effectiveness, the observed efficiency increment needed to achieve BEV
effectiveness will vary with each case. While these increments may
differ, the final effectiveness of a BEV is independent of the
powertrain it replaced. The effectiveness used in the CAFE model
represents the difference between the performance of the full vehicle
models--the full vehicle model representing the baseline vehicle and
the full vehicle model representing the end-state with all additional
fuel economy improving technology applied, as discussed in Section
VI.B.3 Technology Effectiveness Values.
ICCT alleged that the agencies did not assess BEV efficiency
improvements from road load reductions (i.e., from mass reduction, tire
rolling resistance, or aerodynamic improvements) to reduce the battery
and power electronic component sizing costs.\1152\ CARB similarly
commented that battery packs were improperly sized, resulting in
underestimation of electrified vehicle effectiveness. CARB stated that
the NPRM constraints on battery sizing caused electrified vehicles to
end up with oversized, less cost-effective battery packs. CARB further
stated that battery designs are more scalable than engines and could
thus be adjusted by manufacturers even at incremental technology
steps.\1153\
---------------------------------------------------------------------------
\1152\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-82.
\1153\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 145.
---------------------------------------------------------------------------
For reference, battery resizing in the NPRM was constrained in the
same manner as other powertrain components, such as the combustion
engine. Resizing would typically be associated with a major vehicle or
engine redesign, which in turn would justify the high costs of changing
the powertrain. In the NPRM, the battery pack and other powertrain
components were not resized for other improvements in incremental
technologies such as AERO and ROLL. The agencies agree that battery
packs, due to their modularity, should be capable of being resized at
relatively lower cost and complexity, and thus should not be subject to
the same resizing restrictions applied to other powertrain components
such as conventional combustion engines. In consideration of CARB and
ICCT's comments on battery pack resizing, for the final rule, the
agencies allowed SHEV, PHEV, and BEV battery packs to be resized at all
incremental technology steps, including for road load reduction
technology improvements (aerodynamics, rolling resistance reduction,
and low levels of mass reduction). This avoided the additional cost and
range associated with oversized battery packs on BEVs and other
electrified vehicles.
CARB commented that the NPRM analysis oversized battery packs that
targeted 200-mile label range, resulting in exaggerated battery pack
costs. CARB also stated that some MY 2016-2018 BEVs exist that have a
higher efficiency than simulated for BEV200s in Autonomie. They further
argued that although these vehicles were assigned BEV200s, their actual
range was greater than 200 miles.\1154\
---------------------------------------------------------------------------
\1154\ California Air Resources Board, Attachment 2, Docket No.
NHTSA-2018-0067-11873, at 147.
---------------------------------------------------------------------------
We agree with CARB that the NPRM modeled and simulated battery
packs were oversized and that the AERs for BEVs did not match the
current and expected future vehicle AERs. In response to these
comments, for the final rule analysis, the agencies removed certain
constraints from the Autonomie battery sizing algorithm, allowing
batteries to be sized as function of all road load reduction
technologies. As discussed earlier, this additional battery sizing is
feasible due to the modularity of battery pack construction. This
update allowed the battery pack cost and mass to better reflect the
actual required energy capacity and power, and improved the efficiency
of modeled BEVs. The agencies also updated the modeling of electric
machines used in BEVs to reflect improvements in efficiency.
Furthermore, the agencies added the BEV300 (with an AER of 300 miles)
to the final rule analysis, providing a better representation of
production BEVs with more than 200 miles of range. For more discussion
on BEV300 and electrification efficiency improvements, see Sections
VI.C.3.a)(1) Electrification technologies and VI.C.3.d)(1) Electric
Motors, Power Electronics and Accessory Load.
e) Fuel Cell Vehicles
The fuel-cell system in the analysis was modeled to represent
hydrogen consumption as a function of the produced power, assuming
normal-temperature operating conditions with a peak system efficiency
of 60 percent, including the balance of plant.\1155\ The system's
specific power is 650 W/kg. The hydrogen storage technology selected
was a high-pressure tank with a specific weight of 0.04 kg H2/kg, sized
to provide a 320-mile range on the 2-cycle tests on the basis of
adjusted energy values.
---------------------------------------------------------------------------
\1155\ Power needed for supporting components and auxiliary
systems. The balance of plant in a fuel cell system is the auxiliary
equipment required to ensure the fuel cell operates as a reliable
power source. This may include fuel reformers and pumps, for
example.
---------------------------------------------------------------------------
The sizing algorithm for FCVs was similar to PHEVs and BEVs, but
adapted to size the specific components of a FCV powertrain: the
electric motor, fuel-cell, hydrogen (H2) fuel tank, and
battery pack. The electric motor drives the wheels needed to propel the
vehicle. During very low power operation, the battery pack alone powers
the motor/wheels, depleting the battery charge. At moderate driving
loads, the fuel-cell provides electrical power (generated by consuming
stored H2) to the motor and also to charge the battery.
Under heavy loads, both the fuel cell and battery deliver electric
power to the motor. To begin, initial estimates of motor, fuel cell,
and battery powers are calculated based on criteria for acceleration
performance, gradeability performance, and vehicle range. Then, the
algorithm successively runs four simulation loops to finetune
powertrain size, ensuring that all performance and operational criteria
are maintained. First, the FCV motor and battery are sized to power the
vehicle through the US06 cycle. Next, the on-board mass of H2 fuel, as
well as the fuel tank mass are adjusted to ensure the vehicle can
complete a simulated 2-cycle test and meet the range requirement.
Finally, the electric motor and fuel cell powers are sized accordingly
to meet 0-60 and 50-80 mph acceleration targets. If either acceleration
simulation loop results in a change to the electric motor size, the
algorithm repeats all simulation loops. Once the acceleration targets
can be met without any resizing of the electric motor, the algorithm
completes. Figure VI-36 below shows a simplified sizing algorithm for
FCVs.
[[Page 24492]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.215
The agencies did not receive comments on FCV modeling in Autonomie.
For the final rule analysis, the agencies used the same FCV model and
simulations to estimated effectiveness values.
e) Electrification Costs
The primary factors that influence the cost and effectiveness of
hybrid or battery electric vehicles are the cost and efficiency of the
energy storage components and electric machines. Energy storage
components include battery cells, battery management systems, and
thermal management systems. The electric machine components include
electric motors, power electronics, controllers, and other devices that
support thermal management.
Charging infrastructure is an essential component for PHEVs and
BEVs, and may add to the total cost of ownership of the vehicle.
However, most households are equipped with a 110-volt outlet for level
1 charging, for which no additional cost is incurred. Installing a
level 2 charging outlet (220-volt) will add cost to the total ownership
of the vehicle but decreases charging time. The price of level 2
residential charging equipment varies, but typically ranges from $500
to $2,000 before installation and state or utility incentives.\1156\
---------------------------------------------------------------------------
\1156\ U.S. Department of Energy Office of Energy Efficiency and
Renewable Energy, Charging at Home, https://www.energy.gov/eere/electricvehicles/charging-home (last visited March 20, 2020).
---------------------------------------------------------------------------
For this final rule analysis, the agencies used Argonne's BatPaC
modeling tool to develop battery pack manufacturing costs as well as
weight.\1157\ Battery packs were sized in terms of the vehicle's energy
and power requirement and costs were estimate for each of the simulated
technology combinations. The Argonne team used BatPaC to create a
``lookup table'' with battery pack size (energy and power) and cost as
well as weight data for the full vehicle simulations to ``reference,''
to avoid the need for conducting a full BatPaC simulation for each
unique vehicle modeled in the analysis. The table included cost data
for each technology key and vehicle technology classes. As discussed
below, Autonomie runs linearly interpolate between points in the lookup
tables when deriving final values from BatPaC, the differences between
using BatPaC for each configuration and the interpolation using the
lookup table was insignificant.
---------------------------------------------------------------------------
\1157\ The agencies used BatPaC version 3.0 (released in 2015)
for the NPRM and BatPaC version 3.1 (June 2018) for the final rule.
---------------------------------------------------------------------------
The agencies used the cost of electric machines from U.S. DRIVE's
October 2017 report, ``Electrical and Electronics Technical Team
Roadmap.'' In industry, manufacturers use different types of electric
machines resulting in a range of actual costs for the systems. To
capture this range, the agencies considered a single type of high
efficiency electric machine, representative of the range of technology
available in the rulemaking timeframe, uniquely sized for each of the
simulated combinations. For the final rule analysis, the cost of the
electric machine was determined using a dollar-per-kilowatt metric. The
agencies sized the electric machines using the method discussed in
Section VI.C.3.d) Electric Effectiveness Modeling and Resulting
Effectiveness Values.
The following sections discuss the method used for modeling battery
and non-battery component costs, the learning curves applied to those
costs, and the total costs for each type of electrification technology
considered in this final rule analysis.
(l) Battery Pack Modeling
BatPaC is a software designed for policymakers and researchers
interested in estimating the manufacturing cost of lithium-ion
batteries for electric drive
[[Page 24493]]
vehicles.\1158\ BatPaC is used to estimate the cost of manufacturing
lithium-ion batteries and examine trade-offs that result from different
battery performance specifications such as power and energy capacity.
BatPaC includes a library of lithium ion electrode combinations and
inputs for all the parameters associated with materials and
manufacturing operations in a factory.
---------------------------------------------------------------------------
\1158\ BatPaC: Battery Manufacturing Cost Estimation, Argonne
National Laboratory, https://www.anl.gov/tcp/batpac-battery-manufacturing-cost-estimation.
---------------------------------------------------------------------------
Specifically, BatPaC models stiff-pouch, laminated prismatic format
cells, placed in double-seamed, rigid modules. The model supports
liquid- and air-cooling, accounting for the resultant structure,
volume, cost, and heat rejection capacity. The model considers cost of
capital equipment, plant area and labor for each step in the
manufacturing process. The model places relevant limits on electrode
coating thickness, and considers limits applicable to current and near-
term manufacturing processes. The model also considers annual pack
production volumes and economies of scale for high-volume production.
BatPaC calculations are based on a generic pack designs that
reasonably represents the weight and manufacturing cost of batteries
deployed commercially. The advantage of using this approach is the
ability to model wide range of commercial design specifications for the
various classes of vehicles. This modeling approach is particularly
advantageous because the data from commercially available battery packs
is limited and varies widely with respect to the underlying
specifications (power and energy) and constraints (mass, volume,
dimensions, durability) set by the manufacturer.
BatPaC is a Microsoft Office Excel spreadsheets-based model. The
data needed to design and build a battery pack, such as dimensions of
the cell, estimate of materials, and manufacturing cost, are provided
in the model, with the manufacturing costs for the designed battery
based on a ``baseline plant'' designed for a battery of intermediate
size and production scale so as to establish a center-point for other
designs. BatPaC can be configured with alternative chemistries,
charging constraints, battery configurations, production volumes, and
cost factors for other battery designs by customizing these parameters
in the modeling tool.
For this analysis, running individual BatPaC simulations for each
full vehicle simulation requiring an electrified powertrain would have
been computationally intensive and impractical, given that
approximately 750,000 simulated vehicles out of the 1.2 million total
simulated vehicles had an electrified powertrain. Accordingly, staff at
Argonne built ``lookup tables'' with BatPaC to provide battery pack
manufacturing costs, battery pack weights, and battery pack cell
capacities for vehicles modeled in the large-scale simulation runs.
To build the lookup tables, Argonne staff selected a range of
minimum and maximum values for battery pack power (kW) and battery pack
energy (kWh) for each vehicle powertrain based on a combination of
market analysis and analysis of the Autonomie simulations that were run
for the NPRM and final rule. The performance requirements (vehicle
acceleration times, EV range, etc.) were defined from set assumptions
and validated from existing vehicles.\1159\ The range, as well as the
number of power and energy points considered to generate each lookup
table, varies across powertrains. The minimum and maximum power and
energy values have been selected to encompass current designs. For
example, one end of the spectrum is representative of the MY 2016-2017
Tesla Model S 100D (100 kWh total battery energy, 335-mile range),
while the other end of the spectrum is representative of the 2017
Mitsubishi iMiEV (16 kWh total battery energy, 62-mile range). The
components were then sized in Autonomie across all vehicle classes to
define the minimum and maximum values to be considered, as shown in
Table VI-90.
---------------------------------------------------------------------------
\1159\ See Final Rule Argonne Model Documentation Section 5.9,
Battery Performance and Cost Model (BatPaC).
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.216
Figure VI-37 illustrates the inputs generated in Autonomie to
create the BatPaC-based lookup tables, and the outputs characterized in
the BatPaC-based lookup tables that are used to provide estimates
referenced in the agencies' analysis. A linear interpolation was then
performed in MATLAB to determine the associated values for battery pack
manufacturing cost, weight, and cell capacity.
[[Page 24494]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.217
Figure VI-38 shows the linear relationship between cost, power, and
weight used to generate the compact passenger car BEV200 technology
class lookup table presented in Figure VI-39. As seen from the figures
below, the energy values produced by BatPaC consist of a fairly linear
relationship with respect to power and energy for a vehicle class.
Since Autonomie runs would linearly interpolate between the points in
the lookup tables when deriving the final values from BatPaC, the
differences between using BatPaC for each configuration and the
interpolation using the lookup table were insignificant.
[GRAPHIC] [TIFF OMITTED] TR30AP20.218
BILLING CODE 4910-59-C
Figure VI-39 details the estimates of $ per kWh at the pack level
generated from the lookup table for BEV200 compact cars used in the
final rule analysis. As discussed further below, the specific battery
costs for each simulated vehicle were presented for the NPRM (and now
for the final rule) in the docketed Argonne assumptions files and in
the vehicle simulation database included in the CAFE model.
[[Page 24495]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.219
During the Autonomie large-scale simulation runs, calling the
BatPaC model for each individual simulation would have been
computationally intensive. Using the MATLAB lookup tables reduced the
time to run the approximately 750,000 simulations significantly, which
in turn reduced the total simulation run time for all of the technology
combinations by several days with insignificant impact on the
analytical results.
(a) BatPaC Inputs and Assumptions
The Argonne documentation describing the analysis performed for the
NPRM, ``A Detailed Vehicle Simulation Process To Support CAFE
Standards,'' detailed the specific assumptions that Argonne's experts
used to simulate batteries and their associated costs for the full
vehicle simulation modeling.\1160\ In addition, detail on the NPRM
electrification analysis was presented in the PRIA.\1161\ While the
Argonne Summary of Main Component Assumptions Excel file correctly
identified the chemistry used in the NPRM analysis as NMC333,\1162\ the
PRIA inadvertently described that NMC441 was used. The agencies
presented selected lookup table battery cost values in the Argonne
Summary of Main Component Assumptions Excel file,\1163\ as shown above,
and the specific battery costs for each simulated vehicle were
presented for the NPRM and final rule in the vehicle simulation
database included in the CAFE model.
---------------------------------------------------------------------------
\1160\ Islam S. Ehsan. Moawad, Ayman. Kim, Namdoo. Rousseau,
Aymeric. ``A Detailed Vehicle Simulation Process to Support CAFE
Standards.'' ANL/ESD-18/6. Energy Systems Division, Argonne National
Laboratory (2018).
\1161\ PRIA at 362-384.
\1162\ ANL--All Assumptions Summary, NHTSA-2018-0067-0005.
\1163\ ANL--Summary of Main Component Performance Assumptions
NPRM, NHTSA-2018-0067-0003.
---------------------------------------------------------------------------
Several commenters claimed that costs for electrification
technologies were too high, especially regarding battery costs (note
that comments on non-battery component costs are addressed separately
in Section VI.C.3.e)(2) Non-battery Electrification Component Costs,
below).\1164\ Several commenters pointed to text in interagency review
documents that stated the NPRM battery modeling costs were higher than
what EPA recommended,\1165\ and higher than what EPA had obtained from
the most recent version of the BatPaC model.\1166\
---------------------------------------------------------------------------
\1164\ Meszler Engineering Services, NHTSA-2018-0067-11723
Attachment 2; National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969; Workhorse Group Inc., NHTSA-2018-0067-12215;
International Council on Clean Transportation, NHTSA-2018-0067-
11741; California Air Resources Board, NHTSA-2018-0067-11873.
\1165\ California Air Resources Board, NHTSA-2018-0067-11873.
\1166\ Boulder County Public Health et al., NHTSA-2018-0067-
11975; International Council on Clean Transportation, NHTSA-2018-
0067-11741.
---------------------------------------------------------------------------
CARB commented that the agencies incorrectly identified and
assessed existing technologies, improperly oversized components and
batteries for the modeled vehicle classes, and underestimated
technology efficiency through improper modeling.\1167\ CARB also
submitted supplemental comments (discussed further, below) stating that
the PRIA and the underlying modeling were inconsistent regarding which
exact battery chemistries were modeled for every electrified model in
the fleet, which CARB argued was crucial for understanding the battery
compositions and thus their production costs.\1168\
---------------------------------------------------------------------------
\1167\ California Air Resources Board, NHTSA-2018-0067-11873.
\1168\ California Air Resources Board, NHTSA-2018-0067-4166.
---------------------------------------------------------------------------
ICCT stated that the agencies misrepresented the leading research
on both battery and electric vehicle costs, with the result being that
electric vehicles were so costly that they were modeled to remain at
approximately the same penetration in 2025 with the Augural 2025 fuel
economy and adopted 2025 CO2 standards, as they were in mid-
2018 (i.e., between 1.5 percent and 2 percent of new vehicle
sales).\1169\ ICCT stated that the agencies' inputs failed to reflect
the latest industry data on future potential electric vehicle cost
parity with combustion vehicles. ICCT commented that through a
combination of incorrectly high electric vehicle prices (which, they
argue, do not reflect Argonne or other leading battery research groups'
work), and modeling restrictions on electric vehicles, the agencies
unduly inflated technology costs of electric vehicles to comply with
the standards. ICCT argued that although the agencies purported to use
state-of-the-art tools like the BatPaC model for battery costs, the
cost calculations erroneously pushed up electric vehicles' incremental
costs above $10,000 per vehicle. ICCT claimed that the agencies
introduced errors that artificially pushed up the battery costs higher
than indicated by BatPaC and other experts in the field.
---------------------------------------------------------------------------
\1169\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
---------------------------------------------------------------------------
NCAT noted that the PRIA described some ways in which the modeling
increased battery costs, namely, that the battery pack costs were
adjusted
[[Page 24496]]
upwards, the cost of the battery management system increased, and a
cost for a battery automatic and manual disconnect unit was
added.\1170\ Regardless, NCAT stated that the agencies analysis was not
sufficiently transparent, and argued that the battery costs were
significantly overestimated in the modeling supporting the NPRM.
Boulder County Public Health and other Colorado municipal organizations
claimed that overstated battery costs had the effect of
mischaracterizing and downplaying the benefits of increased numbers of
electric vehicles as part of the vehicle fleet.\1171\ Commenters also
argued that discrepancies existed between battery costs used in the
rulemaking documents and battery costs found in the Argonne database,
referring specifically to BISG and CISG costs (discussed further
below).\1172\
---------------------------------------------------------------------------
\1170\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing PRIA at 366-67.
\1171\ Boulder County Public Health et al., NHTSA-2018-0067-
11975.
\1172\ Meszler Engineering Services, NHTSA-2018-0067-11723
Attachment 2; International Council on Clean Transportation, NHTSA-
2018-0067-11741.
---------------------------------------------------------------------------
In addition to comments claiming that the agencies' battery cost
projections were incorrect or difficult to interpret, many commenters
submitted general information about the state of battery technology and
cost advances now and as projected into the future. For example, NCAT
stated that battery technology has improved and battery costs have
fallen dramatically, due in part to reduced material costs,
manufacturing improvements, and higher manufacturing volumes.\1173\ In
compliment, NCAT asserted that the demand for EVs is growing
``dramatically.''
---------------------------------------------------------------------------
\1173\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969. NCAT also stated that the increase in mass
manufacturing of lithium-ion storage is expected to continue to
reduce battery prices.
---------------------------------------------------------------------------
ICCT stated that the agencies' analysis of electric vehicle costs
and the resulting extremely low penetration levels was not in line with
automakers' announcements, which included statements that they would
produce far greater numbers of electric vehicles to comply with
standards around the world.
ICCT summarized projections of electric vehicle battery costs for
2020-2030, and stated that the agencies did not analyze the studies and
automaker announcements they cited to understand the potential for
cost-effective electric drive technology.\1174\ ICCT stated the data
they reviewed included a variety of different technologies, production
volumes, and cost elements, and although there were differences in
methods for each, ``they generally include in some variation of
material, process, overhead, depreciation, warranty, and profit
costs.'' ICCT summarized the results of their review, projecting that
battery pack costs will decline to $150/kWh by 2020-2023 and then to
about $120-$135/kWh by 2025, with the exception of Tesla, which reports
costs of $150 kWh in 2018 and projected costs of $100/kWh by 2022. ICCT
stated that the results of this review were corroborated in the
aforementioned EPA interagency comments on battery costs used in the
proposal.
---------------------------------------------------------------------------
\1174\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
---------------------------------------------------------------------------
NCAT stated that the average price of a battery pack dropped from
$1,000/kWh in 2010 to $209/kWh in 2017, demonstrating a decrease of 79
percent in seven years.\1175\ NCAT stated Tesla is on track to achieve
$100/kWh by the end of 2018, and Audi has been buying batteries at
$114/kWh, according to trade press reports.\1176\ NCAT also cited BNEF
analyses showing that battery costs are projected to continue to
decline substantially,\1177\ specifically projecting a decrease in
battery cost of 77 percent between 2016 and 2030. Accordingly, NCAT
stated that EVs will be less expensive to buy than conventional
gasoline vehicles by 2025 in the United States.\1178\ Workhorse
similarly echoed the assertion that EV costs will reach parity with
conventional vehicle costs before 2025.\1179\
---------------------------------------------------------------------------
\1175\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Bloomberg New Energy Finance, ``Electric
Vehicle Outlook: 2018,'' https://bnef.turtl.co/story/evo2018?teaser=true.
\1176\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Fred Lambert, ``Tesla to achieve leading
$100/kWh battery cell cost this year, says investor after
Gigafactory 1 tour'' (Sept. 11, 2018), https://electrek.co/2018/09/11/tesla-100-kwh-battery-cost-investor-gigafactory-1-tour/.
\1177\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Bloomberg New Energy Finance, ``Electric
Vehicle Outlook: 2018,'' https://bnef.turtl.co/story/evo2018?teaser=true.
\1178\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Jess Shankleman, ``Pretty Soon Electric Cars
Will Cost Less Than Gasoline'' (May 26, 2017), https://www.bloomberg.com/news/articles/2017-05-26/electric-cars-seen-cheaper-than-gasoline-models-within-a-decade; Jess Shankleman, ``The
Electric Car Revolution Is Accelerating'' (July 6, 2017), https://www.bloomberg.com/news/articles/2017-07-06/the-electric-car-revolution-is-accelerating. NCAT also noted that the up-front cost
parity does not take into consideration the fuel savings and
maintenance savings over the lifetime of EV use as compared to
gasoline vehicle use.
\1179\ Workhorse Group Inc., NHTSA-2018-0067-12215.
---------------------------------------------------------------------------
NCAT also cited the ICCT Efficiency Technology and Cost Assessment,
which concluded that, primarily because of rapid developments in
battery pack technologies, EV costs will be reduced by $4,300-$5,300
per vehicle by 2025 compared to EPA's prior estimates in support of the
MY 2017-2025 standards.\1180\ In that report, ICCT concluded that
battery costs of $140/kWh is a realistic estimated value by 2025, as
compared with EPA estimates in the 2016 Mid-Term Evaluation (MTE)
analysis of $180-200/kWh.\1181\
---------------------------------------------------------------------------
\1180\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing ICCT, ``Efficiency Technology and Cost
Assessment for U.S. 2025-2030 Light-duty Vehicles'' (Mar. 2017) at
11, 15, available at http://www.theicct.org/US-2030-technology-cost-assessment.
\1181\ Id.
---------------------------------------------------------------------------
NCAT also cited improvements in manufacturing techniques,
specifically by Tesla, as an example of how batteries are being
manufactured in large volumes with high quality at low cost.\1182\ NCAT
stated that in mid-2018, Tesla was producing batteries at its
Gigafactory 1 facility at an annualized rate of roughly 20 GWh, making
it the highest-volume battery plant in the world.\1183\ NCAT and other
commenters also cited Bloomberg's New Energy Finance research stating
that the average energy density of EV batteries is improving at around
5-7 percent per year.
---------------------------------------------------------------------------
\1182\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Tesla, Inc., S.E.C. Form 10-K (Feb. 22,
2018) at 3-4, available at https://www.sec.gov/Archives/edgar/data/1318605/000156459018002956/tsla-10k-20171231.htm.
\1183\ National Coalition for Advanced Transportation, NHTSA-
2018-0067-11969, citing Tesla, ``Tesla Gigafactory,'' https://www.tesla.com/gigafactory (last visited Oct. 25, 2018).
---------------------------------------------------------------------------
Finally, Workhorse commented that they have more than ten years of
experience in the field of designing and assembling battery packs, and
their business plans are predicated on battery costs much lower than
assumed by the agencies.\1184\
---------------------------------------------------------------------------
\1184\ Workhorse Group Inc., NHTSA-2018-0067-12215.
---------------------------------------------------------------------------
As explained above, the agencies consulted with and relied on
Argonne battery experts to develop inputs to the BatPaC model and
generate the battery cost lookup tables used as references for the
Autonomie full-vehicle simulations, as detailed in Argonne's
documentation supporting the NPRM analysis.\1185\ As explained further
below, the agencies also directed CARB to information about the NPRM
battery cost analysis available
[[Page 24497]]
in the public docket in response to their FOIA request.
---------------------------------------------------------------------------
\1185\ Islam S. Ehsan. Moawad, Ayman. Kim, Namdoo. Rousseau,
Aymeric. ``A Detailed Vehicle Simulation Process to Support CAFE
Standards.'' ANL/ESD-18/6. Energy Systems Division, Argonne National
Laboratory (2018).
---------------------------------------------------------------------------
Commenters are correct that the EPA Draft TAR and Proposed
Determination estimates for battery sizing and cost were different than
the NPRM analysis. For the Draft TAR and in the Proposed Determination,
a separate battery and motor sizing spreadsheet was built to determine
the energy and power requirements for PHEVs and BEVs at different curb
weights, and then BatPaC was used to determine specific energy (kWh/kg)
and the battery pack cost estimate.\1186\ For this NPRM and final rule,
the energy requirements for PHEVs and BEVs were determined using
Autonomie simulations with the integrated BatPaC lookup table to select
the appropriate battery pack size, cost, and weight. As discussed in
Sections VI.B.3.a)(4) How Autonomie Sizes Powertrains for Full Vehicle
Simulation and VI.B.3.a)(6) Performance Neutrality, the Autonomie full-
vehicle simulation modeling assessed metrics to ensure performance
requirements were met for every modeled vehicle. Appropriately
accounting for vehicle metrics and individual vehicle power and weight
requirements resulted in some of the differences observed between the
Draft TAR and Proposed Determination estimates and the estimates
presented in the NPRM and this final rule.
---------------------------------------------------------------------------
\1186\ Draft TAR at 5-315.
---------------------------------------------------------------------------
For the final rule, the agencies considered these public comments,
market observations, literature, industry reports, and additional
research. In addition, as described further below and in the Argonne
documentation accompanying this final rule, Argonne consulted the
A2Mac1 database for additional data points on batteries that were used
to inform the final rule battery cost modeling.
As discussed above, BatPaC version 3.0 was used for the NPRM
analysis because that was the most up-to-date version of BatPaC
available at the time the NPRM analysis was being conducted. BatPaC
version 3.1, released after the NPRM analysis was completed, was used
for this final rule because that was the most up-to-date version of
BatPaC available at the time the final rule analysis was being
conducted.
The agencies note that BatPaC version 4.0 has been released since
the analysis was completed for this final rule. Specifically, that
version was released on January 14, 2020, after the rule had been
submitted for interagency review. The default battery chemistry in
BatPaC version 4.0 continues to be NMC622, which as discussed further
in Section (i) below, reflects the reasonable assumption this chemistry
will likely continue to be used in the rulemaking timeframe based on
its commercial application and market trends towards higher-nickel,
lower-cobalt content chemistries.\1187\ As explained in this section,
and further in Section (c) below, the agencies' modeled costs for
battery packs aligns with current industry estimates and closely tracks
future projections of battery pack costs from the Department of
Energy's Vehicle Technology Office (DOE VTO) lab
targets.1188 1189
---------------------------------------------------------------------------
\1187\ The agencies note that BatPaC version 4.0 provides a new
option to build battery packs with NMC811.
\1188\ Freyermuth, Vincent. Rousseau, Aymeric. ``Impact of
Vehicle Technologies Office Targets on Battery Requirements.'' ANL/
ESD-16/22. Energy Systems Division, Argonne National Laboratory
(2016).
\1189\ Hummel et al., UBS Evidence Lab Electric Car Teardown--
Disruption Ahead?, UBS (May 18, 2017), https://neo.ubs.com/shared/d1ZTxnvF2k/.
---------------------------------------------------------------------------
In addition to using BatPaC version 3.1 for this final rule, BatPaC
assumptions were updated to reflect what the Argonne battery experts
and the agencies believed would be representative and attainable of
battery manufacturing trends in the rulemaking timeframe. Section (ii)
provides additional information on BatPaC inputs and assumptions that
were updated for the final rule based on public comments and the
agencies own market observations and additional research. In addition,
as discussed further below, for the final rule, the calculated battery
pack weight and manufacturing cost was compared with the battery pack
cost and weight data obtained through various benchmarking studies. The
agencies believe that the Argonne methodology for producing the
hundreds of thousands of battery pack cost estimates required for the
full-vehicle modeling and simulation resulted in reasonable estimates
of battery pack costs. The following sections provide additional
context and response to comments on specific BatPaC inputs and
assumptions used in the NPRM and final rule.
(i) Chemistry
The choice of chemistry for battery cells depends on the
application and consideration of cost, energy density, and safety,
among other factors. The PRIA described the battery pack cell chemistry
used for different powertrain types modeled in the NPRM analysis.\1190\
For Micro HEVs, BISG HEVs, CISG HEVs, and Full HEVs, the agencies used
LFP-G, rather than LMO-G, because the latter has a limited lifespan
which is expected to degrade functionality over a vehicle's lifetime,
and has greater limitations on available ranges of battery charge and
discharge rates. As described above, for PHEVs and BEVs, the Argonne
``Summary of Main Component Performance Assumptions'' file correctly
stated that NMC333 was used, however the PRIA misstated that NMC441 was
used.
---------------------------------------------------------------------------
\1190\ PRIA at 373.
---------------------------------------------------------------------------
Both UCS and CARB commented on the agencies' choice of battery
chemistry, with UCS noting that this choice can have a large impact on
performance and materials costs, and therefore on the modeled cost of
drivetrain electrification.
First, both commenters stated that the NPRM documentation was
inconsistent and unclear. UCS noted the discrepancy between the PRIA
and Argonne model documentation, and also that the rulemaking documents
stated the most recent version of Argonne's BatPaC model was used to
estimate battery costs, but the default lithium ion chemistry in the
current BatPaC model is NMC622. UCS stated the choice of NMC variant
effects battery costs, as NMC622 replaces more expensive cobalt with
nickel. UCS further stated it was not possible to determine the
magnitude of the cost error in the PHEV and BEV battery pack costs,
only that the costs were likely higher than current battery cost data
supported.
CARB stated that the agencies' selected battery chemistries
represented a step backward from previous analysis done for the Draft
TAR. CARB claimed that the biggest lithium-ion production companies
have indicated that they will use NMC811 for BEVs, and therefore NMC441
or NMC333 would not represent current technology going into BEVs or
near-future BEV battery technology. CARB stated that NMC811 technology
was expected to come to market in 2019, which is far sooner than
anticipated, even in the agencies' prior analyses.
Commenters also noted that the chemistry chosen for mild and strong
hybrids differed from what is used in current and announced HEVs. UCS
stated that all non-plug-in hybrids in the proposed rule analysis used
lithium iron phosphate (LFP) chemistry, but in practice, most hybrids
on the road did not use this chemistry. UCS referenced the Toyota Prius
and the new RAM 1500 pickup as examples of vehicles that do not use LFP
chemistry. CARB similarly stated that the NPRM battery chemistry
selection for PHEV and strong hybrid batteries does not represent many
of the batteries that are being deployed in the market, nor have been,
for several years now, but did not provide an alternative chemistry
they believed to be better
[[Page 24498]]
represented in the market. CARB stated that this resulted in a
``misappropriation of higher costs for electrification technologies in
the Agencies' analysis, and further highlights the Agencies' sudden
lack of knowledge about electrification, despite the far more
directionally correct projections in previous analysis for the 2016
Draft TAR and EPA's Proposed Determination.''
Similarly, UCS pointed to a discrepancy in strong hybrid battery
costs between the proposed rule estimates (greater than $1,200, even
for the small car classes) and an estimate from Argonne in 2017 ($614),
to argue that the lack of detailed information made it impossible to
determine if the choice of battery chemistry was responsible for the
discrepancy.
The agencies carefully considered these comments. As stated above,
the agencies disagree that the discrepancy in the Argonne Summary of
Main Component Performance Assumptions file and the PRIA over the use
of NMC333 for the NPRM analysis limited commenters ability to comment
on battery chemistry, as both UCS and CARB communicated a belief that
the agencies choice of battery chemistry contributed to the overstated
battery costs in the NPRM. The agencies understand how the choice of
chemistry impacts battery costs, and many of the commenters' concerns
intertwined the NPRM choice of battery chemistry with the NPRM battery
costs. Here, the agencies respond to comments on the choice of
chemistries. The agencies will also discuss costs below.
As stated earlier, although manufacturers use different battery
chemistries in various HEV, PHEV, and BEV applications, the choice of
chemistry for a given application depends on several factors including
safety, stability, and functional requirements (high power or high
energy requirements for performance) of the battery pack. In
determining whether to select one battery chemistry over another, the
agencies concluded that using commercially proven technologies that
represented the current cost of production was more reasonable than
assuming additional technologies would come to fruition during the
rulemaking timeframe, and attempting to project the cost and
effectiveness of such technologies. While there is ongoing research and
development in battery chemistry and in other battery related
technologies that have the potential to reduce costs and increase
battery capacity, these technologies have yet to be proven viable for
commercial use.\1191\
---------------------------------------------------------------------------
\1191\ Recent Advances in Energy Chemical Engineering of Next-
Generation Lithium Batteries, Engineering, Volume 4, Issue 6
(December 2018), at 831-847. Available at https://www.sciencedirect.com/science/article/pii/S2095809918312177. Some
examples include lithium-sulfur battery cell chemistry and solid-
state electrolyte battery cells.
---------------------------------------------------------------------------
In addition, as discussed throughout this document, the agencies
considered technologies that manufacturers could use to comply with
standards in the rulemaking timeframe that reasonably represented the
state of technology across the industry. While the battery chemistries
used in commercial vehicles are largely confidential business
information, proprietary teardown reports are one source of information
used to learn more about the chemistries actually employed in the
market. For both the NPRM and final rule, the agencies consulted
Argonne's battery experts to determine the chemistries that should be
modeled in the BatPaC analysis. Argonne consulted A2Mac1 battery pack
teardown reports, which confirmed that indeed, manufacturers use a
range of chemistries across the electrified vehicle types. Selecting
battery chemistries that can reasonably represent the range employed in
the market ensured that the analysis better captured the average of
costs across the industry.
For example, in addition to the reasons listed in the NPRM, LFP has
been proven in commercial use, as identified in literature and battery
teardown reports.\1192\ This presented a basis for using LFP, as the
chemistry was reasonably representative of chemistries used in mild and
strong hybrids at the time of the analysis. The agencies also
considered that LFP's lower cost compared to other potential HEV
battery chemistries (contrary to commenters' statements) made it more
attractive for vehicles with tight cost constraints, even with the
associated lower energy density.
---------------------------------------------------------------------------
\1192\ Details of cell chemistry and battery cooling system are
described in Nelson, Paul A., Gallagher, Kevin G., Bloom, Ira D.,
and Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles--SECOND EDITION (2012),
available at https://publications.anl.gov/anlpubs/2015/05/75574.pdf.
---------------------------------------------------------------------------
Similarly, although EPA selected NMC622 as the modeled battery
chemistry for the Draft TAR, manufacturers were also using other NMC
chemistries in hybrid and BEV applications in that timeframe depending
on the required application. The chemistry selected for the NPRM,
NMC333, was selected based on proprietary teardown reports that
demonstrated the chemistry's commercial use: a survey of twelve MY 2013
to MY 2018 HEVs, PHEV, and BEVs showed that NMC333 was used in eleven
of those vehicles, and NMC622 was only used in one.\1193\
---------------------------------------------------------------------------
\1193\ A Detailed Vehicle Simulation Process To Support CAFE and
CO2 Standards for the MY 2021--2025 Final Rule Analysis,
Section 5.9 Battery Performance and Cost Model (BatPaC), referencing
A2Mac1 Automotive Benchmarking, https://a2mac1.com.
---------------------------------------------------------------------------
Accordingly, the agencies believe that assuming LFP-G as the
modeled cell chemistry for HEVs and NMC333 as the modeled PHEV and BEV
chemistry for the NPRM analysis of battery costs was not unreasonable,
based on their demonstrated commercial use in a range of electric
vehicle applications. However, employing BatPaC version 3.1 for the
final rule analysis also presented the opportunity to update the
modeled battery chemistry used to assess battery costs.
The agencies similarly consulted Argonne battery experts on battery
chemistry and trends to inform the final rule analysis. Argonne staff
used the A2Mac1 database to determine real-world battery chemistry and
configurations in different electric vehicle applications. As shown in
the Argonne Full Vehicle Modeling documentation for the final rule, the
A2Mac1 battery pack teardown analysis provided an array of data points
on battery chemistries for different electric vehicle applications,
among other relevant battery pack data, that informed the final rule
battery analysis.\1194\
---------------------------------------------------------------------------
\1194\ Id.
---------------------------------------------------------------------------
In determining which of these chemistries would best represent the
range of chemistries demonstrated in the market, the agencies
considered several issues. Due to the increasing manufacturing volume
of battery packs with NMC, it is expected that NMC battery cells will
continue to be used in battery packs across different electric vehicle
applications in the future. The agencies considered concerns about NMC
formulations with varying cobalt content, and issues including the
current and future cost of cobalt,\1195\
[[Page 24499]]
and the cobalt supply chain.\1196\ These concerns, among others, have
led to the market shift towards cathode active materials with a higher
fraction of nickel and less cobalt.\1197\ Manufacturers have
demonstrated the use of NMC622, which contains more nickel and less
cobalt than NMC333, in different electric vehicle applications. In
addition, as CARB noted and has been reported in the news for some
time, the expected next step in battery chemistries using even less
cobalt is NMC811. However, the shift to higher-nickel-content
chemistries is not without challenges; increasing nickel content
results in lower thermal stability, leading to safety concerns.\1198\
---------------------------------------------------------------------------
\1195\ See, e.g., MIT Energy Initiative. 2019. Insights into
Future Mobility, at 78. Cambridge, MA: MIT Energy Initiative (``. .
. significant uncertainty remains about the steady-state price of
cobalt in the future as demand and supply continues to increase
[internal citation omitted]. Under our base case scenario, global
demand for cobalt in 2030 from new EV sales (even if all EVs use
batteries with the high nickel content of NMC811) would reach
approximately 80% of the world's total cobalt output in 2016.
Considering that only 15% of the worldwide demand for cobalt in 2017
was used in EV batteries (Jackson 2019), an increase in demand of
this magnitude might result in higher prices for cobalt. Thus,
automakers may need to move to different battery chemistries that
are less reliant on cobalt to avoid raw materials shortages and
price volatility.'').
\1196\ See, e.g., Todd C. Frankel, The Cobalt Pipeline: Tracing
the path from deadly hand-dug mines in Congo to consumers' phones
and laptops, Washington Post (Sept. 30, 2016), https://www.washingtonpost.com/graphics/business/batteries/congo-cobalt-mining-for-lithium-ion-battery/?itid=lk_inline_manual_9&tid=lk_inline_manual_9; Peter Whoriskey and
Todd C. Frankel, Tech giants pledge to keep children out of cobalt
mines that supply smartphone and electric-car batteries, Washington
Post (Dec. 20, 2016), https://www.washingtonpost.com/news/the-switch/wp/2016/12/20/tech-giants-pledge-to-keep-children-out-of-cobalt-mines-that-supply-smartphone-and-electric-car-batteries/.
\1197\ See, e.g., Gohlke, David, and Zhou, Yan. Assessment of
Light-Duty Plug-In Electric Vehicles in the United States, 2010-
2018. United States: N. p., 2019. Web. doi:10.2172/1506474 (citing
Berman, Kimberly, Jared Dziuba, Colin Hamilton, Richard Carlson,
Joel Jackson, and Peter Sklar, 2018. ``The Lithium Ion Battery and
the EV Market: The Science Behind What You Can't See.'' BMO Capital
Markets, February 2018. https://bmo.bluematrix.com/docs/pdf/079c275e-3540-4826-b143-84741aa3ebf9.pdf); MIT Energy Initiative.
2019. Insights into Future Mobility, at 77. Cambridge, MA: MIT
Energy Initiative. http://energy.mit.edu/insightsintofuturemobility.
\1198\ Schipper, Florian, Evan M. Erickson, Christoph Erk, Ji-
Yong Shin, Frederick Francois Chesneau, and Doron Aurbach. 2017.
``Review--Recent Advances and Remaining Challenges for Lithium Ion
Battery Cathodes I. Nickel-Rich, LiNixCoyMnzO2.'' Journal of the
Electrochemical Society 164, no. 1 (1): A6220-A6228. https://doi.org/10.1149/2.0351701jes.
---------------------------------------------------------------------------
For the final rule analysis, based on these considerations, the
agencies in consult with Argonne determined that it was reasonable to
model HEV, PHEV, and BEV batteries using NMC622 as the cathode active
material, as shown in Table VI-91 below.
[GRAPHIC] [TIFF OMITTED] TR30AP20.220
The agencies recognize that there will be advancements in battery
chemistries during the rulemaking timeframe. As discussed further in
Section (3), below, the analysis accounts for the potential that
battery costs will decrease, but in a technology-agnostic manner. The
agencies used BatPaC to model battery costs for the analysis by
modeling battery prices in a specific year--in this case, MY 2020--and
then used learning curves to reduce the cost of batteries over time.
The learning curves act as a proxy for potential future improvements in
battery chemistry and other battery-related advancements that would
reduce costs. Using the learning curves in this way makes it
unnecessary to make inherently uncertain projections of potential
future improvements in battery chemistry over time.
BatPaC version 4.0, which contains NMC811 as a chemistry option,
was released after the analysis for this rule was completed. However,
the cost estimates generated in BatPaC version 3.1 using NMC622, with
discussed learning curves applied resulted in estimated $/kWh battery
pack costs, during the rule making time frame within a reasonable range
of other estimated projections that considered NMC811 as the
predominant battery chemistry. As discussed further in Section (3), a
significant shift in battery chemistry alone is only one factor
required to significantly lower battery costs; other developments like
increases in battery pack production quantities and cell yield (plant
efficiencies) would be required to reach the commonly-cited $100/kWh
target.
The agencies recognize that the specific chemistries manufacturers
may choose for future model years may or may not be the same as the
chemistries selected by the agencies for the analysis. However, this
approach mirrors the approach taken to modeling technology
effectiveness and cost used across the analysis; the modeled technology
effectiveness and cost represents a level of performance representative
of the typical range of performance across industry. If the agencies
modeled pre-production battery chemistries unlikely to be widely
adopted by the industry for several years, the analysis would likely
under-predict the actual cost and effectiveness of electrification
technology application. Accordingly, the agencies determined that using
LFP-G as the modeled chemistry of choice for mild hybrids and NMC622 as
the modeled chemistry of choice for strong HEVs, PHEVs, and BEVs was
reasonable.
The agencies also refined other inputs and assumptions used for
modeling battery costs in BatPaC, based on a review of public comments
and subsequent review of market research, technical publications, and
other information.
Argonne continuously studies the battery pack designs of existing
electrified vehicles in the market, using, among other information,
detailed battery pack teardown analysis reports spanning a range of
electrified vehicle types and vehicle classes produced over a range of
MYs. For the final rule, Argonne utilized detailed battery pack
teardown analysis reports for 10 MY
[[Page 24500]]
2013 to MY 2018 vehicles from A2mac1,\1199\ as shown in the Table VI-92
below.
---------------------------------------------------------------------------
\1199\ Argonne Vehicle Modeling for Safer Affordable Fuel
Efficient (SAFE) Vehicles Final Rulemaking, Section 5.9 Battery
Performance and Cost Model (BatPaC), referencing A2Mac1 Automotive
Benchmarking, https://a2mac1.com.
[GRAPHIC] [TIFF OMITTED] TR30AP20.221
The teardown analysis reports were used to evaluate different
battery pack design criteria, including battery pack power, battery
pack energy, battery pack configuration, total number of cells per
module, number of modules per pack, battery pack mass, energy density
(cell/pack), cell voltage, battery pack voltage, cathode chemistry,
cell capacity, and pack capacity. The metrics data collected from
teardown analysis were used to estimate the battery pack manufacturing
cost and mass (energy density-Wh/kg) in BatPaC for these exemplar
vehicles from the A2Mac1 database. The data collected was also used to
validate the battery pack design assumptions in BatPaC for the final
rule. The four metrics that BatPaC provides are: Battery pack
manufacturing cost, battery pack weight (energy density-Wh/kg), battery
pack capacity (Ah) and nominal battery pack voltage. Since the A2mac1
teardown reports do not avail the manufacturing costs of these battery
packs, the analyses and comparisons were limited to the scope of the
other three criteria.
For the NPRM, Argonne used the U.S. Department of Energy VTO
targets for battery energy density (Wh/kg) for high energy and power
density-(W/kg) for high powered batteries.\1200\ As a result of the
analysis discussed above Argonne updated the method of estimating
battery pack weight for each battery pack design in the final rule
analysis. The analysis revealed greater influences on battery pack
design by usable energy density characteristics then was initially
assumed for the NPRM. For the final rule analysis BatPaC was used for
battery pack weight estimates along with manufacturing cost estimates.
---------------------------------------------------------------------------
\1200\ Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, ANL/CSE-19/2.
---------------------------------------------------------------------------
As discussed further in Section VI.C.3.e)(1)(c) Battery Pack Costs,
the number of cells per pack influenced total battery pack costs for
the final rule. As result of the analysis discussed above Argonne
updated the number of cells in each battery. For the final rule
analysis battery cell counts increased or decreased for some battery
pack designs, while battery counts for some designs remained the same.
Argonne's process for evaluating different design criteria for
electrified vehicles is detailed further in the Argonne model
documentation.\1201\
---------------------------------------------------------------------------
\1201\ A Detailed Vehicle Simulation Process To Support CAFE and
CO2 Standards for the MY 2021-2026 Final Rule Analysis, Section 5.9
Battery Performance and Cost Model (BatPaC).
---------------------------------------------------------------------------
The agencies also updated other BatPaC inputs and assumptions based
on additional market information or research. For the NPRM, the
agencies modeled battery packs in BatPaC using the default values
associated with the baseline manufacturing plant, including an annual
production rate of 100,000 batteries.\1202\
---------------------------------------------------------------------------
\1202\ See Nelson, Paul A., Gallagher, Kevin G., Bloom, Ira D.,
and Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles--SECOND EDITION (2012), at 62.
Available at https://publications.anl.gov/anlpubs/2015/05/75574.pdf.
---------------------------------------------------------------------------
The estimate for battery pack costs incorporates an assumption of
the battery pack production volume. Both BatPaC version 3.0, used in
the NPRM, and BatPaC version 3.1, used in the final rule, include a
default value assumption of 100,000 battery pack units manufactured per
year per manufacturing plant as well as the plant efficiency (cell
yield) of 95 percent. For the final rule, the agencies adjusted the
production volume assumption used in BatPaC version 3.1 to 25,000
battery pack units, based on the analysis presented below.
As described in the BatPaC model documentation, the BatPaC models
the differences in pack designs and how they affect the costs of one or
more steps in the battery production process and the physical plant
layout.\1203\ For example, increasing the power of the battery packs
without increasing the number of cells, or cell capacity, results in
the model increasing the area of the cells and decreasing the electrode
coating thickness. This results in an increased cost of the coating
equipment, the floor area occupied by the equipment, and the direct
labor for the process.1204 1205 The agencies are aware that
each manufacturer (not brand) has a unique battery pack design that
differs from other manufacturers. Accordingly, it is likely that each
manufacturer's BEV models had distinct characteristics, such as unique
battery packaging space, energy requirements, thermal control systems,
and safety systems, which cause battery pack designs to vary between
each manufacturer.
---------------------------------------------------------------------------
\1203\ Nelson, Paul A., Ahmed, Shabbir, Gallagher, Kevin G., and
Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, Third Edition (2019), at 100.
Available at https://publications.anl.gov/anlpubs/2019/03/150624.pdf.
\1204\ Kupper et al, The Future of Battery Production for
Electric Vehicles, Boston Consulting Group, (Sept. 11, 2018),
https://www.bcg.com/publications/2018/future-battery-production-electric-vehicles.aspx.
\1205\ Id.
---------------------------------------------------------------------------
Thus, the agencies determined that even though one battery
manufacturer
[[Page 24501]]
might manufacture batteries for multiple vehicle manufacturers, the
default BatPaC assumption of 100,000 battery pack units manufactured
per plant likely did not account for all of the cost differences in
pack designs between manufacturers. Therefore, the agencies assumed the
production volume of each battery pack type was reasonably represented
by the BEV production volume for each manufacturer. The agencies also
assumed that battery pack manufacturing plants operated at reasonable
capacity during that timeframe, which would produce the lowest cost
assumption.
The agencies analyzed BEV sales for MYs 2016-2019, referencing data
collected by the Department of Energy.\1206\ Table VI-93 shows that
individual manufacturer U.S. BEV sales are substantially below 100,000
units per year except for Tesla, beginning in MY 2018 Tesla is a
vertically integrated battery and BEV manufacturer, which is not the
model the remainder of the industry has implemented, or intends to,
based on the agencies current understanding. More specifically, Tesla
sold more BEVs than all manufacturers combined in MYs 2016, 2018, and
2019. 2017 was the only year in which all other manufacturers combined
sold more BEVs than Tesla. Ultimately, in selecting a battery pack
volume estimates for an industry-wide assessment, the agencies sought
to accurately account for both the representative production volumes
and representative practices applicable to the industry. As such, the
agencies evaluated the average per manufacturer volumes, less the
outlying and vertically integrated volumes of Tesla (shown in Table VI-
94). As depicted in Table VI-93 and Table VI-94, the data show that the
average annual sales of BEVs for individual manufacturers, excluding
Tesla, is just 5% of the default battery pack production volume in
BatPaC.
---------------------------------------------------------------------------
\1206\ Light Duty Electric Drive Vehicles Monthly Sales Updates,
Argonne National Laboratory Energy Systems Division, https://www.anl.gov/es/light-duty-electric-drive-vehicles-monthly-sales-updates (last visited March 2, 2020); Maps and Data, Alternative
Fuels Data Center, https://afdc.energy.gov/data/ (last visited March
2, 2020).
[GRAPHIC] [TIFF OMITTED] TR30AP20.222
[[Page 24502]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.223
In consideration of this data, when estimating the production
volume in the final rule analysis, the agencies selected a value of
25,000 units per year per manufacturer as a reasonable estimate for the
average industry for MY 2020, which is the base model year for
estimated battery pack costs using BatPaC version 3.1. As discussed in
Section VI.C.3.e)(3) Electrification Learning Curves, other model year
battery pack costs are estimated using cost learning. Using the default
production volume of 100,000 units per year per manufacturer, the
agencies would have underestimated the actual cost of battery pack
production for MY 2020, as the model assumes that production costs
decrease as production volumes increase. By selecting the value of
25,000 units per year per manufacturing plant, the battery cost
estimate from the BatPaC model better aligned with the cost estimate
published in industry-recognized reports such as the UBS MY 2016 Chevy
teardown report.1208 1209 1210
---------------------------------------------------------------------------
\1207\ Note, for the assessment, Nissan and Mitsubishi are
considered a single manufacturer.
\1208\ Proposed Determination TSD at 2-127.
\1209\ Based on the battery cell to battery pack ratio of 1.3 to
1.5, the 2015-2019 cell-level figure of $145 per kWh used in the MY
2016 Chevy Bolt would translate to approximately $190 to $220 per
kWh on a pack level.
\1210\ Hummel et al., UBS Evidence Lab Electric Car Teardown--
Disruption Ahead?, UBS (May 18, 2017), https://neo.ubs.com/shared/d1ZTxnvF2k/.
---------------------------------------------------------------------------
The agencies performed a sensitivity study for production volume
using BatPaC version 3.1. The cost of the battery pack dropped by 15
percent on average when the production volume was changed from 25,000
to 100,000 units per year. The sensitivity analysis showed that
manufacturing plant volume has a significant impact on battery pack
costs and therefore it is important to use realistic production volume
estimates for the battery pack cost analysis.
Manufacturing plant efficiency is another parameter important to
estimate battery pack costs. BatPaC version 3.1 defines manufacturing
plant efficiency in terms of cell yield, or the number of cells that
are usable out of the total number of cells that the plant
produced.\1211\ Since battery pack technology and battery pack
manufacturing processes are proprietary, the data on plant efficiencies
are not widely reported. While BatPaC uses a default cell yield (plant
efficiency) value of 95 percent, Argonne battery experts have used an
85 percent cell yield value to represent the current production yield
for internal DOE studies.\1212\ By selecting an 85 percent cell yield
value for the final rule analysis, the agencies aligned the cell yield
value assumption with internal DOE studies.
---------------------------------------------------------------------------
\1211\ Cells might not be usable because of, for example,
manufacturing defects, among other reasons.
\1212\ Argonne National Laboratory, BatPaC Model Software,
https://www.anl.gov/cse/batpac-model-software (last visited March
19, 2020). Argonne used an 85% cell yield assumption in its
Estimated Cost of EV Batteries 2018-19 analysis.
---------------------------------------------------------------------------
In addition, as discussed in detail above, the final rule analysis
was performed using BatPaC version 3.1, with NMC622 assumed as the
battery chemistry for HEVs, PHEVs, and BEVs. Separate from the inputs
and assumptions discussed here, the Argonne battery experts made a
number of changes to BatPaC version 3.1, and these are extensively
documented in the BatPaC manual,\1213\ as well as in Argonne model
documentation for final rule.
---------------------------------------------------------------------------
\1213\ Nelson, Paul A., Ahmed, Shabbir, Gallagher, Kevin G., and
Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, Third Edition (2019),
available at https://publications.anl.gov/anlpubs/2019/03/150624.pdf.
---------------------------------------------------------------------------
(b) Comments on Information Availability
In addition to comments that the agencies' battery pack costs were
too high, the agencies received comments that the analysis for battery
pack costs was unclear and not well documented. ICCT stated that the
agencies largely obscured the BEV cost sources and calculations, which
made it ``nearly impossible for even very interested researchers to
understand how all the BatPaC costs translate into BEV costs that can
be compared with other full-BEV costs in the literature.'' \1214\ ICCT
stated that to enable meaningful public comments, the sources and cost
calculations must be made explicit and the agencies must provide an
additional public comment opportunity.\1215\
---------------------------------------------------------------------------
\1214\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1215\ Id.
---------------------------------------------------------------------------
CARB claimed that it could not comment meaningfully on the battery
modeling for the NPRM analysis without extensive additional
information.\1216\ As such, CARB submitted a letter to the agencies'
NPRM docket posing, under FOIA, a number of questions pertaining to
battery assumptions used for the modeling. This requested information
concerned what version of BatPaC was used in the NPRM analysis, inputs
incorporated into the BatPaC model; and information about how battery
costs were generated for the analysis.
---------------------------------------------------------------------------
\1216\ California Air Resources Board, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
Specifically, CARB's initial comments alleged that the agencies had
not disclosed the exact version of BatPaC used, and had simply claimed
to use the ``most up-to-date'' version of BatPaC,
[[Page 24503]]
and further that the agencies had not disclosed ``the BatPaC modeling
files that were used, clear statements about what version of the model
was used, or thorough descriptions of the inputs to those modeling
runs.'' CARB claimed that without that information, ``there is no way
to know what assumptions were made for raw material pricing, battery
cell yields, pack electrical connection topology, battery production
volume assumptions, or if any additional parameters were modeled, like
rapid charging capability.'' CARB argued that these pieces were
critical to understanding whether the BatPaC model was estimating
proper battery pack cost values.
In a subsequent docketed comment submitted as an administrative
appeal to NHTSA's FOIA response, CARB reasserted that, in fact, the
``most recent version'' of BatPaC had not been used, because the FOIA
response stated clearly that version 3.0 had been used and Argonne had
updated to version 3.1 in October 2017, which was the last version
released before the NPRM was published. CARB further argued that NHTSA
was ``choosing to withhold information about battery pack
configurations,'' and that the agencies had not posted the BatPaC model
version and files used for the NPRM to the agencies' dockets,
inhibiting meaningful comment.
The majority of information sought by CARB's comment was already
published in supporting documents and materials posted to the agencies'
dockets and online websites for the NPRM. Nevertheless, in an effort to
answer CARB's specific questions, NHTSA also processed the initial
comment as a FOIA request and provided a written response directly to
CARB within the comment period. This response both pointed CARB to the
locations where the sought material could be located among the
published NPRM materials, and expressly answered several of CARB's
questions for clarification, such as identifying the specific version
of BatPaC utilized in the NPRM analysis. For example, although the
Argonne model documentation describing the battery modeling for the
NPRM was included in the docket, the agencies' response directed CARB
to the precise location in the docket where it could be found.
The agencies believe that the NPRM docket contained enough
information for stakeholders to comment meaningfully. This is apparent
from the voluminous comments the agencies received regarding the NPRM's
electrification analysis--including from CARB. For example, as
discussed above, CARB submitted extensive comments on each element of
the battery cost modeling that CARB claimed the agencies did not
adequately explain. As discussed above, CARB stated that the agencies'
selected battery chemistries represented a step backward from previous
analysis done for the Draft TAR. CARB noted that regardless of whether
NMC441 or NMC333 was chosen for PHEVs and BEVs in the NPRM analysis,
the biggest lithium-ion production companies have indicated that they
will use NMC811 for BEVs, and therefore neither NMC441 nor NMC333 would
represent current technology going into BEVs or near-future BEV battery
technology. CARB stated that NMC811 technology is expected to come to
market in 2019, which, the agencies note, is far sooner than
anticipated, even in the agencies' prior analyses. CARB was accordingly
able to communicate its opinion that NMC881 should have been used to
model battery chemistries for the NPRM analysis, and that NMC441 or
NMC333 should not be used.
As these comments demonstrate, in addition to the extensive
comments listed above, the expansive information, data, and
documentation concerning the Argonne BatPaC modeling analysis for the
NPRM sufficiently enabled commenters to submit voluminous technical
analysis regarding the electrification analysis. Moreover, while the
docketed and published NPRM materials themselves afforded sufficient
notice on these topics, the agencies even undertook the additional step
of directly responding to CARB in writing in an attempt to address
specific questions raised by CARB. This written correspondence both
directed CARB to specific locations on the rulemaking dockets and
agencies' websites where information CARB was seeking could be
accessed, and even directly answered several of CARB's questions
through narrative responses. Both CARB and other commenters submitted
subsequent comments, which referenced the material described in this
written response. Accordingly, the agencies consider the information
provided with the NPRM sufficient to enable meaningful comment, which
is underscored by the voluminous technical comments received on the
electrification issues.
For this final rule, the BatPaC model version 3.1 (June 2018) model
documentation has been included in the docket for this
rulemaking.\1217\ Furthermore, Argonne's detailed documentation
describing the modeling process used to support this final rule
provides information and specific assumptions that Argonne's experts
used to simulate batteries and their associated costs for the full
vehicle simulation modeling.\1218\ These resources, in addition to the
detailed description of the battery cost modeling process provided here
and in the FRIA provide interested stakeholders the necessary tools to
understand the battery cost modeling analysis.
---------------------------------------------------------------------------
\1217\ Nelson, Paul A., Ahmed, Shabbir, Gallagher, Kevin G., and
Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, Third Edition (ANL/CSE-19/2),
available at https://publications.anl.gov/anlpubs/2019/03/150624.pdf.
\1218\ A Detailed Vehicle Simulation Process To Support CAFE and
CO2 Standards for the MY 2021-2026 Final Rule Analysis.
---------------------------------------------------------------------------
c) Final Rule Battery Pack Costs
As discussed above, based on comments and additional research, the
agencies updated the battery cost analysis for the final rule by
relying on BatPaC version 3.1.\1219\ In addition, as outlined above and
explained in more detail in the Argonne Model Documentation for this
final rule, several inputs and assumptions were updated based on public
comments, market research, and additional literature review. The
agencies computed the average battery pack cost across all road load
combinations for electrification technologies that could be reasonably
compared between the NPRM and final rule.\1220\
---------------------------------------------------------------------------
\1219\ Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, Third Edition (ANL/CSE-19/2)
provides a complete list of changes and assumptions incorporated in
BatPaC version 3.1.
\1220\ Costs data is from the CAFE Model core file
Battery_Costs.csv.
---------------------------------------------------------------------------
Table VI-95 to Table VI-99 show the differences between battery
pack costs presented in the NPRM and final rule.\1221\ The tables show
absolute cost differences between battery packs, which can vary for
battery packs with different energy and power combinations. For
example, as shown in Table VI-96, the cost difference between the NPRM
and final rule for a Mild HEV battery pack with a 1kWh energy and 10kW
power rating is -28 percent. Similarly, the cost difference in an HEV
battery pack with a 1kWh battery energy and 40kW power rating is 5
percent. In summary, the percentage increase or decrease in the table
represents the
[[Page 24504]]
absolute cost differences between the battery packs used in NPRM and in
final rule.
---------------------------------------------------------------------------
\1221\ The absolute cost differences shown here is by comparing
the cost of battery pack with similar number of cells in the NPRM to
the final rule cost lookup tables for compact and medium car. The
cost differences between the NPRM and the final rule cost lookup
tables for small SUV, medium SUV and Pickup trucks will be different
from the table shown here.
---------------------------------------------------------------------------
Figure VI-40 to Figure VI-42 shows the average battery pack costs
across all road load combinations for each applicable vehicle
technology class for SHEVPS, PHEV50, and BEV200s between the NPRM and
final rule.\1222\ Since the battery pack size varies for different road
load combinations, the battery pack cost across different road load
combinations varies as well. For example, there are 105 combinations of
different mass reduction, aerodynamic improvements and rolling
resistance improvements. The battery pack size for an initial road load
condition that includes MR0, AERO0 and ROLL0 is larger, and therefore,
the cost of the battery pack is higher as well. The battery pack size
is smaller for the highest level of road load reduction such as in MR6,
AERO20 and ROLL20, and the cost of battery pack is less as well.
---------------------------------------------------------------------------
\1222\ The agencies did not simulate SHEVPS and BEV200
powertrain architectures on pickup trucks in the NPRM, so those are
not included in the comparison.
---------------------------------------------------------------------------
Table VI-95 shows the cost difference in Micro HEV battery packs.
The cost reduction is from the reduced number of cells in the battery
pack.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.224
Table VI-96 shows percentage cost differences for mild hybrid
(BISG) battery packs. The cost difference is due, in part, to
accounting for BISG-related hardware costs, such as the battery
management system, as part of the electric machine costs in this final
rule.\1223\
---------------------------------------------------------------------------
\1223\ In the NPRM, additional hardware component costs were
included as part of the battery pack cost.
[GRAPHIC] [TIFF OMITTED] TR30AP20.225
Table VI-97 shows the percentage cost differences for HEV battery
packs. Even as the battery chemistry changed to NMC622, the cost
increase is from the different battery pack production volume and plant
efficiency assumptions used in the final rule.
[[Page 24505]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.226
Figure VI-40 shows the difference in battery pack costs for SHEVPS
applications between the NPRM and final rule. Power-split hybrids could
not be used in pickup trucks due to their unique power and towing
requirements, so those technology classes are not shown. In general,
the cost of the battery pack in the final rule analysis increased due
to the updated battery pack production volume and plant efficiency
assumptions.
[GRAPHIC] [TIFF OMITTED] TR30AP20.227
Table VI-98 shows the percentage cost differences between the NPRM
and final rule for PHEV50 battery packs. The cost increase in the
PHEV50 battery pack shown here is mainly due to the increase in number
of cells per pack as well as the other updated BatPaC assumptions.
[[Page 24506]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.228
Table VI-94 shows the difference in average PHEV50 battery pack
costs between the NPRM and final rule for all technology combinations.
[GRAPHIC] [TIFF OMITTED] TR30AP20.229
Table VI-99 shows the percentage cost differences for BEV battery
packs. In the example shown in Table VI-99, the agencies compared the
cost lookup table from the NPRM with 300 cells to the cost lookup table
in the final rule analysis with 320 cells. The cost increase in the
higher energy packs is due to the different battery pack production
volume and plant efficiency value assumptions, along with the different
battery chemistry assumption.
[[Page 24507]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.230
BILLING CODE 4910-59-C
Figure VI-42 shows the average cost of BEV200 battery packs across
all technology combinations for technology classes that could be
compared between the NPRM and final rule. As shown, for the final rule
analysis, the average cost of a BEV200 battery pack is lower than the
average cost of the NPRM BEV200 battery pack. For the final rule
analysis, the agencies updated the motor efficiency map for BEVs (as
explained in Section VI.C.3.d) Electrification Technology
Effectiveness) and updated the glider share of the vehicles from 50
percent of the curb weight to 71 percent of the vehicle curb weight (as
explained in Section VI.C.4 Mass Reduction). In addition, the updated
motor weight resulted in further reduced vehicle weights. This
combination of improved vehicle assumptions resulted in reduced energy
and power requirements in BEVs.
The agencies also observed that even as the number of cells in the
battery pack increased from 300 to 320, and changes in production
volume and plant efficiency values resulted in marginal cost increases
for higher energy packs, the overall battery capacity requirement went
down due to overall reduction in power and energy demand from electric
vehicles.\1224\ A reduction in battery capacity leads to reduced cell
size in a pack with number of cells and voltage. A reduction in cell
size leads to cost reductions at the cell level and at the pack level.
In general, a higher capacity battery pack is more expensive than a
lower capacity battery pack due to the increase in cell size for a
given number of cells and voltage.1225 1226
---------------------------------------------------------------------------
\1224\ As explained above, the energy density values in the NPRM
were kept constant. For the final rule analysis, the power density
varied to meet different power and energy requirements, as was
observed through market research.
\1225\ Nelson, Paul A., Ahmed, Shabbir, Gallagher, Kevin G., and
Dees, Dennis W. Modeling the Performance and Cost of Lithium-Ion
Batteries for Electric-Drive Vehicles, Third Edition (ANL/CSE-19/2),
at 15 (battery design worksheet). Available at https://publications.anl.gov/anlpubs/2019/03/150624.pdf.
\1226\ The amount of electrode materials and electrode area of
the cells are determining cost factors in the battery. Higher
capacity battery packs require additional manufacturing steps to
increase the energy density of the pack.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24508]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.231
The graphs demonstrate the range of cost changes observed, with the
other electrification technologies falling somewhere in between the
extremes. In summary, the agencies observed that the BEV200 technology
showed a cost reduction in battery packs across all vehicle platforms
with the largest reductions occurring for the largest battery packs. In
contrast the PHEV50 technology showed a cost increase in battery packs
across all vehicle platforms with the smallest increase for the largest
battery packs and the largest increase for the smallest battery packs.
It is worth noting the cost decreases seen across the technologies are
generally larger than the cost increases.
For the final rule, when possible, the calculated battery pack
weight and manufacturing cost was also compared with the battery pack
cost and weight data obtained through various benchmarking studies. For
example, UBS reported a battery pack manufacturing cost of $12,500 from
its 2017 Chevrolet Bolt teardown analysis.\1227\ Using a production
volume of 25,000 packs per year per plant and similar battery pack
design, BatPaC estimated a manufacturing cost of $10,680.\1228\ These
comparisons were used to verify the different assumptions used in
BatPaC and helps represent the battery packs for electrified vehicles
used in representative market volume. Table VI-100 shows a comparison
of specifications estimates for 60 kWh and 160 kW battery packs from
the 2016 DOE VTO report 1229 1230 and BatPaC version 3.1
(June 2018), and the Chevrolet Bolt. The comparison shows modeled and
actual battery packs are in close agreement.
---------------------------------------------------------------------------
\1227\ Hummel et al., UBS Evidence Lab Electric Car Teardown--
Disruption Ahead?, UBS (May 18, 2017), https://neo.ubs.com/shared/d1ZTxnvF2k/.
\1228\ $178/kWh x 60kWh = $10,680.
\1229\ Peter Faguy, Overview of the DOE Advanced Battery R&D
Program (June 2015), https://www.energy.gov/sites/prod/files/2015/06/f23/es000_faguy_2015_o.pdf.
\1230\ Freyermuth, Vincent. Rousseau, Aymeric. ``Impact of
Vehicle Technologies Office Targets on Battery Requirements.'' ANL/
ESD-16/22. Energy Systems Division, Argonne National Laboratory
(2016).
---------------------------------------------------------------------------
[[Page 24509]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.232
In addition, the agencies compared the battery pack cost estimates
generated using BatPaC to other current studies or studies cited by
commenters. Table VI-101 summarizes battery pack estimates from
selected studies in MYs for which that information was available.
---------------------------------------------------------------------------
\1231\ Not each study distinguished a DMC source year, so these
values vary slightly based on inflation.
\1232\ Sources generally provided estimates for 2018 or 2020.
\1233\ Hummel et al., UBS Evidence Lab Electric Car Teardown--
Disruption Ahead?, UBS (May 18, 2017), https://neo.ubs.com/shared/d1ZTxnvF2k/.
\1234\ Mosquet et al., The Electric Car Tipping Point, BCG (Jan.
11, 2018), https://www.bcg.com/publications/2018/electric-car-tipping-point.aspx. This study provided cell cost estimates that the
agencies converted to pack cost estimates using a multiplier of 1.3,
as outlined in the Draft TAR at 5-124.
\1235\ Nic Lutsey and Michael Nicholas, Update on electric
vehicle costs in the United States through 2030, ICCT (April 2,
2019), available at https://theicct.org/publications/update-US-2030-electric-vehicle-cost. The presented values are $/kWh pack costs for
mid-range electric cars/crossovers and SUVs.
\1236\ McKerracher et al., Electric Vehicle Outlook 2019--Free
Interactive Report, Bloomberg New Energy Finance (May 2019), https://about.bnef.com/electric-vehicle-outlook/.
\1237\ Logan Goldie-Scot, A Behind the Scenes Take on Lithium-
ion Battery Prices, Bloomberg New Energy Finance (March 5, 2019),
https://about.bnef.com/blog/behind-scenes-take-lithium-ion-battery-prices/. BNEF projected the pack costs in 2018$ for 2018 as $176,
and used the same value in the Electric Vehicle Outlook 2019 to
describe pack cost levels ``today.''
\1238\ MIT Energy Initiative. 2019. Insights into Future
Mobility. Cambridge, MA: MIT Energy Initiative. Available at http://energy.mit.edu/insightsintofuturemobility.
\1239\ Islam, E., Kim, N., Moawad, A., Rousseau, A., ``A Large-
Scale Vehicle Simulation Study To Quantify Benefits & Analysis of
U.S. Department of Energy VTO & FCTO R&D Goals.'' Report to U.S.
Department of Energy. Contract ANL/ESD-19/10 (forthcoming).
---------------------------------------------------------------------------
[[Page 24510]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.233
As shown in the table above, there are a range of cost estimates
for battery packs. Each individual cost estimate is derived based on
certain set of assumptions to arrive at a rate of cost reduction. Among
all the different cost estimates, Bloomberg New Energy Finance (BNEF)
has the most aggressive year-over-year cost reductions, based on the
historical learning rate of 18% and their battery demand
forecast.\1240\ Similar to other sources of cost estimates BNEF assumes
improved battery chemistry and battery density increasing greater than
200Wh/kg by 2030. In order for the battery manufacturer to achieve
economies of scale, BNEF assumes a global battery manufacturing
facility capable of producing battery packs for both stationary energy
storage and vehicle applications.
---------------------------------------------------------------------------
\1240\ Logan Goldie-Scot, A Behind the Scenes Take on Lithium-
ion Battery Prices, Bloomberg New Energy Finance (March 5, 2019),
https://about.bnef.com/blog/behind-scenes-take-lithium-ion-battery-prices/.
---------------------------------------------------------------------------
A recent report from the Massachusetts Institute of Technology
(MIT), the MIT Energy Initiative's Insights into Future Mobility, has
the most conservative estimate among all the cost sources listed the
Table VI-101. The authors use a more rigorous two-stage method of
estimating composite battery learning curves independently for (a)
battery material synthesis and minerals costs, and (b) battery pack
production processes. The learning rates are defined as the cost
reduction that results from cumulative volume doubling, and produce
separate cost learning rates for the two stages of 3.5 percent and 16.5
percent, respectively. The study argues that there are greater
opportunities for cost learning in the production stage than the
chemical synthesis stage, which is more mature. These cost estimates
produce global EV fleet penetration rates that may not be as aggressive
as other estimates, reaching only 33 percent by 2050. This study also
assumes NMC811 will be available by 2030.
The cost estimates from other sources referenced above also include
assumptions about higher levels of battery pack production and higher
density battery cells. Most cost estimates assume improved battery
chemistry, such as NMC811. As discussed above, the agencies determined
that modeling assuming NMC622 was reasonable, based on current
production vehicles, the relative uncertainty surrounding large-scale
NMC811 deployment in the rulemaking timeframe, and the ability to
account for lower battery pack costs over time with cost learning. The
agencies also believe that, based on the market analysis and from the
teardown analysis, improvements in battery chemistry may be slow to be
applied in a widespread manner, and therefore the economies of scale
required to achieve considerable cost reductions solely from
improvements in chemistry may remain effusive during the rulemaking
timeframe.
For these reasons, the agencies believe that the BatPaC-generated
battery cost estimates using the updated inputs and assumptions are
reasonable.
2) Non-Battery Electrification Component Costs
Battery components are the biggest driver of the cost of
electrification, however, non-battery electrification components also
add to the total cost required to electrify a vehicle. In this
analysis, the agencies accounted for the following non-battery
component costs: Electric motor(s), inverter, and other power
electronics including a bi-directional DC/DC converter, a voltage step
down DC/DC converter, and an on-board charger. Collectively, these
components (except for the on-board charger) are referred to as the
electric traction drive systems (ETDS), or the electric machine. Non-
plug-in hybrid electric vehicles include all of the listed components
except for an on-board charger; PHEVs include all of the listed
components; and BEVs include all of the listed components except, in
some cases, a second motor.
For the NPRM, the agencies accounted for battery pack costs and
ETDS costs independently.\1241\ The Alliance commented broadly in
support of separating electrification hardware costs and battery costs,
and stated that it was a positive change to the modeling.\1242\ The
Alliance correctly noted that the separation allowed for separate
learning rates and cost differentiation between the two distinct pieces
of electrification technologies.
---------------------------------------------------------------------------
\1241\ PRIA at 362.
\1242\ Alliance of Automobile Manufacturers, NHTSA-2018-0067-
12073, at 140.
---------------------------------------------------------------------------
[[Page 24511]]
As stated in the PRIA,\1243\ the agencies derived the cost values
for the EDTS using Argonne National Laboratory's ``Assessment of
Vehicle Sizing, Energy Consumption, and Cost through Large-Scale
Simulation of Advanced Vehicle Technologies'' report.\1244\ Generally,
the agencies referred to this report in the PRIA as the DOE VTO report,
as it was a report that reviewed results of the DOE VTO. Some
commenters seemed confused by this alternative reference--even
questioning why the agencies didn't rely on recent Argonne National
Laboratory reports.\1245\ To clarify, this report was written by
Argonne National Laboratory, and to avoid further confusion it is
referred to using the full title throughout this rule.
---------------------------------------------------------------------------
\1243\ 83 FR 43047; PRIA at 362.
\1244\ Moawad, Ayman, Kim, Namdoo, Shidore, Neeraj, and
Rousseau, Aymeric. Assessment of Vehicle Sizing, Energy Consumption
and Cost Through Large Scale Simulation of Advanced Vehicle
Technologies (ANL/ESD-15/28). United States (2016), available at
https://www.autonomie.net/pdfs/Report%20ANL%20ESD-1528%20-%20Assessment%20of%20Vehicle%20Sizing,%20Energy%20Consumption%20and%20Cost%20through%20Large%20Scale%20Simulation%20of%20Advanced%20Vehicle%20Technologies%20-%201603.pdf.
\1245\ California Air Resources Board, NHTSA-2018-0067-11973, at
130-31.
---------------------------------------------------------------------------
CARB expressed concerns with non-battery component effectiveness
values, arguing that the agencies inappropriately relied on outdated
data for electric machines and inverter efficiencies across all
electrification applications, and further claiming that the agencies
did not project any efficiency gains in those components over
time.\1246\ Broadly, as these comments on effectiveness related to the
NPRM non-battery component cost estimates, CARB claimed that the
agencies failed to consider new data, including the 2015 ORNL Annual
Progress Report for the Power Electronics and Electric Motors Program,
and two Argonne studies, which rendered the analysis unrepresentative
of actual technology costs.
---------------------------------------------------------------------------
\1246\ California Air Resources Board, NHTSA-2018-0067-11973, at
130.
---------------------------------------------------------------------------
CARB also commented that the agencies did not provide any
substantive discussion or documentation of how non-battery component
costs were developed for the NPRM analysis. CARB claimed that
dissonance existed between the PRIA description of voltage systems and
associated costs needed for different performance classes, the
Autonomie files, and the technologies input file, and that this served
as an example of how the agencies failed to include information
regarding how costs and cost differences were derived, or any component
changes from previous analyses.
CARB also commented that the lack of disclosure of non-battery cost
development information was an issue for other electrification
technologies. CARB cited the increase in parallel (P2) and power-split
(PS) hybrid systems costs relative to costs used in past agency
analyses, noting that there was no discussion on what changed from the
past analyses. CARB referenced a 2010 FEV teardown (Light Duty
Technology Cost Analysis, Power-Split and P2 HEV Case Studies, EPA-420-
R-11-015) study that the agencies had previously relied on for
component costs, noting that not only did the agencies ignore that
study in the NPRM, but that ICCT had commented 2010 FEV report
overstated strong hybrid costs at the time of the study, making it
likely that costs are likely to be lower now and even more so in the
future. CARB claimed that the agencies provided no justification or
rationale for the increases in strong hybrid modeled costs for the
proposal, and that there was no meaningful way to comment on the exact
components or cost changes that the agencies relied upon. Similarly,
CARB cited EPA's 2016 Proposed Determination and associated public
comments from Ford and Tesla on the Draft TAR for the proposition that
non-battery costs, which were lower in the Draft TAR than the NPRM,
were conservative and not overly optimistic.
Finally, in addition to the ORNL and Autonomie group studies that
CARB referenced as examples of sources that provided updated data on
non-battery component effectiveness and costs, CARB claimed that newer
data existed from a UBS Global Research report that examined the
component costs of a MY 2016 Chevrolet Bolt, and the agencies did not
discuss why the newer data was not used in the NPRM analysis. CARB
stated the significant upward adjustment in non-battery costs from
previous analyses was not supported by industry input, analysis
conducted by other outside sources, or by the agencies' previous
analyses.
As explained above, for the NPRM the agencies relied on Argonne's
``Assessment of Vehicle Sizing, Energy Consumption, and Cost through
Large-Scale Simulation of Advanced Vehicle Technologies'' for EDTS
costs. In turn, the Assessment of Vehicle Sizing, Energy Consumption,
and Cost through Large-Scale Simulation of Advanced Vehicle
Technologies report referenced electric machine data provided by OEMs,
suppliers, and Oak Ridge National Laboratory.\1247\ Regarding CARB's
assertion that the agencies did not refer to the UBS Global Research
report on the MY 2016 Chevy Bolt teardown for the NPRM, the agencies
agree. The UBS Global Research report was not available at the time the
CAFE model inputs were finalized for the NPRM analysis. That study,
among others, was considered for the final rule.
---------------------------------------------------------------------------
\1247\ Moawad, Ayman, Kim, Namdoo, Shidore, Neeraj, and
Rousseau, Aymeric. Assessment of Vehicle Sizing, Energy Consumption
and Cost Through Large Scale Simulation of Advanced Vehicle
Technologies (ANL/ESD-15/28), at 32.
---------------------------------------------------------------------------
For the final rule analysis, the agencies carefully considered
comments and the referenced studies, as well as other studies. The
agencies determined the cost and component efficiency estimates from
U.S. DRIVE's October 2017 report, Electrical and Electronics Technical
Team (EETT) Roadmap,\1248\ provided reasonable estimates to use in the
final rule. The EETT Roadmap report reflected considerable work by the
DOE VTO collaboratively with U.S. DRIVE, a government-industry
partnership. The EETT Roadmap report estimated the 2017 manufacturing
cost of a commercial on-road 100kW ETDS consisting of a single electric
traction motor and inverter. The reported costs were approximately
$1,800, with the cost of the electric motor accounting for $800, and
approximately $1,000 for the inverter, equaling $18/kW for the ETDS.
---------------------------------------------------------------------------
\1248\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
---------------------------------------------------------------------------
The agencies also referenced the UBS MY 2016 Chevy Bolt teardown
report to compare the cost of the ETDS.\1249\ To compare the costs, the
agencies applied the $18/kW metric for ETDS as determined by EETT
Roadmap report to the 150kW ETDS used in the MY 2016 Chevy Bolt ($18kW
x 150kW = $2700). As shown in Table VI-102, the cost estimate from the
above computation aligned with UBS MY 2016 Chevy Bolt teardown cost
estimate. As a result, the agencies determined that it was appropriate
to use $18/kW to estimate the cost of the ETDS for all hybrid and
electric vehicle architectures for the final rule.
---------------------------------------------------------------------------
\1249\ Hummel et al., UBS Evidence Lab Electric Car Teardown--
Disruption Ahead?, UBS (May 18, 2017), https://neo.ubs.com/shared/d1ZTxnvF2k/.
---------------------------------------------------------------------------
The EETT Roadmap report did not explicitly estimate the cost of
other electrical equipment present in PHEVs and BEVs, such as on-board
chargers, DC to DC converters, and charging cables, but recommended
cost targets for the years 2020 and 2025. As a consequence, the
agencies relied on the
[[Page 24512]]
UBS MY 2016 Chevy Bolt teardown report to estimate the cost of on-board
chargers, DC to DC converters, and charging cables. Table VI-102 shows
the cost estimate for the ETDS from the EETT Roadmap report and from
the UBS MY 2016 Chevy Bolt teardown report, and the cost estimate for
other electrical equipment from the same UBS report.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.234
[GRAPHIC] [TIFF OMITTED] TR30AP20.235
While the EETT Roadmap report estimated the cost of the ETDS at the
system level, the report did not itemize the cost of individual
components in electric motor and inverter in 2017. However, the EETT
Roadmap report provided target cost estimates for the motor and
inverter system for the year 2025. As shown in Table VI-104, the EETT
Roadmap report estimated a cost reduction of 73 percent for the
inverter and 59 percent for the motor relative to 2017. Using the
percentage cost reductions from 2025 to the on-road status as defined
in the EETT Roadmap report, the agencies developed an estimated motor
and inverter component cost for 2017. The resulting cost estimate for
2017 using the scaling factor matches the $18/kW for motor and inverter
($10/kW for Inverter + $8/kW for motor). Since the motor and inverter
component costs are developed based on a $/kW basis, the agencies
applied the same $/kW metric for all hybrid and electric vehicle
applications for the final rule analysis.
---------------------------------------------------------------------------
\1250\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap, at 12 (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
\1251\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap, at 12 (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
\1252\ T U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap, at 12 (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
\1253\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap, at 18 (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
---------------------------------------------------------------------------
[[Page 24513]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.236
[GRAPHIC] [TIFF OMITTED] TR30AP20.237
[GRAPHIC] [TIFF OMITTED] TR30AP20.238
BILLING CODE 4910-59-C
In addition, the EETT Roadmap report provided notably newer data
than the 2010 FEV teardown study referenced by commenters. Based on
these considerations, the agencies determined that the EETT Roadmap
report provided reasonable costs to estimate the cost of EDTS
components in the rulemaking timeframe.
---------------------------------------------------------------------------
\1254\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap, at 23 (Oct. 2017), available at https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
---------------------------------------------------------------------------
(3) Electrification Learning Curves
The total incremental costs of electrification powertrain
technologies are comprised of the DMC as modified by the learning
curves for each individual powertrain component, which include
batteries, non-battery components, and IC engines and transmissions
(for hybrids and PHEVs). The PRIA showed the learning curves for
battery and non-battery electrification technologies,\1255\ and listed
the sources used to develop those curves, including the 2015 NAS
report,
[[Page 24514]]
Wright-based learning curves,\1256\ and Argonne's 2016 Assessment of
Vehicle Sizing, Energy Consumption, and Cost through Large-Scale
Simulation of Advanced Vehicle Technologies.\1257\ Learning rates for
batteries were also derived using Argonne's BatPaC model.
---------------------------------------------------------------------------
\1255\ PRIA at 380.
\1256\ Wright, T. P. (1936). Factors Affecting the Cost of
Airplanes. Journal of Aeronautical Sciences, vol. 3 124-125. http://www.uvm.edu/pdodds/research/papers/others/1936/wright1936a.pdf.
\1257\ Moawad, Ayman, Kim, Namdoo, Shidore, Neeraj, and
Rousseau, Aymeric. Assessment of Vehicle Sizing, Energy Consumption
and Cost Through Large Scale Simulation of Advanced Vehicle
Technologies (ANL/ESD-15/28). United States (2016). Available at
https://www.autonomie.net/pdfs/Report%20ANL%20ESD-1528%20-%20Assessment%20of%20Vehicle%20Sizing,%20Energy%20Consumption%20and%20Cost%20through%20Large%20Scale%20Simulation%20of%20Advanced%20Vehicle%20Technologies%20-%201603.pdf.
---------------------------------------------------------------------------
For the NPRM, to develop the learning curves for non-battery
components, the agencies consulted Argonne's 2016 Assessment of Vehicle
Sizing, Energy Consumption, and Cost through Large-Scale Simulation of
Advanced Vehicle Technologies report. The report provided estimated
cost projections from the 2010 lab year to the 2045 lab year for
individual vehicle components.1258 1259 The agencies
considered the component costs used in electrified vehicles, and
determined the learning curve by evaluating the year over year cost
change for those components.
---------------------------------------------------------------------------
\1258\ ANL/ESD-15/28 at 116.
\1259\ DOE's lab year equates to five years after a model year,
e.g., DOE's 2010 lab year equates to MY 2015.
---------------------------------------------------------------------------
The agencies used BatPaC version 3.0 to develop the NPRM learning
curves for batteries. As discussed above, BatPaC calculations are based
on generic pack design for a given set of inputs that could reasonably
represent potential current and future designs. Because BatPaC does not
simulate battery costs as a function of time, the agencies modified the
battery volume inputs for MY 2015, MY 2020, MY 2025 to show costs in
each of those MYs. Like the non-battery component analysis, a learning
curve was developed from the year over year cost change, and this rate
was used to develop the learning curves used in the NPRM.
CARB stated that publicly available data supported lower costs in
the near term than what the applied learning curve rates would do to
the battery costs developed by the agencies, and the agencies failed to
consider new information or data to adjust battery costs.\1260\ CARB
stated that considering the substantial volume of publicly available
information and public input to the agencies' previous analysis,
projected battery costs should have been adjusted even further downward
for the NPRM. CARB stated that instead, the agencies moved costs upward
without sufficient justification, and in contrast, the analysis for the
Proposed Determination and 2016 Draft TAR provided far more
justification for those modeled battery costs.
---------------------------------------------------------------------------
\1260\ California Air Resources Board, NHTSA-2018-0067-11873, at
142-43.
---------------------------------------------------------------------------
As discussed in Section VI.B.4.d) Cost Learning, above, ICCT
commented broadly on the change in approach to learning curves since
the Draft TAR, stating that this change in approach led to lower
decreases in costs over time in the NPRM than the Draft TAR analysis.
ICCT compared EPA's Draft TAR learning curves and NPRM learning curves
for batteries in MYs 2016-2025, concluding that there was a 29%
reduction in learning for batteries from EPA's Draft TAR analysis to
the NPRM analysis.
The agencies considered an array of both present and future cost
estimates from various public and private sector organizations to
validate the rate at which battery pack costs declined over time. These
estimates, in addition to estimates submitted by commenters as
discussed in BatPaC Inputs and Assumptions and Final Rule Battery Pack
Costs are shown in Table VI-101. In addition, the agencies had to
consider how to project learning rates out through 2050, as discussed
in Section VI.B.4.d) Cost Learning and Section VI.C.3.e)(3)
Electrification Learning Curves.
The agencies also assessed and reviewed literature evaluating more
recent battery technology development.1261 1262 The NPRM
analysis used a three percent learning rate per year from MY 2033 to MY
2050. Learning rate forecasts from MY 2033 to MY 2050 for this final
rule analysis were scaled down in steps from the previous analysis
based on literature, market research, and Wright's learning curve
assumptions.
---------------------------------------------------------------------------
\1261\ MIT Energy Initiative. 2019. Insights into Future
Mobility. Cambridge, MA: MIT Energy Initiative. Available at http://energy.mit.edu/insightsintofuturemobility.
\1262\ Islam, E., Kim, N., Moawad, A., Rousseau, A., ``A Large-
Scale Vehicle Simulation Study To Quantify Benefits & Analysis of
U.S. Department of Energy VTO & FCTO R&D Goals.'' Report to U.S.
Department of Energy. Contract ANL/ESD-19/10. (forthcoming).
---------------------------------------------------------------------------
It is difficult to predict which battery chemistry and production
processes will be prevalent for electrified vehicles in MY 2030, let
alone for MY 2050. The agencies reviewed potential battery chemistries
that could come into readiness for adoption at different timeframes,
such as MY 2030s to MY 2039, and MY 2040 to MY 2050.\1263\ It is
possible that costs based on other lithium-ion based chemistries will
learn at the same rate as lithium-ion NMC development. However, the
same learning effect in battery production may not be additive across
different chemistries, especially in learning effects related to
battery production. Accordingly, the learning rates applied between MY
2030 to MY 2039 considered development and increased volume for the
same or similar battery chemistries as an NMC battery platform.\1264\
Learning curves beyond MY 2040 were flattened further to ensure that
the cost of batteries did not lower beyond the projected price of the
raw materials. Further, new chemistries introduced in later years may
learn at different rates than the curve identified for NMC-based
chemistries. The battery pack cost learning rate that resulted from
this exercise produced the schedule that appears in Table VI-96, which
shows this final rule analysis battery pack cost reduction as function
of time. By MY 2040, the pack cost has reduced by 54 percent.
Accordingly, the estimated battery pack cost between MY 2040 and MY
2050 as shown in Figure VI-43 below shows flatter curve.
---------------------------------------------------------------------------
\1263\ MIT Energy Initiative. 2019. Insights into Future
Mobility. Cambridge, MA: MIT Energy Initiative, at p. 79. Available
at http://energy.mit.edu/insightsintofuturemobility.
\1264\ For example, an NMC lithium-ion-based platform could move
from a cathode composition of NMC622 to NMC811.
---------------------------------------------------------------------------
[[Page 24515]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.239
The reference cost is defined for MY 2020 vehicles, and vehicles
produced in subsequent years (as well as earlier years) use a per kWh
cost that is a percentage of the 2020 cost. As the figure shows, the
cost reduction is rapid through MY 2030, after which cost reductions
slow considerably. As discussed above, the cost projections assumed
different battery chemistries and different rates of cost learning.
The agencies expect there will be incremental improvements in
battery chemistry, energy density, plant efficiency, and production
volume over the timeframe modeled in the analysis. While each of these
factors may have an impact on the rate at which battery costs decline
over time, the agencies determined that using the same cost learning
projection method from the NPRM to project learning rates out through
2050 provided a reasonable method for accounting for something that is
inherently uncertain. Accordingly, the learning curve used in the NPRM
and in the final rule represent a composite learning curve irrespective
of the type of battery chemistry, the production volume necessary to
achieve economies of scale, or energy density of the battery pack. For
the final rule, the agencies have performed sensitivity analyses
varying the battery pack learning rate, and these analyses are
presented in FRIA Chapter VII.E Sensitivity cases.
(4) Electrified Powertrain Costs
For the NPRM analysis and carried forward for the final rule
analysis, the total electrified powertrain costs were developed by
summing individual component costs. The costs associated with the IC
engine, transmissions, electric machines, and battery packs were
combined to create a full-system cost, per Section VI.C.3.e)(2) Non-
battery Electrification Component Costs, Section VI.C.3.e)(1) Battery
Pack Modeling, Section VI.C.1.g) Engine Costs, and VI.C.2.e)
Transmissions Costs. This approach assured all technologies
appropriately contributed to the total system cost.
The Alliance commented in support of the agencies' accounting
separately for the subsystems' costs and benefits for CISG, BISG, P2
hybrid, power split hybrid (PS), and PHEV technologies.\1265\ The
Alliance noted that these distinctions are important to capture the
differences between various technologies, which can have separate
packaging requirements, efficiency potentials, and vehicle
applications. Ford echoed the Alliance comments on the modeling of
electric vehicles in the NPRM, stating they supported the use of
separate cost and benefits modeling for P2 and power split strong
hybrid technologies.\1266\ Additionally, Ford commented that the
modeling ``better reflects market realities by recognizing that
manufacturers cannot simply pass on the entire incremental costs of
hybrid, plug-in hybrid, and battery electric vehicles to the
customers.''
---------------------------------------------------------------------------
\1265\ Alliance of Automobile Manufacturers, NHTSA-2018-0067-
12073, at 140.
\1266\ Ford Motor Company, NHTSA-2018-0067-11928, at 10.
---------------------------------------------------------------------------
Comments from other stakeholders generally stated that the NPRM
powertrain sizing approach resulted in costs for complete powertrains
that were too high compared to other studies or market observations. In
addition, as discussed in Section VI.C.1.g) Engine Costs, CARB also
commented that the costs associated with IC engines were not excluded
from the final costs of BEV vehicles.\1267\ CARB continued, stating
that ``the final costs of BEV vehicles are higher due to the inclusion
of the base absolute costs, to which the assigned BEV incremental cost
would be added.'' The agencies agreed with CARB that inclusion of IC
engine costs in the BEV cost was an error in the analysis.
---------------------------------------------------------------------------
\1267\ NHTSA-2018-0067-11873 at p.122.
---------------------------------------------------------------------------
In response to this comment, the agencies developed absolute costs
for baseline engines for the CAFE Model so the absolute costs for IC
engines could be removed from BEVs. In the final rule analysis, when a
vehicle adopted BEV technology, the costs associated with IC powertrain
systems were removed. As the vehicle walks through the technology tree,
becoming a battery electric vehicle, the motor and inverter (ETDS)
costs replaced the internal IC engine costs. Since the cost of the ETDS
accounted for significant portion of the
[[Page 24516]]
total cost of electrification, it was important to accurately
characterize the motor size (motor rating). To do this, the agencies
used the MY 2017 market data file to compute the average engine power
for each technology class.
For SHEVPS and SHEVP2 vehicles, as explained further in Section
VI.C.3.e)(4)(c) Strong Hybrid Costs, the agencies computed the average
rating for traction and generator motors across all road load
combinations using Autonomie simulation runs. Since motor sizing varies
based on road load levels, the average motor sizes acted as a mid-range
representation for motor ratings across all road load combinations. The
full range of motor sizes are driven by road load limits; the motor
size for initial road load levels (MR0, AERO0 and ROLL0) would be
larger compared to the motor size for highest level of road load
reduction (MR6, AERO20 and ROLL20). After calculating the average motor
size, the agencies applied the $18/kW metric (derived from the EETT
Roadmap report) for both traction motors and generator motors. As
discussed earlier, the agencies also used the cost of the CVTL2 as
proxy to represent the cost of the eCVT used in power-split hybrid
vehicle systems, and used the cost of the AT8L2 as proxy for the cost
of the planetary gear set used in the P2 parallel hybrid system. The
total cost of electrification for power-split hybrid vehicles includes
the cost of the eCVT transmission, and the total cost of
electrification for the P2 parallel hybrid vehicles includes the cost
of the planetary gear set transmission.
CARB also submitted supplemental comments attempting a cost walk
for electrified powertrain technologies, stating that inconsistencies
in the model files and PRIA and lack of documentation about how the
costs were derived ``[left] the public without the ability to
understand why the costs are what they are and what should be
applied.'' \1268\ Accordingly, a cost walk for a vehicle adopting an
electrified powertrain is shown below. Additional comments on
electrified powertrain costs are discussed in each individual
technology section below, along with a discussion of changes made for
the final rule in response to these comments.
---------------------------------------------------------------------------
\1268\ California Air Resources Board, NHTSA-2018-0067-12428, at
25.
---------------------------------------------------------------------------
For the final rule analysis, the agencies have updated several
electrification inputs and assumptions in response to these comments,
as discussed in the previous sections. An example of how the costs are
applied to a simulated vehicle platform's technology cost is discussed
here, to assist CARB and other stakeholders in assessing
electrification technology costs for the final rule analysis. The
example shows the costs for a vehicle with conventional engine and
transmission technology as it adds electrification technology.
The application of the electrification costs to an existing
platform follows the same basic process for each technology on the
electrification path. All technology costs used are for the model year
of the electrification technology application. The first step is the
process is the removal of the costs associated with the conventional
drivetrain technologies. The next step is the application of the costs
associated with the electrification technology. The costs include the
cost of the engine, if applicable, transmission, non-battery
components, and the battery pack. After the electrification costs are
applied, other technology costs, such as aerodynamic or rolling
resistance technologies are applied.
The specific example is the Toyota Rav4 LE AWD/XLE AWD simulated
platform. The platform data were used from the reference run CAFE model
standard setting vehicle_report.csv result file, augural standards
results. The change in technology for the simulated platform was
between MY 2023 and MY 2024. Table VI-107 shows the costing change
between the MYs.
[GRAPHIC] [TIFF OMITTED] TR30AP20.240
Table VI-108 shows the costs, and where to find them, for the
drivetrain components subtracted from the MY 2023 version of the
platform. The costs for current engine and transmission were
subtracted. To properly cost the engine it is important to note the
engine was designated as a 4C1B engine, or, 4 cylinder 1 bank engine
type. For more information about engine geometry designation in the
technology input file please see Section VI.A.7 Structure of Model
Inputs and Outputs.
[[Page 24517]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.241
The costs for the new electrification technology were then applied.
For the specific example the simulated vehicle platform is being
converted to a PHEV20 powertrain. For all the technologies in the
electrification path two major component groups were always added, the
battery pack and the non-battery components. Hybrid electric
technologies will also include the cost for an engine. Table VI-109
shows the costing data for the non-battery pack electrification
technology components, and where the cost data can be found.
[GRAPHIC] [TIFF OMITTED] TR30AP20.242
The battery pack is cost is determined by multiplying the baseline
battery pack cost by the learn curve factor. Table VI-110 shows the
calculation of the battery pack costs. The baseline battery costs are
determined per discussions in Section VI.C.3.e)(1) Battery Pack
Modeling.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.243
Table VI-111 shows a summary of the total cost application for the
technology transition of the Rav4 example platform. The added costs of
the addition of the LDB technology, improvement from AERO15 to AERO20,
improvement from MR0 to MR1 are summarized. However, the costing data
for these technologies can be found in the Technology Input file on the
`SmallSUV' tab under each technology's respective rows.
[[Page 24518]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.244
The following sections discuss specific electrification component
cost comments on the NPRM, responses, and any relevant assumptions for
the final rule analysis.
a) Micro Hybrid Cost
As stated in PRIA, the cost of SS12V in NPRM included the cost of
the battery, learning rate and retail price equivalent.\1269\ The
assumed direct manufacturing cost (DMC) was the same as was used for
the Draft TAR and the Proposed Determination,\1270\ but adjusted for
learning and updated from 2013 to 2016 dollars. Cost learning made the
cost of SS12V presented in the NPRM slightly lower than the Proposed
Determination.
---------------------------------------------------------------------------
\1269\ Footnote n. 364 in PRIA; Table 6-32 and Table 6-33.
\1270\ Draft TAR Table 5.210.
---------------------------------------------------------------------------
ICCT compared the agencies' NPRM cost effectiveness estimate for
SS12V with EPA's Proposed and Final Determination analyses, and
concluded that the latter analyses found SS12V cost nearly $100 less
than the agencies found in the NPRM, with a higher effectiveness
benefit.\1271\ ICCT noted its difficulty in evaluating whether SS12V
technology was actually cost-effective, since the NPRM CAFE model added
the incremental cost of BISG over SS12V. ICCT stated that because SS12V
is not as cost effective as other technologies in the electrification
technology pathway, such as BISG, the analysis' estimate of SS12V costs
was exaggerated and resulted in an unrealistic increase in compliance
costs.
---------------------------------------------------------------------------
\1271\ International Council on Clean Transportation,
``Attachment 3_ICCT 15page summary and full comments appendix,''
NHTSA-2018-0067-11741, at I-63.
---------------------------------------------------------------------------
While BISG is more expensive than the SS12V, BISG provides
additional benefits such as smoother start-stop (reduced vibration
during each start-stop event), launch assist and/or torque assist
(during certain sudden acceleration while passing or load at low speed
for short burst of time). Therefore, the effectiveness of SS12V should
not be compared to BISG. The agencies have always considered BISG as a
separate technology. Also, the effectiveness of SS12V in the Proposed
Determination was determined using ALPHA modeling. A peer reviewer
noted that ``[a]ccording to the documentation review, ALPHA's stop/
start modeling appears to be very simplistic.'' \1272\ As discussed in
Section VI.B.3 Autonomie model, the Autonomie tool simulates the
technology as part of the full vehicle system, accounting for
interactions with other technologies, and therefore the agencies
believe the full-vehicle simulations provide more realistic
effectiveness estimates than the value from the Proposed Determination.
For these reasons, the agencies disagree with ICCT's assertions. For
SS12V, the agencies continued to use the costs from the NPRM, which are
consistent with the Draft TAR and Proposed Determination. The ETDS
costs presented in the final rule do not include the cost of the
battery.
---------------------------------------------------------------------------
\1272\ Peer Review of ALPHA Full Vehicle Simulation Model, at C-
4, available at https://nepis.epa.gov/Exe/ZyPdf.cgi?Dockey=P100PUKT.pdf.
---------------------------------------------------------------------------
b) Mild Hybrid Cost
The belt integrated starter generator (BISG) and crank integrated
starter generator (CISG), sometimes referred to as mild hybrid systems,
provide idle-stop capability and use a higher voltage battery with
increased energy capacity over typical automotive batteries. The higher
voltage allows the use of a smaller, more powerful and efficient
electric motor/generator which replaces the standard alternator. For
the NPRM the agencies developed the costs for the mild hybrid systems
assuming the use of a 115V system. The battery, motor, and supporting
components were sized and costed based on this voltage level.
Many commenters asserted that the costs presented in the NPRM
analysis for BISG and CISG systems were inflated or incorrect.\1273\
ICCT noted that because mild hybrid systems were
[[Page 24519]]
widely adopted by the fleet under the augural standards, the high cost
of those systems had a significant impact on the costs of the
standards.\1274\
---------------------------------------------------------------------------
\1273\ International Council on Clean Transportation, NHTSA-
2018-0067-11741; Union of Concerned Scientists, NHTSA-2018-0067-
12039; Fiat Chrysler Automobiles, NHTSA-2018-0067-11943; Alliance of
Automobile Manufacturers, NHTSA-2018-0067-12073; California Air
Resources Board, NHTSA-2018-0067-11873.
\1274\ International Council on Clean Transportation, NHTSA-
2018-0067-11741, at I-24.
---------------------------------------------------------------------------
Meszler Engineering Services noted that the NPRM documentation
presented BISG/CISG battery costs that were ``not unreasonable,'' and
that the CAFE model database of battery costs used for NPRM analysis
included estimates for those electrification technologies that were
$259 higher than those presented in the NPRM documentation.\1275\
Meszler surmised that it initially appeared as if the model may have
been applying a redundant RPE factor to BISG/CISG costs, but noted that
the determination that the costs differed from those documented by a
constant absolute offset made that assumption an unlikely possibility.
---------------------------------------------------------------------------
\1275\ Meszler Engineering Services, NHTSA-2018-0067-11723
Attachment 2.
---------------------------------------------------------------------------
ICCT and UCS both noted the discrepancy between the reported
battery costs in the PRIA and costs reported in the NPRM Autonomie
simulation databases.\1276\ ICCT disagreed with the agencies' approach
to modeling batteries in the NPRM analysis, stating that ``[n]ot only
is [the Argonne] database exceedingly difficult to access to modify
battery costs (as battery costs should be a user input), but it makes
it much harder to see how battery costs affect mild hybrid costs over
time.'' \1277\ Claimed difficulties aside, ICCT concluded that the
battery costs were outdated and grossly overstated, based on the tables
in section 6.3.9.12 of the PRIA and the outputs of the low battery cost
sensitivity case, which ICCT stated were more closely aligned with EPA
and other research on battery costs. ICCT presented its own best
estimate of NPRM BISG costs, stating that they were not able to make
the PRIA and datafile costs match up.
---------------------------------------------------------------------------
\1276\ International Council on Clean Transportation, NHTSA-
2018-0067-11741; Union of Concerned Scientists, NHTSA-2018-0067-
12039.
\1277\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
---------------------------------------------------------------------------
Several commenters noted that the costs of BISG/CISG systems were
higher for Small Cars/SUVs and Medium Cars than for Medium SUVs and
Pickup trucks, which the Alliance and FCA described as ``implausible''
and ``misaligned with industry understanding,'' and which ICCT
described as ``contrary to basic engineering logic, which holds that a
system which would be smaller and have lower energy and power
requirements would be less expensive, not more.'' \1278\ Both ICCT and
UCS stated that regardless of alleged errors in costs between
technology classes, even the lower of the values presented in the PRIA
overestimated the cost of mild hybrid batteries.\1279\
---------------------------------------------------------------------------
\1278\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1279\ International Council on Clean Transportation, NHTSA-
2018-0067-11741; Union of Concerned Scientists, NHTSA-2018-0067-
12039.
---------------------------------------------------------------------------
The Alliance and FCA urged the agencies to update the CAFE model to
address this issue so that the cost of compliance was properly
reflected in the results. To estimate the impact of the error, the
Alliance and FCA modified the technology input file so that the Medium
SUV and Pickup truck electrification costs were changed to be identical
to the Small Car/SUV and Medium Car costs for SS12V, BISG, and CISG,
and re-ran the CAFE model to show an estimated $13 billion increase in
compliance costs under the augural standards with the error
corrected.\1280\
---------------------------------------------------------------------------
\1280\ Fiat Chrysler Automobiles, NHTSA-2018-0067-11943;
Alliance of Automobile Manufacturers, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Conversely, CARB modified the fuel consumption improvement
estimates for BISG systems to match those predicted by Argonne in a
recent report after calculating the smallest modified improvement from
MYs 2015-2025 for five vehicle classes, resulting in efficiency
improvements of 8.5-11 percent.\1281\ CARB also reduced the non-battery
costs for Small Car/SUVs to match the non-battery costs for Medium SUV
and Pickup trucks, which CARB stated still reflected higher costs than
those previously used by EPA in the Proposed Determination. CARB did
not modify the battery costs, but did comment that they were overstated
by approximately 50 percent ``due to the erroneous oversizing of the
battery.'' CARB's modified run decreased average vehicle technology
costs by a range of $300-$500 per year, ``reflecting an approximate 25
percent drop in 2029 model year incremental technology costs to meet
the existing standards relative to the rollback standards.''
---------------------------------------------------------------------------
\1281\ California Air Resources Board, NHTSA-2018-0067-11873
(``Specifically, the fuel consumption improvements modeled by ANL in
the most recent report for DOE were utilized in place of the
assumptions used for the Agencies' analysis. As noted above, ANL,
via Autonomie modeling, identified efficiencies between 8.5 percent
to 12.7 percent for mild hybrids, relative to both gasoline spark
ignited and relative to turbocharged gasoline spark ignited across
five different vehicle classes. Using approximately the smallest
modeled improvement across the 2015 to 2025 model years for each of
the five classes, improvements of 8.5 percent-11 percent were
utilized for a modified CAFE Model run.'').
---------------------------------------------------------------------------
Commenters also pointed to prior agency analyses, studies, and
applications of BISG systems to provide examples of what they believed
BISG system costs should be, with ICCT arguing that the agencies' cost
values for BISG/CISG systems were contrary to the research and
evidence.\1282\ HDS noted that the 2018 PRIA estimate was approximately
double the estimate from the 2016 Draft TAR, that the difference in
battery costs between those two analyses did not explain the
difference, and that there was no discussion in the PRIA that did
so.\1283\
---------------------------------------------------------------------------
\1282\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1283\ H-D Systems, NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
UCS stated that BISG system costs have already reached that which
was predicted in EPA's first Final Determination, published in 2017,
for 2025, and would decline further because of continued volume-based
learning.\1284\ UCS also cited a 2018 Argonne report that estimated the
battery component cost for a mild hybrid system to be $159.35, and a
Chevrolet Malibu eAssist teardown study that estimated total battery
subsystem direct costs at $166, and battery modules, power
distribution, and covers at $120 in direct manufacturing costs.\1285\
UCS summarized that the aforementioned costs are less than half the
costs listed in the PRIA and approximately one quarter of the
``BatPaCCost'' value given in the Argonne input files. UCS also cited
cost estimates from the 2015 NAS report and two EPA reports, and
concluded that the agencies did not sufficiently explain why the NPRM
cost data differed so substantially from this other available
information.
---------------------------------------------------------------------------
\1284\ Union of Concerned Scientists, NHTSA-2018-0067-12039.
\1285\ Id. (citing [Component Cost, ANL 2017k]).
---------------------------------------------------------------------------
ICCT cited its own 2016 study of supplier costs with estimates for
48V mild hybrid systems, estimating the system cost at $600-$1,000
(with costs on the lower side for cars and the higher side for light
trucks) in the 2025 timeframe.\1286\ ICCT pointed to the RAM 1500
pickup truck as an example of a vehicle with a BISG system that ``has
already validated the ICCT figures in 2019.'' ICCT noted that the BISG
system, branded as eTorque, was first offered as a ``free standing''
option on the RAM 1500 truck for $800, and that price was recently
raised to $1,450. ICCT stated that even with the higher price, applying
the agencies' RPE of 1.5 means
[[Page 24520]]
that the direct manufacturing cost is less than $1,000, which is less
than the $1,616 direct manufacturing cost estimate in the NPRM for 2016
pickup trucks.\1287\ Similarly, UCS cited the $500 premium that General
Motors charged for the technology on its Chevrolet Silverado pickup
trucks with eAssist.\1288\
---------------------------------------------------------------------------
\1286\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1287\ ICCT also stated that the eTorque system offered improved
performance and driveability and contributes to higher payload and
towing ratings for 2019 compared with 2018, and noted that the
agencies ``have completely failed to account for the consumer value
of the utility benefits'' from the system. The agencies' approach to
simulating performance neutrality and the consumer benefit of
increased performance are discussed in Section VI.B.3.a)(6)
Performance Neutrality.
\1288\ Union of Concerned Scientists, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
The agencies reviewed all of the comments and information provided.
It appears there may have been confusion about what costs were used for
the Draft TAR and NPRM. For the Draft TAR, non-battery BISG costs,
including learning and RPE, were $1,701 compared to $1,186 for the NPRM
(both costs in 2018 dollars). Therefore, the costs for the NPRM were
lower than for the Draft TAR when cost accounting is on an equivalent
basis.
---------------------------------------------------------------------------
\1289\ Table 5.131 in Draft TAR ($1,045 x 1.5 = $1567.5 in
2013$. (Absolute cost, without batteries. This includes learning and
Retail Price Equivalent).
\1290\ Table 6-32 in PRIA (Absolute Electrification Cost without
batteries. This includes learning and Retail Price Equivalent).
\1291\ See Table I 19--Cost and Mass Estimate of BISG
components.
[GRAPHIC] [TIFF OMITTED] TR30AP20.245
The agencies also determined the cost presented by EPA in Draft TAR
(see Table 5.131 in Draft TAR) was the direct manufacturing cost of the
BISG system, and not the retail price equivalent. The Draft TAR cost
estimate in Table VI-112 includes the RPE and costs updated from 2013
to 2018 dollars. The agencies agree with the commenters about the
discrepancy in the cost of the battery pack for the BISG system
presented in PRIA and in CAFE model. To avoid any confusion, Table VI-
112 shows the non-battery costs of the BISG system.
After considering the comments and reviewing the approach used in
the NPRM, the agencies agreed updating the cost of the BISG system was
appropriate for the final rule analysis. Adjustments were based on
using a 48V BISG system instead of the 115V system used for the NPRM.
For the final rule, the agencies considered several cost sources,
including the EPA-sponsored FEV report titled: Light-Duty Vehicle
Technology Cost Analysis on 2013 Chevrolet Malibu ECO with eAssist BAS
Technology Study.\1292\ Based on the teardown study, EPA estimated the
direct manufacturing cost of the BISG system (without batteries) to be
$1,045 in 2013 dollars. This included a cost adjustment for reduced
voltage insulation. The agencies also considered the 2019 Dodge Ram
eTorque system retail price. A cost of $1,195 for water-cooled system
and $1,450 for air-cooled system in 2018 dollars was deduced from the
retail price of eTorque assist (BISG) system. The 2015 NAS report
estimated the cost range of BISG technology at $888 to $1,164 in 2010
dollars in 2025.\1293\ This is equivalent to a range of $1,020 to
$1,337.27 in 2018 dollars in 2025. The agencies also reviewed
confidential business information on BISG cost and mass estimates
provided by manufacturers.
---------------------------------------------------------------------------
\1292\ Light Duty Vehicle Technology Cost Analysis 2013
Chevrolet Malibu ECO with eAssist BAS Technology Study, FEV P311264
(Contract no. EP-C-12-014, WA 1-9).
\1293\ Cost, Effectiveness and Deployment of Fuel Economy
Technologies for Light-Duty Vehicles, National Academy of Sciences,
2015.
---------------------------------------------------------------------------
For the final rule analysis, the agencies used the A2Mac1 database
to develop a bill of materials for BISG systems. The agencies sourced
cost estimates for the motor, inverter and DC-DC converter from the
2017 EETT roadmap report.\1294\ The agencies used BatPaC model version
3.1 to perform a standalone analysis determining the cost of a battery
pack for the 48V system.1295 1296 Table VI-113 shows the
cost and mass estimates for BISG components used in the final rule.
---------------------------------------------------------------------------
\1294\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap (October 2017), https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
\1295\ A Detailed Vehicle Simulation Process To Support CAFE and
CO2 Standards for the MY 2021--2026 Final Rule Analysis, at Table
50.
\1296\ BatPac 10032018 BISG Version 3.1--28June2018_FINAL.
---------------------------------------------------------------------------
[[Page 24521]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.246
The agencies compared the cost estimates in the 2017 EETT roadmap
report and found they aligned well with cost estimates from sources
cited by commenters. For reference, Table VI-113 above showed the cost
estimate for BISG system (without the battery) used in Draft TAR, NPRM
and in Final Rule. Furthermore, the agencies considered the Alliance
and FCA analysis, provided in their respective comments, recommending
the use of the same BISG system cost for both cars and
trucks.1297 1298 This analysis, supplemented with CBI data,
demonstrated that the costs for implementing BISG systems on different
vehicle classes was not appreciably different. The agencies agree with
this assessment. For the final rule analysis, the cost of the BISG
system is the same for cars, SUVs, and pickups.
---------------------------------------------------------------------------
\1297\ Fiat Chrysler Automobiles, NHTSA-2018-0067-11943, at 85.
\1298\ Alliance of Automobile Manufacturers, NHTSA-2018-0067-
12073, at 140-42.
---------------------------------------------------------------------------
(c) Strong Hybrid Cost
In the NPRM and this final rule analysis, the total cost for strong
hybrids (SHEVP2 and SHEVPS) included the electric machine, battery
pack, IC engine, and transmission. Discussed earlier in Section
VI.C.3.d) Electrification Effectiveness Modeling, each strong hybrid
powertrain is optimized for the given vehicle class by appropriate
sizing of the electric machine, IC engine and battery pack.
Accordingly, the costs represent the optimized system. For the NPRM,
the agencies referred to the ``Assessment of vehicle sizing, energy
consumption, and cost through large-scale simulation of advanced engine
technologies'' report to estimate the cost and effectiveness for
different hybrid systems for the NPRM.\1299\ For the final rule, as
discussed in Section 2) and further below, the agencies sourced cost
estimates from the October 2017 U.S. DRIVE report, ``Electrical and
Electronics Technical Team Roadmap.'' \1300\
---------------------------------------------------------------------------
\1299\ Moawad, Ayman, Kim, Namdoo, Shidore, Neeraj, and
Rousseau, Aymeric. Assessment of Vehicle Sizing, Energy Consumption
and Cost Through Large Scale Simulation of Advanced Vehicle
Technologies (ANL/ESD-15/28). United States (2016), available at
https://www.autonomie.net/pdfs/Report%20ANL%20ESD-1528%20-%20Assessment%20of%20Vehicle%20Sizing,%20Energy%20Consumption%20and%20Cost%20through%20Large%20Scale%20Simulation%20of%20Advanced%20Vehicle%20Technologies%20-%201603.pdf.
\1300\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap (October 2017), https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
---------------------------------------------------------------------------
SHEVP2 and SHEVPS have different characteristics and in turn have
different costs, as reflected in both the NPRM and this final rule
analysis. The cost for engines and transmissions for SHEVP2s are based
on estimates discussed further in Sections VI.C.1 Engine Path and
VI.C.2 Transmission Path, respectively. The cost for SHEVP2 electric
machines and battery packs were dependent on their sizes, which were
optimized by the Autonomie sizing algorithm. SHEVPS total powertrain
costs includes the optimized battery pack, electric machine, an
Atkinson engine, and the CVT.
Many commenters generally stated that the costs of hybrid
technology were overestimated in comparison to prior agency estimates
and other publicly available sources, and that the agencies'
documentation of hybrid system costs was unclear.
Meszler Engineering Services commented that the net costs of
vehicles that apply SHEVP2 technology were in error, resulting from the
way that the CAFE model applied HCR, CEGR and TURBO technology in
combination with the SHEVP2 strong hybrid system.\1301\
---------------------------------------------------------------------------
\1301\ Meszler Engineering Services, NHTSA-2018-0067-11723.
---------------------------------------------------------------------------
HDS claimed that cost estimates for both SHEVP2 and SHEVPS were
significantly higher than the Draft TAR estimates, differing by a
factor of about 2 for SHEVP2 and by a factor of 2.5 for SHEVPS, with no
justification given for the increase in costs.\1302\ HDS noted that the
SHEVPS cost estimates were particularly surprising since the costs have
been investigated extensively since that technology was introduced to
the market over a decade ago. HDS stated that the 2016 TAR estimates
were in line with other analyses like the NAS
[[Page 24522]]
estimate, and consistent with actual retail price increments observed
in the market.
---------------------------------------------------------------------------
\1302\ H-D Systems, NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
HDS also pointed to cost estimates based on teardown studies
sponsored by EPA and the European Union,\1303\ public cost data
disclosed by suppliers of hybrid systems, and the retail prices of
available hybrid vehicles as estimates that contradict the agencies'
NPRM cost estimates. HDS compared the European Vehicle Market Phase 1
FEV cost analysis to the costs published by EPA in the TAR, concluding
that the EU costs ``even at [levels adjusted for the strength of the
Euro] are quite similar to EPA estimates of $2,650 to $3,300 (depending
on vehicle size) published in the TAR for the P2 hybrid, and also shows
that the PS hybrid is just 7 percent more expensive than the P2
hybrid.'' HDS stated that battery costs have also certainly decreased
since 2012 when the report was written, so current costs are estimated
to be approximately $400 less than the values cited above.
---------------------------------------------------------------------------
\1303\ Id., citing FEV, Light-Duty Vehicle Technology Cost
Analysis-European Vehicle Market (Phase 1), (2012, updated 2013),
available at https://www.theicct.org/.
---------------------------------------------------------------------------
HDS also cited a methodology to estimate costs from retail price
increments in the market,\1304\ stating that a typical cost-to-retail
price ratio is 1.5. Applying this methodology, the cost of the SHEVPS
hybrid as used by Ford and Toyota would be in the $2,500 to $3,000
range, the cost of a SHEVP2 as used by Hyundai Kia would be $2,250, and
the cost of a low volume and/or luxury model system would be estimated
at $3,300 for a SHEVP2.
---------------------------------------------------------------------------
\1304\ Id. (citing Vincentric Hybrid Analysis, executive
summary, www.vincentric.com/Home/IndustryReports/HybridAnalysis
October2014.aspx.).
---------------------------------------------------------------------------
Similarly, ICCT stated that the agencies failed to analyze properly
the dozens of hybrid vehicles in the marketplace, their costs which
were lower than the agencies assumed, and their rapid improvements from
automakers and suppliers competitively developing lower cost components
for those vehicles.\1305\ ICCT observed an incremental price increase
in the analysis for hybrid vehicles under the augural standards of
approximately $6,600 per hybrid vehicle in 2017 and $4,800 in 2025, and
concluded that this was not a plausible result considering hybrid
component costs and full-vehicle prices in the marketplace in 2016 as
well as the technology improvement that continues to enter the fleet.
ICCT stated that the agencies must set a maximum cost premium for full
hybrids of $2,500 in 2017, declining linearly to $1,400 by 2025 for
mid-size cars and crossovers, with cost components likely scaling by
vehicle power requirements (up for pickups, down for smaller cars),
which it stated the agencies must also account for in the modeling.
---------------------------------------------------------------------------
\1305\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
---------------------------------------------------------------------------
ICCT stated that the agencies must disclose the basis for the
``unrealistically high'' hybrid system cost estimates, such that the
public can clearly connect the bottom-up cost components to full
vehicle costs for all vehicle models that have hybrid cost
applied.\1306\ ICCT stated that hybrid system cost estimates are ``one
of the most important technology cost estimations to assess the Augural
standards' compliance cost, as the NPRM projects that 22 percent of
vehicles will need full hybrid systems to meet the augural standards,''
and accordingly after disclosing those costs, the agencies must provide
another opportunity for public comment. Similarly, CARB stated that it
was unable to decipher the hybrid cost components, and without that
information could only guess as to why the costs increased relative to
costs in the Draft TAR and EPA's Proposed Determination.\1307\ As such,
CARB stated they could not make a conclusion as to whether improper
battery resizing, incorrectly modeled batteries, or oversized electric
motors contributed to the overestimation of costs for strong hybrid
systems.
---------------------------------------------------------------------------
\1306\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1307\ California Air Resources Board, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
The agencies believe comparing the retail price of P2 or PS hybrid
to conventional vehicles could be misleading. Even though hybrid
vehicles may have higher direct manufacturing costs, manufacturers may
choose not to price it higher than the conventional version of the
vehicle. In other words, manufacturers may choose to subsidize the cost
of hybrid technologies to gain overall credit for fleetwide compliance.
Therefore, the agencies believe that comparing retail price between
hybrid and conventional vehicles should be done only when other sources
of information are available to corroborate the differences in retail
price.
The agencies also referred to an EPA-sponsored teardown and cost
estimate report as suggested by HDS. Table VI-114 shows the absolute
cost of P2 and PS hybrid systems as estimated in the EPA sponsored
teardown report and the absolute cost estimated in the final rule in
2018$. As indicated above, the absolute cost in the final rule includes
the cost of transmissions for the PS and P2 hybrid systems. The EPA
teardown cost estimate includes the cost of the eCVT for the PS hybrid
systems only. The P2 hybrid system costs do not include the cost of
engine and transmission in the table below.
Although ICCT suggested that the agencies cap the maximum cost
premium for full hybrids of $2,500 in 2017 and linearly decrease the
cost to $1,400 by 2025, ICCT did not provide any supporting material to
suggest that maximum upper limit of $2,500 for full hybrid is
economically feasible, nor did they provide an example of an existing
full hybrid vehicle in the marketplace with a technology increase of
$2,500 in 2017. ICCT also did not make it clear if the costs suggested
would be applicable to P2 or PS hybrid architecture.
Based on the comments, the agencies reassessed SHEVP2 and SHEVPS
cost estimates for the final rule. As discussed above, the agencies
referred to U.S. DRIVE's October 2017 report, ``Electrical and
Electronics Technical Team Roadmap'' \1308\ to estimate the cost of
motors and inverters. The agencies also agreed with commenters and
referenced the MY 2016 Chevrolet teardown report by UBS to estimate the
cost of other hybrid components such as wiring harness, cables,
voltage-step-down DC to DC converters, and on-board chargers. Per
Section VI.C.3.e)(2) Non-battery Electrification Component Costs, for
the final rule, the cost of non-battery hybrid system components
includes the cost of traction motor, motor/generators, high voltage
cables and connectors, charging cord, and on-board chargers. The cost
of the planetary gear set is also included in the cost of non-battery
components. Per Section VI.B.4 Technology Costs, for the final rule,
the cost of hybrid systems is presented as absolute cost, and not as an
incremental to some previous technology (absolute cost includes the
retail price equivalent). The agencies used the cost of the AT8L2
transmission as a cost proxy for the planetary gear set in P2 hybrid
systems, and used the cost of CVTL2 transmission as a cost proxy for
planetary gear set for PS hybrid systems. It should also be noted the
costs shown here do not include the cost of engine coupled to the
hybrid system.
---------------------------------------------------------------------------
\1308\ U.S. DRIVE, Electrical and Electronics Technical Team
Roadmap (October 2017), https://www.energy.gov/sites/prod/files/2017/11/f39/EETT%20Roadmap%2010-27-17.pdf.
---------------------------------------------------------------------------
The agencies reviewed the FEV 2010 Ford Fusion HEV teardown report,
Light Duty Technology Cost Analysis, Power-
[[Page 24523]]
Split and P2 HEV Case Studies.\1309\ In a Split-HEV architecture, there
are two motors; one motor provides torque while the other motor act as
a generator to recapture the energy during regenerative braking. The
report does not capture the cost of motor-generator and the cost of the
DC to DC converter. The report did not include an extensive teardown of
a P2 hybrid vehicle, but rather made a cost adjustment for the PS motor
and inverter to reflect additional cost. Table VI-114 shows the
breakdown of cost estimates for the electric machine in the 2010 Ford
Fusion HEV.\1310\ Since the costs were developed in 2009$, the cost
estimates for the same components are presented in 2018$. Table VI-115
shows the cost estimate for electric machines for a midsize passenger
car for MY 2017 in 2018$.\1311\ The cost is estimated using the EETT
Roadmap report as explained earlier. Since EPA uses indirect cost
multiplier (ICM) to determine the final retail price, and ICMs vary for
different technologies, the agencies compared the direct manufacturing
cost from report to the direct cost estimate in the final rule.
---------------------------------------------------------------------------
\1309\ Light Duty Technology Cost Analysis, Power-Split and P2
HEV Case Studies, EPA-420-R-11-015 (November 2011), available at
https://nepis.epa.gov/Exe/ZyPDF.cgi/P100EG1R.PDF?Dockey=P100EG1R.PDF.
\1310\ Table D-4 (components considered are transmission, power
distribution cables and Inverter). The cost of inverter is from
Table D-11.
\1311\ Average peak power for the traction motor used in this
final rule is 72kW, and 37kW continuous power for the generation
motor.
---------------------------------------------------------------------------
The direct manufacturing cost estimated in the Light Duty
Technology Cost Analysis, Power-Split and P2 HEV Case Studies published
for EPA is $3,689.28 in 2018$, and direct manufacturing cost estimated
for electric machines in this final rule is $4,355.82. As mentioned
before, the cost of the motor-generator and the cost of the DC to DC
converter is not captured in that report.
[GRAPHIC] [TIFF OMITTED] TR30AP20.247
[GRAPHIC] [TIFF OMITTED] TR30AP20.248
(d) PHEV Cost
Plug-in hybrid vehicles' costs were developed similar to strong
hybrids for the NPRM analysis and the final rule analysis. The plug-in-
hybrid system components were optimized, per Section VI.C.3.d)(2)
Modeling and Simulating Vehicles with Electrified Powertrains in
Autonomie and the resultant systems were used to determine costs, per
Battery Pack Modeling and Non-battery Electrification Component Costs.
Per Section VI.C.3.c) Electrification Adoption Features, the agencies
used one engine technology and one transmission technology per plug-in
hybrid architecture type.
For PHEVs following SHEVP2 on the hybrid/electric architecture
path, per Section VI.C.3.a)(1) Electrification technologies, the total
cost of the technology package was determined from summing the costs of
the TURBO1 engine, the AT8L2 transmission, and the battery and non-
battery electrification technology components. For PHEVs following
SHEVPS on the hybrid/electric architecture path, per Section
VI.C.3.a)(1) Electrification technologies, the total cost of the
technology package was determined from summing the costs of the
Atkinson engine, the CVT transmission, and the battery and non-battery
electrification technology components.
CARB provided observations about non-battery component costs for
PHEVs,
[[Page 24524]]
arguing that what the agencies asserted for the incremental costs of a
PHEV over a strong hybrid vehicle are not supported in the
market.\1312\ CARB cited the Toyota Prius Prime and Hyundai Sonata as
examples of vehicles that share most of their components with their
non-plug-in hybrid counterparts, with components like the on-board
charger and higher voltage, larger energy capacity battery pack
excepted. CARB stated the agencies' lack of discussion about how non-
battery component costs were developed made it ``virtually impossible
to understand what the drivers are for the increases in costs relative
to the Agencies' previous analysis for the 2016 Draft TAR and EPA's
Proposed Determination.'' CARB concluded that the available PHEV market
offerings do not support the higher costs relative to the Draft TAR and
EPA's Proposed Determination analyses, and no justification was
provided for the change.
---------------------------------------------------------------------------
\1312\ California Air Resources Board, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
The agencies agree with CARB that the incremental costs of PHEV
over strong hybrid costs were too high, and that values were not
supported by the market. In response to this comment, the agencies
updated the non-battery component costs as well as the battery costs to
better reflect the market values. In addition, the agencies have
optimized the Autonomie modeling in a way to maintain the same engine,
transmission and other components from a SHEVP2 or SHEVPS moving to a
PHEV20/50 or PHEV20T/50T.\1313\ For further discussions on PHEV
modeling and updates, see Section VI.C.3.a)(1) Electrification
technologies and Section VI.C.3.d) Modeling and Simulating Vehicles
with Electrified Powertrains in Autonomie. The updates discussed here
and applied to the final analysis resulted in values that more
accurately represented PHEV technology costs.
---------------------------------------------------------------------------
\1313\ I.e., a SHEVP2 with a turbocharged engine may adopt
PHEV20T or PHEV50T technology, but a SHEVPS will only ever adopt
PHEV20 or PHEV50 technology, as the SHEVPS do not use turbocharged
engines.
---------------------------------------------------------------------------
(e) BEV Cost
For the NPRM and this final rule analysis, the total costs of BEVs
included optimized battery pack and electric machine costs. Like the
other electrified powertrains, Autonomie optimized both the size of the
battery pack and electric machine to fulfill the performance neutrality
requirements for each vehicle. Further discussion on electrification
technology component sizing and optimization is provided in Section
VI.C.3.d) Modeling and Simulating Vehicles with Electrified Powertrains
in Autonomie. Discussion on electrification component costing is
provided in Battery Pack Modeling and Non-battery Electrification
Component Costs. When computing the total cost of a vehicle, the
agencies remove the costs of the IC engines and transmission when a
conventional or hybridized powertrain adopts BEV technologies. In
Section VI.C.1 Engines Path and Section VI.C.22 Transmission, the
agencies discussed the absolute costs used for engine and transmission
technologies in the final rule analysis.
ICCT stated that if the agencies had considered BEV battery and
other component costs correctly, cost parity would be reached with
conventional combustion vehicles in the 2025-2027 timeframe.\1314\ ICCT
went on to allege that if the agencies removed all constraints on
electric vehicles,\1315\ they would appropriately realize that the 2025
standards are more cost-effective if electric vehicles are included.
---------------------------------------------------------------------------
\1314\ International Council on Clean Transportation, NHTSA-
2018-0067-11741.
\1315\ As discussed above, the agencies believe that ICCT
misunderstood the agencies' statutory obligations and the
differences between the standard setting modeling scenario and the
``real-world'' modeling scenario. The agencies did not apply
additional constraints on BEVs in the NPRM analysis.
---------------------------------------------------------------------------
The agencies disagree with ICCT's statement that BEVs would reach
parity to IC engines by the 2025-2027 timeframe. For this final rule
analysis, the agencies have updated the battery pack costs, electric
machine costs, and excluded costs of IC engines and transmission when a
vehicle was converted to a BEV. However, the costs still did not reach
parity within the rulemaking time frame. Furthermore, NHTSA notes that
the decision to exclude BEV technology from the CAFE program standard-
setting analysis is not a choice made by the agency, but a statutory
requirement.\1316\
---------------------------------------------------------------------------
\1316\ See 49 U.S.C. 32902(h).
---------------------------------------------------------------------------
(f) FCV Cost
For the NPRM and the final rule analysis the agencies considered
fuel cell vehicle technology advancements in hydrogen storage tanks,
sensors and control systems, and market penetration.\1317\ The agencies
are also considered the availability of hydrogen refueling stations
across the country and cost of compressed hydrogen.1318 1319
Although the agencies did not receive any comments on the cost of fuel
cell vehicles, the agencies updated the cost of hydrogen storage tanks
and fuel cells based on a cost analysis from Department of Energy
(DOE), Office of Energy Efficiency and Renewable Energy (EERE), Fuel
Cell Technologies Office.\1320\
---------------------------------------------------------------------------
\1317\ The agencies referenced EPA's 2018 Automotive Trends
Report, available at https://nepis.epa.gov/Exe/ZyPDF.cgi/P100W5C2.PDF?Dockey=P100W5C2.PDF, for information about FCV market
penetration.
\1318\ MIT Energy Initiative. Insights into Future Mobility
(2019). Cambridge, MA: MIT Energy Initiative. http://energy.mit.edu/insightsintofuturemobility.
\1319\ U.S. Department of Energy, Alternative Fuels Data Center:
Alternative Fueling Station Counts by State: https://afdc.energy.gov/stations/states (last visited January 3, 2020).
\1320\ James et al., Final Report: Hydrogen Storage System Cost
Analysis (September 2016), available at https://www.osti.gov/servlets/purl/1343975.
---------------------------------------------------------------------------
The DOE estimates that the cost of a compressed gas storage system
is around $28/kWh (assumed rate of production of 10,000 units per
year). The hydrogen fuel price ranges from $12.85 to $16 per kilogram,
which translates to approximately $5.60 per gallon on an equivalent
energy basis.\1321\
---------------------------------------------------------------------------
\1321\ California Fuel Cell Partnership: https://cafcp.org/content/cost-refill (last visited January 3, 2020).
---------------------------------------------------------------------------
Table VI-116 shows the evolution of the fuel cell vehicle costs
from the Draft TAR to final rule (costs include the fuel cell, control
systems, motors, inverters, hydrogen storage tanks, wiring harness,
hydrogen fuel sending lines, safety systems, sensors and hardware for
mounting and installation). The cost of the battery pack and battery
management system is not included in the cost of the fuel cell vehicle.
[GRAPHIC] [TIFF OMITTED] TR30AP20.249
[[Page 24525]]
4. Mass Reduction
Mass reduction is a relatively cost-effective means of improving
fuel economy and reducing CO2 emissions, and vehicle
manufacturers are expected to apply various mass reduction technologies
to meet fuel economy and CO2 standards. Reducing vehicle
mass can be accomplished through several different techniques, such as
modifying and optimizing vehicle component and system designs, part
consolidation, and adopting lighter weight materials (advanced high
strength steel, aluminum, magnesium, and plastics including carbon
fiber reinforced plastics). The cost for mass reduction depends on the
type and amount of materials used, the manufacturing and assembly
processes required, and the degree to which changes to plants and new
manufacturing and assembly equipment is needed. In addition,
manufacturers may develop expertise and invest in certain mass
reduction strategies that may affect the approaches for mass reduction
they consider and the associated costs. Manufacturers may also consider
vehicle attributes like noise-vibration-harshness (NVH), ride quality,
handling, and various acceleration metrics when considering how to
implement any mass reduction strategy. See Section VI.B.3.a)(5)
Maintaining Vehicle Attributes for more details.
The automotive industry uses different metrics to measure vehicle
weight. Some commonly used measurements are vehicle curb weight,\1322\
gross vehicle weight (GVW),\1323\ gross vehicle weight rating
(GVWR),\1324\ gross combined weight (GCVW),\1325\ and equivalent test
weight (ETW),\1326\ among others.
---------------------------------------------------------------------------
\1322\ This is the weight of the vehicle with all fluids and
components but without the drivers, passengers, and cargo.
\1323\ This weight includes all cargo, extra added equipment,
and passengers aboard.
\1324\ This is the maximum total weight of the vehicle,
passengers, and cargo to avoid damaging the vehicle or compromising
safety.
\1325\ This weight includes the vehicle and a trailer attached
to the vehicle, if used.
\1326\ For the EPA two-cycle regulatory test on a dynamometer,
an additional weight of 300 lbs. is added to the vehicle curb
weight. This additional 300 lbs. represents the weight of the
driver, passenger, and luggage. Depending on the final test weight
of the vehicle (vehicle curb weight plus 300 lbs.), a test weight
category is identified using the table published by EPA according to
40 CFR 1066.805. This test weight category is called ``Equivalent
Test Weight'' (ETW).
---------------------------------------------------------------------------
The vehicle curb weight is the most commonly used measurement when
comparing vehicles. A vehicle's curb weight is the weight of the
vehicle including fluids, but without a driver, passengers, and cargo.
A vehicle's glider weight, which is vehicle curb weight minus the
powertrain weight, is used to track the potential opportunities for
weight reduction not including the powertrain. A glider's subsystems
may consist of the vehicle body, chassis, interior, steering,
electrical accessory, brake, and wheels systems. However, as noted in
the PRIA, the definition of a glider may vary from study to study (or
even simulation to simulation).
Each of the subsystems presents an opportunity for weight
reduction; however, some weight reduction is dependent on the weight
reduction of other subsystems. The agencies characterize mass reduction
as either primary mass reduction or secondary mass reduction. Primary
mass reduction involves reducing mass of components that can occur
independent from the mass of other components. For example, reducing
the mass of a hood (e.g., replacing a steel hood with an aluminum hood)
or reducing the mass of a seat are examples of primary mass reduction
because each can be implemented independently. Other components and
systems that may contribute to primary mass reduction include the
vehicle body, chassis, and interior components.
When significant primary mass reduction occurs, other components
designed based on the mass of primary components may be redesigned as
well. An example of a subsystem where secondary mass reduction can be
applied is the brake system. If the mass of primary components is
reduced sufficiently, the resulting lighter weight vehicle could safely
maintain braking performance and attributes with a lighter weight brake
system. Other examples of components where secondary mass reduction can
be applied are wheels and tires.
For this analysis, the agencies consider mass reduction
opportunities from the glider subsystems of a vehicle first, and then
consider associated opportunities to downsize the powertrain, which are
accounted for separately.\1327\ As explained later, in the Autonomie
simulations, the glider system includes both primary and secondary
systems from which a percentage of mass is reduced for different glider
weight reduction levels; specifically, the glider includes the body,
chassis, interior, electrical accessories, steering, brakes and wheels.
The model sizes the powertrain based on the glider weight and the mass
of some of the powertrain components in an iterative process. The mass
of the powertrain depends on the powertrain size. Therefore, the weight
of the glider impacts the weight of the powertrain.\1328\ See Section
VI.B.3.a)(3) Vehicle models for Autonomie and Section VI.B.3.a)(4) How
Autonomie Sizes Powertrains for Full Vehicle Simulation for more
details.
---------------------------------------------------------------------------
\1327\ When the mass of the vehicle is reduced by an appropriate
amount, the engine may be downsized to maintain performance. See
Section VI.B.3.a)(5) Maintaining Vehicle Attributes] and Section
VI.B.3.a)(6) Performance Neutrality for more details.
\1328\ Since powertrains are sized based on the glider weight
for the analysis, glider weight reduction beyond a threshold amount
during a redesign will lead to re-sizing of the powertrain. For the
analysis, the glider was used as a base for the application of any
type of powertrain. A conventional powertrain consists of an engine,
transmission, exhaust system, fuel tank, radiator and associated
components. A hybrid powertrain also includes a battery pack,
electric motor(s), generator, high voltage wiring harness, high
voltage connectors, inverter, battery management system(s), battery
pack thermal system, and electric motor thermal system.
---------------------------------------------------------------------------
The agencies use glider weight to apply non-powertrain mass
reduction technology, and use Autonomie simulations to determine the
size of the powertrain and corresponding powertrain weight for the
respective glider weight. The combination of glider weight (after mass
reduction) and re-sized powertrain weight equal the vehicle curb
weight. See Section VI.C.4.d)(1) glider mass and mass reduction
subsection below for more detail on glider mass and glider mass
reduction.
(a) Mass Reduction in the CAFE Model
Several studies have explored the amount of vehicle mass reduction
that is feasible in the rulemaking timeframe and the cost for that mass
reduction.1329 1330 1331 1332 Those studies were sponsored
by the agencies, CARB, ICCT, the automotive industry, and material
manufacturers, and are discussed in Section VI.C.4.e)(1), below. All of
the studies showed that the maximum feasible amount of mass reduction
that can be applied in the rulemaking timeframe is around 20 percent of
a baseline vehicle's curb weight. The National Academies of Sciences
similarly concluded, based on some of these same studies along with
other information, that it is feasible to
[[Page 24526]]
reduce up to 20 percent of the mass of the vehicle.\1333\
---------------------------------------------------------------------------
\1329\ DOT HS 811 692: Investigation of Opportunities for
Lightweight Vehicles Using Advanced Plastics and Composites.
\1330\ A Review of the Safety of Reduced Weight Passenger Cars
and Light Duty Trucks by Michigan Manufacturing Technology Center,
October 2018.
\1331\ ATG Silverado Body Light weighting Study, Aluminum
Technology Group, January 2017.
\1332\ 2013 NanoSteel Intensive Body-In-White, EDAG and
NanoSteel Company Inc.
\1333\ Cost, Effectiveness and Deployment of Fuel Economy
Technologies for Light-Duty Vehicles, National Academy of Sciences,
2015, at 212 .
---------------------------------------------------------------------------
As discussed in Section VI.C.4.e), the mass reduction studies show
that the cost for mass reduction increases progressively as the amount
of mass reduction increases. In other words, lower levels of mass
reduction are more cost effective than higher levels of mass reduction.
As in past rulemakings, the agencies have considered multiple levels of
mass reduction to provide options similar to what manufacturers could
consider at vehicle redesigns.
For the NPRM, the agencies included five levels of mass reduction
with a maximum of 20 percent glider mass reduction, corresponding to 10
percent curb mass reduction, using the assumption that the glider was
50 percent of curb weight. Table VI-117 shows the glider and curb
weight mass reduction levels for each level of mass reduction
considered in the NPRM analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.250
The agencies received a number of comments suggesting that the
amount of mass reduction allowed should be 20 percent of curb weight,
as well as suggestions that the agencies should assume the glider
represents 75 percent of the vehicle's curb weight. These comments are
addressed in more detail in Section VI.C.4.d) below, but some
understanding of how the glider share assumption affects the maximum
amount of mass reduction allowed in the CAFE model is required here.
Several commenters stated that the agencies should allow further
levels of mass reduction technology improvements in the CAFE model. For
example, ICCT commented that the agencies must revise their treatment
of mass reduction because studies have demonstrated that at least 20%
mass reduction of curb weight is available for adoption across vehicle
classes by 2025. \1334\ ICCT stated that based on these studies, the
agencies must increase the maximum available mass reduction potential
levels to include up to 20% and 25% mass reduction of curb weight, as
the industry ``will cost-effectively deploy at least 15% vehicle curb
mass reduction in the 2025 timeframe at net zero cost.'' ICCT caveated
that amount of mass reduction seems less likely in smaller cars, which
typically employ lower levels of mass reduction, so a constraint of 7.5
percent mass reduction as was applied in the Draft TAR would be
appropriate for those vehicles.
---------------------------------------------------------------------------
\1334\ NHTSA-2018-0067-11741. ICCT also alleged that the
agencies intentionally disregarded the studies that presented this
result; those comments are discussed in Section VI.C.4.e) Mass
Reduction Costs, below.
---------------------------------------------------------------------------
ICCT also commented that there were numerous material improvements
in development that were not considered in the rule, including but not
limited to higher strength aluminum, improved joining techniques for
mixed materials, third-generation steels with higher strength and
enhanced ductility, a new generation of ultra-high strength steel cast
components, and metal/plastic hybrid components, among other
technologies mentioned in ICCT's working paper on light-weighting.
In assessing these comments, the agencies reconsidered the mass
reduction studies and available reports and agreed that additional
levels of mass reduction should be available for the final rule
analysis. In response to comments, the agencies made two adjustments to
allow higher levels of mass reduction in the analysis. First, as
explained in Section VI.C.4.d)(1), below, the agencies increased the
glider percentage of vehicle curb weight used for the analysis from 50
percent to 71 percent. As explained in that section, increasing the
glider percentage also increases the amount of curb weight reduction
for all levels of mass reduction. Second, the agencies created another
level of mass reduction (MR6) in the CAFE model, which represents a
significant application of carbon fiber in the vehicle to achieve
nearly 30 percent reduction in glider weight (which approximately
translates to 20 percent reduction in vehicle curb weight). For
example, incorporating a carbon fiber tub,\1335\ or a carbon fiber
monocoque with aluminum sub frame in the front and back,\1336\ or a
carbon fiber splitter and carbon fiber wheels,\1337\ allows for greater
levels of mass reduction, albeit at a very high cost. These
technologies are not ready for high volume production vehicles.
---------------------------------------------------------------------------
\1335\ The BMW i3 and BMW i8, which are about 20 percent lighter
than an average MY 2017 vehicle, use a carbon fiber tub.
\1336\ The Alfa Romeo 4c/4c Spider, which is about 20 percent
lighter than an average MY 2017 vehicle, uses this design.
\1337\ The Ford Shelby GT350R which is about 20 percent lighter
than an average MY 2017 vehicle, uses this design.
---------------------------------------------------------------------------
Table VI-118 shows the levels of mass reduction technology
available for application in the final rule analysis, with the
associated glider weight percentage reduction and the percentage curb
weight reductions for passenger cars and light trucks. As discussed in
Section VI.C.4.c) below, the agencies declined to place a constraint on
the amount of mass reduction technology that smaller cars could adopt.
[[Page 24527]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.251
The agencies continue to believe the maximum feasible mass
reduction levels identified in comprehensive design studies, such as
those discussed in Section VI.C.4 Mass Reduction Costs are the most
reliable for projecting the maximum amount of mass reduction in the
rulemaking timeframe, and therefore have determined MR6 is the highest
level that should be used for the final rule analysis. While the
information provided by ICCT on newer materials and manufacturing and
assembly methodology is interesting and relevant, this information, by
itself, is insufficient to assess the amount of mass reduction that is
feasible and the cost for the mass reduction. ICCT did not provide a
comprehensive analysis showing a design concept that maintains vehicle
attributes and performance, such as noise, vibration and harshness,
stiffness, handling, compliance with NHTSA safety standards, good
performance under NHTSA NCAP and IIHS rating systems, and other
criteria. The various studies in Section VI.C.4.e) Mass Reduction
considered those factors to varying degrees. Without that rigorous
analysis, the actual amount of mass reduction that could be enabled
through the use of those materials and methods described by ICCT, and
the cost of achieving that mass reduction, would be highly speculative.
As explained in Section VI.C.4.e) Mass Reduction below, the agencies
determined the NHTSA-sponsored design studies remain a reasonable basis
for estimating a feasible amount of mass reduction and the cost for
mass reduction in the rulemaking timeframe, because those studies
considered a wide range of materials (including advanced materials) and
design solutions.
(b) Analysis Fleet Mass Reduction Assignments
The agencies included an estimated level of mass reduction
technology for each vehicle model in the MY 2016 analysis fleet for the
NPRM, and have updated the estimates for the MY 2017 analysis fleet for
the final rule analysis. The methodology used to provide each vehicle
model an appropriate initial mass reduction technology level for
further improvements was described in detail in the Draft TAR (when
NHTSA first employed this methodology), in the PRIA accompanying the
NPRM, and is reproduced here, in part, to provide additional context to
the agencies' responses to comments on analysis fleet mass reduction
assignments. The methodology used in this final rule was unchanged from
the NPRM.
For the Draft TAR, NHTSA/Volpe Center staff developed regression
models to estimate curb weights based on other observable attributes.
With regression outputs in hand, Volpe evaluated the distribution of
vehicles in the analysis fleet. In addition, vehicle platforms were
evaluated based on the sales-weighted residual of actual vehicle curb
weights versus predicted vehicle curb weights. Based on the actual curb
weights relative to predicted curb weights, platforms (and the
subsequent vehicles) were assigned a baseline mass reduction level (MR0
through MR6). For the NPRM and final rule analysis, the agencies
followed a similar procedure for the MY 2016 and MY 2017 analysis
fleets.
To develop the curb weight regressions, the agencies grouped
vehicles into three separate body design categories for analysis: 3-
Box, 2-Box, and Pick-up.
[GRAPHIC] [TIFF OMITTED] TR30AP20.252
For the NPRM and final rule analysis, the agencies retained the MY
2015 regressions for 3-Box and 2-Box vehicles, however the pickup
category regression was updated in response to comments on the Draft
TAR. The
[[Page 24528]]
agencies trained a new regression with EPA MY 2014 data and added pick-
up bed length as an independent variable. As a result of stepping back
to MY 2014 data for the pick-up regression, the training data did not
include the all-aluminum body Ford F-150 in the calculation of the
baseline. The advanced F-150 in the MY 2015 pick-up regression
meaningfully affected Draft TAR regression statistics because the F-150
accounted for a large portion of observations in the analysis fleet,
and the F-150 included advanced weight savings technology.
The agencies leveraged many documented variables in the analysis
fleet as independent variables in the regressions. Continuous
independent variables included footprint (wheelbase x track width) and
powertrain peak power. Binary independent variables included strong HEV
(yes or no), PHEV (yes or no), BEV or FCV (yes or no), all-wheel drive
(yes or no), rear-wheel drive (yes or no), and convertible (yes or no).
In addition, for PHEV and BEV/FCV vehicles, the capacity of the battery
pack was included in the regression as a continuous independent
variable. In some body design categories, the analysis fleet did not
cover the full spectrum of independent variables. For instance, in the
pickup body style regression, there were no front-wheel drive vehicles
in the analysis fleet, so the regression defaulted to all-wheel drive
and left an independent variable for rear-wheel drive.
Furthermore, the agencies evaluated alternative regression
variables in response to comments from vehicle manufacturers on the
NHTSA/Volpe analysis in the Draft TAR.\1338\ The agencies evaluated
regressions including overall dimensions of vehicles, such as height,
width, and length, instead of and in addition to just wheelbase and
track width. The experimental regression variables only marginally
changed predicted curb weight residuals as a percentage of predicted
curb weight, at an industry level and for most manufacturers. The
results were not significantly different, and therefore the agencies
opted not to add these variables to regressions or replace independent
variables presented in Draft TAR with new variables.
---------------------------------------------------------------------------
\1338\ PRIA at 407.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.253
[[Page 24529]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.254
[[Page 24530]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.255
BILLING CODE 4910-59-C
Each of the three regressions produced outputs effective for
identifying vehicles with a significant amount of mass reduction
technology in the analysis fleet. Many coefficients for independent
variables provided clear insight into the average weight penalty for
the utility feature. In some cases, like battery size, the relatively
small sub-sample size and high collinearity with other variables
confounded coefficients.
By design, no independent variable directly accounted for the
degree of weight savings technology applied to the vehicle. Residuals
of the regression captured weight reduction efforts and noise from
other sources.
The agencies received many comments on the Draft TAR encouraging
the use of observed technologies in each vehicle, and in each vehicle
subsystem to assign levels of mass reduction technology. As a practical
matter, the agencies cannot conduct a tear down study and detailed cost
assessment for every vehicle in every model year. However, upon review
of many vehicles and their subsystems, the agencies recognized a few
vehicles with MR0 or MR1 assignments in NHTSA's analysis of the Draft
TAR that contained some advanced weight savings technologies, yet these
vehicles and their platforms still produced ordinary residuals.
Engineers from industry confirmed important factors other than glider
weight savings and the independent variables considered in the
regressions may factor into the use of lightweight technologies. Such
factors included the desire to lower the center of gravity of a
vehicle, improve the vehicle weight distribution for handling, optimize
noise-vibration-and-harshness, increase torsional rigidity of the
platform, offset increased vehicle content, and many other factors. In
addition, engineers highlighted the importance of sizing shared
components for the most demanding applications on the vehicle platform;
optimum weight savings for one platform application may not be suitable
for all platform applications. For future analysis, the agencies will
look for practical ways to improve the assessment of mass reduction
content and the forecast of incremental mass reduction costs for each
vehicle.
Figure VI-44 below shows results from the pickup truck regression
on predicted curb weight versus actual curb weight. Points above the
solid regression line represent vehicles heavier than predicted (with
lower mass reduction technology levels); points below the solid
regression line represent vehicles lighter than predicted (with higher
mass reduction technology levels). The dashed lines in the Figure VI-44
show the thresholds (5, 7.5, 10, 15, 20 and 28 percent of glider
weight). Final rule glider weight assumption is 71 percent of vehicle
curb weight.
BILLING CODE 4910-59-P
[[Page 24531]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.256
BILLING CODE 4910-59-C
For points with actual curb weight below the predicted curb weight,
the agencies used the residual as a percent of predicted weight to get
a sense for the level of current mass reduction technology used in the
vehicle. Notably, vehicles approaching -20% curb weight widely use
advanced composites throughout major vehicle systems, and few examples
exist in the MY 2016 fleet.\1339\
---------------------------------------------------------------------------
\1339\ This evidence suggests that achieving a 20% curb weight
reduction for a production vehicle with a baseline defined with this
methodology is extremely challenging, and requires very advanced
materials and disciplined design.
---------------------------------------------------------------------------
Generally, residuals of regressions as a percent of predicted
weight appropriately stratified vehicles by mass reduction level. Most
vehicles showed near zero residuals or had actual curb weights close to
the predicted curb weight. Few vehicles in the analysis fleet were
identified with the highest levels of mass reduction. Most vehicles
with the largest negative residuals have demonstrably adopted advanced
weight savings technologies at the most expensive end of the cost
curve.
To validate the residuals, the agencies estimated the mass
reduction technology level for several vehicle models in the analysis
fleet and compared those estimates to the numerical results from the
regression analysis. To estimate the mass reduction technology level
for the selected vehicles, the agencies conducted an in-depth review of
available information on the materials, design, and last redesign year
for those vehicle models, and compared that information with the
designs and materials used in the mass reduction feasibility and cost
studies summarized in Section VI.C.4.e), below.
[[Page 24532]]
That comparison showed good agreement with the technology levels from
the regression analysis.
The agencies believe the regression methodology is a technically
sound methodology for estimating mass reduction levels in the analysis
fleet.
As part of their comments stating the NPRM modeling reflected
reality better than the Draft TAR and Proposed Determination analyses,
Toyota commented broadly that the MY 2016 baseline fleet used in the
NPRM encompassed powertrain and tractive energy (including mass
reduction) improvements more representative of vehicles on the road
today.\1340\ Toyota noted that the 2016 baseline fleet generally
contained higher levels of technology compared to the MY 2014 and MY
2015 baseline fleets, and included a comparison of its initial fleet
mass reduction assignments in the Draft TAR and the NPRM. Toyota showed
how moving further up the technology tree (e.g., starting with a
baseline that includes higher levels of technology) for certain
pathways such as mass reduction increased costs exponentially. Toyota
stated that the NPRM underestimated mass reduction cost values.
---------------------------------------------------------------------------
\1340\ NHTSA-2018-0067-12098.
---------------------------------------------------------------------------
While a more specific discussion of costs is located in Section
VI.C.4.e), the agencies agree with Toyota's assessment that the costs
for mass reduction technology increase exponentially as progressively
higher levels of mass reduction are incorporated. Having an accurate
assessment of baseline technology levels ensures that the subsequent
application of technology and its associated costs is correctly
accounted for.
C.A.R produced a report in response to the Draft TAR that generally
agreed with the regression methodology of using observed vehicle
attributes for estimating mass reduction levels, as opposed to
comparing vehicle curb weight from a newer model year to a previous
generation of the same vehicle, pointing to several of the limitations
discussed above.\1341\
---------------------------------------------------------------------------
\1341\ EPA Mass Reduction Analysis--Observations and
Recommendations, Center for Automotive Research, October 2017 (page
15), available at https://www.cargroup.org/wp-content/uploads/2017/10/EPA-MR-Analysis-Critique_Oct-5_final.pdf.
---------------------------------------------------------------------------
Both ICCT and H-D Systems commented on the methodology for
identifying mass reduction technology levels in the analysis fleet,
with ICCT broadly stating that by placing additional mass reduction
technology in the baseline, the agencies artificially removed ``the
most cost-effective lightweighting from future use, which incorrectly
increases the costs of all subsequent mass-reduction in the compliance
modeling.'' \1342\
---------------------------------------------------------------------------
\1342\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
ICCT claimed that the agencies unjustifiably increased the amount
of vehicle mass reduction technology present in the 2016 baseline fleet
from the 2015 baseline used in the Draft TAR, stating that the 2015
Draft TAR fleet had 26 percent of vehicles sold with some level of mass
reduction applied (MR1 or a higher level), whereas the 2016 NPRM fleet
had 47 percent of vehicles sold with some level of mass reduction
applied. In addition to faulting the agencies for not acknowledging the
change and not attempting to justify it, ICCT stated that the 2016
analysis fleet mass reduction assignments were overstated, as ``it
appears that the agencies have applied mass reduction technology to
vehicles in the model that did not have mass reduction applied in the
real world.'' ICCT stated that the effect of this change was to
``render[] unavailable mass reduction technologies for these vehicles
in the model,'' causing the model to select less cost-effective
technologies instead and driving the modeled compliance costs higher.
ICCT argued that to substantiate the changes made to the baseline
fleet mass reduction assignments, the agencies must show data on how
these improvements are evident in the fleet and to quantify and include
their realized benefits in the analysis, including a detailed and
justified explanation of all mass reduction technologies deemed already
to have been applied to the MY 2016 analysis fleet. More specifically,
ICCT stated that the agencies ``must clearly and precisely share their
estimated percent (and absolute pounds) mass reduction amount for each
vehicle make and model in the baseline fleet (rather than simply
showing binned categories), and their technical justification for each
value,'' and ``[t]o not do so obscures the agencies' new methods and
data sources from public view, rendering their lightweighting
calculations a black box.''
In addition, ICCT recommended that the agencies conduct two
sensitivity analyses, one assuming that every baseline make and model
has not yet applied any lightweighting (setting the baseline to 0% mass
reduction), and one assuming that each vehicle model has applied Draft
TAR baseline mass reduction assignments, to demonstrate how much the
agencies' decision to load up more baseline technology affects the
compliance scenarios.
ICCT concluded that because the changes in baseline mass reduction
assignments from prior analyses to the NPRM ``are opaquely buried in
the agencies' datafiles and unexplained, we believe the agencies have
to reissue a new regulatory analysis and allow an additional comment
period for review of their methods and analysis.''
To address ICCT's comment, it is important to understand the mass
reduction baseline technology assignment methodology previously used by
EPA in the Draft TAR and Proposed Determination.\1343\ As stated in the
Draft TAR, the curb weight of each vehicle model in the MY 2008
analysis fleet (used for the 2012 rulemaking to establish MYs 2017-2025
standards) was assumed to be at a baseline MR0 level. The mass
reduction technology level in the MY 2014 analysis fleet was determined
by comparing the curb weight of the MY 2014 vehicle to the most similar
vehicle in the MY 2008 analysis fleet.\1344\ The curb weight of the
newer model year vehicle was adjusted to account for changes in the
vehicle footprint and changes in mass due to added safety technology.
If a vehicle did not have a previous generation vehicle, then the sales
weighted average percent mass reduction over the manufacturer's name
plate product line was used to represent the expectation of mass
reduction technology available within the vehicle.
---------------------------------------------------------------------------
\1343\ Draft TAR at 5-395.
\1344\ Draft TAR at 5-395.
---------------------------------------------------------------------------
EPA listed some limitations to this methodology in the Draft
TAR,\1345\ and others are also addressed here. First, assuming that
every vehicle started with MR0 technology did not account for the
actual varying levels of mass reduction technology that existed in the
MY 2008 fleet. Second, for each vehicle model, there was no accounting
for the mass associated with different powertrain configurations. This
was particularly problematic because the method did not account for
light weight technology already available in the vehicle structure to
counter the increased mass associated with more advanced powertrains,
such as HEV, PHEV, and EV technologies.\1346\ Third, there was no
sales-weight accounting for the various configurations in estimating
the vehicle model mass reduction technology level, meaning that if a
high-sales-volume vehicle employed significant mass reduction
technology, that vehicle was not credited as such in the analysis
[[Page 24533]]
fleet. Fourth, there was no accounting for mass increases due to the
addition of future regulatory requirements like potential safety
regulations. Fifth, there was no accounting for mass associated with
changes in vehicle attributes and utility, such as the addition of
infotainment systems and crash avoidance technologies. These
limitations all individually had the effect of overestimating mass
reduction technology effectiveness and undercounting mass reduction
technology costs across the fleet, and accordingly their combined
effect was significant. The lack of controls for these items introduced
errors into the mass reduction technology level effectiveness
estimates.
---------------------------------------------------------------------------
\1345\ Draft TAR at 5-395.
\1346\ PRIA at 413.
---------------------------------------------------------------------------
After considering the comments, the agencies determined the use of
the regression method, based on observable attributes, is the best
available methodology to provide a reasonable estimate of mass
reduction technology for the analysis fleet. The agencies believe that,
contrary to ICCT's assertion, the regression methodology used in the
NHTSA Draft TAR, NPRM, and final rule analyses provides a more
transparent method for calculating baseline mass reduction technology
assignments. The methodology was fully explained in the Draft TAR and
PRIA, and avoided the limitations identified by EPA by using data from
the analysis fleet, and not requiring the use of or assumptions about
the exact mass reduction levels of vehicles in a prior model year
fleet. In addition, the regression accounted for differences in
powertrains between trim levels, including non-ICE powertrains by
accounting for these factors in the regression analysis.
Also, because manufacturers generally apply mass reduction
technology at a vehicle platform level (i.e., using the same components
across multiple vehicle models that share a common platform) to
leverage economies of scale and to manage component and manufacturing
complexity, conducting the regression analysis at the platform level
leads to more accurate estimates for the real-world vehicle platform
mass reduction levels. The platform approach also addresses the impact
of potential weight variations that might exist for specific vehicle
models, as all of the individual vehicle models are aggregated into the
platform group, and are effectively averaged using sales weighting,
which minimizes the impact of any outlier vehicle configurations.
The agencies also disagree that the changes in baseline mass
reduction assignments were unexplained. The PRIA discussed reasons that
baseline mass reduction assignments differed from prior analyses,
including that, ``[s]ince the Draft TAR, many platforms have not been
redesigned, but in some cases the sales-weighted residuals for
carryover platforms have moved. In the case of 2-Box and 3-Box
vehicles, the analysis attributes such changes to differences in sales
mix year-over-year and other updates to reported curb weights and
platform designations. In the case of platforms with pick-up trucks,
the analysis updated the pick-up regression since the Draft TAR, so
that may be a contributing factor.'' \1347\
---------------------------------------------------------------------------
\1347\ PRIA at 424.
---------------------------------------------------------------------------
To the extent that the NPRM glider weight assumption impacted the
NPRM MY 2016 analysis fleet baseline mass reduction assignment values,
the agencies presented a table in the PRIA showing how different glider
weight assumptions impacted mass reduction technology levels for the
analysis fleet.\1348\ The following Table VI-123 recreates that table
in part, with updates based on the glider weight values used for the
final rule.
---------------------------------------------------------------------------
\1348\ PRIA at 422.
---------------------------------------------------------------------------
For example, from the regression analysis, the Ford F-150 has a
predicted curb weight (residual) of 12.4 percent of the actual curb
weight. If the glider weight assumption is 50 percent of the vehicle
curb weight (like in NPRM), then the agencies would assign MR5 as an
initial mass reduction assignment in the analysis fleet. With this high
level of mass reduction technology already applied, the opportunity for
further mass reduction would be limited. However, if the glider weight
is assumed to be 71 percent of the vehicle curb weight, then Ford F-150
would be assigned MR4, and would have an opportunity to apply another
level of mass reduction albeit at higher cost.
[[Page 24534]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.257
The agencies also disagree that the amount of vehicle mass
reduction technology present in the 2016 baseline fleet was
``unjustifiably increased'' from the 2015 baseline used in the Draft
TAR. Table VI-124 shows the percent mass reduction technology used in
Draft TAR, NPRM, and in final rule. It is clear from the table below
that total percentage of MY 2016 vehicle fleet used in the NPRM had
nearly the same level of some mass reduction technology applied
compared to the Draft TAR. Similar to ICCT's observations, 28 percent
of the MY 2015 vehicle fleet used in the Draft TAR had some level of
mass reduction technology (MR1 to MR5) and 26 percent of MY 2016
vehicle fleet had some mass reduction technology applied. Since the
agencies assumed a reduced glider share in the NPRM, the percentage of
vehicles assigned a MR4 or MR5 technology level increased compared to
Draft TAR. In addition, for this final rule, the agencies observed that
many of the vehicles in the MY 2017 fleet had been redesigned, which
provided the opportunity to incorporate additional mass reduction
technologies.
[GRAPHIC] [TIFF OMITTED] TR30AP20.258
The agencies considered a sensitivity case that assumed no mass
reduction, rolling resistance, or aerodynamic improvements had been
made to the MY 2017 fleet (i.e., setting all vehicle road levels to
zero--MRO, AERO and
[[Page 24535]]
ROLL0), in response to ICCT's comment. While this is an unrealistic
characterization of the initial fleet, the agencies conducted a
sensitivity analysis to understand any affect it may have on technology
penetration along other paths (e.g., engine and hybrid technology).
Under the CAFE program, the sensitivity analysis shows a slight
decrease in reliance on engine technologies (HCR engines, turbocharge
engines, and engines utilizing cylinder deactivation) and hybridization
(strong hybrids and plug-in hybrids) in the baseline (relative to the
central analysis). The consequence of this shift to reliance on lower-
level road load technologies is a reduction in compliance cost in the
baseline of about $300 per vehicle (in MY 2026). As a result, cost
savings in the preferred alternative are reduced by about $200 per
vehicle. Under the CO2 program, the general trend in
technology shift is less dramatic (though the change in BEVs is larger)
than the CAFE results. The cost change is also comparable, but slightly
smaller ($200 per vehicle in the baseline) than the CAFE program
results. Cost savings under the preferred alternative are further
reduced by about $100. With the lower technology costs in all cases,
the consumer payback periods decreased as well. These results are
consistent with the approach taken by manufacturers who have already
deployed many of the low-level road load reduction opportunities to
improve fuel economy.
Second, as discussed above, EPA's Draft TAR baseline mass reduction
assignments had identified limitations that the regression methodology
has addressed. Moreover, as discussed above, the regression methodology
was updated from the Draft TAR to characterize data better on pickup
trucks. The agencies do not believe that conducting sensitivity
analyses using these outdated or limited assumptions would be useful
for this final rule.
More narrowly, HDS commented that while the regression coefficients
between 2-box and 3-box vehicles for footprint seemed consistent, the
regression coefficients for horsepower between the 2-box and 3-box
vehicles seemed incorrect because both types of vehicles use similar
engines.\1349\ HDS stated that ``[c]ollinearity between footprint and
HP or other effects caused by having electric vehicles (with electric
motor HP ratings) in the regression data is the probable cause of these
inconsistent coefficients for HP, but this cannot be confirmed without
access to the same database used by NHTSA.'' HDS concluded that
``[r]evisions to the regression could have a significant effect on the
baseline assignment of vehicles, as the current assignment for vehicles
like the 2016 Mazda MX5 as having the highest level of weight reduction
technology (MR5) and the 2016 Chevy Malibu as having MR4 technology
appear incorrect as their curb weights are comparable to other similar
MY 2016 vehicles in their respective class.''
---------------------------------------------------------------------------
\1349\ H-D Systems, NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
While many of the vehicles share same the same powertrain for
passenger cars and SUVs or for cars and pickup trucks, the utility and
functionality of the vehicle in SUVs and pickup trucks (2-box) is
different than passenger cars (3-box). The presence of additional
structure for towing or higher capacity towing, rear cross member,
higher capacity suspension, and other differences, enable SUVs and
pickup trucks to have towing and heavier payload capability. For
example, Ford uses the nearly similar displacement and horsepower
engines in Mustang Ecoboost Coupe and in F150 2WD XL, Regular Cab, Long
Box. However, the curb weight for the pickup truck is higher than the
Mustang. Directionally, this supports that the 2-box weight per
horsepower coefficient should be greater than the 3-box coefficient,
just as it is in the for the regression. The coefficient for passenger
cars and SUVs has not changed since the Draft TAR (based on MY2015
vehicle fleet). Based on the comments to Draft TAR, for the NPRM, a new
set of coefficients were generated for pickups using the MY 2014
vehicle fleet. This was done so that coefficients were not skewed due
to presence of the aluminum intensive Ford F150 pickup truck. Hence,
the agencies believe the coefficients used in the regression analysis
are directionally correct and disagree with HDS's assertion. The
agencies further note that HDS did not suggest any alternate
methodology or specific coefficients to use in the regression analysis.
(c) Mass Reduction Technology Adoption Features
The agencies described in the NPRM that given the degree of
commonality among the vehicle models built on a single platform,
manufacturers do not have complete freedom to apply unique technologies
to each vehicle that shares the platform: while some technologies
(e.g., low rolling resistance tires) are very nearly ``bolt-on''
technologies, others involve substantial changes to the structure and
design of the vehicle, and therefore often necessarily affect all of
the vehicle models that share that platform. In most cases, mass
reduction technologies are applied to platform level components and
therefore the same design and components are used on all of the vehicle
models that share the platform.
As discussed in Section Analysis Fleet, above, each vehicle in the
analysis fleet is associated with a specific platform. Similar to the
application of engine and transmission technologies, the CAFE model
defines a platform ``leader'' as the vehicle variant of a given
platform that has the highest level of observed mass reduction present
in the analysis fleet. If there is a tie, the CAFE model begins mass
reduction technology on the vehicle with the highest sales in model
year 2017. If there remains a tie, the model begins by choosing the
vehicle with the highest Manufacturer Suggested Retail Price (MSRP) in
MY 2017. As the model applies technologies, it effectively levels up
all variants on a platform to the highest level of mass reduction
technology on the platform. So, if the platform leader is already at
MR3 in MY 2017, and a ``follower'' starts at MR0 in MY 2017, the
follower will get MR3 at its next redesign (unless the leader is
redesigned again before that time, and further increases the mass
reduction level associated with that platform, then the follower would
receive the new mass reduction level).
Important for analysis fleet mass reduction assignments (discussed
above), and for understanding adoption features as well, is the
agencies' handling of vehicles that traditionally operated on the same
platform but had a mix of old and new platforms in production when the
analysis fleet was created. As described in the PRIA, the Honda Civic
and Honda CR-V traditionally share the same platform. In MY 2016, Honda
redesigned the Civic and updated the platform to include many mass
reduction technologies. Also in MY 2016, Honda continued to build the
CR-V on the previous generation platform--a platform that did not
include many of the mass reduction technologies on the all new MY 2016
Civic. In MY 2017, Honda launched the new CR-V that incorporated
changes to the Civic platform, and the Civic and CR-V again shared the
same platform with common mass reduction technologies. The NPRM and
final rule analyses treat the old and new platforms separately to
assign technology levels in the baseline, and the CAFE model brings
vehicles on the old platform up to the level of mass reduction
technology on the new shared platform at the first available redesign
year.
Furthermore, as stated in the NPRM and PRIA, unlike the analysis
presented in the Draft TAR that restricted high
[[Page 24536]]
levels of mass reduction for cars to show a safety neutral pathway to
compliance, the NPRM analysis did not artificially restrict mass
reduction to achieve a safety neutral outcome.\1350\ The NPRM CAFE
model considered MR0 through MR5 for all vehicles at redesign, and
similarly for the final rule, the CAFE model considers MR0 through MR6
for all vehicles at redesign.
---------------------------------------------------------------------------
\1350\ PRIA at 494.
---------------------------------------------------------------------------
Ford commented in support of the removal of ``previously applied
modeling rules that disallowed the mass reduction technology pathway
for certain vehicle classes since this restriction was not supported by
an adequate technical justification.'' \1351\ ICCT commented that a
constraint of 7.5 percent mass reduction to smaller cars, as was
applied in the Draft TAR, would be appropriate for those vehicles.
---------------------------------------------------------------------------
\1351\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
The agencies considered ICCT's comment that mass reduction on small
passenger cars should be limited to 7.5 percent, and Ford's comment
supporting the removal of ``previously applied modeling rules that
disallowed the mass reduction technology pathway for certain vehicle
classes.'' Neither CAFE standards nor this analysis mandate mass
reduction, or mandate that mass reduction occur in any specific manner.
The mass reduction cost subsection below shows mass reduction is a
cost-effective technology for improving fuel economy and CO2
emissions. The steel, aluminum, plastics, composite, and other material
industries are developing new materials and manufacturing equipment and
facilities to produce those materials. In addition, suppliers and
manufacturers are optimizing designs to maintain or improve functional
performance with lower mass. Manufacturers have stated that they will
continue to reduce vehicle mass to meet more stringent standards, and
therefore, this expectation is incorporated into the modeling analysis
supporting the standards to: (1) Determine capabilities of
manufacturers; and (2) predict costs and fuel consumption effects of
CAFE standards. The CAFE and CO2 rulemakings in 2012, and
the Draft TAR and EPA Proposed Determination, imposed an artificial
constraint that limited vehicle mass reduction in some small vehicles
to achieve a desired safety-neutral outcome. For the current
rulemaking, this artificial constraint is eliminated so the analysis
reflects manufacturers' applying the most cost effective technologies
to achieve compliance with the regulatory alternatives and the final
standards; this approach allows mass reduction to be applied across the
fleet. This approach is consistent with industry trends. To the extent
that mass reduction is only cost-effective for the heaviest vehicles,
the CAFE model would create the outcome predicted by commenters. In
reality, however, mass reduction is a cost-effective means of improving
fuel economy and does take place across vehicles of all sizes and
weights. Accordingly, the model reflects that manufacturers may reduce
vehicle mass--regardless of vehicle class--when doing so is cost
effective.
The agencies have included one additional mass reduction level for
the final rule in response to comments by ICCT and others, and to
account for carbon fiber use in vehicles. For the NPRM, the maximum
level of mass reduction was limited to 10 percent of a vehicle's curb
weight, and that amount of mass reduction could be applied during the
rulemaking timeframe. For the final rule, based on the current state of
mass reduction technology and the application rate of different levels
of mass reduction technologies, the agencies applied phase-in caps for
MR5 and MR6 (15 percent and 20 percent reduction of a vehicle's curb
weight, respectively). The agencies applied a phase-in cap for MR5
level technology so that 15 percent of the vehicle fleet starting in
2016 employed the technology, and the technology could be applied to
100 percent of the fleet by MY 2022. This cap is consistent with the
NHTSA lightweighting study that found that a 15 percent curb weight
reduction for the fleet is possible within the rulemaking
timeframe.\1352\ The agencies also applied a phase in cap for MR6
technology so that one percent of the vehicle fleet starting in MY2016
employed the technology, and the technology could be applied to 13
percent of the fleet by MY2025. The agencies believe that this phase-in
cap appropriately functions as a proxy for the cost and complexity
currently required (and that likely will continue to be required until
manufacturing process evolve) to produce carbon fiber components.
Again, MR6 technology in this analysis reflects the use of a
significant share of carbon fiber content, as seen through the BMW i3
and Alfa Romeo 4c as discussed above.
---------------------------------------------------------------------------
\1352\ DOT HS 811 666: Mass Reduction for Light Duty Vehicles
for Model Years 2017-2025: Figure 397 at page 356.
---------------------------------------------------------------------------
(d) Mass Reduction Technology Effectiveness
As discussed in Section VI.B.3, Argonne developed a database of
vehicle attributes and characteristics for each vehicle technology
class that included over 100 different attributes like frontal area,
drag coefficient, fuel tank weight, transmission housing weight,
transmission clutch weight, hybrid vehicle component weights, and
weights for components that comprise engines and electric machines,
tire rolling resistance, transmission gear ratios, and final drive
ratio. Argonne used these attributes to ``build'' each vehicle that it
used for the effectiveness modeling and simulation. Important for
precisely estimating the effectiveness of different levels of mass
reduction is an accurate list of initial component weights that make up
each vehicle subsystem, from which Autonomie considered potential mass
reduction opportunities.
As stated above, glider weight, or the vehicle curb weight minus
the powertrain weight, is used to determine the potential opportunities
for weight reduction irrespective of the type of powertrain.\1353\ This
is because weight reduction can vary depending on the type of
powertrain. For example, an 8-speed transmission may weigh more than a
6-speed transmission, and a basic engine without variable valve timing
may weigh more than an advanced engine with variable valve timing.
Autonomie simulations account for the weight of the powertrain system
inherently as part of the analysis, and the powertrain mass accounting
is separate from the application and accounting for mass reduction
technology levels (MR0-MR6) that are applied to the glider in the
simulations. Similarly, Autonomie also accounts for battery and motor
mass used in hybrid and electric vehicles separately. This secondary
mass reduction is discussed further, below.
---------------------------------------------------------------------------
\1353\ Depending on the powertrain combination, the total curb
weight of the vehicle includes glider, engine, transmission and/or
battery pack and motor(s).
---------------------------------------------------------------------------
Accordingly, in the Autonomie simulation, mass reduction technology
is simulated as a percentage of mass removed from the specific
subsystems that make up the glider, as defined for that set of
simulations (including the non-powertrain secondary mass systems such
as the brake system).
(1) Glider Mass and Mass Reduction
Autonomie accounts for the mass of each subsystem that comprises
the glider. For the NPRM, the glider subsystems included the vehicle
body and the chassis, but did not include mass from subsystems such as
the interior system, brake system, electrical accessory system, and
steering and
[[Page 24537]]
wheel systems. The agencies described in the PRIA that based on
advances in active and passive safety technologies that add some mass
to the interior system, certain subsystems were not considered for
potential light-weighting to maintain safety performance.\1354\ For the
NPRM, the A2Mac1 database was used to estimate the average mass of each
subsystem considered as part of the glider based on the subsystem
assumptions, and to compute the average glider share of vehicle curb
weight.\1355\ That analysis showed the glider accounted for 50 percent
of the vehicle curb weight. The agencies solicited comment on whether
systems or components beyond the vehicle body and chassis should be
included as part of the glider, and also indicated that the glider
weight assumption might increase for the final rule based on further
research.
---------------------------------------------------------------------------
\1354\ PRIA at 411-12.
\1355\ The A2Mac1 database was used and this analysis was
presented in ANL report docketed here: NHTSA-2018-0067-1490. The
mass data in the database were obtained from vehicle teardown
studies.
---------------------------------------------------------------------------
The agencies received several comments on the NPRM glider weight
assumptions, with the overarching theme of the comments being that the
NPRM did not include all systems and components that should be
included, and if those systems and components were included, the glider
share would be higher. Commenters also stated that the 50 percent
glider share value used for the NPRM reduced the amount of mass
reduction that could be applied to vehicles in the analysis.
UCS stated that representing the glider as a reduced fraction of
the curb weight caused the agencies significantly to underestimate the
potential for mass reduction. UCS noted that because mass reduction is
applied at the glider level, reducing the share of the glider
inherently caps the potential reduction in the curb weight, and this
single change cut the potential improvement from mass reduction by one-
third. Similarly, CARB stated that the updated glider weight assumption
severely limited the effectiveness of mass reduction, as the most
aggressive mass reduction category of 15 to 20 percent mass reduction
can only reduce the vehicle curb weight by 10 percent.
UCS cited previous agency analyses and analyses from other
organizations that stated the total potential for mass reduction by
2025 is between 15.8 and 32 percent of curb weight, contrasted to the
NPRM assumption of a maximum 10 percent reduction.\1356\ UCS also cited
industry data which showed that the glider represented a higher share
of vehicle curb weight than was assumed in the Draft TAR analysis, and
both UCS and CARB cited to industry data from vehicles like the Ford F-
150, which UCS stated was able to achieve the NPRM maximum achievable
mass reduction through the deployment of aluminum alone.\1357\ UCS
concluded that by capping the total potential for mass reduction at
such a low level, the agencies artificially reduced the potential for
the cost-effective technology, which increased the use of more
expensive and more advanced technologies. CARB concluded that the
agencies' 10 percent restriction means that real-world improvements
that have already happened on production vehicles were not considered
feasible in the NPRM analysis.
---------------------------------------------------------------------------
\1356\ NHTSA-2018-0067-12039 (citing Caffrey et al. 2013,
Caffrey et al. 2015, Lotus 2012, NAS 2015, Singh et al. 2012, Singh
et al. 2016, Singh et al. 2018).
\1357\ NHTSA-2018-0067-12039. See also NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
Several commenters also stated that the 50 percent glider weight
assumption was unexplained and unjustified, and argued that the
agencies' own studies showed that the glider weight percentage should
range from 75-80 percent.\1358\ UCS stated that both the NHTSA-
sponsored 2011 Honda Accord study, which showed the glider making up 79
percent of the vehicle, and the NHTSA-sponsored 2014 Chevrolet
Silverado study, which showed the glider making up 73.6 percent, showed
values substantially higher than the 50 percent value, and were in line
with the agencies' prior analyses.\1359\ As part of its comments that
key assumptions about mass reduction changed from the Draft TAR without
any supporting rationale, CARB stated that EPA had previously relied on
four studies (two contracted for by EPA and two contracted for by
NHTSA), and for the NPRM analysis the agencies only cited two of those
studies.\1360\ Moreover, ICCT commented that the agencies' previous
studies showed a glider fraction greater than 75 percent even with
numerous safety features considered. Accordingly, ICCT stated that the
agencies must specifically identify the ``safety components'' referred
to in the NPRM and justify the limitations placed on light weighting in
response. ICCT affirmatively concluded that the agencies must re-adopt
the Draft TAR methodology in which glider mass is assumed to be 75
percent of vehicle mass, or provide detailed justification and evidence
supporting the new value of 50 percent.\1361\
---------------------------------------------------------------------------
\1358\ NHTSA-2018-0067-11985; NHTSA-2018-0067-12039; NHTSA-2018-
0067-11873.
\1359\ NHTSA-2018-0067-12039.
\1360\ NHTSA-2018-0067-11873.
\1361\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
The agencies carefully considered these comments and reexamined
available data and information. The NHTSA-sponsored passenger car light
weighting study showed a glider mass of 79 percent, and the NHTSA-
sponsored light duty truck light weighting study showed a glider mass
of 73.6 percent, and the 75 percent value used for the Draft TAR was a
value between the values from these two studies. The agencies
determined it would be more rigorous to consider data from a broader
array of vehicles with various powertrain combinations and trim levels
to assess the glider share for the final rule, considering that the
vehicle fleet analyzed in this rule consists of over 2900 vehicle
models.
The agencies examined glider weight data available in the A2Mac1
database.\1362\ The A2Mac1 database tool is widely used by industry and
academia to determine the bill of materials and mass of each component
in the vehicle system.\1363\ The A2Mac1 database has been used by the
agencies to inform past CAFE and CO2 rulemakings. The
agencies analyzed a total of 147 MY 2014 to 2016 vehicles, covering 35
vehicle brands with different powertrain options representing a wide
array of vehicle classes to determine the glider weight for the final
rule analysis.\1364\
---------------------------------------------------------------------------
\1362\ A2Mac1: Automotive Benchmarking. (n.d.). Retrieved from
https://a2mac1.com.
\1363\ Bill of material (BOM) is a list of the raw materials,
sub-assemblies, parts and quantities needed to manufacture an end
product.
\1364\ The agencies presented this material for comments in the
ANL report posted in the docket NHTSA-2018-0067-1490.
---------------------------------------------------------------------------
The agencies also considered that the NHTSA passenger car and light
truck light-weighting studies examined mass reduction in the body,
chassis, interior, brakes, steering, electrical accessory, and wheels
subsystems and had developed costs for light weighted components in
those subsystems. As a result, the agencies determined it is
appropriate to include all of those subsystems as available for mass
reduction as part of the glider. Therefore, all of these systems were
included for the analysis of glider weight using the A2Mac1 database.
Table VI-125 shows the average mass for each subsystem and the glider
share for each of the vehicle classes for all powertrain combinations.
BILLING CODE 4910-59-P
[[Page 24538]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.259
This data was also compared with the glider weight measured in the
NHTSA MY 2014 Chevrolet Silverado light weighting study,\1365\ and the
glider weight data range was similar to the analysis results. Based on
the comments and the agencies' updated assessment, the agencies have
increased the glider weight assumption to 71 percent of the vehicle
curb weight for the final rule.
---------------------------------------------------------------------------
\1365\ DOT HS 812 487: Mass Reduction for Light-Duty Vehicles
for Model Years 2017-2025.
---------------------------------------------------------------------------
As stated above, for the NPRM, the interior, brake system,
electrical accessory system, and steering and wheel systems were not
included as part of the glider. The decision not to include the
interior system was based on an assumption at that time that interior
system mass reduction might adversely impact safety. In addition, the
decision not to include the brake system was based on an assumption at
that time that there would be little or no opportunity for downsizing
and reducing mass based on the reduced weight from body and chassis
only. As a result, brake systems were not considered as part of the
glider in the NPRM. For the final rule, the agencies included the
interior system based on market observations that light-weighted seats,
side door trim, frontal dash, and others interior components have been
incorporated on production vehicles that meet FMVSSs and perform well
on voluntary NCAP and IIHS safety tests. The agencies also considered
that interior, brakes, steering, wheel and electrical subsystems were
included in the NHTSA light weighting studies. By adding the interior,
steering, wheel subsystems and electrical subsystems as part of glider,
the agencies believe light weighting the glider increases the
opportunity for brake system optimization and mass reduction.
Similarly, there is increased opportunity for mass reduction for wheels
using gauge optimization, resulting from including more subsystems in
the glider.
By including the interior, brake, steering, electrical accessory,
and wheel subsystems in addition to the body and chassis subsystems in
the definition of what subsystems comprise the glider, the agencies
increased the glider weight from 50 percent of the vehicle curb weight
to 71 percent of the vehicle curb weight. This increase in turn means
that the potential for vehicle mass reduction was increased from 10
percent of the vehicle curb weight to 20 percent of the vehicle curb
weight. Table VI-126 shows the percent of light truck glider weight
reduction and the corresponding vehicle curb weight reduction for each
level of mass reduction for the glider shares used in the Draft TAR (75
percent), NPRM (50 percent), and final rule (71 percent)
analyses.\1366\
---------------------------------------------------------------------------
\1366\ Table 6-57 in PRIA showed the vehicle curb weight changes
for different glider weight assumptions.
---------------------------------------------------------------------------
BILLING CODE 4910-59-C
[[Page 24539]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.260
2) Powertrain Mass Reduction
As explained above, any mass reduction due to powertrain
improvements is accounted for separately from glider mass reduction.
Autonomie considers several components for powertrain mass reduction,
including engine downsizing, and transmission, fuel tank, exhaust
systems, and cooling system lightweighting.
The 2015 NAS report suggested an engine downsizing opportunity
exists when the glider mass is lightweighted by at least 10%. The 2015
NAS report also suggested that 10% lightweighting of the glider mass
alone would boost fuel economy by 3% and any engine downsizing
following the 10% glider mass reduction would provide an additional 3%
increase in fuel economy.\1367\ The agencies' lightweighting studies
applied engine downsizing (for some vehicle types but not all) when the
glider weight was reduced by 10 percent. Accordingly, the NPRM analysis
limited engine resizing to several specific incremental technology
steps; \1368\ important for this discussion, engines in the analysis
were only resized when mass reduction of 10% or greater was applied to
the glider mass, or when one powertrain architecture was replaced with
another architecture.
---------------------------------------------------------------------------
\1367\ National Research Council. 2015. Cost, Effectiveness, and
Deployment of Fuel Economy Technologies for Light-Duty Vehicles.
Washington, DC--The National Academies Press. https://doi.org/10.17226/21744.
\1368\ 83 FR 43027.
---------------------------------------------------------------------------
Argonne performed a regression analysis of engine peak power versus
weight for the NPRM based on attribute data taken from the A2Mac1
benchmarking database, to account for the difference in weight for
different engine types. For example, to account for weight of different
engine sizes like 4-cylinder versus 8-cylinder, Argonne developed a
relationship curve between peak power and engine weight based on the
A2Mac1 benchmarking data. For the NPRM analysis, this relationship was
used to estimate mass for all engine types regardless of technology
type (e.g., variable valve lift and direct injection). Weight
associated with changes in engine technology was applied by using this
linear relationship between engine power and engine weight from the
A2Mac1 benchmarking database. When a vehicle in the analysis fleet with
an 8-cylinder engine adopted a more fuel efficient 6-cylinder engine,
the total vehicle weight would reflect the updated engine weight with
two less cylinders based on the peak power versus engine weight
relationship.
When Autonomie selects a powertrain combination for a lightweighted
glider, the engine and transmission are selected such that there is no
degradation in the performance of the vehicle relative to the baseline
vehicle. The resulting curb weight is a combination of the
lightweighted glider with the resized and potentially new engine and
transmission. This methodology also helps in accurately accounting for
the cost of the glider and cost of the engine and transmission in the
CAFE model. This is one of the fundamental differences between the
analysis for this rulemaking the analysis for the Proposed
Determination. For the Proposed Determination, the cost for mass
reduction included mass reduction and cost reduction for one specific
engine downsizing, and applied it to all vehicle classes without regard
for performance and utility. There also was no accounting for the mass
of other applied powertrains and the associated effectiveness impacts.
As explained in the introduction, secondary mass reduction is
possible from some of the components in the glider after mass reduction
has been incorporated in primary subsystems (body, chassis, and
interior). Similarly, engine downsizing and powertrain secondary mass
reduction is possible after certain level of mass reduction is
incorporated in the glider. For the analysis, the agencies include both
primary mass reduction, and when there is sufficient primary mass
reduction, additional secondary mass reduction. The Autonomie
simulations account for the aggregate of both primary and secondary
glider mass reduction, and separately for powertrain mass.
The agencies received several comments about secondary mass
reduction and powertrain mass reduction. Broadly, CARB commented that
the agencies did not include powertrain downsizing and associated
secondary mass reduction, which was a departure from the analysis done
by
[[Page 24540]]
EPA for the Draft TAR.\1369\ CARB stated that the agencies
``inexplicably'' did not consider secondary mass reduction
opportunities ``including but not limited to drive axles, suspension,
and braking components (as a result of the overall vehicle being
lighter); fuel tank (and corresponding weight of fuel during
certification testing); powertrain (lighter engine and transmission
needed to power the lighter vehicle); and thermal systems.'' CARB cited
both EPA and NHTSA light weighting studies for the proposition that
there are significant opportunities for secondary mass reduction that
lead to additional cost savings. As a result, CARB stated that the
agencies inflated the cost of mass reduction as well as the amount of
mass reduction that is feasible and cost-effective, leading to an over
estimate in the technology costs to meet the existing standards.
---------------------------------------------------------------------------
\1369\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
As CARB correctly noted, the NHTSA-sponsored studies have taken
into consideration secondary mass reduction benefits such as radiator
engine support, and optimized engine cradles, wheels, and suspension
systems. As discussed above, in response to comments, the agencies have
included additional subsystems such as brakes, wheels, steering,
electrical, and interior systems to the glider for the final rule
analysis, thereby accounting for mass reduction opportunities for these
systems.
Also, as discussed further in Section VI.C.4.e), below, secondary
mass reduction is integrated into the mass reduction cost curves.
Specifically, the NHTSA studies, upon which the cost curves were built,
first generated costs for lightweighting the vehicle body, chassis,
interior, and other primary components, and then calculated costs for
lightweighting secondary components. Accordingly, the cost curves
reflect that, for example, secondary mass reduction for the brake
system is only applied after there has been sufficient primary mass
reduction to allow the smaller brake system to provide safe braking
performance and to maintain mechanical functionality.
CARB appears to have misunderstood how the analysis accounts for
powertrain mass reduction. The agencies described in the PRIA that the
Autonomie simulations recognize that many powertrain packages have
different weights for each vehicle class; for example, an eight-speed
transmission may weigh more than a six-speed transmission, and a basic
engine with variable valve timing may weigh more than a basic engine
without variable valve timing.\1370\ Autonomie varies the weight of
these powertrain systems as part of the analysis, and these changes are
done separately from the glider mass reduction technology levels (MR0
to MR6) in the simulations. Accordingly, accounting for powertrain mass
reduction as part of the mass reduction technology analysis would
double count impacts. The use of separate accounting assures that the
analysis accounts for mass associated with secondary mass reduction
from glider, and engine downsizing, as well as mass associated with
each individual engine, transmission, and electrification technology.
These mass changes were not accounted for in the Draft TAR and Proposed
Determination analyses. Moreover, these are accounted for separately in
the cost accounting, which is discussed further in the Section
VI.C.4.e), below.
---------------------------------------------------------------------------
\1370\ PRIA at 418.
---------------------------------------------------------------------------
HDS commented that some assumptions in the Autonomie modeling
related to engine weight appeared incorrect, such as the assumption
that a turbocharged 4-cylinder engine weighed the same as a DOHV V6
engine with 1.5 times the 4-cylinder's displacement, when in fact that
engine is often 75 to 100 lbs. lighter.\1371\
---------------------------------------------------------------------------
\1371\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
HDS also noted that ``mass reduction assumes no reduction of
powertrain weight for mass reduction levels of 2.5% and 5%. Mass
reduction effectiveness therefore are somewhat more appropriate for
reductions over 5% which apparently include some powertrain weight
reduction. More transparency in the PRIA regarding powertrain weight
changes will allow more detailed comment on engine weight assumptions
used.''
We agree with the comment that certain advanced engines could be
lighter than a basic engine. For the final rule, the estimated mass
levels for engines were updated, as discussed in Section VI.B.3 Tech
Effectiveness, based on the A2Mac1 database and other sources that
provided more precise mass data for powertrain technologies. Also, the
agencies improved upon the precision of estimated engine weights by
creating two curves to represent separately naturally aspirated engine
designs and turbocharged engine designs.\1372\ This update resulted in
two benefits. First, small naturally aspirated 4-cylinder engines that
adopted turbocharging technology reflected the increased weight of
associated components like ducting, clamps, the turbocharger itself, a
charged air cooler, wiring, fasteners, and a modified exhaust manifold.
Second, larger cylinder count engines like naturally aspirated 8-
cylinder and 6-cylinder engines that adopted turbocharging and
downsized technologies would have lower weight due to having fewer
engine cylinders. For the final rule analysis, a naturally aspirated 8-
cylinder engine that adopts turbocharging technology and is downsized
to a 6-cylinder turbocharged engine appropriately reflects the added
weight of the turbocharging components, and the lower weight of fewer
cylinders. These refinements address the issues identified in HDS's
comments.
---------------------------------------------------------------------------
\1372\ ANL Final Model Documentation for final rule analysis
Chapter 5.2.9 Engine Weight Determination.
---------------------------------------------------------------------------
Regarding HDS's second comment, as discussed in the NPRM, to
address product complexity and economies of scale, engine resizing is
limited to specific incremental technology changes that would typically
be associated with a major vehicle or engine redesign.\1373\ As
discussed further in Section VI.B.3.a)(6) Performance Neutrality, the
NPRM also referred to the 2015 NAS report conclusion that ``[f]or small
(under 5 percent [of curb weight]) changes in mass, resizing the engine
may not be justified, but as the reduction in mass increases (greater
than 10 percent [of curb weight]), it becomes more important for
certain vehicles to resize the engine and seek secondary mass reduction
opportunities.'' \1374\ In consideration of both the NAS report and
comments received from manufacturers, the agencies determined it would
be reasonable to allow allows engine resizing upon adoption of 7.1%,
10.7%, 14.2%, and 20% curb weight reduction, but not at 3.6% and
5.3%.\1375\ Resizing is also allowed upon changes in powertrain type or
the inheritance of a powertrain from another vehicle in the same
platform. The increments of these higher levels of mass reduction, or
complete powertrain changes, more appropriately match the typical
engine displacement increments that are available in a manufacturer's
engine portfolio.
---------------------------------------------------------------------------
\1373\ See 83 FR 43027 (Aug. 24, 2018).
\1374\ National Research Council. 2011. Assessment of Fuel
Economy Technologies for Light-Duty Vehicles. Washington, DC--The
National Academies Press. http://nap.edu/12924.
\1375\ These curb weight reductions equate to the following
levels of mass reduction as defined in the analysis: MR3, MR4, MR5
and MR6, but not MR1 and MR2; additional discussion of engine
resizing for mass reduction can be found in Section VI.B.3
Technology Effectiveness.
---------------------------------------------------------------------------
[[Page 24541]]
3) Summary of Final Rule Mass Reduction Technology Effectiveness
Figure VI-45 below shows the range of incremental effectiveness
used for the NPRM analysis. The chart lumps all of the vehicle classes
for each of the technology types.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.261
BILLING CODE 4910-59-C
Figure VI-46 below shows the range of incremental effectiveness
improvement from full vehicle modeling when mass reduction technologies
were applied to vehicles for the final rule analysis.
BILLING CODE 4910-59-P
[[Page 24542]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.262
BILLING CODE 4910-59-C
e) Mass Reduction Costs
The PRIA described the decision to use NHTSA's passenger car light
weighting study based on a MY 2011 Honda Accord and NHTSA's full-size
pickup truck light weighting study based on a MY 2014 Chevrolet
Silverado to derive the estimated cost for each of the mass reduction
technology levels.\1376\ The agencies relied on the results of those
studies because they considered an extensive range of material types,
material gauge, and component redesign while taking into account real
world constraints such as manufacturing and assembly methods and
complexity, platform-sharing, and maintaining vehicle utility,
functionality and attributes, including safety, performance, payload
capacity, towing capacity, handling, NVH, and other characteristics. In
addition, the agencies described that the baseline vehicles assessed in
the NHTSA-sponsored studies were reasonably representative of baseline
vehicles in the MY 2016 analysis fleet.\1377\ The agencies also noted
they made the decision to rely on these studies after reviewing other
agency, CARB, ICCT and industry studies.\1378\ The other studies often
did not consider important factors, made unrealistic assumptions about
key vehicle systems, and/or applied secondary mass reduction
inappropriately, resulting in unrealistically low costs. The PRIA also
described how the cost estimates derived from the NHTSA lightweighting
studies were adjusted to reflect the NPRM glider share
assumption.\1379\
---------------------------------------------------------------------------
\1376\ PRIA at 391; Table 6-38 and Table 6-41 in PRIA.
\1377\ PRIA at 403.
\1378\ As described in the PRIA at 390-91, studies by EPA, CARB,
Transport Canada, the American Iron and Steel Institute (AISI), the
Aluminum Association, and the American Chemistry Council were all
reviewed for potential incorporation into the analysis.
\1379\ See PRIA at 396, Tables 6-38 and 6-39; PRIA at 401,
Tables 6-41 and 6-42. See also PRIA at 391 (``While the definitions
of glider may vary from study to study (or even simulation to
simulation), the agencies referenced the same dollar per pound of
curb weight to develop costs for different glider definitions. In
translating these values, the agencies took care to track units ($/
kg vs. $/lb.) and the reference for percentage improvements (glider
vs. curb weight).'').
---------------------------------------------------------------------------
Furthermore, the agencies changed the cost of mass reduction
accounting from a curb weight basis in the Draft TAR to glider weight
basis in the NPRM.\1380\ Because the mass reduction studies provide
mass reduction costs for the glider, this change enabled more direct
use of cost curve data from the studies in the CAFE model. This change
also allowed independent accounting for powertrain mass, which enabled
the CAFE model to account more accurately for the unique mass of each
of the powertrains that are available in each vehicle model. The cost
of the engine, transmission, and electrification are accounted for
separately from the glider in the CAFE model.
---------------------------------------------------------------------------
\1380\ In the Draft TAR, the agencies presented the cost
estimates from mass reduction studies sponsored by both NHTSA and
EPA. EPA presented the cost of mass reduction as function of vehicle
curb weight. To harmonize the cost estimates with EPA, NHTSA also
presented the cost of mass reduction as a function of vehicle curb
weight.
---------------------------------------------------------------------------
The agencies received several comments on the mass reduction costs
used in the NPRM. FCA commented that the costs and benefits used the
CAFE model were overly optimistic,
[[Page 24543]]
stating that although its Ram 1500 pickup truck achieved several
hundred pounds of weight reduction, the cost of achieving that weight
reduction was greater than that used in the CAFE model.\1381\
Similarly, as mentioned above, Toyota commented that mass reduction
cost values were underestimated.\1382\ Conversely, CARB, UCS, and the
City of Oakland in California commented that the costs used for mass
reduction in the NPRM overstated the cost of mass reduction. The
agencies also received several comments relating to the studies used to
develop the mass reduction cost curves, how the values from those
curves were applied in the CAFE model, and costs for secondary mass
reduction; those comments are discussed in turn.
---------------------------------------------------------------------------
\1381\ NHTSA-2018-0067-11943.
\1382\ NHTSA-2018-0067-12098.
---------------------------------------------------------------------------
(1) Studies Used To Develop Mass Reduction Cost Curves
The agencies described in the PRIA that since the 2012 final rule,
both agencies conducted lightweighting studies to assess the technical
feasibility and cost of mass reduction.\1383\ The agencies also stayed
apprised of studies performed by other agencies, manufacturers, and
industry trade associations, and reviewed them in development of
lightweighting assumptions used in the NPRM and final rule
analysis.\1384\ Among the several lightweighting studies, the agencies
used NHTSA's passenger car lightweighting study, based on a MY 2011
Honda Accord, and NHTSA's full-size pickup truck lightweighting study,
based on a MY 2014 Chevrolet Silverado, to derive the cost estimates to
achieve different levels of mass reduction for the NPRM and final rule.
---------------------------------------------------------------------------
\1383\ PRIA at 390.
\1384\ PRIA at 403.
---------------------------------------------------------------------------
The agencies described that the decision to rely on those studies
included that those studies considered materials, manufacturing,
platform-sharing, functional attribute, performance, and noise-
vibration- and harshness (NVH), among other constraints pertaining to
cost, effectiveness, and safety considerations, in addition to that
these vehicles were a reasonable representation of the baseline
vehicles in the MY 2016 compliance simulation.\1385\ Specifically in
regards to safety, the agencies described a preference to use studies
that considered small overlap impact tests conducted by the Insurance
Institute for Highway Safety (IIHS) and not all studies took that test
into account. In regards to maintaining vehicle functionality, the
agencies described that the NHTSA pickup truck study accounted for
vehicle functional performance for attributes including towing, noise
and vibration, and gradeability, in addition to considering platform
sharing constraints.
---------------------------------------------------------------------------
\1385\ PRIA at 403.
---------------------------------------------------------------------------
In contrast, the agencies explained that the other studies often
did not consider many important factors, or those studies made
unrealistic assumptions about key vehicle systems through secondary
downsizing, resulting in unrealistically low costs. Specifically, the
agencies referenced EPA's past analysis of a MY 2010 Toyota Venza as an
example of a study that employed overly aggressive secondary mass
reduction, which translated into cost savings for the initial 10% mass
reduction.\1386\
---------------------------------------------------------------------------
\1386\ PRIA at 391.
---------------------------------------------------------------------------
The agencies received several comments on the studies used to
generate the mass reduction cost curves. Ford commented in support of
the agencies' decision to exclude mass reduction studies that were
misaligned with tear-down studies.\1387\ Ford cited the MY 2010 Toyota
Venza Phase II study used to establish the mass reduction cost values
used for the Draft TAR and Proposed Determination that suggested the
first 7-10% of mass reduction could be accomplished with zero or
reduced cost,\1388\ which Ford characterized as ``a gross
underestimation of industry investment and material costs associated
with any weight reduction.''
---------------------------------------------------------------------------
\1387\ NHTSA-2018-0067-11928.
\1388\ EPA-420-R-16-021: Proposed Determination Technical
Support Document at 2-158, November 2016.
---------------------------------------------------------------------------
ICCT commented that The National Academies of Science
``specifically endorsed tear-down studies as the most appropriate way
to get at vehicle technology costs, [as those] studies are typically
more accurate and far more transparent than the older method of
surveying manufacturers, and such whole-vehicle studies are key to
capturing holistic vehicle level mass-reduction technology costs.''
ICCT noted that there are many peer-reviewed tear-down studies that
demonstrate that at least 20 percent mass reduction is available for
adoption across vehicle classes by 2025, including studies by EDAG,
FEV, Ford, and Lotus Engineering; however, ICCT alleged that the
agencies ``have either incorrectly interpreted or invalidly nullified
the most relevant detailed engineering teardown studies on mass-
reduction technology.'' ICCT noted that the agencies were ``well
aware'' of these studies, as they were performed by CARB in conjunction
with the agencies, however, ICCT alleged that the agencies
``reinterpreted the results of the main study relied upon in the TAR in
order to inflate costs,'' and that the ``technical assessment by the
agencies has a clear technical bias towards reducing CAFE and GHG
standards.'' ICCT concluded that ``[e]xcluding these studies amounted
to intentionally disregarding the most pertinent and rigorous
engineering studies that are applicable to the rulemaking timeframe.''
ICCT recommended the agencies adjust their technology cost inputs
to reflect the ``best-available technology studies.'' ICCT stated that
the correct cost assumption from these studies is that ``a 5-10% mass
reduction by 2025 reduces vehicle cost, and the auto industry will
cost-effectively deploy at least 15% vehicle curb mass reduction in the
2025 timeframe at near zero net cost (and consistently less than
$500).''
CARB asserted that the agencies inflated the costs of mass
reduction in the NPRM analysis by only considering NHTSA-sponsored
studies and improperly excluding the effects of secondary mass
reduction as documented in those studies.\1389\ CARB provided a table
of studies that largely mirrored the tables of studies the agencies
considered in the Draft TAR and PRIA,\1390\ and also included the
associated mass reduction costs in $/kg included in each study, noting
that for all excluded studies cited in the table, all mass reduction
costs were substantially lower than the values used in the agencies'
analysis.\1391\ Similarly, UCS commented that while the PRIA did state
that additional studies ``often did not consider many important factors
or . . . made unrealistic assumptions about key vehicle systems,'' the
agencies did not specifically identify the factors and assumptions that
merited disregarding those studies, which were included previously in
agency analysis as part of the record when deriving previous estimates
for the costs of mass reduction.\1392\
---------------------------------------------------------------------------
\1389\ NHTSA-2018-0067-11873.
\1390\ Draft TAR at 5-168; PRIA at 404-05.
\1391\ NHTSA-2018-0067-11873.
\1392\ NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
The agencies agree with ICCT that peer-reviewed tear-down studies
present an appropriate method to capture holistic vehicle-level mass
reduction technology costs. The agencies also agree with ICCT that the
agencies were well aware of studies conducted by EDAG, FEV, Ford, and
Lotus Engineering; in fact, the agencies
[[Page 24544]]
presented a table listing several of those studies in the PRIA with the
qualification that those studies were reviewed in developing
lightweight assumptions for the analysis, but those studies did not
consider important factors, or those studies made unrealistic
assumptions about key vehicle systems through secondary downsizing,
resulting in unrealistically low costs.
The agencies also agree with UCS' comment that the language could
have been more specific about identifying the factors and assumptions
that merited disregarding studies that were previously included as part
of the record when deriving previous estimates for the costs of mass
reduction. The following discussion briefly summarizes the record since
the Draft TAR and differences between NHTSA's and other lightweighting
studies' approach to factors listed in the PRIA. Important for this
discussion is an understanding of primary versus secondary mass
reduction; as described above, when there is sufficient primary mass
reduction, other components that are designed based on the mass of
primary components may be redesigned and have lower mass. Recall the
braking system example used throughout this section; mass reduction in
the braking system is secondary mass reduction because it requires
primary mass reduction before it can be incorporated. If the mass of
primary components is reduced sufficiently, the resulting lighter
weight vehicle could maintain braking performance, attributes, and
safety with a lighter weight brake system.
Several studies were referenced in the Draft TAR that either used
tear-down analyses and computer-aided engineering (CAE) to produce a
future engineered lightweight vehicle, or considered future
technologies and processes for lightweighting vehicle components.\1393\
---------------------------------------------------------------------------
\1393\ Draft TAR at 5-158 through 5-197.
---------------------------------------------------------------------------
EPA developed cost curves for cars and CUVs based on the MY 2010
Toyota Venza study, and pickup truck cost curves based on the MY 2011
Chevrolet Silverado study.\1394\ The other studies were considered by
EPA, but not used to develop the Draft TAR, Proposed Determination and
Final Determination cost curves. In brief, EPA described that the
Toyota Venza Phase I was a mass reduction opportunity study only, and
the Phase II study was a holistic vehicle study that examined nearly
every component in the vehicle for mass reduction potential and
calculated a related cost and mass saved for each. For the cost curve,
EPA applied the individual components in sequence from largest cost per
kilogram savings to largest cost per kilogram increase. For example,
the cost curve for the Draft TAR and Proposed Determination applied
engine downsizing and transmission system mass reduction first, and
before lightweighting the body, chassis, doors and other
components.\1395\ EPA stated this methodology was chosen based on the
understanding that OEMs will choose the cost saving technologies first
and that some cost mass reduction technologies will be paid for by the
cost save mass reduction technologies, citing a 2016 publication by CAR
and a GM presentation that stated over $2,000,000,000 was saved in
material costs through various lightweighting approaches.\1396\
---------------------------------------------------------------------------
\1394\ Draft TAR at 5-367.
\1395\ EPA-420-R-16-021: Proposed Determination Technical
Support Document at 2-161 and 2-162
\1396\ Draft TAR at 5-172 (citing ``Identifying Real world
Barriers to Implementing Lightweighting Technologies and Challenges
in Estimating the Increase in Costs,'' Center for Automotive
Research, Jay Baron, Ph.D., January 2016 http://www.cargroup.org/?module=Publications&event=View&pubID=128; General Motors, ``General
Motors 2015 Global Business Conference,'' Presentation, October 1,
2015, Slides 43-45 in document, https://www.gm.com/content/dam/gm/events/docs/5194074-596155-ChartSet-10-1-2015.).
---------------------------------------------------------------------------
The NHTSA cost curves were developed by rearranging the
lightweighted components from the MY 2011 Honda Accord and MY 2014
Chevrolet Silverado studies based on cost effectiveness, assuming the
vehicle body, chassis, interior, and other primary components were
lightweighted first, followed then by lightweighting powertrain
components and other secondary systems.\1397\ The cost curves based on
the NHTSA studies reflect that, returning to this example, secondary
mass reduction for the brake system is only applied after there has
been sufficient primary mass reduction to allow the smaller brake
system to provide safe braking performance and to maintain mechanical
functionality.\1398\
---------------------------------------------------------------------------
\1397\ Draft TAR at 5-421 (``The powertrain components which
include engine, transmission, and fuel systems such as fuel filler
pipe, fuel tank, fuel pump, etc., exhaust systems and cooling
systems were not considered for application of primary mass
reduction but benefits of secondary mass reduction were accounted
for. These powertrain components are assumed to be downsized only
after the primary vehicle structural components (Body-In-White)
achieve certain level of mass reduction.'').
\1398\ Draft TAR at 5-422.
---------------------------------------------------------------------------
The EPA and NHTSA studies took fundamentally different approaches
to accounting for the costs of mass reduction technology, and
accordingly, EPA needed to translate the cost curves from the NHTSA
studies to use a similar methodology as the cost curves from the EPA
studies.\1399\ To ``normalize'' the NHTSA studies with the EPA's
studies, EPA listed components identified for lightweighting in the
NHTSA studies and reorganized those components from the lowest cost to
highest cost along with associated mass reduction per the ``whole
vehicle'' approach mentioned above, distributed mass savings from
secondary mass reduction to all points along the cost curve, and
included the mass saved from engine downsizing without taking into
consideration the cost of added engine technology. This resulted in
lower-cost secondary mass reduction opportunities being considered
before primary mass reduction opportunities, which in turn resulted in
artificially low $/kg costs for mass reduction.
---------------------------------------------------------------------------
\1399\ Draft TAR at 5-369.
---------------------------------------------------------------------------
For the NPRM and final rule, the agencies simply used the original
ordered list of components from the MY 2011 Honda Accord study and MY
2014 Chevrolet Silverado study, arranged sequentially for cost
effectiveness based on primary then secondary mass reduction
opportunities, to generate the cost curves for passenger cars and light
trucks. Accordingly, the agencies did not ``reinterpret'' the results
of studies used in the Draft TAR in the NPRM, as ICCT alleged, but
rather appropriately represented how primary and secondary mass
reduction opportunities are implemented in the real world (to the
extent that ICCT is referring to the translation of the study costs to
the NPRM glider weight assumptions, that is discussed in Section
VI.C.4.e)(1), below). To maintain utility and performance in the real
world, primary components must be lightweighted first before the engine
and transmission can be resized. Moreover, as described in the Draft
TAR, NHTSA's mass reduction studies did not ``improperly exclude'' the
effects of secondary mass reduction, rather those effects were
appropriately accounted for after primary components achieved certain
levels of mass reduction. As discussed in Section VI.B.3.a)(6)
Performance Neutrality, this approach aligned with the NAS approach to
consider powertrain downsizing only after the vehicle structural
components achieved 10 percent mass reduction.
OEMs have also disagreed with the conclusion that mass reduction
could come at a cost savings. For instance, Ford characterized the
Toyota Venza studies, which concluded the first 7-10% of mass reduction
could come at a negative cost as ``a gross
[[Page 24545]]
underestimation of industry investment and material costs associated
with any weight reduction.'' The agencies believe that the approach to
secondary mass reduction followed in the NHTSA passenger car and pickup
truck lightweighting studies appropriately incorporated both the costs
and real-world constraints associated with employing primary and
secondary mass reduction technologies.
Aside from the differences in how studies treated secondary mass
reduction, the agencies opted not to use, or could not use, other
studies either previously considered in the rulemaking record or
mentioned by commenters for several reasons:
Studies were not comprehensive, and therefore could not be used to
develop a comprehensive cost curve: Some studies narrowly assessed
lightweighting of a portion of vehicle, such as the body in white
subsystem, or as stated in the PRIA,\1400\ were limited to material
substitution of the vehicle components, such as replacing steel with
aluminum or replacing mild steel with AHSS or replacing mild steel with
CFRP in selective components. Factors important to vehicle
functionality, like material joining techniques and the feasibility of
manufacturing processes or necessary retooling requirements were not
considered, and therefore could not be used to develop a comprehensive
cost curve representative of the costs required to reduce mass in a
vehicle.\1401\
---------------------------------------------------------------------------
\1400\ PRIA at 391.
\1401\ An Assessment of Mass Reduction Opportunities for a 2017-
2020 Model Year Vehicle Program, March 2010, Lotus Engineering, at
p. 6.
---------------------------------------------------------------------------
Cost curves were not developed or no cost analysis was performed:
For the CARB Holistic Vehicle Mass Reduction/Cost Study, a cost curve
was not developed, and the resulting cost per kilogram data points were
point estimates. The calculated cost per kilogram was used as one data
point of several to indicate the direction for mass reduction beyond
EPA's original passenger car/CUV curve.\1402\ Or, as in the case of the
DOE/Ford/Magna Multi Material Lightweight Vehicle (MMLV) project, no
cost analysis was performed for the initial project, and later
project(s) concluded that ``a 37% to 45% mass reduction in a standard
mid-sized vehicle is within reach if carbon fiber composite materials
and manufacturing processes are available and if customers are willing
to accept a reduction in vehicle features and content, as demonstrated
with the Multi-Materials and Carbon Fiber Composite-Intensive vehicle
scenarios.'' \1403\
---------------------------------------------------------------------------
\1402\ Draft TAR at 5-185.
\1403\ Draft TAR at 5-194.
---------------------------------------------------------------------------
Engineered vehicles did not meet functional design or manufacturing
requirements: As noted in the update to EPA's Light-Duty Vehicle Mass
Reduction and Cost Analysis for the Toyota Venza, the Phase I
engineered Venza did not meet the design target of no expected NVH
degradation.\1404\ The Phase II (High Development) study assumed
significant cost savings from reduced parts manufacturing, but did not
appropriately explain the methodology used to conclude that the part
count reduction was feasible.\1405\
---------------------------------------------------------------------------
\1404\ Light-Duty Vehicle Mass Reduction and Cost Analysis--
Midsize Crossover Utility Vehicle, EPA-420-R-12-026 (August 2012).
\1405\ Peer Review of Demonstrating the Safety and
Crashworthiness of a 2020 Model-Year, Mass-Reduced Crossover Vehicle
(Lotus Phase 2 Report), EPA-420-R-12-028 (September 2012).
---------------------------------------------------------------------------
In addition, the agencies qualified in the PRIA a preference to use
studies that considered the small overlap impact test conducted by
IIHS, and not all studies took that test into account.\1406\ NHTSA's
``Update to Future Midsize Lightweight Vehicle Findings in Response to
Manufacturer Review and IIHS Small-Overlap Testing'' based on the MY
2011 Honda Accord presented results incorporating suggestions from
Honda regarding NVH and durability, and updating the engineered vehicle
to achieve a ``good'' rating in seven crash safety tests.\1407\ EPA
studies also accounted for the IIHS small overlap test through an ad
hoc estimate of mass and cost, unlike the NHTSA update, which
explicitly modeled to account for NVH performance and to comply with
the IIHS small overlap test.
---------------------------------------------------------------------------
\1406\ PRIA at 391.
\1407\ Singh, H., Kan, C-D., Marzougui, D., & Quong, S. (2016,
February). Update to future midsize lightweight vehicle findings in
response to manufacturer review and IIHS small-overlap testing
(Report No. DOT HS 812 237). Washington, DC: National Highway
Traffic Safety Administration.
---------------------------------------------------------------------------
The agencies continue to believe that the MY 2011 Honda Accord and
MY 2014 Chevrolet Silverado lightweighting studies are the best studies
upon which to estimate the costs of mass reduction in the rulemaking
timeframe.
(2) How the Cost Curves Are Applied in the Model
Commenters also submitted comments on how the cost curves were
applied in the model, including that the studies the agencies relied
upon to generate cost curves, discussed above, did not support the 50
percent glider share assumption used in the NPRM, and the agencies did
not correctly scale the costs to match the glider share assumption.
UCS commented that the agencies based the costs for mass reduction
on glider weight reduction, however, the need for more expensive
materials and more advanced engineering and design strategies only
results from the need for greater levels of absolute mass reduction on
the vehicle.\1408\ UCS stated that the cost curves had effectively been
derived from the assumption of reductions as great as 16.8 percent
reduction in curb weight in the case of the Silverado (Singh et al.
2018) and as great as 18 percent reduction in curb weight in the case
of the Honda Accord (Singh et al. 2016), but applied to curb weight
reductions approximately two-thirds that magnitude. UCS stated that
approach was ``completely invalid and significantly overstates the
costs for mass reduction.'' UCS also commented that the agencies
incorrectly scaled the cost curves based on the agencies' mass
reduction studies, which refer to direct manufacturing costs as a
function of vehicle curb weight, not just glider weight. UCS stated
this incorrectly yielded the same costs for two-thirds the amount of
mass reduction.
---------------------------------------------------------------------------
\1408\ NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
CARB similarly commented that the mass reduction costs assigned to
both passenger cars and light trucks in the CAFE model were
inappropriately inflated based on incorrect scaling from the glider
share assumptions used in the Honda Accord and Chevy Silverado studies
to the NPRM glider share value.\1409\ CARB analyzed two tables in the
PRIA that showed the agencies' translation of cost numbers derived from
the two studies to the cost numbers used in the CAFE model, and
asserted that the agencies improperly used costs from the upper end of
the mass reduction range rather than the midpoint of the range, leading
to cost overestimation.
---------------------------------------------------------------------------
\1409\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
Similarly, HDS commented that the PRIA passenger car cost curves
used data that were not in agreement with the study that they were
based upon, noting that the Honda Accord study showed the glider
accounting for 78% of curb weight, and this limited absolute weight
reduction.\1410\ HDS noted that the truck weight reduction cost data
were closer to those cited in the Chevy Silverado teardown study,
although the glider share for that study was also 73.6% of vehicle curb
weight.
---------------------------------------------------------------------------
\1410\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
HDS also commented that although the agencies relied on the same
Honda Accord study that was used in the Draft
[[Page 24546]]
TAR, ``the costs have been changed significantly [from the Draft TAR]
for unexplained reasons.'' \1411\ HDS stated that the PRIA showed
average costs for mass reduction, whereas earlier studies showed the
cost increment for each 5% mass reduction, noting that with increasing
incremental cost with increased mass reduction, average cost will
always be lower than incremental cost. HDS claimed that it was
``unusual'' for the Draft TAR incremental costs to decrease between 11%
and 19% mass reduction but increase elsewhere, but also noted the
unexplained increase in cost, specifically a $536 cost for 175kg weight
reduction, shown in the PRIA.
---------------------------------------------------------------------------
\1411\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
HDS also compared manufacturing costs from the Draft TAR to the
PRIA analysis, noting that the direct manufacturing cost was found to
be negative (i.e., a cost saving) in the Draft TAR analysis for mass
reduction up to 15 percent, but EPA assumed the indirect costs were
positive so that the total cost was a sum of positive and negative
costs--meaning the total cost could be positive or negative. In
contrast, HDS noted there were no negative costs in the cost curves
used for the PRIA analysis, resulting in a very large differential
between the costs of mass reduction, with the 2018 average cost being
higher than even the 2016 marginal costs.
Three notable changes from the NHTSA Draft TAR to NPRM and final
rule analysis impacted how the cost curves for mass reduction are
applied in the CAFE Model.
First, the Draft TAR considered mass reduction in the glider and
powertrain together, and calculated the percentage mass reduction on a
vehicle curb weight basis. In the Draft TAR, only one engine and
transmission combination were considered to account for the mass change
associated with downsizing the engine, and the cost estimates for mass
reduction for this one powertrain combination was applied to all
powertrain combinations. This approach did not account for the mass
changes associated with the application of powertrain technologies
(engine, transmission and electrification) technologies, and did not
account for the corresponding change in glider mass needed to offset
the powertrain mass change and to achieve the specified curb weight
mass reduction level. This approach did not reflect the real world,
where there are many vehicles with different body styles and powertrain
combinations, and therefore did not account for differences in mass for
different engines, transmissions, or electrification.
Accordingly, for the NPRM and final rule, the cost of mass
reduction was calculated on a glider weight basis so that the weight of
each powertrain configuration could be directly and separately
accounted for. This approach provides the true cost of mass reduction
without conflating the mass change and costs associated with downsizing
a powertrain or adding additional advanced powertrain technologies.
Hence, the mass reduction costs in the NPRM reflect the cost of mass
reduction in the glider and do not include the mass reduction
associated with engine downsizing, and therefore appear to be higher
than the cost estimates in the Draft TAR.
Second, the glider share of curb weight changes from the Draft TAR
to NPRM and from the NPRM to the final rule analysis also affected the
absolute amount of curb weight reduction that was applied, and
therefore for cost per pound for the mass reduction changes with the
change in the glider share. The cost for removing 20 percent of the
glider weight when the glider represents 75% of a vehicle's curb weight
is not the same as the cost for removing 20 percent of the glider
weight when the glider represents 50% of the vehicle's curb weight. For
example, the glider share of 79 percent of a 3,000-pound curb weight
vehicle is 2,370 pounds, while the glider share of 50 percent of a
3,000-pound curb weight vehicle is 1,500 pounds, and the glider share
of 71 percent of a 3,000-pound curb weight vehicle is 2,130 pounds. The
mass change associated with 20 percent mass reduction is 474 pounds for
79 percent glider share (=[3,000 pounds x 79% x 20%]), 300 pounds for
50 percent glider share (=[3,000 pounds x 50% x 20%]), and 426 pounds
for 71 percent glider share (=[3,000 pounds x 71% x 20%]). The mass
reduction cost studies show that the cost for mass reduction varies
with the amount of mass reduction. Therefore, for a fixed glider mass
reduction percentage, different glider share assumptions will have
different costs.
To further illustrate, Table VI-127 and Table VI-128 below shows
the associated curb weight percentage mass reduction and the associated
average cost per pound for different glider weight assumptions for each
glider mass reduction technology level used in the final rule analysis.
For reference, the costs from the passenger car light weighting study
are presented.\1412\ These costs were the basis for deriving the costs
for each mass reduction technology level in the Draft TAR, NPRM, and
final rule analyses, using the unique glider share values for each of
those analyses. In the light weighting study, NHTSA applied the mass
reduction technologies identified for the exemplar vehicle on other
vehicle(s) and vehicle types to understand the level of mass reduction
that could be achieved. In the case of passenger cars, the maximum
level of mass reduction was around 15% of the vehicle curb weight if
all the mass reduction technologies are applied. In other words,
achieving mass reduction greater than 10% of the curb weight for
passenger cars will require extensive use of advanced materials such as
high strength aluminum and carbon fiber composite material.
---------------------------------------------------------------------------
\1412\ Table 6-39 in PRIA.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24547]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.263
[GRAPHIC] [TIFF OMITTED] TR30AP20.264
BILLING CODE 4910-59-C
Finally, as explained earlier, to determine the mass reduction
technology levels for the NPRM 2016 analysis fleet, a distribution of
the residuals from the regression using 50 percent glider weight
generally showed a greater percentage of vehicles achieving higher
levels of mass reduction. With this high level of mass reduction
already achieved, the opportunities for further mass reduction would be
limited and have higher costs. For the final rule, since the agencies
updated the glider share to 71 percent of the vehicle curb weight, the
distribution of residuals from the regression shifted some vehicles to
lower baseline mass reduction
[[Page 24548]]
technology levels, providing more opportunity for further mass
reduction, on average. Even as some of the vehicles start further up on
the mass reduction cost curve due to higher levels of mass reduction
technology (MR3, MR4) already present in the vehicles, there are
additional opportunities for further mass reduction to achieve MR5 and
above.
Table VI-127 and Table VI-128 show that for the final rule, cost
estimates with the 71 percent glider share come closer to the cost
estimates used in Draft TAR, which assumed a 79 percent glider share.
(3) Secondary Mass Reduction Costs
As discussed above, the agencies changed the cost of mass reduction
calculation from a curb weight basis in the Draft TAR to a glider
weight basis in the NPRM.\1413\ This change allowed us to estimate the
cost of mass reduction independently of the cost associated with
downsized advanced engines and advanced transmissions, as the cost of
downsized advanced engines and transmissions are accounted for
separately in the CAFE model.
---------------------------------------------------------------------------
\1413\ In the Draft TAR, the agencies presented the cost
estimates from mass reduction studies sponsored by both NHTSA and
EPA. EPA presented the cost of mass reduction as function of vehicle
curb weight. To harmonize the cost estimates with EPA, NHTSA also
presented the cost of mass reduction as a function of vehicle curb
weight.
---------------------------------------------------------------------------
The MY 2011 Honda Accord and MY 2014 Chevy Silverado studies used
to develop the NPRM and final rule cost curves for mass reduction
technologies include some non-powertrain secondary mass reduction
technologies such as brakes and wheels. The agencies presented the list
of mass reduction technologies in NPRM.\1414\ Following the publication
of NHTSA's light weighting studies, peer reviewers and manufacturers
commented that many components such as drive axles, engine cradles, and
radiator engine support that are considered to be non-powertrain
secondary mass reduction opportunities cannot be downsized, as the same
components are used across many vehicles with different powertrain
options. Even though some of these components may provide opportunities
for additional mass reduction, NHTSA agreed with peer reviewers and
manufacturers that retaining a common design for all powertrain options
provides for cost reductions due to economies of scale.
---------------------------------------------------------------------------
\1414\ Table 6-37 and Table 6-40 in PRIA.
---------------------------------------------------------------------------
Commenters faulted the agencies for a perceived lack of accounting
for the cost decreases from secondary mass reduction. ICCT commented
although the agencies relied on the Honda Accord study, which
considered cost savings from downsizing the powertrain, in the NPRM
only glider weight reduction was ever considered without the cost-
offsetting engine downsizing.\1415\ ICCT stated that this omission had
two effects, first that accounting for associated powertrain weight
reductions would have allowed for more mass reduction, thus allowing
for greater efficiency benefits at a lower cost, and second, that
vehicle performance was erroneously improved, contrary to the agencies'
assertion that the analysis assumed a level of performance neutrality.
ICCT concluded that it was unclear if and how costs were reduced for
powertrain downsizing, as well as the precise changes to fuel
efficiency.
---------------------------------------------------------------------------
\1415\ NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
CARB faulted the agencies for not including secondary mass
reduction in the NPRM analysis, and stated that by failing to account
for secondary mass reduction as was done in the Draft TAR, the agencies
inflated the costs for mass reduction as well as the amount of mass
reduction that is feasible and cost-effective leading to an
overestimate in the technology costs needed to meet the existing
standards.
The agencies note that the cost curves used for the NPRM and this
final rule do in fact include secondary mass reduction. The cost curves
reflect secondary mass reduction applied when there is sufficient
primary mass reduction to implement secondary mass reduction without
degrading function and safety. Specifically, the NHTSA studies, upon
which the cost curves were built, first generated costs for
lightweighting the vehicle body, chassis, interior, and other primary
components, and then calculated costs for lightweighting secondary
components. Accordingly, the cost curves reflect that, for example,
secondary mass reduction for the brake system is only applied after
there has been sufficient primary mass reduction to allow the smaller
brake system to provide safe braking performance and to maintain
mechanical functionality.
In addition, CARB stated that the 2011 Honda Accord and the 2014
Chevrolet Silverado studies had ``markedly'' lower costs than the
proposal when secondary mass reduction is included. Again, the agencies
believe these comments resulted from a lack of understanding about how
the analysis considers primary and secondary mass reduction, even
though the NPRM and PRIA explicitly stated how costs are accounted for
separately.\1416\ Also, as discussed above, engine mass reduction
enabled by mass reduction in the glider is accounted for separately and
therefore not included as part of glider mass reduction technology, as
doing so would result in double counting the impacts.
---------------------------------------------------------------------------
\1416\ PRIA at 413.
---------------------------------------------------------------------------
(4) Summary of Final Rule Mass Reduction Costs
For the final rule, the agencies continue to use multiple mass
reduction technology levels and costs based on the lightweighting
studies that were presented in PRIA.\1417\ Since the agencies have
changed the glider share of curb weight assumption from 50 percent in
NPRM to 71 percent in the final rule, the mass reduction costs reflect
the updated glider share. Table VI-129 and Table VI-130 show mass
reduction costs used in the CAFE model for passenger car and light
trucks.
---------------------------------------------------------------------------
\1417\ Table 6-37 and 6-40 in PRIA.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24549]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.265
[GRAPHIC] [TIFF OMITTED] TR30AP20.266
BILLING CODE 4910-59-C
5. Aerodynamics
The energy required to overcome aerodynamic drag accounts for a
significant portion of the energy consumed by a vehicle, and can become
the dominant factor for a vehicle's energy consumption at high speeds.
Reducing aerodynamic drag can, therefore, be an effective way to reduce
fuel consumption and emissions.
Aerodynamic drag is proportional to the frontal area (A) of the
vehicle and coefficient of drag (Cd), such that aerodynamic
performance is often expressed as the product of the two values,
CdA, which is also known as the drag area of a vehicle. The
coefficient of drag (Cd) is a dimensionless value that
essentially represents the aerodynamic efficiency of the vehicle shape.
The
[[Page 24550]]
frontal area (A) is the cross-sectional area of the vehicle as viewed
from the front. It acts with the coefficient of drag as a sort of
scaling factor, representing the relative size of the vehicle shape
that the coefficient of drag describes. The force imposed by
aerodynamic drag increases with the square of vehicle velocity,
accounting for the largest contribution to road loads' higher speeds.
Aerodynamic drag reduction can be achieved via two approaches,
either by reducing the drag coefficient or reducing vehicle frontal
area, with two different categories of technologies, passive and active
aerodynamic technologies. Passive aerodynamics refers to aerodynamic
attributes that are inherent to the shape and size of the vehicle,
including any components of a fixed nature. Active aerodynamics refers
to technologies that variably deploy in response to driving conditions.
These include technologies such as active grille shutters, active air
dams, and active ride height adjustment. It is important to note that
manufacturers may employ both passive and active aerodynamic
technologies to achieve aerodynamic drag values.
The greatest opportunity for improving aerodynamic performance is
during a vehicle redesign cycle when significant changes to the shape
and size of the vehicle can be made. Incremental improvements may also
be achieved during mid-cycle vehicle refresh using restyled exterior
components and add-on devices. Some examples of potential technologies
applied during mid-cycle refresh are restyled front and rear fascia,
modified front air dams and rear valances, addition of rear deck lips
and underbody panels, and low-drag exterior mirrors. While
manufacturers may nudge the frontal area of the vehicle during
redesigns, large changes in frontal area are typically not possible
without impacting the utility and interior space of the vehicle.
Similarly, manufacturers may improve Cd by changing the
frontal shape of the vehicle or lowering the height of the vehicle,
among other approaches, but the form drag of certain body styles and
airflow needs for engine cooling often limit how much Cd may
be improved.
During the vehicle development process, manufacturers use various
tools such as Computational Fluid Dynamics (CFD), scaled clay models,
and full size physical prototypes for wind tunnel testing and
measurements to determine aerodynamic drag values and to evaluate
alternate vehicle designs to improve those values.
The agencies presented a table in the PRIA showing aerodynamic drag
improvements from individual technologies based on wind-tunnel testing
for a study commissioned by Transport Canada, which is reproduced in
Table VI-131 below.\1418\ The individual technologies are present in
many of the 2016 and 2017 vehicles in the fleet. Table VI-131 shows the
list of aerodynamic technologies and corresponding aero drag
improvements.
---------------------------------------------------------------------------
\1418\ Table 6-63 in PRIA.
[GRAPHIC] [TIFF OMITTED] TR30AP20.267
As discussed in the PRIA and further below, the agencies made
several notable changes for modeling aerodynamic improvement
technologies from the Draft TAR to the NPRM. First, the agencies
revised the aerodynamic
[[Page 24551]]
improvements from two levels in the Draft TAR (10% and 20% improvement
over the baseline) to four levels (5%, 10%, 15% and 20% aerodynamic
drag improvement values over the baseline). This change provided the
improved granularity to bin the vehicles with different aerodynamic
improvements more appropriately. Next, the agencies assigned levels of
aerodynamic technology to the MY 2016 fleet on a relative basis based
on confidential business information submitted by the manufacturers,
taking steps to verify information submitted by manufactures with other
sources, and making changes particularly for vehicles that showed large
improvements over baseline values. Third, the agencies limited the
maximum level of aerodynamic improvements that certain body styles
(pickup trucks, minivans) could achieve and limited the maximum level
of improvements that cars and SUVs with more than 405 horsepower could
achieve, based on the agencies' assessment of industry comments.
Finally, the agencies updated the cost for aerodynamic improvements
based on the assessment of comments that the National Academy of
Sciences (NAS) cost estimates used in the Draft TAR underestimated the
cost for aerodynamic improvements.
Broadly, Ford commented in support of the approach to aerodynamic
improvement modeling in the NPRM, stating that the rule recognized
potential constraints like consumer needs and preferences regarding
vehicle styling, vehicle utility, and interior space, by among other
things, recognizing that the potential for aerodynamic drag differs
among different vehicle body styles and vehicle classes.\1419\ Ford
stated that these are major factors considered by customers when
comparing competing vehicles, and the failure of a manufacturer to
deliver in these areas can lead to the production of non-competitive,
poor-selling vehicles.
---------------------------------------------------------------------------
\1419\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
On the other hand, ICCT claimed that the agencies greatly limited
the availability of many load reduction technologies (i.e., mass
reduction improvements, aerodynamic improvements, and rolling
resistance improvements) by pushing very large amounts of these
technologies into the 2016 model year baseline fleet, thereby making
the technologies unavailable for use in future years.\1420\ ICCT
commented that these improvements in the analysis fleet would
ostensibly amount to massive efficiency improvements, however, these
assumed changes were not substantiated as resulting in any test-cycle
efficiency improvements in the model year 2016 fleet versus the 2015
fleet. ICCT concluded that the adjusted baseline had been developed and
presented opaquely, apparently based primarily upon estimations from
automaker-supplied data, without critical analysis, vetting, or sharing
of the necessary data to substantiate the changes and real-world
benefits by the agencies.
---------------------------------------------------------------------------
\1420\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
As discussed further in Section VI.C.5.b) AERO drag analysis fleet
assignments below, the agencies believe the updated analysis fleet
aerodynamic technology level assignments in the NPRM analysis represent
an improvement over the MY 2015 assignments in the Draft TAR, as the
updated assignments are based on precise values, not estimated from
road load coefficients, and have been corroborated by observed
improvements on actual production vehicles. Accordingly, the agencies
carried over the NPRM approach for determining the aerodynamic
technology levels for the analysis fleet to the final rule.
a) Aerodynamics Drag Reduction Modeling in the CAFE Model
The agencies summarized in the PRIA that the Draft TAR aerodynamic
improvement levels were binned into two groups, AERO1 and AERO2.
However, market observations showed that many vehicles had aero
improvements from 0% to 10%, and some vehicles showed improvements from
10% to 20%.\1421\ Based on industry feedback and market observations,
the agencies revised the aerodynamic improvements from two levels in
the Draft TAR (10% and 20% improvement over the baseline) to four
levels (5%, 10%, 15% and 20% aerodynamic drag improvement values over
the baseline). This revision provided the necessary granularity to bin
the vehicles with different aerodynamic improvements appropriately.
---------------------------------------------------------------------------
\1421\ PRIA at 437.
---------------------------------------------------------------------------
ICCT commented that to model appropriately the baseline standards,
the agencies would need to include increasing use of aerodynamic off-
cycle technology credits across all companies through 2025. ICCT stated
that it appeared that the agencies did not use EPA's engineering
expertise or compliance data, where EPA would be able to advise better
based on their certification data from the off-cycle program.
As discussed further in Sections VI.A and VI.C.8, the NPRM analysis
carried forward manufacturers' off-cycle fuel consumption improvement
values (FCIVs) at MY 2016 levels unless an explicitly simulated off-
cycle technology, like start-stop systems, was added to a vehicle in
the simulation modeling. Specific to aerodynamic improvements, active
grille shutters were assumed to be applied at the 20 percent
aerodynamic improvement (AERO20) level. For the final rule analysis,
based on the assessment of comments that the application of off-cycle
technologies in the analysis was too conservative, the agencies agreed
and increased each manufacturers' application of off-cycle technologies
so that 10 g/mi of technology was applied by 2023, using an
extrapolated increase in levels in MYs 2017-2023 based on EPA
compliance data.\1422\ This approach did not assume any specific mix of
off-cycle technologies that would be used by manufacturers to achieve
the 10 g/mi off-cycle improvement, because manufactures currently use a
variety of technologies, and different manufacturers likely would
implement unique combinations of technologies. It is expected that
aerodynamic off-cycle technologies would be included in the mix of off-
cycle technologies.
---------------------------------------------------------------------------
\1422\ The 2018 EPA Automotive Trends Report, https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends.
---------------------------------------------------------------------------
Table VI-132 and Table VI-133 show aerodynamic technologies that
could be used to achieve 5%, 10%, 15% and 20% aero improvements in
passenger cars, SUVs, and pickup trucks.\1423\ The agencies developed
these potential combinations of technologies using aerodynamic data
from a National Research Council (NRC) of Canada sponsored wind tunnel
testing program that included an extensive review of production
vehicles utilizing these technologies, and industry
comments.1424 1425 These technology combinations are
intended to show a potential way for a manufacturer to achieve each
aerodynamic improvement level; however, in the real world,
manufacturers may implement different combinations of aerodynamic
technologies to achieve a percentage
[[Page 24552]]
improvement over their baseline vehicles.
---------------------------------------------------------------------------
\1423\ Table 6-67 and Table 6-68 in PRIA.
\1424\ Larose, G., Belluz, L., Whittal, I., Belzile, M. et al.,
``Evaluation of the Aerodynamics of Drag Reduction Technologies for
Light-duty Vehicles--a Comprehensive Wind Tunnel Study,'' SAE Int.
J. Passeng. Cars--Mech. Syst. 9(2):772-784, 2016, https://doi.org/10.4271/2016-01-1613.
\1425\ Larose, Guy & Belluz, Leanna & Whittal, Ian & Belzile,
Marc & Klomp, Ryan & Schmitt, Andreas. (2016). Evaluation of the
Aerodynamics of Drag Reduction Technologies for Light-duty
Vehicles--a Comprehensive Wind Tunnel Study. SAE International
Journal of Passenger Cars--Mechanical Systems. 9. 10.4271/2016-01-
1613.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.268
[[Page 24553]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.269
BILLING CODE 4910-59-C
b) Aerodynamic Drag Reduction Analysis Fleet Assignments
The agencies described in the PRIA that for the 2015 analysis fleet
used in the Draft TAR, the agencies received Cd values for
the MY 2015 vehicles' baseline assignments from manufacturers, or used
estimated Cd values. In response, the industry commented
that Cd values often varied by measurement approach and,
therefore, it was important to account for differences in the
methodologies used to estimate those values. For instance, aerodynamic
drag coefficients for the same vehicle often vary significantly from
wind-tunnel to wind-tunnel, complicating cross-comparison and cross-
referencing.\1426\ The industry commented that, on average, the
manufacturer-reported Cd values are nine percent lower than
the values reported by USCAR.\1427\ For reference, USCAR follows the
SAE J2881 test procedure. However, because Cd values are not
required to be reported for compliance, manufacturers can and do choose
different methods to estimate the Cd values. Therefore, the
industry commented that assigning baseline aerodynamic improvement
levels should not simply be comparing the lowest reported Cd
value in a vehicle segment to other reported Cd values. The
industry commented that such a comparison would not reflect the
plausible amount of aerodynamic drag improvement that could be
achieved. Accordingly, the industry suggested that the analysis should
normalize manufacturer-reported Cd values using SAE J2881.
---------------------------------------------------------------------------
\1426\ PRIA at 435.
\1427\ Footnote in PRIA at 435: FCA Draft TAR comments. Docket
ID: NHTSA-2016-0068-0082.
---------------------------------------------------------------------------
The commenters stated manufacturers have the option to use other
methods (apart from coast down testing) to estimate the Cd
values such as wind tunnel testing, cross referencing the Cd
value from other vehicles with similar frontal design and aero
technologies deployed. Since manufacturers do not have to specify the
methodology used to estimate the Cd value, the agencies have
limited capability to make accurate comparisons of the Cd
value estimates from different testing methods. As a result, the
agencies determined using average(s) of the fleet provide a better
estimate of Cd levels than using the lowest Cd
value in the fleet to assign aerodynamic improvement levels. The
agencies determined it is appropriate to continue to use the NPRM
approach for the final rule.
The NPRM and final rule analysis used a relative performance
approach to assign the current aerodynamic technology level to a
vehicle. Different body styles offer different utility and have varying
levels of baseline form drag. In addition, frontal area is a major
factor in aerodynamic forces, and the frontal area varies by vehicle.
This analysis considered both frontal area and body style as utility
factors affecting aerodynamic forces; therefore, the analysis assumed
all reduction in aerodynamic drag forces come from improvement in the
Cd. Per the process outlined in NHTSA's section of the Draft
TAR,\1428\ the agencies computed an average Cd for each body
style segment in the MY 2015 analysis fleet from drag coefficients
published by manufacturers. By comparing the Cd among
vehicles sharing body styles, this allowed the agencies to estimate the
level of aerodynamic improvement present on specific vehicles.
---------------------------------------------------------------------------
\1428\ Draft TAR at 4-80.
---------------------------------------------------------------------------
While some small differences existed between the aggregate MY 2015
and MY 2016 data, the agencies retained the NHTSA-calculated MY 2015
average Cd as the baseline drag coefficient for nearly all
body styles. For pickup trucks, the agencies assigned a baseline drag
coefficient of 0.42, considering that a large portion of the pickups
sold in MY 2015 already included aerodynamic features assumed for
advanced levels of aero. The agencies harmonized the Autonomie
simulation baselines with
[[Page 24554]]
the analysis fleet assignment baselines to the fullest extent
possible.\1429\
---------------------------------------------------------------------------
\1429\ Often, vehicles assigned to technology classes do not
perfectly match up with simulated vehicles, but in most cases this
analysis assumed the aerodynamic effects and other specifications
were comparable and appropriate for use as proxies.
---------------------------------------------------------------------------
The agencies assigned levels of aerodynamic technology to the MY
2016 fleet based on confidential business information submitted by
manufacturers on aerodynamic drag coefficients, and from other
information sources such as in product release information. The
analysis referenced manufacturer-submitted data (if that data was
supplied), and the agencies took industry comments to Draft TAR into
account and closely reviewed the manufacturer-submitted Cd
data. In the few cases that manufacturers did not submit Cd
values as confidential business information, the agencies estimated the
Cd based vehicle attributes, design, and aero technologies
applied to that vehicle. The agencies noted that the Cd
values reported by some manufacturers showed high levels of improvement
relative to the previous model year or previous generation. In some
cases, the agencies contacted the manufacturers to further discuss
differences in Cd estimation methodologies. Where
appropriate, the agencies adjusted MY 2016 fleet Cd values
after consultation with the manufacturers and used these values to
assign baseline technology levels for each vehicle in the NPRM CAFE
model simulation.
The Alliance commented that the NPRM analysis fleet had more
appropriately assigned aerodynamic technology levels, and the
assignments were more accurate than the Draft TAR, where vehicles were
generally considered to have little aerodynamic improvement technology,
and the CAFE model would add aerodynamic improvement despite the fact
that manufacturers had already made significant improvements and there
was little opportunity remaining for more.\1430\ The Alliance concluded
that the Draft TAR approach ultimately led the CAFE model to under-
predict how much powertrain technology was required for compliance. The
Alliance also commented that it is possible to estimate aerodynamic
features of a vehicle using road load coefficients, but the process
requires various assumptions and is not very accurate. The Alliance
concluded that the agencies' use of CBI to assign initial aerodynamic
improvement values is an accurate and practical solution to support
correct baseline assignments.
---------------------------------------------------------------------------
\1430\ NHTSA-2018-0067-12039 at 136.
---------------------------------------------------------------------------
Ford commented that the use of actual data, like manufacturer
confidential information or other sources, to characterize better the
aerodynamic improvements already incorporated into the baseline fleet
is a substantial improvement over previous analyses that either assumed
no aero improvement due to insufficient data, or attempted to infer
Cd from the road load coefficients.\1431\ Ford stated that
attempting to infer Cd from road load coefficients is not
sufficiently accurate for a vehicle-level determination since the
aerodynamic component of the road load coefficients is inextricably
confounded with tire, transmission, and other parasitic losses. As part
of its comments that the proposed rule analysis recognized constraints
like consumer needs and preferences regarding vehicle styling and
utility, Ford stated that the baseline Cd for pickup trucks
properly recognized that these vehicles already include many advanced-
level aerodynamic technologies. Ford concluded that an accurate
assessment of the current technological state of the baseline fleet is
critical to ensuring that the benefits of technological improvements
are not ``double-counted'' in the modeling.
---------------------------------------------------------------------------
\1431\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
On the other hand, ICCT commented that the agencies artificially
limited the availability of aerodynamic technologies in the CAFE model
in future years by assigning approximately three times as many
aerodynamic technology packages in the 2016 analysis fleet as they did
in the 2015 baseline fleet used in the Draft TAR.\1432\ ICCT noted that
the 2015 Draft TAR fleet had about 8 percent vehicles with one of the
aerodynamic packages, whereas the NPRM's 2016 fleet had about 53
percent, and argued that the agencies did not justify the increase with
data to show that automakers actually deployed the technology. ICCT
pointed to the agencies' introduction of intermediate aerodynamic
improvement steps as the justification for the change, which ICCT
argued ``redistributes the baseline fleet into more advanced
aerodynamic levels without observing or verifying real-world
aerodynamic improvements.''
---------------------------------------------------------------------------
\1432\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
ICCT argued that if an improvement of this magnitude were true, it
would be evident in fleet level miles-per-gallon and CO2
levels (e.g., in EPA's Trends and Manufacturer Performance reports),
but none of the quantifiable mpg or CO2 benefits that would
be associated with these additional aerodynamic improvements were
reflected in any real-world evidence in the model year 2016 fleet. ICCT
stated that to show the automakers deployed this level of aerodynamic
improvements, the agencies needed to show data on how these
improvements are evident in the fleet and delivering benefits.
Specifically, ICCT stated that the agencies must share the basis for
any aerodynamic calculation and exact estimated percent improvement
(rather than binned percentage categories) for each vehicle make and
model in the baseline and future modeled fleet, and their technical
justification for each value, arguing that not doing so would obscure
the agencies' methods. In addition, ICCT stated that the agencies must
conduct two sensitivity analysis cases that assume that every baseline
make and model is set to 0 percent aerodynamic improvement and set to
the previous baseline aerodynamic levels (i.e., from TAR) to
demonstrate how much the agencies' decision to load up more baseline
technology affects the compliance scenarios. ICCT concluded that
because changes in aerodynamic improvement assumptions ``are opaquely
buried in the agencies' datafiles and unexplained,'' the agencies must
issue a new regulatory analysis and allow an additional comment period
for review of the methods and analysis.
ACEEE asserted, as part of its comments that the MY 2016 analysis
fleet assignments appeared to contain errors, that the assignment of
AERO10 for the MY 2016 Toyota Tundra pickup truck was in error.\1433\
ACEEE stated that Tundra pickup trucks have had similar specs from MY
2011 to today, and the Cd for all Tundra models has been
0.37 or 0.38 for 2WD and 4WD, respectively, since MY 2011. ACEEE noted
that this is higher than the AERO10 Cd cut off value of
0.355 for pickups, as shown in the 2016 Draft TAR and referenced in the
PRIA.
---------------------------------------------------------------------------
\1433\ NHTSA-2018-0067-12122, at 6.
---------------------------------------------------------------------------
As described above, the agencies assigned levels of aerodynamic
technology to the NPRM MY 2016 analysis fleet on a relative basis based
on confidential business information submitted by the manufacturers on
aerodynamic drag coefficients and other information sources such as in
product release information. In addition, based on the Draft TAR
comments, the agencies verified wherever possible the information
submitted by manufactures with other sources (product release
information and cross referencing with vehicles with similar design
features and aero technologies), and made
[[Page 24555]]
changes particularly for vehicles which showed large improvements over
baseline values. Figure 6-175 in PRIA presented the distribution of
different levels of aerodynamic drag improvements in MY 2016 vehicle
fleet in NPRM relative to MY 2015 vehicle fleet used in Draft TAR. The
distribution shows that 46 percent of the MY 2016 vehicle fleet was
assigned AERO0 (0 percent improvement), 31 percent of the fleet was
assigned AERO5 (5% improvement), and 15 percent of the vehicle fleet
was assigned AERO10 (10 percent improvement). This distribution clearly
shows that there is substantial opportunity for additional aerodynamic
drag improvements in the vehicle fleet.
Regarding comments by ACEEE on Toyota Tundra pickup trucks, as just
stated, the agencies used manufacturer submitted information and other
available information to assign aerodynamic technology levels and the
agencies applied the same process for all of the manufacturers for the
NPRM and for the final rule. The agencies did assign AERO10 for some
Toyota Tundra pickups, but not for all as asserted by ACEEE. Some of
the Toyota Tundra pickups with 2WD and short bed and crew cab or double
cab were assigned AERO5 and other configurations were assigned
AER10.\1434\ For reference, the baseline Cd value used in
the NPRM for pickups is 0.395; a 5 percent improvement in Cd value is
0.375 and 10 percent improvement in Cd value is 0.355. The agencies
considered the ACEEE comment and available information and determined
the aerodynamic assignments for the Toyota Tundra were reasonable for
the final rule analysis.
---------------------------------------------------------------------------
\1434\ The variations could be from coast down testing with
different powertrains and with different pickup bed length and crew
cab configurations.
---------------------------------------------------------------------------
Table VI-134 below shows the percentage aerodynamic drag
improvement assigned to the MY 2015 (Draft TAR), MY 2016 (NPRM) and MY
2017 (final rule) analysis fleets. It is clear from this table that
there is natural progression of aero technologies being adopted and the
vast majority of the MY 2017 vehicle fleet is at or below AERO10
(81percent).
[GRAPHIC] [TIFF OMITTED] TR30AP20.270
Moreover, notable aerodynamic improvements have actually been
observed on production vehicles. As described in PRIA, EPA observed 76
vehicles at the 2015 North American International Auto Show in Detroit
(2015 NAIAS).\1435\ EPA's observations showed that manufacturers have
widely deployed both active and passive aerodynamic drag reduction
technologies with significant opportunity remaining to apply aero
technologies further in more optimized fashion as vehicles enter
redesign cycles in the future.\1436\ Although EPA did not identify the
aerodynamic drag coefficient values for these vehicles, Figure 6-167 in
PRIA showed the distribution of some aero technologies identified by
EPA during this informal survey.
---------------------------------------------------------------------------
\1435\ PRIA at 432. See also Docket No. EPA-HQ-OAR-2015-0827.
\1436\ Draft TAR at 5-363.
---------------------------------------------------------------------------
The survey showed that wheel dams and underbody panels are the most
widely used aero technologies, followed by front bumper air dams and
active grill shutters. Since this survey, many pickup trucks and
passenger cars have active grill shutters installed to improve
aerodynamic drag, and to get off-cycle credit. Table 6-67 in PRIA shows
the ``active grill shutter'' by itself will improve aerodynamic drag
reduction improvement by 3 percent. Combined with other aero
technologies, this can improve the aerodynamic drag reduction values
significantly in pickup trucks and SUVs. As a result, there has been
overall fleet wide aerodynamic drag reduction improvement; however, the
above Table VI-134 shows that only 19 percent (13 percent from AERO10,
5 percent from AERO15 and 1 percent from AERO20) of the MY 2017 vehicle
fleet has aerodynamic drag reduction improvement greater than 10
percent. This shows that there is significant opportunity for the
vehicle fleet to improve aero technologies by MY 2025.
The agencies also described examples of how production vehicles in
different technology classes improved aerodynamic drag reduction values
relative to their previous generation model since the 2012 final
rule.\1437\ The PRIA described how aerodynamic technologies were being
deployed on production vehicles, using the MY 2015 Nissan Murano and MY
2015 Ford F150 as examples. For example, MY 2015 Ford F150 has the
passive and active aerodynamic technologies as shown in Table VI-135.
---------------------------------------------------------------------------
\1437\ PRIA at 433.
---------------------------------------------------------------------------
The air curtain technology in the MY 2015 F150 guides the air flow
across the front wheels to reduce wind turbulence.\1438\ For reference,
the wind tunnel testing by NRC of the MY 2015 Ford F150 showed a drag
coefficient value of 0.37 while the coast down testing by EPA pegged
the drag coefficient value between 0.35 to 0.40 depending on the type
of powertrain, cab and cargo box combination. The prior generation F150
was released in 2008 as a MY 2009 and this vehicle had
[[Page 24556]]
very few aerodynamic technologies applied. The agencies do not have the
MY 2009 Cd value to estimate the percentage improvement.
Since the F150 also included significant light weighting and powertrain
improvements including a downsized turbocharged engine, the
effectiveness improvement attributable to aerodynamic technologies is
uncertain.
---------------------------------------------------------------------------
\1438\ Ford, How Air Curtains on F-150 Help Reduce Aerodynamic
Drag and Aid Fuel Efficiency (July 15, 2015), https://media.ford.com/content/fordmedia/fna/us/en/news/2015/07/15/how-air-curtains-on-f-150-help-reduce-aerodynamic-drag.html.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.271
The Nissan Murano is an example of a mid-size SUV with greater than
fifteen percent improvement in aerodynamic drag values compared to the
previous generation. The SAE paper published in 2015 outlines the
specifics of aerodynamics in the Nissan Murano,\1439\ and they include
those listed in Table VI-136 below.
---------------------------------------------------------------------------
\1439\ Arai, M., Tone, K., Taniguchi, K., Murakami, M. et al.,
``Development of the Aerodynamics of the New Nissan Murano,'' SAE
Technical Paper 2015-01-1542, 2015, https://doi.org/10.4271/2015-01-1542.
---------------------------------------------------------------------------
The exterior of this vehicle was completely redesigned from the MY
2013-2014 generation with the goal of minimizing aerodynamic drag by
combining passive aerodynamic devices with an optimized vehicle shape.
The primary passive devices employed include optimization of the rear
end shape to reduce rear end drag, and addition of a large front
spoiler to reduce underbody air flow and redirect it toward the roof of
the vehicle, thus augmenting the rear end drag improvements. Other
passive improvements include plastic fillet moldings at the wheel
arches, raising the rear edge of the hood, shaping the windshield
molding and front pillars, engine under-cover and floor cover, and air
deflectors at the rear wheel wells. An active lower grille shutter also
redirects air over the body when closed. Together, these measures for
the MY 2015 model achieved a drag coefficient of 0.31, representing a
16 to 17 percent improvement over the 0.37 Cd of the
previous model.
[[Page 24557]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.272
[[Page 24558]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.273
BILLING CODE 4910-59-C
A combination of a slightly lighter MY 2015 Nissan Murano (on
average lighter by 94 lbs. considering all trim levels), relative to
the previous generation, and engine improvements (comparing 3.5L V6 in
MY 2014 to 3.5L V6 in MY 2015), and transmission improvements resulted
in an overall improvement in fuel economy.\1440\ Accordingly, the real-
world fuel economy improvement directly attributable to the package of
aerodynamic technologies included on either vehicle is uncertain, as
each vehicle included other fuel economy improving technologies along
with the improvements in aerodynamic technologies.
---------------------------------------------------------------------------
\1440\ https://www.fueleconomy.gov/feg/Find.do?action=sbs&id=34457&id=37198 (last visited 12.12.2019) shows
20 mpg (combined) in MY2014 Nissan Murano (3.5L VQ35DE V6 with
Variable gear ratio transmission) and 24 mpg (combined in MY2015
Nissan Murano (3.5L VQ35DE V6 with Automatic AV S7 transmission)).
---------------------------------------------------------------------------
The agencies considered a sensitivity case that assumed no mass
reduction, rolling resistance, or aerodynamic improvements had been
made to the MY 2017 fleet (i.e., setting all vehicle road levels to
zero--MRO, AERO and ROLL0), in response to ICCT's comment. While this
is an unrealistic characterization of the initial fleet, the agencies
conducted a sensitivity analysis to understand any affect it may have
on technology penetration along other paths (e.g., engine and hybrid
technology). Under the CAFE program, the sensitivity analysis shows a
slight decrease in reliance on engine technologies (HCR engines,
turbocharge engines, and engines utilizing cylinder deactivation) and
hybridization (strong hybrids and plug-in hybrids) in the baseline
(relative to the central analysis). The consequence of this shift to
reliance on lower-level road load technologies is a reduction in
compliance cost in the baseline of about $300 per vehicle (in MY 2026).
As a result, cost savings in the preferred alternative are reduced by
about $200 per vehicle. Under the CO2 program, the general
trend in technology shift is less dramatic (though the change in BEVs
is larger) than the CAFE results. The cost change is also comparable,
but slightly smaller ($200 per vehicle in the baseline) than the CAFE
program results. Cost savings under the preferred alternative are
further reduced by about $100. With the lower technology costs in all
cases, the consumer payback periods decreased as well. These results
are consistent with the approach taken by manufacturers who have
already deployed many of the low-level road load reduction
opportunities to improve fuel economy.
Second, as discussed above, EPA's baseline aerodynamic levels in
the Draft TAR were based on road load coefficients, leading to baseline
assignments that were not accurate. In the NPRM, the agencies discussed
in the tradeoffs between building the analysis fleet using confidential
information from manufacturers and publicly available data on the
vehicles.\1441\ In the case of drag coefficient values, which cannot be
gleaned from publicly available information, except in cases where a
manufacturer chooses to publicly release that data, or by simply
observing a vehicle, the agencies decided that the improved accuracy
associated with using manufacturer-provided Cd values
outweighed the benefits of using publicly releasable Cd
estimates based on road load coefficients, especially as manufacturer-
provided Cd values are only used to assign initial
aerodynamic improvement levels relative to Cd values for
each body style segment in the analysis fleet.
---------------------------------------------------------------------------
\1441\ 83 FR 43004.
---------------------------------------------------------------------------
In addition, manufacturers had submitted comments that the Draft
TAR approach to baseline fleet assignments had underestimated
technology already present on vehicles, leading the analysis to apply
more aerodynamic drag reduction technology than could be applied in the
real world. In response to those comments, as described in the Proposed
Determination TSD, EPA stated that they ``agree[ ] with the commenters
that it is appropriate to account for aerodynamic drag reductions
already present in the baseline fleet in order to avoid overestimating
the amount of additional improvement that can be achieved at a given
cost.'' \1442\ Accordingly, EPA ``applied some level of aerodynamic
drag reduction to a significant portion of the MY2015 baseline fleet.''
\1443\ Consequently, the agencies believe that ICCT's statement that if
aerodynamic improvements between the MY 2015 analysis fleet used in the
Draft TAR and the MY 2016 analysis fleet were true it would be evident
in the fleet is incorrect. It is inappropriate to compare the Draft TAR
MY 2015 analysis fleet, which notably included too few aerodynamic
technology assignments, with the fleet's achieved fuel economy in the
real world. The agencies disagree
[[Page 24559]]
with ICCT that the availability of aerodynamic technologies was
artificially limited by appropriately assigning baseline aerodynamic
technology levels in the analysis fleet.
---------------------------------------------------------------------------
\1442\ Proposed Determination TSD at 2-406.
\1443\ Proposed Determination TSD at 2-408.
---------------------------------------------------------------------------
This also relates to ICCT's comment that the agencies must share
the basis for any aerodynamic calculation and exact estimated percent
improvement (rather than binned percentage categories) for each vehicle
make and model in the baseline and future modeled fleet, and their
technical justification for each value. As discussed above, the
agencies shared the relative performance approach methodology for
assigning baseline aerodynamic levels to vehicles in the analysis fleet
in detail in the PRIA,\1444\ and this approach was the basis for the
aerodynamic calculation performed for every vehicle make and model in
the analysis fleet. The agencies provided the summary of aerodynamic
drag coefficients (including averages for MY 2016 vehicles) by vehicle
body style,\1445\ and the baseline aerodynamic improvement assignments
for each vehicle model were included in the
2018_NPRM_market_inputs_ref.xlsx. In addition, because aerodynamic drag
information from manufacturers is provided as confidential business
information, the agencies are unable to disclose that specific
information. However, as discussed above, the agencies are closely
examining the data provided and comparing it to other available
information to assess the best estimate for aerodynamic technology for
each vehicle in the analysis fleet.
---------------------------------------------------------------------------
\1444\ PRIA at 441.
\1445\ PRIA at 443.
---------------------------------------------------------------------------
For these reasons, the agencies continued to use the NPRM
methodology to assign aerodynamic drag reduction improvements for the
MY 2017 vehicle fleet for this final rule.
c) Aerodynamic Drag Technology Adoption Features
As discussed above, the agencies used a relative performance
approach to assign current aerodynamic technology level to a vehicle.
For some body styles with different utility, such as pickup trucks,
SUVs and minivans, frontal area can vary, and this can affect the
overall aerodynamic drag forces. In order to maintain vehicle utility
and functionality related to passenger space and cargo space, the
agencies assumed all technologies that improve aerodynamic drag forces
would do so through reducing the Cd while maintaining
frontal area.
In the NPRM, the agencies noted that the Proposed Determination
analysis assumed that some vehicles from all body styles could (and
would) reduce aerodynamic forces by 20 percent, which in some cases led
to future pickup trucks having aerodynamic drag coefficients better
than some of today's typical cars, if frontal area were held constant
in order to preserve interior space and cargo space. The agencies
further noted that for some vehicle types, there was limited practical
capability to significantly improve aerodynamic drag coefficients over
baseline levels. In those cases, the agencies deemed the most advanced
levels of aerodynamic drag simulated as not technically practicable
given the need to maintain vehicle functionality and utility, such as
interior volume, cargo area, and ground clearance.
The industry had also commented in response to EPA's Proposed
Determination on the difficulty to achieve AERO20 improvements for
certain body styles. In the NPRM, the agencies considered the industry
comments along with the observations made in the MY 2016 fleet, and
tentatively determined the maximum feasible improvement in
Cd that could be achieved for pickup trucks is AERO15.\1446\
Similarly, the agencies determined the maximum feasible improvement in
Cd that could be achieved for minivans is AERO10. Next, the
NPRM analysis did not apply 15 percent or 20 percent aerodynamic drag
coefficient reduction to cars and SUVs with more than 405 horsepower.
The agencies noted that many high-performance vehicles already include
advanced aerodynamic features despite middling aerodynamic drag
coefficients. In these high-performance vehicle cases, the agencies
recognized that manufacturers tune aerodynamic features to provide
desirable downforce at high speeds and to provide sufficient cooling
for the powertrain, and, therefore, manufacturers may have limited
ability to improve aerodynamic drag coefficients for high performance
vehicles with internal combustion engines without reducing horsepower.
Accordingly, the agencies did not allow application of AERO15 and
AERO20 technology for all vehicles with more than 405 HP. Approximately
400,000 units of volume in the MY 2016 market data file included
limited application of aerodynamic technologies because of vehicle
performance. The agencies sought comment on limiting the Cd
improvement in these circumstances.
---------------------------------------------------------------------------
\1446\ The agencies noted in the NPRM that although ANL created
full-vehicle simulations for trucks with 20 percent drag reduction,
those simulations were not used in the CAFE modeling. The agencies
concluded that level of drag reduction was likely not
technologically feasible with today's technology, and the analysis
accordingly restricted the application of advanced levels of
aerodynamics in some instances, such as in that case, due to
bodystyle form drag limitations.
---------------------------------------------------------------------------
Ford commented in support of the agencies' decision to limit the
application of AERO20 on pickup trucks, noting that limiting AERO20 on
pickups is appropriate given the high inherent form drag associated
with pickups' aerodynamic profile.\1447\
---------------------------------------------------------------------------
\1447\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
CARB commented that the agencies excluded AERO20 inconsistently
across the fleet, noting that while some of the restrictions may be
valid, the broad rule the agencies used resulted in technology being
inappropriately excluded from too many vehicles.\1448\ Specifically,
CARB took issue with the majority of luxury sedans and SUVs being
excluded from AERO20 because they had high horsepower engines, while
the agencies did assign AERO20 to vehicles like the Tesla Model S and
Model X SUVs, which have horsepower in excess of 405. CARB stated that
while electrification provides a higher motivation to minimize road
load through technologies such as aerodynamic reductions, implementing
AERO20 reductions on high horsepower sedans and SUVs is clearly
feasible and should not be artificially restricted in the CAFE model.
---------------------------------------------------------------------------
\1448\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
In addressing these comments, the agencies considered the relative
cooling requirements for all electric powertrains and for high
performance internal combustion engine powertrains since airflow
diverted for cooling adversely impacts a vehicle's Cd. The
peak heat rejection and engine cooling needs for high performance
internal combustion engines is significantly higher than for all
electric powertrains. Internal combustion engines convert a lower
percentage of energy contained in gasoline into mechanical work (and
other useful work, such as lighting and sound), and the energy not
converted into mechanical work (or other useful work) is converted into
heat. A significant amount of the waste heat must be handled by the
cooling systems. Battery electric vehicles convert most of the
electrical energy stored in the battery into mechanical work and other
useful work, and therefore convert less energy into heat that must be
handled by the cooling system. Also, electric powertrains can provide a
degree of electric braking, whereas internal combustion engines
exclusively use friction braking, which generates heat and requires
greater cooling,
[[Page 24560]]
particularly on vehicles with substantial braking performance
capabilities. In the case of high-performance BEVs, since the cooling
needs are not as demanding as with high-performance vehicles that use
internal combustion engines, manufacturers can (and do, as can be
observed in the fleet) apply higher levels of aerodynamic technologies.
The agencies believe it is appropriate to account for these differences
in considering the amount of aerodynamic improvement that can be
implemented, and determined there are valid technical reasons for
allowing BEVs with greater than 405 horsepower to adopt AERO20
technology.
d) Aerodynamic Drag Technology Effectiveness
The NPRM analysis included four levels of aerodynamic improvements,
AERO5, AERO10, AERO15, and AERO20, representing 5, 10, 15, and 20
percent Cd improvements, respectively. Notably, the NPRM
analysis assumed that aerodynamic drag reduction could only come from
reduction in the aerodynamic drag coefficient and not from reduction of
frontal area, to maintain vehicle functionality and utility, such as
passenger space, ingress/egress ergonomics, and cargo space.\1449\
---------------------------------------------------------------------------
\1449\ 83 FR 43047.
---------------------------------------------------------------------------
Ford commented in support of the agencies' decision to consider the
frontal area and body style as ``utility factors'' and requiring that
aerodynamic improvements come from reductions in Coefficient of Drag
(Cd) and not from reductions in frontal area.\1450\
---------------------------------------------------------------------------
\1450\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
CBD commented that EPA staff had critiqued NHTSA's characterization
of research on aerodynamic drag coefficients and the NPRM did not
appear to incorporate or respond to this input.1451 1452
Specifically, CBD stated that EPA staff had commented in response to
the characterization that ``[f]or some bodystyles, the agencies have no
evidence that manufacturers may be able to achieve 15 percent or 20
percent aerodynamic drag coefficient reduction relative to baseline
(for instance, with pickup trucks'' and noted that ``[i]n the past, EPA
has assigned aero tech in the baseline relative to a ``Null'' and then
applied drag reduction level against that Null in order to ensure that
the maximum aero level (i.e., 15 or 20 percent) would always be
achievable for all body styles.'' This comment reflects deliberative,
in-process input from EPA staff. In fact, the NPRM text was developed
by the agencies with the benefit of this and other input from EPA
staff, and the NPRM clarified that reducing frontal area would likely
degrade other utility features like interior volume or ingress/egress.
---------------------------------------------------------------------------
\1451\ NHTSA-2018-0067-12000, at 188.
\1452\ Docket No. EPA-HQ-OAR-2018-0283-0453, June 29, 2018
Comments at 93.
---------------------------------------------------------------------------
CARB commented, as part of its broader comments, that the agencies'
effectiveness values were reduced relative to what EPA's LPM
calculated, that the benefits of aerodynamic improvements were
underestimated.\1453\ Specifically, CARB cited the H-D Systems
comparison of LPM benefits for AERO10 and AERO20 of 2.1 percent and 4.3
percent, respectively, compared with Autonomie benefits of 1.51 percent
and 3.03 percent, respectively, and stated that the agencies' analysis
provided no description or cited any new data or evidence as to why
they reduced the projected assumptions compared to what EPA's Lumped
Parameter Model calculated.
---------------------------------------------------------------------------
\1453\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
HDS also commented that the Autonomie modeling assumed no engine
change when aerodynamic drag and rolling resistance reductions were
implemented, as well as no changes to the transmission gear ratios and
axle ratios, which vary by transmission type but not by the tractive
load.\1454\ HDS stated that the EPA ALPHA model adjusted for this
effect, which accounted for the difference in technology effectiveness
estimates that HDS characterized between the Draft TAR and NPRM. HDS
provided a ``correct estimate'' for AERO20 effectiveness improvements
of 4.3 percent, with the justification that there was no gear/axle
ratio adjustment in the Autonomie analysis.
---------------------------------------------------------------------------
\1454\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
In response to HDS's comment, the Alliance submitted supplemental
comments questioning the extent to which aerodynamics (and changes in
top gear ratio) affect performance metrics held constant in the
analysis, like low- and high-speed acceleration performance and
gradeability.\1455\ The Alliance cited a study for the proposition that
vehicle acceleration is most influenced by engine power and weight, and
also that bodystyle differences have a lesser impact on acceleration
performance. The Alliance further commented that ``[r]egarding changes
in top gear ratios in response to aerodynamic changes, the Alliance is
not aware of any examples in which a top gear ratio was changed solely
due to aerodynamic improvements. There may be examples where a
vehicle's top gear ratio was changed at the same time aerodynamic
changes were made, but such changes would be made in response to the
cumulative changes across the entire vehicle, not just aerodynamic
improvements.'' The Alliance concluded that ``[t]here are also
practical manufacturing and investment constraints which limit the
potential for applying engine changes in response to improved vehicle
aerodynamics,'' citing the agencies decision to only resize engines
with significant design changes, to account for product complexity and
economies of scale.
---------------------------------------------------------------------------
\1455\ NHTSA-2018-0067-12385, at 31-32.
---------------------------------------------------------------------------
In response to the Alliance's supplemental comment, HDS submitted
supplemental comments stating that ``[d]rag reduction is usually
accomplished when a vehicle body is redesigned, so gear and axle ratios
are typically re-optimized for the entire set of changes, but these
changes include the drag reduction.'' \1456\ HDS commented that the
Alliance's comments acknowledged that calibration changes are made in
response to tractive load changes, while the Autonomie analysis
recalibrates the powertrain in response only to large mass reduction
improvements, and not any other vehicle changes that reduce tractive
load, like aerodynamic improvements, even when those changes would
result in a greater tractive load reduction than a 10 percent mass
reduction. HDS reiterated its statement that ``[i]n the real world (and
as captured in EPA's prior ALPHA model), automakers typically alter
many vehicle attributes affecting tractive load simultaneously,
including aerodynamics,'' and the Autonomie outputs underrepresent the
benefit of tractive load reduction strategies by not optimizing engine
efficiency after most changes in tractive load because the model
employees fixed shift points, gear ratios, and axle ratios when drag or
tire rolling resistance is reduced.
---------------------------------------------------------------------------
\1456\ NHTSA-2018-0067-12395, at 4-5.
---------------------------------------------------------------------------
Regarding the first set of comments that the aerodynamic
effectiveness values were reduced from EPA's values presented in the
Draft TAR, that results from differences in the two modeling
approaches. As discussed above, for this analysis the agencies decided
that aerodynamic drag reduction could only come from reduction in the
aerodynamic drag coefficient, and not from a reduction in vehicle
frontal area, at least without reducing other attributes of the
vehicle. EPA's process for assigning road load technologies to baseline
vehicles used road load coefficients from coast downs, which aggregated
individual aero, mass and tire reduction technologies. In contrast,
[[Page 24561]]
the CAFE Model and Autonomie used individually assigned road load
technologies for each vehicle to appropriately assign initial road load
and to appropriately capture benefits of subsequent individual road
load technologies. The differences in using road load coefficients from
coast downs and individually isolating the improvements from existing
and future road load technologies in the Autonomie modeling resulted in
the differences noted by commenters. And so, the resulting
effectiveness from the incremental adoption of individual technologies
to a newer analysis fleet will have different result than what was
estimated by the previous analyses. For further discussion of the
analysis fleet see Section VI.B.1.
In Section VI.B.3 Tech Effectiveness and Modeling and Section
VI.C.2 Transmissions, the agencies provide a full discussion of the
issues associated with assuming the engine and transmission can be
optimized for every combination of technologies. It would be
unreasonable and unaffordable to resize powertrains, including engines
and transmission and axle ratios, for every unique combination of
technologies, and exceedingly so for every unique combination
technologies across every vehicle model due to the extreme
manufacturing complexity that would be required to do so. Product
complexity and economies of scale are real, and in the NPRM, engine
resizing was limited to specific incremental technology changes that
would typically be associated with a major vehicle or engine
redesign.\1457\ As noted by HDS, the EPA Draft TAR and Proposed
Determination analyses adjusted the effectiveness of every technology
combination, including for aerodynamics technologies, assuming
performance could be held constant for every combination. However,
those analyses did not recognize or account for the extreme complexity
nor the associated costs for that impractical assumption. The NPRM and
final rule analyses account for these real-world practicalities and
constraints, and doing so explains some of the effectiveness and cost
differences between the Draft TAR/Proposed Determination and the NPRM/
final rule. The agencies believe the NPRM and the final rule approach
appropriately resizes powertrain components for specific incremental
technology changes that would typically be associated with a major
vehicle or engine redesign.
---------------------------------------------------------------------------
\1457\ See 83 FR 43027 (Aug. 24, 2018).
---------------------------------------------------------------------------
For the NPRM, and carried into the final rule analysis, Autonomie
simulates all road load conditions (e.g., MR, AERO, and ROLL technology
levels) for each engine and transmission combination. In addition,
engines are resized for appropriate specific technology changes that
would be associated with a major vehicle or engine redesign. Also, as
discussed further in Section VI.C.2 Transmissions, many commenters
seemed to conflate the practice in the analysis of using a common
(same) gear set across vehicle configurations (to address manufacturing
complexity) with using the same shift maps. As commenters stated, they
assumed the same shift maps were applied across vehicle models.
However, the shift initializer routine was run for every unique
Autonomie full vehicle model configuration and generated customized
shifting maps. The algorithms' optimization was designed to balance
minimization of energy consumption and vehicle performance. This
balance was necessary to achieve the best fuel efficiency while
maintaining customer acceptability by meeting performance neutrality
requirements. The agencies believe the level of optimization of engine
size, transmissions, gear ratios and shift schedules reasonably
approximate what is achievable and what manufacturers actually do.
Figure VI-47 below shows the range effectiveness used for AERO
technologies for the NPRM analysis.
BILLING CODE 4910-59-P
[[Page 24562]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.274
Figure VI-48 below shows the range of aero effectiveness used for
the final rule analysis.
[[Page 24563]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.276
BILLING CODE 4910-59-C
e) Aerodynamic Drag Technology Cost
For the Draft TAR, the agencies relied on the 2015 NAS report to
estimate the cost of AERO1 and AERO2 levels of aerodynamic drag
coefficient improvements. The agencies received several comments
related to the cost assumptions used in the Draft TAR, mainly that they
were too low to meet AERO1 and AERO2 levels. The industry submitted
confidential business information on the costs of passive aerodynamic
technologies needed to achieve AERO1 (10 percent improvement in drag
improvement), which showed a significantly higher estimated costs than
assumed for the Draft TAR. Similarly, the industry submitted
confidential business information on the costs of active aerodynamic
technologies, including some high cost technologies. The industry also
commented that some active aerodynamic technologies could only be
implemented during vehicle redesigns and not during a mid-cycle vehicle
refresh.
The agencies considered these comments and performed additional
research to assess the costs for passive and active aerodynamic
technologies. The agencies revised the cost estimates for the NPRM,
based in part on confidential information from the automotive industry,
and from the agencies' own assessment of manufacturing costs for
specific aerodynamic technologies from available sources. In general,
the NPRM cost estimates were higher than Draft TAR cost estimates. The
agencies included a high-level discussion in the PRIA that the cost to
achieve AERO5 is relatively low, as most of the improvements can be
made through body styling changes. The cost to achieve AERO10 is higher
than AERO5, due to the addition of several passive aero technologies,
and the cost to achieve AERO15 and AERO20 is higher than AERO10 due to
use of both passive and active aero technologies.
The agencies did not receive any comments on the costs of
aerodynamic improvements, and accordingly, for the final rule, as shown
in Table VI-137 and Table VI-138 below, the agencies used the same
aerodynamic improvement costs presented in NPRM.
[[Page 24564]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.277
[GRAPHIC] [TIFF OMITTED] TR30AP20.278
6. Tire Rolling Resistance
Tire rolling resistance is a road load force that arises primarily
from the energy dissipated by elastic deformation of the tires as they
roll. Tire design characteristics (for example, materials,
construction, and tread design) have a strong influence on the amount
and type of deformation and the energy it dissipates. Designers can
select these characteristics to minimize rolling resistance. However,
these characteristics may also influence other performance attributes,
such as durability, wet and dry traction, handling, and ride comfort.
Low rolling resistance tires are increasingly specified by OEMs in
new vehicles and are also increasingly available from aftermarket tire
vendors. They commonly include attributes such as higher inflation
pressure, material changes, tire construction optimized for lower
hysteresis, geometry changes (e.g., reduced aspect ratios), and reduced
sidewall and tread deflection. These changes are commonly accompanied
by additional changes to vehicle suspension tuning and/or suspension
design to mitigate any potential impact on other performance attributes
of the vehicle.
Lower-rolling-resistance tires have characteristics that reduce
frictional losses associated with the energy dissipated mainly in the
deformation of the tires under load, thereby improving fuel economy and
reducing CO2 emissions. The agencies considered two levels
of improvement for low rolling resistance tires in the analysis: The
first level of low rolling resistance tires considered reduced rolling
resistance 10 percent from an industry-average baseline, while the
second level reduced rolling resistance 20 percent from the baseline.
Walter Kreucher commented that the agencies should eliminate low
rolling resistance tires from the list of viable technologies, in
recognition of the safety impacts of low rolling resistance tires in
relation to stopping distance and accident rates.\1458\ Separately, Mr.
Kreucher argued that the model should reflect the safety impact of low
rolling resistance tires.
---------------------------------------------------------------------------
\1458\ NHTSA-2018-0067-0444.
---------------------------------------------------------------------------
The agencies have been following the industry developments and
trends in application of rolling resistance technologies to light duty
vehicles. As stated in the NAP special report on Tires and Passenger
Vehicle Fuel Economy,\1459\ cited by Mr. Kreucher, national crash data
does not provide data about tire structural failures specifically
related to tire rolling resistance, because the rolling resistance of a
tire at a crash scene cannot be determined. However, other metrics like
brake performance compliance test data
[[Page 24565]]
are helpful to show trends like that stopping distance has not changed
in the last ten years,\1460\ during which time many manufacturers have
installed low rolling resistance tires in their fleet--meaning that
manufacturers were successful in improving rolling resistance while
maintaining stopping distances through tire design, tire materials,
and/or braking system improvements. In addition, NHTSA has addressed
other tire-related issues through rulemaking,\1461\ and continues to
research tire problems such as blowouts, flat tires, tire or wheel
deficiency, tire or wheel failure, and tire degradation.\1462\ However,
there are currently no data connecting low rolling resistance tires to
accident or fatality rates.
---------------------------------------------------------------------------
\1459\ Tires and Passenger Vehicle Fuel Economy: Informing
Consumers, Improving Performance--Special Report 286 (2006),
available at https://www.nap.edu/read/11620/chapter/6.
\1460\ https://one.nhtsa.gov/cars/problems/comply/index.cfm.
\1461\ 49 CFR 571.138, Tire pressure monitoring systems.
\1462\ Tire-Related Factors in the Pre-Crash Phase, DOT HS 811
617 (April 2012), available at https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/811617.
---------------------------------------------------------------------------
With better tire design, tire compound formulations and improved
tread design, tire manufacturers have tools to balance stopping
distance and reduced rolling resistance. As stated in one article
referenced by Mr. Kreucher, tire manufacturers can use ``higher
performance materials in the tread compound, more silica as reinforcing
fillers and advanced tread design features'' to mitigate issues related
to stopping distance.\1463\ The agencies do not believe that there is
sufficient data or other information to support removing low rolling
resistance tires as a viable technology considered in the CAFE and
CO2 analysis at this time.
---------------------------------------------------------------------------
\1463\ Jesse Snyder, A big fuel saver: Easy-rolling tires (but
watch braking) (July 21, 2008), https://www.autonews.com/article/20080721/OEM01/307219960/a-big-fuel-saver-easy-rolling-tires-but-watch-braking. Last visited December 3, 2019.
---------------------------------------------------------------------------
HDS argued, as discussed further below, that based on available
data on current vehicle models and the likely possibility that there
would be additional tire improvements over the next decade, the
agencies should consider ROLL30 technology, or a 30 percent reduction
of tire rolling resistance over the baseline.\1464\
---------------------------------------------------------------------------
\1464\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
As stated in Joint TSD for the 2017-2025 final rule, tire
technologies that enable rolling resistance improvements of 10 and 20
percent have been in existence for many years.\1465\ Achieving
improvements of up to 20 percent involves optimizing and integrating
multiple technologies, with a primary contributor being the adoption of
a silica tread technology. Tire suppliers have indicated that
additional innovations are necessary to achieve the next level of low
rolling resistance technology on a commercial basis, such as
improvements in material to retain tire pressure, tread design to
manage both stopping distance and wet traction, and development of
carbon black material for low rolling resistance without the use of
silica to reduce cost and weight.\1466\ The agencies are continuously
monitoring these and other tire technology improvements. The agencies
believe that the tire industry is in the process of moving automotive
manufacturers towards the first level of low rolling resistance
technology across the vehicle fleet (10 percent reduction in rolling
resistance), and that 20 percent improvement is achievable in the
rulemaking timeframe. However, the agencies believe that at this time,
the emerging tire technologies that would achieve 30 percent
improvement in rolling resistance, like changing tire profile,
strengthening tire walls, or adopting improved tires along with active
chassis control,\1467\ among other technologies, will not be available
for commercial adoption in the fleet during the rulemaking timeframe.
As a result, the agencies decided not to incorporate 30 percent
reduction in rolling resistance technology for this final rule.
---------------------------------------------------------------------------
\1465\ EPA-420-R-12-901, at page 3-210.
\1466\ Assessment of Fuel Economy Technologies for Light-Duty
Vehicles (2011) at page 103.
\1467\ Mohammad Mehdi Davari, Rolling resistance and energy loss
in tyres (May 20, 2015), available at https://www.sveafordon.com/media/42060/SVEA-Presentation_Davari_public.pdf. Last visited
December 30, 2019.
---------------------------------------------------------------------------
a) Rolling Resistance Modeling in the CAFE Model
The two levels of rolling resistance technology considered in the
analysis include ROLL10 and ROLL20, which represent a 10 percent and 20
percent rolling resistance reduction from the baseline (ROLL0),
respectively.
To understand the following discussions about rolling resistance
analysis fleet assignments and effectiveness values, it is important to
understand how the agencies developed the baseline value (ROLL0) used
in prior analyses, and how the agencies developed the baseline value
used in the NPRM and final rule. In the Draft TAR, the agencies used
unique baseline rolling resistance coefficients for each vehicle class.
Specifically, the compact car class value was 0.0075, the midsize car
value was 0.008, the small SUV value was 0.0084, the midsize SUV value
was 0.0084, and the pickup truck value was 0.009. The PRIA described
that since the Draft TAR, the agencies had reassessed rolling
resistance values for contemporary tires through discussions with
vehicle manufacturers, tire manufactures, and independent bench
testing. Based on a thorough review of confidential business
information submitted by industry, and a review of other literature,
including the CARB/CONTROLTEC study mentioned below, the baseline
rolling resistance coefficient for all vehicle classes was updated to
0.009 for the NPRM analysis. The agencies concluded that the updated
baseline value brought the NPRM simulations into better alignment with
tires in the MY 2016 analysis fleet. The agencies also discussed that
updated value was consistent with the findings of the CONTROLTEC study
on vehicle road loads, sponsored by CARB.\1468\ The following figure
shows the distribution of estimated tire rolling resistance coefficient
values for the 1,358 MY 2014 vehicles evaluated in the CONTROLTEC/CARB
study.
---------------------------------------------------------------------------
\1468\ Technical Analysis of Vehicle Load Reduction Potential
for Advanced Clean Cars, https://www.arb.ca.gov/research/apr/past/13-313.pdf, page 39.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24566]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.279
BILLING CODE 4910-59-C
ICCT commented that it was ``quite confusing and perhaps
troubling'' that the agencies adopted a higher average rolling
resistance coefficient than that of the Draft TAR, ``as it would imply
that the fleet rolling resistance got worse, but the agencies are
deciding to provide baseline credit as if there was more rolling
resistance technology deployed.'' \1469\ ICCT stated that the change
appeared to be attributed to the agencies' use of CBI on tire rolling
resistance received since the Draft TAR.
---------------------------------------------------------------------------
\1469\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
As described in the PRIA, the values used in the Draft TAR
represented the ``Best in Class'' values in each of the vehicle classes
and this did not necessarily reflect the average ``Rolling Resistance
Coefficient'' (RRC) of the fleet. For the Draft TAR, the agencies did
not have access to manufacturer confidential business information and
relied on estimates from CONTROLTEC. As stated earlier, Figure VI-49
shows the distribution of the estimated RRC for 1,358 vehicles models.
The average RRC from the CONTROLTEC study (0.009) aligned with the NPRM
estimate which was based in part on manufacturer submitted confidential
business information. CONTROLTEC compared the estimated RRC data with
the values provided by Rubber Manufacturers Association (renamed as
USTMA-U.S. Tire Manufacturers Association) for original equipment
tires. The average RRC from the data provided by RMA was 0.0092,\1470\
compared to average of 0.009 from CONTROLTEC. CONTROLTEC attributed the
difference due to analysis assumption, tire loading during coast down
vs. load during tire testing, inflation pressure during coast down vs.
inflation pressure during tire testing, coast down test reporting
issues, tire types represented in the sample, tire break-in, and
advancement in tire rolling resistance since the time RMA collected the
data.
---------------------------------------------------------------------------
\1470\ Technical Analysis of Vehicle Load Reduction by
CONTROLTEC for California Air Resources Board (April 29, 2015) at
page 40.
---------------------------------------------------------------------------
CONTROLTEC also stated that RRC values for some vehicles fell below
the average RRC (indicating better performance) due to estimation
assumptions for vehicles where manufacturer data was not available, and
coast down test reporting issues.\1471\ Further, CONTROLTEC performed a
sensitivity study by mathematically removing aerodynamic contribution
from the coast down coefficients. It was observed that the average RRC
without the aerodynamic contribution is around 0.011. Accordingly, the
agencies believe that it was reasonable to use 0.009 as the average RRC
for the fleet for the NPRM and to continue to use that value for the
final rule, based on the latest available data from manufacturers and
alignment with the average RRC to the CONTROLTEC study estimate.
---------------------------------------------------------------------------
\1471\ Technical Analysis of Vehicle Load Reduction by
CONTROLTEC for California Air Resources Board (April 29, 2015) at
page 38.
---------------------------------------------------------------------------
H-D Systems (HDS) commented that the CONTROLTEC/CARB study showed
that there is a very significant fraction of the fleet with tire
rolling resistance coefficients above 10kg/1000 kg, and a small
percentage of vehicles with rolling resistance coefficients already at
0.05 or 0.06. HDS stated that NHTSA's baseline of 0.09 appeared ``a
little low but may be appropriate if the distribution was sales
weighted.'' HDS argued that a number of vehicle models already have
tires below 0.07, and the likelihood that there would be additional
tire improvements over the next decade are likely, meaning that ROLL30
technology--or a 30 percent reduction of the tire rolling resistance
coefficient to 0.063--is possible and appropriate for MY 2025.
Roush commented that rolling resistance is erroneously assumed to
be the same across different vehicle classes, and that rolling
resistance would vary depending upon the vehicle size, power,
acceleration and performance package.\1472\
---------------------------------------------------------------------------
\1472\ NHTSA-2018-0067-11984.
---------------------------------------------------------------------------
As explained earlier, the RRC values used in the CONTROLTEC study
were a combination of manufacturer information, estimates from coast
down tests for some vehicles, and application of tire RRC values across
other vehicles on the same platform. CONTROLTEC stated that some RRC
values were below the estimated average (showing significant
improvement from the baseline) due to assumptions that were
[[Page 24567]]
applied to some vehicles when manufacturer data was not available.
Further, some of the RRC estimates were based on vehicle coast down
tests which had errors.\1473\ As a result, some of the RRC values used
in the Draft TAR showed significant improvements (30 percent reduction
in rolling resistance relative to baseline), as observed by HDS. Based
on a review of manufacturer-submitted confidential business information
and other sources, the agencies are unaware of any tires in production
which have 30 percent reduction in rolling resistance relative to
baseline values.
---------------------------------------------------------------------------
\1473\ Technical Analysis of Vehicle Load Reduction by
CONTROLTEC for California Air Resources Board (April 29, 2015) at
page 38.
---------------------------------------------------------------------------
As stated earlier, the baseline values used for the Draft TAR
analysis were ``Best in Class'' values from the estimates developed by
CONTROLTEC and not representative of the average of the fleet or
average for the vehicle classes. For the NPRM, the agencies revisited
the ROLL technology assignments based on the RRC values provided by
manufacturers, and the average RRC for each of the vehicle class was
near the fleet average (RRC = 0.009). As shown in Figure VI-50, a vast
majority of the vehicles in the fleet are in the ROLL0 bin across the
different vehicle class, vehicle size, power, acceleration and
performance configurations. For these reasons, the agencies will
continue to use the fleet average of RRC = 0.009 as the baseline value
to assess ROLL technology improvements.
b) Rolling Resistance Analysis Fleet Assignments
As discussed above, NHTSA's Draft TAR analysis showed little
rolling resistance technology in the baseline fleet for three reasons:
the simulations used baseline values already reflecting best-in-class
tire rolling resistance, credible tire rolling resistance values for
all vehicles from bench data were not available to the agencies at the
time of Draft TAR, and few manufacturers submitted rolling resistance
values for the Draft TAR analysis.
For the NPRM, baseline (ROLL0) rolling resistance values were
updated to 0.009, and any better rolling resistance values were
assigned based on whether information indicated that vehicle had
technology at least 10 percent better than baseline (.0081 or better
for ROLL10), or at least 20 percent better than baseline (.0072 or
better for ROLL20). The agencies used confidential business information
provided by manufacturers to assign initial rolling resistance values
for each vehicle make and model.
The Alliance commented that the NPRM MY 2016 analysis fleet had
been updated with appropriate ratings of rolling resistance
improvements, compared to the Draft TAR where vehicles were generally
considered to have unimproved tires (meaning the Draft TAR assumed
additional improvements were more achievable than in reality).\1474\
The Alliance noted that the Draft TAR approach led to the CAFE model
adding additional tire rolling resistance improvements even though
manufacturers had already made significant improvements with that
technology. This meant that the real-world fleet had little remaining
opportunity for additional tire-related improvements, ultimately
leading to the Draft TAR analysis underpredicting the amount of
powertrain technology required for compliance.
---------------------------------------------------------------------------
\1474\ NHTSA-2018-006712039 at 136.
---------------------------------------------------------------------------
The Alliance noted that it is possible to estimate rolling
resistance features of a vehicle using road load coefficients, but the
process requires various assumptions and is not very accurate. The
Alliance concluded that the agencies' use of CBI to assign baseline
technology levels correctly was an accurate and practical solution.
Similarly, Ford commented in support of the agencies' low rolling
resistance tire assignments in the baseline fleet, stating that the
accuracy of the baseline fleet assessment had been considerably
improved using actual tire rolling resistance data.\1475\
---------------------------------------------------------------------------
\1475\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
HDS commented that the analysis fleet ``accounts for the
distribution of tires below 0.09 as 19% of vehicles in MY 2016 are
modeled as having used ROLL10 and 25% of vehicles as having used ROLL20
in the base year, but there is no accounting for the ~25% of vehicles
having RRC values 10 to 20% above the 0.09 RRC average.'' \1476\ HDS
concluded that ``[a] stricter accounting of the baseline and, possibly
setting specific lower limits for 2025 RRC by vehicle type (as done for
aero drag in the PRIA) will show significant additional fleetwide
effectiveness from RRC reduction which is a very cost-effective
technology.''
---------------------------------------------------------------------------
\1476\ NHTSA-2018-0067-11985 at 49.
---------------------------------------------------------------------------
ICCT commented that the agencies made a ``dramatic and
unjustified'' shift in baseline tire rolling resistance assignments
from the 2015 fleet used in the Draft TAR to the 2016 fleet used in the
NPRM.\1477\ ICCT noted that per the agencies' updated baseline value,
nearly 20 percent of all vehicles in the MY 2016 analysis fleet
achieved 0.0081 (or better) rolling resistance value, and more than 26
percent achieve 0.0072 (or better). ICCT argued that rather than
changing the definition of rolling resistance technology to include
improvements beyond the baseline, the agencies instead redefined the
technology available, reducing the number of vehicles that can use tire
improvements in future compliance years within the modeling framework,
which artificially forced companies to use other, more expensive
technologies.
---------------------------------------------------------------------------
\1477\ NHTSA-2018-0067-11741 full comments.
---------------------------------------------------------------------------
ICCT stated that to substantiate the baseline rolling resistance
assignments, the agencies need to show data on how these improvements
are evident in the fleet and delivering benefits. ICCT alleged that if
an improvement of that magnitude were true, it would be evident in
fleet level miles-per-gallon and CO2 levels; however, ``none
of the quantifiable mpg or CO2 benefits that would be
associated with these additional rolling resistance improvements were
reflected with any real-world evidence in the model year 2016 fleet.''
ICCT stated this seemed to be a case of the agencies ``artificially
burying efficiency technology in the baseline, rendering it unusable in
the post model year 2016 compliance scenarios.''
ICCT also stated that the agencies must share absolute road load
coefficients for each vehicle make and model in the baseline fleet, and
the technical justification for each value, in addition to conducting
two sensitivity analysis cases ``assum[ing] that every baseline make
and model is set to 0% rolling resistance improvement and set to the
previous baseline rolling resistance (from the Draft TAR) to
demonstrate how much the agencies' decision to load up more baseline
technology affects the compliance scenarios, as it appears that the
agencies may have made a unsupportable and non-rigorous assumption
about rolling resistance technology across the models.'' ICCT concluded
that because the changes were buried in the datafiles and unexplained,
the agencies must issue a new regulatory analysis and allow an
additional comment period for review of the methods and analysis.
Based on the comments from HDS and ICCT, the agencies reexamined
available tire rolling resistance data. The assignment of ROLL20
technology was revised for some vehicle models based on information on
the use of common tires across vehicles that shared a platform. As a
consequence, for the final rule, only 20 percent of the MY2017 vehicle
fleet is assigned ROLL20. The
[[Page 24568]]
agencies will continue to investigate additional methods to improve the
accuracy of this method, however as the Alliance and Ford noted, the
accuracy of the baseline levels had been significantly improved over
prior analyses by using actual tire RRC data. The agencies approach is
consistent with the NAS recommendation to have two ROLL technology
levels. The agencies determined that 30 percent rolling resistance
improvement while maintaining other tire characteristics is unlikely to
be available in the rulemaking timeframe.
The agencies considered a sensitivity case that assumed no mass
reduction, rolling resistance, or aerodynamic improvements had been
made to the MY 2017 fleet (i.e., setting all vehicle road levels to
zero--MRO, AERO and ROLL0), in response to ICCT's comment. While this
is an unrealistic characterization of the initial fleet, the agencies
conducted a sensitivity analysis to understand any affect it may have
on technology penetration along other paths (e.g. engine and hybrid
technology). Under the CAFE program, the sensitivity analysis shows a
slight decrease in reliance on engine technologies (HCR engines,
turbocharge engines, and engines utilizing cylinder deactivation) and
hybridization (strong hybrids and plug-in hybrids) in the baseline
(relative to the central analysis). The consequence of this shift to
reliance on lower-level road load technologies is a reduction in
compliance cost in the baseline of about $300 per vehicle (in MY 2026).
As a result, cost savings in the preferred alternative are reduced by
about $200 per vehicle. Under the CO2 program, the general
trend in technology shift is less dramatic (though the change in BEVs
is larger) than the CAFE results. The cost change is also comparable,
but slightly smaller ($200 per vehicle in the baseline) than the CAFE
program results. Cost savings under the preferred alternative are
further reduced by about $100. With the lower technology costs in all
cases, the consumer payback periods decreased as well. These results
are consistent with the approach taken by manufacturers who have
already deployed many of the low-level road load reduction
opportunities to improve fuel economy.
Figure VI-50 shows the distribution of ROLL technology for the
Draft TAR, NPRM and final rule. For the NPRM, 64 percent of the MY 2016
vehicle fleet was assigned ROLL0 and for the final rule, 59 percent of
the MY2017 vehicle fleet is assigned ROLL0. This shows that the
majority of the fleet is still at the ROLL0 technology level and there
is still significant opportunity for the vehicle fleet to improve ROLL
technology.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.280
BILLING CODE 4910-59-C
c) Rolling Resistance Adoption Features
In some cases, low rolling resistance tires can affect traction,
which may adversely impact acceleration, braking and handling
characteristics for some high-performance vehicles. Similar to past
rulemakings, the agencies recognized in the NPRM that to maintain
performance, braking and handling functionality, some high-performance
vehicles would not adopt low rolling resistance tire technology. For
cars and SUVs with more than 405 horsepower (hp), the agencies
restricted the application of ROLL20. For cars and SUVs with more than
500 hp, the agencies restricted the application of any additional
rolling resistance technology (ROLL10 or ROLL20). The agencies
developed these cutoffs based on a review of confidential business
information and the distribution of rolling resistance values in the
fleet.
Ford commented that the NPRM analysis appropriately limited the
application of ROLL technology where it would be infeasible or would be
at odds with the vehicles' intended function, characterizing that the
decision to restrict application of ROLL10 and ROLL20 for high
performance vehicles as reasonable.\1478\
---------------------------------------------------------------------------
\1478\ NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
Accordingly, the agencies continued with the NPRM methodology of
restricting certain ROLL technology for high performance vehicles. In
the final rule, the agencies restricted the ROLL technology to ROLL0
and ROLL10 for vehicles with greater than 405 hp and below 505hp. For
vehicles greater than 505hp, the agencies restricted the ROLL
technology to ROLL0.
d) Rolling Resistance Effectiveness Modeling and Resulting
Effectiveness Values
As discussed above, the agencies updated the baseline rolling
resistance value to 0.009, based on a thorough review of confidential
business information submitted by industry, and a review of other
literature. To achieve ROLL10 in the NPRM and for the final rule
analysis, the tire rolling resistance must be at least 10 percent
better than baseline (.0081 or better). To achieve ROLL20, the tire
rolling resistance must be at least 20 percent better than baseline
(.0072 or better).
HDS commented that the Autonomie modeling assumed no engine change
when drag and rolling resistance reductions were implemented, as well
as no change to the transmission gear ratios and axle ratios, which
vary by transmission type but not by the tractive
[[Page 24569]]
load.\1479\ HDS stated that ``reduction in rolling resistance is
accompanied by axle ratio adjustments so that the engine operates at
about the same load but at lower RPM. The EPA ALPHA model adjusts for
this effect, which accounts for the difference in benefit estimates''
between Autonomie and the ALPHA model simulations.
---------------------------------------------------------------------------
\1479\ NHTSA-2018-0067-11985.
---------------------------------------------------------------------------
As stated in Section VI.B.3 Tech Effectiveness and Modeling,
Autonomie builds performance-neutral vehicle models by resizing
engines, electric machines, and hybrid electric vehicle battery packs
only at specific incremental technology steps. To address product
complexity and economies of scale, engine resizing is limited to
specific incremental technology changes that would typically be
associated with a major vehicle or engine redesign.\1480\ Manufacturers
have repeatedly told the agencies that the high costs for redesign and
the increased manufacturing complexity that would result from resizing
engines for small technology changes preclude them from doing so. It
would be unreasonable and unaffordable to resize powertrains for every
unique combination of technologies, and exceedingly so for every unique
combination technologies across every vehicle model due to the extreme
manufacturing complexity that would be required to do so. The agencies
explained in the NPRM that the analysis should not include engine
resizing with the application of every technology or for combinations
of technologies that drive small performance changes to reflect better
what is feasible for manufacturers.\1481\
---------------------------------------------------------------------------
\1480\ See 83 FR 43027 (Aug. 24, 2018).
\1481\ For instance, a vehicle would not get a modestly bigger
engine if the vehicle comes with floor mats, nor would the vehicle
get a modestly smaller engine without floor mats. This example
demonstrates small levels of mass reduction. If manufacturers
resized engines for small changes, manufacturers would have
dramatically more part complexity, losing economies of scale.
---------------------------------------------------------------------------
Compliance modeling in the CAFE model also accounts for the
industry practice of platform, engine, and transmission sharing to
manage component complexity and associated costs.\1482\ At a vehicle
refresh cycle, a vehicle may inherit an already resized powertrain from
another vehicle within the same engine-sharing platform that adopted
the powertrain in an earlier model year. In the Autonomie modeling,
when a new vehicle adopts fuel saving technologies (such as ROLL
technology) that are inherited, the engine is not resized (the
properties from the baseline reference vehicle are used directly and
unchanged) and there may be a small change in vehicle performance.
---------------------------------------------------------------------------
\1482\ Ford EcoBoost Engines are shared across ten different
models in MY 2019. https://www.ford.com/powertrains/ecoboost/. Last
accessed Nov. 05, 2019.
---------------------------------------------------------------------------
Regarding customizing transmission gear ratios as rolling
resistance changes are implemented, the agencies explained in Section
VI.C.2 Transmissions that it is an observable practice in industry to
use a common gear set across multiple platforms and applications. The
most recent example is the GM 10L90, a 10-speed automatic transmission
that used the same gear set in both pick-up truck and passenger car
applications.\1483\ In Autonomie, optimization of transmission
performance is achieved through shift control logic rather than
customized hardware (e.g., gear ratios) for each vehicle line. The
shift initializer routine was run for every unique Autonomie full
vehicle model configuration to generate customized shifting maps. The
algorithms' optimization was designed to balance minimization of energy
consumption against vehicle performance.\1484\ This balance was
necessary to achieve the best fuel efficiency while maintaining
customer acceptability by meeting performance neutrality requirements.
See Section VI.B.3.a)(6) Performance Neutrality for more details. If
the systems were over-optimized for the agencies' modeling, such as
applying a unique gear set for each individual vehicle configuration,
the analysis would likely over-predict the reasonably achievable fuel
economy improvement for the technology. Over-prediction would be
exaggerated when applied under real-world large-scale manufacturing
constraints necessary to achieve the estimated costs for the
transmission technologies.
---------------------------------------------------------------------------
\1483\ ``GM Global Propulsion Systems--USA Information Guide
Model Year 2018'' (PDF). General Motors Powertrain. Retrieved
September 26, 2019. https://www.gmpowertrain.com/assets/docs/2018R_F3F_Information_Guide_031918.pdf.
\1484\ See ANL model documentation for final rule.
---------------------------------------------------------------------------
As HDS noted, the EPA Draft TAR and Proposed Determination analyses
performed using the ALPHA model adjusted the effectiveness of every
technology combination assuming performance could be held constant for
every combination, and did not recognize or account for the extreme
complexity nor the associated costs for that impractical assumption.
The NPRM and final rule analyses account for real-world practicalities
and constraints related to both engine adoption and transmission
adoption when other vehicle technologies are implemented, which
explains some of the effectiveness and cost differences between the
Draft TAR/Proposed Determination and the NPRM/final rule.
Figure VI-51 below shows the range of effectiveness used for the
NPRM analysis for ROLL technologies.
BILLING CODE 4910-59-P
[[Page 24570]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.281
BILLING CODE 4910-59-C
Figure VI-52 below shows the range of effectiveness values used for
the final rule analysis.
BILLING CODE 4910-59-P
[[Page 24571]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.282
BILLING CODE 4910-59-C
e) Rolling Resistance Cost
For the NPRM, the analysis used DMC for ROLL technology from the
Draft TAR and updated the values to reflect 2016$ dollars. The agencies
continued to use the same cost assumptions presented in the NPRM for
the final rule, and updated the values to 2018$ dollars. Table VI-139
and Figure VI-53 show the different levels of tire rolling resistance
technology cost.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.283
[[Page 24572]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.284
BILLING CODE 4910-59-C
7. Other Vehicle Technologies
Four other vehicle technologies were included in the analysis--
electric power steering (EPS), improved accessory devices (IACC), low
drag brakes (LDB), and secondary axle disconnect (SAX) (which may only
be applied to vehicles with all-wheel-drive or four-wheel-drive). The
effectiveness of these technologies was applied directly by the CAFE
model, with unique effectiveness values for each technology and for
each technology class. This methodology was used in these four cases
because the effectiveness of these technologies varies little with
combinations of other technologies. Also, applying these technologies
directly in the CAFE model significantly reduces the number of
Autonomie simulations that are needed.
a) Electric Power Steering (EPS)
Electric power steering reduces fuel consumption and CO2
emissions by reducing load on the engine. Specifically, it reduces or
eliminates the parasitic losses associated with engine-driven power
steering pumps, which pump hydraulic fluid continuously through the
steering actuation system even when no steering input is present. By
selectively powering the electric assist only when steering input is
applied, the power consumption of the system is reduced in comparison
to the traditional ``always-on'' hydraulic steering system. Power
steering may be electrified on light duty vehicles with standard 12V
electrical systems and is also an enabler for vehicle electrification
because it provides power steering when the engine is off (or when no
combustion engine is present).
Power steering systems can be electrified in two ways.
Manufacturers may choose to eliminate the hydraulic portion of the
steering system and provide electric-only power steering (EPS) driven
by an independent electric motor, or they may choose to move the
hydraulic pump from a belt-driven configuration to a stand-alone
electrically driven hydraulic pump. The latter system is commonly
referred to as electro-hydraulic power steering (EHPS). As discussed in
the NPRM, manufacturers have informed the agencies that full EPS
systems are being developed for all types of light-duty vehicles,
including large trucks.
EPS is also discussed in Section VI.C.3.a) Electrification Modeling
in the CAFE model.
b) Improved Accessories (IACC)
Engine accessories typically include the alternator, coolant pump,
cooling fan, and oil pump, and are traditionally mechanically-driven
via belts, gears, or directly by other rotating engine components such
as camshafts or the crankshaft. These can be replaced with improved
accessories (IACC) which may include high efficiency alternators,
electrically driven (i.e., on-demand) coolant pumps, electric cooling
fans, variable geometry oil pumps, and a mild regeneration
strategy.\1485\ Replacing lower-efficiency and/or mechanically-driven
components with these improved accessories results in a reduction in
fuel consumption, as the improved accessories can conserve energy by
being turned on/off ``on demand'' in some cases, driven at partial load
as needed, or by operating more efficiently.
---------------------------------------------------------------------------
\1485\ IACC in this analysis excludes other electrical
accessories such as electric oil pumps and electrically driven air
conditioner compressors.
---------------------------------------------------------------------------
For example, electric coolant pumps and electric powertrain cooling
fans provide better control of engine cooling. Flow from an electric
coolant pump can be varied, and the cooling fan can be shut off during
engine warm-up or cold ambient temperature conditions,
[[Page 24573]]
reducing warm-up time, fuel enrichment requirements, and, ultimately
reducing parasitic losses.
IACC is also discussed in Section VI.C.3.a) Electrification
Modeling in the CAFE model.
c) Low Drag Brakes (LDB)
Low or zero drag brakes reduce or eliminate brake drag force by
separating the brake pad from the rotor, either by mechanical or
electric methods. Conventional disc brake systems are designed such
that the brake pad is in contact with the brake rotor at all times.
This is true even when the brakes are not being applied, and although
the contact pressure is light in this case, this still produces some
drag force on the vehicle.
LDBs have historically employed a caliper and rotor system that
allows the piston in the caliper to retract,\1486\ in turn pulling the
brake pads away from the rotor. However, if pads are allowed to move
too far away from the rotor, the first pedal application made by the
vehicle operator can feel spongy and have excessive travel. This can
lead to customer dissatisfaction regarding braking performance and
pedal feel. For this reason, in conventional hydraulic-only brake
systems, manufacturers are limited by how much they can allow pads to
move away from the rotor.
---------------------------------------------------------------------------
\1486\ The brake caliper pistons are used to push the brake pad
against the brake rotor, or disc.
---------------------------------------------------------------------------
Recent developments in braking systems have resulted in brakes with
the potential for zero drag. In these systems, the pedal feel is
separated from hydraulics by a pedal simulator. This system is similar
to the brake systems designed for hybrid and electric vehicles, where
some of the primary braking is done through the recuperation of kinetic
energy in the drive system. However, the pedal feel and the
deceleration the operator experiences is tuned to provide a braking
experience equivalent to that of a conventional hydraulic brake system.
These ``brake-by-wire'' systems have highly tuned pedal simulators that
feel like typical hydraulic brakes and seamlessly transition to a
conventional system as required by different braking conditions. The
application of a pedal simulator and brake-by-wire system is new to
non-electrified vehicle applications. By using this type of system,
vehicle manufacturers can allow brake pads to move farther away from
the rotor and still maintain the initial pedal feel and deceleration
associated with a conventional brake system.
In addition to reducing brake drag, the zero drag brake system
provides ancillary benefits. It allows for a faster brake application
and greater deceleration than is normally applied by the average
vehicle operator. It also allows manufacturers to tune the braking for
different customer preferences within the same vehicle. This means
manufacturers can provide a ``sport'' mode, which provides greater
deceleration with less pedal displacement and a ``normal'' mode, which
might be more appropriate for day-to-day driving.
The zero drag brake system also eliminates the need for a brake
booster. This saves cost and weight in the system. Elimination of the
conventional vacuum brake booster could also improve the effectiveness
of stop-start systems. Typical stop-start systems need to restart the
engine if the brake pedal is cycled because the action drains the
vacuum stored in the booster. Because the zero drag brake system
provides braking assistance electrically, there is no need to
supplement lost vacuum during an engine off event.
Finally, many engine technologies being considered to improve
efficiency also reduce pumping losses through reduced throttling, and
in turn there is less engine vacuum available to power-assist a
conventional brake system. The reduction in throttling could require a
supplemental vacuum pump to provide vacuum for a conventional brake
system. This is the situation in many diesel-powered vehicles. Diesel
engines have no throttling and require a supplemental vacuum for
conventional brake systems. A zero drag brake system both eliminates
brake drag and avoids the need for a supplemental vacuum pump.
d) Secondary Axle Disconnect (SAX)
All-wheel drive (AWD) and four-wheel drive (4WD) vehicles provide
improved traction by delivering torque to the front and rear axles,
rather than just one axle. When a second axle is rotating, it tends to
consume more energy because of additional losses related to lubricant
churning, seal friction, bearing friction, and gear train
inefficiencies.\1487\ \1488\ Some of these losses may be reduced by
providing a secondary axle disconnect function that disconnects one of
the axles when driving conditions do not call for torque to be
delivered to both.
---------------------------------------------------------------------------
\1487\ Phelps, P. ``EcoTrac Disconnecting AWD System,''
presented at 7th International CTI Symposium North America 2013,
Rochester MI.
\1488\ Pilot Systems, ``AWD Component Analysis,'' Project
Report, performed for Transport Canada, Contract T8080-150132, May
31, 2016.
---------------------------------------------------------------------------
The terms AWD and 4WD are often used interchangeably, although they
have also developed a colloquial distinction, and are two separate
systems. The term AWD has come to be associated with light-duty
passenger vehicles providing variable operation of one or both axles on
ordinary roads. The term 4WD is often associated with larger truck-
based vehicle platforms providing a locked driveline configuration and/
or a low range gearing meant primarily for off-road use.
Many 4WD vehicles provide for a single-axle (or two-wheel) drive
mode that may be manually selected by the user. In this mode, a primary
axle (usually the rear axle) will be powered, while the other axle
(known as the secondary axle) is not. However, even though the
secondary axle and associated driveline components are not receiving
engine power, they are still connected to the non-driven wheels and
will rotate when the vehicle is in motion. This unnecessary rotation
consumes energy,\1489\ and leads to increased fuel consumption and CO2
emissions that could be avoided if the secondary axle components were
completely disconnected and not rotating.
---------------------------------------------------------------------------
\1489\ Any time a drivetrain component spins it consumes some
energy, primarily to overcome frictional forces.
---------------------------------------------------------------------------
Light-duty AWD systems are often designed to divide variably torque
between the front and rear axles in normal driving to optimize traction
and handling in response to driving conditions. However, even when the
secondary axle is not necessary for enhanced traction or handling, in
traditional AWD systems it typically remains engaged with the driveline
and continues to generate losses that could be avoided if the axle was
instead disconnected. The SAX technology observed in the marketplace
disengages one axle (typically the rear axle) for 2WD operation, but
detects changes in driving conditions and automatically engages AWD
mode when it is necessary. The operation in 2WD can result in reduced
fuel consumption. For example, Chrysler has estimated the secondary
axle disconnect feature in the Jeep Cherokee reduces friction and drag
attributable to the secondary axle by 80% when in disconnect
mode.\1490\
---------------------------------------------------------------------------
\1490\ Brooke, L. ``Systems Engineering a new 4x4 benchmark,''
SAE Automotive Engineering, June 2, 2014.
---------------------------------------------------------------------------
e) Analysis Fleet Assignments for Other Vehicle Technologies
The agencies described in the PRIA that the aforementioned
technologies have been applied, to some extent, in the MY 2016 fleet.
However, these technologies are difficult to observe and
[[Page 24574]]
assign to the analysis fleet, and the agencies relied heavily on
industry engagement and feedback to assign the technologies properly to
the NPRM analysis fleet vehicles. In the NPRM, the agencies noted that
the Draft TAR analysis did not properly account for the presence of
these technologies in the analysis fleet, and far too few were
assigned. Accordingly, the NPRM analysis reflected higher EPS and IACC
application rates than the Draft TAR analysis.
The agencies received a handful of comments stating that the
additional technologies were incorrectly applied to the analysis fleet.
ICCT stated that the inclusion of EPS, IACC, and LDB in the analysis
fleet was unsubstantiated, and removed the technologies from potential
use during the subsequent simulated years.\1491\ ACEEE commented that
IACC should not have been applied to certain vehicles in the analysis
fleet because those vehicles do not in actuality display the fuel
consumption reduction that would confirm the presence of these
additional technologies.\1492\ In addition, ACEEE commented that the
CAFE model assumes significant baseline SAX penetration that they could
not corroborate from Ford F-150 product information brochures.\1493\
HDS compared the available levels of IACC improvements from the Draft
TAR to the NPRM analysis, noting that the NPRM only employed one level
of improved accessory technologies.\1494\ HDS stated that this implied
the effectiveness of what was previously considered IACC1 (the first
level of IACC technology improvement available in the Draft TAR) was
completely used up in the 2016 analysis fleet for this rule.
---------------------------------------------------------------------------
\1491\ International Council on Clean Transportation, Attachment
3, Docket No. NHTSA-2018-0067-11741, at I-37.
\1492\ American Council for an Energy-Efficient Economy,
Attachment 6, Docket No. NHTSA-2018-0067-12122, at 6.
\1493\ American Council for an Energy-Efficient Economy,
Attachment 6, Docket No. NHTSA-2018-0067-12122, at 7.
\1494\ H-D Systems, ``HDS final report,'' Docket No. NHTSA-2018-
0067-11985, at 21.
---------------------------------------------------------------------------
As the agencies stated in the PRIA, in part because of the
difficulty in observing EPS, IACC, LDB, and SAX on actual vehicles, far
too few of those technologies were assigned to vehicles in the Draft
TAR analysis fleets. For the final rule, each vehicle in the MY 2017
analysis fleet was studied using confidential and publicly available
information to determine whether, as commenters suggested, the agencies
had improperly applied any of these additional vehicle technologies.
This resulted in some adjustments in the application of the
technologies in the analysis fleet. In regard to ACEEE's comment on SAX
penetration in the analysis fleet, for the NPRM and final rule
analysis, the agencies considered all 4WD vehicles to have the
capability manually to disconnect either the front or rear wheel axle
and associated rotating components, thus shifting to a 2WD mode. When
4WD operation is required for safety and utility, the consumer can
enable this feature. As stated above, this capacity to shift between
2WD and 4WD modes is another form of SAX. For AWD vehicles, publicly
available manufacturer information was reviewed to identify the
specific vehicles that have SAX technology. Based on market
observations and feedback from OEMs, the entire analysis fleet for NPRM
and the final rule was considered to have a basic level of improved
accessories (comparable to what Draft TAR referred to as IACC1). The
application of IACC in the NPRM and final rule analysis fleets
represents further improvements to accessories such as electric water
pumps and higher efficiency alternators with mild regeneration
capacity.
The following distribution of technologies in the analysis fleet
from the NPRM to the final rule analysis shows a slight decrease in the
portion of total vehicles produced that have EPS and IACC, a very
slight increase in the portion of total vehicle production that have
LDB, and a slight increase in the portion of 4WD/AWD vehicles with SAX
technology.
BILLING CODE 4910-59-P
[[Page 24575]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.285
BILLING CODE 4910-59-C
f) Effectiveness Estimates for Other Vehicle Technologies
The effectiveness estimates for these four technologies rely on
previous work published as part of the rulemaking process, both for the
2012 rule for MYs 2017-2025 and the Draft TAR. The effectiveness values
are unchanged from the Draft TAR.
The effectiveness of both EPS and EHPS is derived from the
decoupling of the pump from the crankshaft, and is considered to be
practically the same for both. Thus, a single effectiveness value is
assigned to all vehicles in the analysis fleet that possess either EPS
or EHPS, and the ``EPS'' designation is applied.
For the Draft TAR analysis, two levels of IACC were offered as a
technology path (a low improvement level and a high improvement level).
Since much of the market has incorporated some of these technologies in
the baseline MY 2016 and 2017 fleets, the NPRM and final rule analyses
assumed all vehicles have incorporated what was previously the low
level, so only the high level remained as an option for vehicles. The
figure above shows the distribution of IACC for NPRM and FRM, which is
the equivalent type of technology as the high-level IACC in the DRAFT
TAR.
The NPRM analysis carried forward work on the effectiveness of SAX
systems conducted in the Draft TAR and EPA Proposed Determination. This
work involved gathering information by monitoring press reports,
holding meetings with suppliers and OEMs, and attending industry
technical conferences. The resulting effectiveness estimates used in
the Draft TAR, NPRM, and this final rule are shown below.
BILLING CODE 4910-59-P
[[Page 24576]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.286
BILLING CODE 4910-59-C
g) Cost Estimates and Learning Rates for Other Vehicle Technologies
The cost estimates for these technologies rely on previous work
published as part of the rulemaking process, both for the 2012 rule for
MYs 2017-2027 and the Draft TAR. The cost values are from the same
sources as the Draft TAR and were updated to 2016 dollars for the NPRM
and 2018 dollars for the final rule analysis. Learning rates for these
technologies are also unchanged since the NPRM, and can be seen in
Section VI.B.4.d)(4) Cost Learning as Applied in the CAFE Model.
CARB noted that the IACC costs in Tables 6-32 and 6-33 of the PRIA
did not align with the Technologies central analysis input file.\1495\
HDS commented, as part of its comparison of IACC penetration in the
analysis fleet from the Draft TAR to NPRM, that IACC costs were based
on the difference between IACC1 and IACC2 costs and this appeared to be
inconsistent with the cost of accessory electrification which is more
expensive.\1496\
---------------------------------------------------------------------------
\1495\ CARB, Docket No. NHTSA-2018-0067-12428, at 21.
\1496\ H-D Systems, ``HDS final report,'' Docket No. NHTSA-2018-
0067-11985, at 21.
---------------------------------------------------------------------------
In the PRIA, the cost of IACC was reported in some tables as an
absolute cost (the cost of adding IACC to a base vehicle), while the
NPRM Technologies central analysis input file showed IACC cost
incremental to EPS. This was necessary in the model input file because
the accounting method of the NPRM CAFE model utilized incremental
costs. In contrast, a change in the CAFE model accounting method for
this final rule allows all costs in the input file to be reported as
absolute costs, incremental to a base vehicle. It was assumed that EPS
must be present on a vehicle in order for it to adopt IACC, and as such
the cost of IACC includes the cost of EPS. For further detail on the
use of absolute costs in place of incremental costs, see Section
VI.C.7.g). Although HDS commented that accessory electrification has a
higher cost than what is being used in the analysis, no specific
additional input was given; the cost of IACC, as was done for Draft TAR
(where it was referred to as IACC2), was taken from the 2015 NAS
Report.\1497\
---------------------------------------------------------------------------
\1497\ National Research Council. 2015. Cost, Effectiveness, and
Deployment of Fuel Economy Technologies for Light-Duty Vehicles.
Washington, DC--The National Academies Press, Table 8A.2a, available
at https://www.nap.edu/catalog/21744/cost-effectiveness-and-deployment-of-fuel-economy-technologies-for-light-duty-vehicles.
---------------------------------------------------------------------------
Table VI-141 below shows the absolute costs for these technologies
for select model years. The FRM Technologies central analysis input
file shows the costs for all model years.
[[Page 24577]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.287
8. Simulating Off-Cycle and A/C Efficiency Technology Adjustments
Off-cycle and air conditioning (A/C) efficiency technologies can
provide fuel economy improvements in real-world vehicle operation, but
that benefit cannot be adequately captured by the 2-cycle test
procedures used to demonstrate compliance with fuel economy and
CO2 emissions standards.\1498\ Off-cycle technologies
include technologies like high efficiency alternators and high
efficiency exterior lighting.\1499\ A/C efficiency technologies operate
mainly by reducing the operation of the compressor, which pumps A/C
refrigerant around the system loop. The less the compressor operates or
the more efficiently it operates, the less load the compressor places
on the engine, resulting in better fuel efficiency and lower
CO2 emissions.
---------------------------------------------------------------------------
\1498\ See 49 U.S.C 32904(c) (``The Administrator shall measure
fuel economy for each model and calculate average fuel economy for a
manufacturer under testing and calculation procedures prescribed by
the Administrator. . . . the Administrator shall use the same
procedures for passenger automobiles the Administrator used for
model year 1975 (weighted 55 percent urban cycle and 45 percent
highway cycle), or procedures that give comparable results.'').
\1499\ See 83 FR 43057. A partial list of off-cycle technologies
is included in Tables II-21 and II-22 of the NPRM.
---------------------------------------------------------------------------
Vehicle manufacturers have the option to generate credits for off-
cycle technologies and improved A/C systems under the EPA's
CO2 program and receive a fuel consumption improvement value
(FCIV) equal to the value of the benefit not captured on the 2-cycle
test under NHTSA's CAFE program. The FCIV is not a credit in the NHTSA
CAFE program, but the FCIVs increase the reported fuel economy of a
manufacturer's fleet, which is used to determine compliance. EPA
applies FCIVs during determination of a fleet's final average fuel
economy reported to NHTSA.\1500\ FCIVs are only calculated and applied
at a fleet level for a manufacturer and are based on the volume of the
manufacturer's fleet that contain qualifying technologies.\1501\
---------------------------------------------------------------------------
\1500\ 49 U.S.C. 32904(c)-(e). EPCA granted EPA authority to
establish fuel economy testing and calculation procedures. See
Section IX for more information.
\1501\ 40 CFR 600.510-12(c)
---------------------------------------------------------------------------
As discussed further in Section IX.D Compliance Issues that Affect
Both the CO2 and CAFE Programs, three pathways can be used
to determine the value of A/C efficiency and off-cycle adjustments.
First, manufacturers can use a predetermined list or ``menu'' of credit
values established by EPA for specific off-cycle technologies.\1502\
Second, manufacturers can use 5-cycle testing to demonstrate and
justify off-cycle CO2 credits; \1503\ the additional tests
allow emission benefits to be demonstrated over some elements of real-
world driving not captured by the 2-cycle compliance tests, including
high speeds, rapid accelerations, and cold temperatures. Third,
manufacturers can seek EPA approval, through a notice and comment
process, to use an alternative methodology other than the menu or 5-
cycle methodology for determining the off-cycle technology improvement
values.\1504\
---------------------------------------------------------------------------
\1502\ See 40 CFR 86.1869-12(b). The Technical Support Document
(TSD) for the 2012 final rule for MYs 2017 and beyond provides
technology examples and guidance with respect to the potential
pathways to achieve the desired physical impact of a specific off-
cycle technology from the menu and provides the foundation for the
analysis justifying the credits provided by the menu. The
expectation is that manufacturers will use the information in the
TSD to design and implement off-cycle technologies that meet or
exceed those expectations in order to achieve the real-world
benefits of off-cycle technologies from the menu.
\1503\ See 40 CFR 86.1869-12(c). EPA proposed a correction for
the 5-cycle pathway in a separate technical amendments rulemaking.
See 83 FR 49344 (Oct. 1, 2019). EPA is not approving credits based
on the 5-cycle pathway pending the finalization of the technical
amendments rule.
\1504\ See 40 CFR 86.1869-12(d).
---------------------------------------------------------------------------
The agencies have been collecting data on the application of these
technologies since implementing the programs.\1505\ Most manufacturers
are generating A/C efficiency and off-cycle credits; in MY 2017, 15
manufacturers generated A/C efficiency credits and 15 manufacturers
generated off-cycle credits, through the level of deployment varies by
manufacturer.\1506\
---------------------------------------------------------------------------
\1505\ See 77 FR at 62832, 62839 (Oct. 15, 2012). EPA introduced
A/C and off-cycle technology credits for the CO2 program
in the MYs 2012-2016 rule and revised the program in the MY 2017-
2025 rule and NHTSA adopted equivalent provisions for MYs 2017 and
later in the MY 2017-2025 rule.
\1506\ The 2018 EPA Automotive Trends Report, EPA-420-R-19-002,
March 2019 at Chapter 5.B., Figures 5.10 and 5.11.
---------------------------------------------------------------------------
a) A/C and Off-Cycle Effectiveness Modeling
The NPRM analysis used the off-cycle FCIVs and credits earned by
each manufacturer in MY 2016 and carried these forward at the same
levels for future years for the CO2 analysis and beginning
in MY 2017 for the CAFE analysis. The 2016 values for off-cycle FCIVs
for each manufacturer and fleet, denominated in grams CO2
per mile,\1507\ are provided in Table VI-142.\1508\ Additional off-
cycle FCIVs were added in future years if a manufacturer applied a
technology that was explicitly simulated in the analysis and also was
an off-cycle technology listed on the predefined menu.\1509\
Technologies explicitly simulated in the analysis that are also on the
off-cycle menu include start-stop systems that reduce fuel consumption
during idle and active grille shutters that improve aerodynamic drag at
highway speeds,
[[Page 24578]]
among others. Any off-cycle adjustments that accrued as the result of
applying these technologies were calculated dynamically in each model
year the technology was applied, with adjustments accumulating up to
the 10 g/mi cap. As a practical matter, most of the adjustments for
which manufacturers can claim off-cycle FCIVs exist outside of the CAFE
model technology tree so the off-cycle menu cap was rarely reached for
the NPRM analysis.
---------------------------------------------------------------------------
\1507\ For the purpose of estimating their contribution to CAFE
compliance, the grams CO2/mile values in Table I-1 are
converted to gallons/mile and applied to a manufacturer's 2-cycle
CAFE performance. When calculating compliance with EPA's
CO2 program, there is no conversion necessary (as
standards are also denominated in grams/mile).
\1508\ 2016 GHG Manufacturer Performance Report. EPA-420-R-18-
002. January 2018. https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100TGIA.pdf. Last Accessed Nov. 14, 2019. 2016
Report Tables for the GHG Manufacturer Performance Report. January
2018. https://www.epa.gov/sites/production/files/2018-01/ghg-report-2016-data-tables.xlsx. Last Accessed Nov. 14, 2019.
\1509\ For more details, see Section IX.D Compliance Issues that
Affect Both the CO2 and CAFE Programs and Section IX.D.3
Flexibilities for Off-Cycle Technologies.
---------------------------------------------------------------------------
The agencies sought comment on both the A/C and off-cycle data that
was used for the NPRM analysis as well as the assumptions for applying
those technologies.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.288
BILLING CODE 4910-59-C
Universally, stakeholders believed the application of off-cycle
adjustments in the analysis was too conservative. Stakeholders believed
the A/C and off-cycle technologies would be rapidly deployed and
manufacturers would reach the cap values within the rulemaking
timeframe.
---------------------------------------------------------------------------
\1510\ See 83 FR 43159-60 (``. . . this analysis uses the off-
cycle credits submitted by each manufacturer for MY 2017 compliance
and carries these forward to future years with a few exceptions.'').
---------------------------------------------------------------------------
The Institute for Policy Integrity (IPI) questioned the position
the agencies assumed in the NPRM analysis, and suggested the agencies
``assume that manufacturers will efficiently deploy all cost-saving
offset opportunities, especially in the face of increasingly stringent
standards.'' \1511\
---------------------------------------------------------------------------
\1511\ Comments from Institute from Policy Integrity, Attachment
1, NPRM Docket No. NHTSA-2018-0067-12213, at 20-21.
---------------------------------------------------------------------------
ICCT stated ``far greater use of the off-cycle provisions will
occur by 2025'' and emphasized that off-cycle technologies are ``highly
cost-effective and being deployed in greater sales penetrations than
many of the test-cycle efficiency technologies that the agencies are
analyzing.'' \1512\ ICCT supported manufacturers maximizing the use of
off-cycle technologies, and supported the analysis estimating
``fleetwide off-cycle credit use at over 10 g/mile by 2020,'' and
further suggested fleetwide achievement of 15 g/mile by 2025.\1513\
---------------------------------------------------------------------------
\1512\ Comments from ICCT, Attachment 1, NPRM Docket No. NHTSA-
2018-0067-11741, at I40--I41.
\1513\ Note there is a regulatory ``cap'' on menu technologies
of 10 g/mi (see Section IX for further discussion of the cap),
however a manufacturer can receive additional off-cycle credit/FCIV
by using the pathways described above to petition for off-menu
technologies. ICCT's comment suggests that manufacturers will reach
the regulatory menu cap and apply additional technologies to get an
additional 5 g/mi credit above the menu cap.
---------------------------------------------------------------------------
FCA, General Motors and the Auto Alliance all provided similar
observations, stating ``[m]anufacturers have rapidly deployed
technology in response to this all new regulatory
[[Page 24579]]
mechanism.'' Each of the commenters provided support for an argument of
rapid off-cycle technology adoption, stating ``[i]n the MY2021-2026
timeframe of the proposed rule, it is likely that manufacturers will
hit the existing 10 g/mi cap.'' \1514\
---------------------------------------------------------------------------
\1514\ Comments from Automotive Alliance, Appendix 1, NPRM
Docket No. NHTSA-2018-0067-12073, at 92; Comments from Fiat Chrysler
Automobiles, Attachment1, NPRM Docket No. NHTSA-2018-0067-11943, at
8; Comments from General Motors, Appendix 4--Comments to Technical
Issues, NPRM Docket No. NHTSA-2018-0067-11858, at 1.
---------------------------------------------------------------------------
The DENSO Corporation further supported the increased use of off-
cycle technologies, commenting that ``[a]vailable data on OEM off-cycle
technology credit utilization within the past few years demonstrates
that the use of off-cycle technologies is expected to grow--
particularly technologies on the credit menus.'' \1515\
---------------------------------------------------------------------------
\1515\ Comments from DENSO Corporation, Attachment 1, NPRM
Docket No. NHTSA-2018-0067-11880, at 6.
---------------------------------------------------------------------------
However, Toyota Motors North America asked for constraints on
considerations of off-cycle technology in the analysis.\1516\ Toyota
expressed concern for over-reliance on off-cycle technologies to
provide flexibilities for compliance, as ``most of the technologies
provide little tangible value proposition for customers.'' In
additional comments, Toyota repeated the concern noting, ``most of
these technologies lack consumer demand.'' Finally, Toyota specifically
cautioned against overusing off-cycle technologies in the analysis,
stating ``[t]he suggested pursuit of maximum credits overlooks the
associated costs and market acceptance challenge for certain off-cycle
technologies.'' Toyota listed costs versus risk of customer acceptance
and agency approval as factors that ``introduce a high level of
uncertainty for an auto manufacturer's planning and make investments in
off-cycle technologies risky and less appealing.''
---------------------------------------------------------------------------
\1516\ Comments from Toyota Motors North America, Attachment 1,
NHTSA Docket No. NHTSA-2018-0067-130798, at 9-10; Supplemental
Comments from Toyota Motors North America, Attachment 1, NHTSA
Docket No. NHTSA-2018-0067-12150, at 24; Supplemental Comments from
Toyota Motors North America, Attachment 1, NHTSA Docket No. NHTSA-
2018-0067-12376, at 4-5.
---------------------------------------------------------------------------
After carefully considering the comments, the agencies agree that
A/C and off-cycle technologies are likely to be more broadly applied by
manufacturers within the rulemaking timeframe. The final rule analysis
has been updated to reflect an increased application of the
technologies. Similar to the NPRM, the final rule analysis used the A/C
and off-cycle FCIVs earned by each manufacturer in the baseline fleet
(MY 2017 for the final rule analysis) as a starting point. However, the
final rule analysis increased these values in subsequent model years.
In addition to the dynamic application of off-cycle FCIVs, as in the
NPRM, each manufacturer's fleet FCIVs were increased by extrapolating
the manufacturers' historical rate of FCIV application through
2017.\1517\ In line with most commenters, the agencies increased the
FCIVs for each manufacturer such that the maximum value of 10 g/mi will
be reached by MY 2023. For manufacturers who did not reach maximum
values prior to 2023 through data extrapolation, a linear increase to
the cap was assumed. The agencies believe this approach balances a
greater application of FCIV technologies across the fleet, while
avoiding uncertain over-reliance on flexibilities for the analysis.
---------------------------------------------------------------------------
\1517\ The 2018 EPA Automotive Trends Report, https://www.epa.gov/fuel-economy-trends/download-report-co2-and-fuel-economy-trends. Accessed Aug 23, 2019.
---------------------------------------------------------------------------
The agencies disagreed with the proposal to model the application
of 15 g/mi of FCIVs universally in the rulemaking timeframe. Based on
historical data and industry comments from both manufacturers and
suppliers, the agencies expect there will be an increase in off-cycle
technology application. However, there are two issues with assuming
manufacturers will exceed the existing off-cycle caps. First, only a
few manufacturers approached the cap limit in MY 2018, and the fleet
average menu credit was 4.7 grams/mile, less than half the cap
value.\1518\ Second, new off-cycle technologies may address the same
inefficiencies as menu technologies, rather than work in conjunction.
Accordingly, the agencies believe there is a reasonable basis for
assuming manufacturers could, and would only achieve 10 g/mi on average
by MY 2023, and used that assumption for the final rule analysis.
---------------------------------------------------------------------------
\1518\ The 2018 EPA Automotive Trends Report, Greenhouse Gas
Emissions, Fuel Economy, and Technology since 1975, EPA-420-R-19-002
(Mar. 2019).
---------------------------------------------------------------------------
Table VI-143 shows passenger car values for FCIVs and Table VI-144
shows light truck values for FCIVs applied for the final rule analysis.
BILLING CODE 4910-59-P
[[Page 24580]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.289
[[Page 24581]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.290
[GRAPHIC] [TIFF OMITTED] TR30AP20.291
[[Page 24582]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.292
[[Page 24583]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.293
BILLING CODE 4910-59-C
[[Page 24584]]
A/C Efficiency, A/C Leakage and Off-Cycle Costs
As discussed above, the only A/C efficiency and off-cycle
technologies applied dynamically in the NPRM analysis were explicitly
simulated technologies like stop-start systems and active aerodynamic
technologies. The NPRM analysis fully accounted for both the
effectiveness and cost of these technologies and therefore separate
cost accounting was not needed. For example, when stop-start or active
aerodynamics technology was added by the model to a vehicle, the
corresponding off-cycle FCIVs were applied and the technology costs
were captured the same as every other technology on the decision trees.
For the final rule analysis, A/C and off-cycle technologies are
applied independently of the decision trees using the extrapolated
values, so it is necessary to account for the costs of those
technologies independently. Table VI-145 shows the costs used for A/C
and off-cycle FCIVs the final rule analysis. The costs are shown in
dollars per gram of CO2 per mile ($ per g/mile). The A/C
costs and off-cycle technology costs are the same costs used in the EPA
Proposed Determination and described in the EPA Proposed Determination
TSD.\1519\
---------------------------------------------------------------------------
\1519\ EPA PD TSD. EPA-420-R-16-021. November 2016. At 2-423-2-
245. https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100Q3L4.pdf. Last
accessed Nov.14, 2019.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.294
BILLING CODE 4910-59-C
D. Impacts that Result From Simulating Manufacturer Compliance with
Regulatory Alternatives
1. Simulating Economic Impacts of Regulatory Alternatives
a) What Economic Impacts Occur When Vehicle Manufacturers Comply With
Different CAFE and CO2 Standards?
1) The NPRM Framework for Analyzing Economic Impacts
In the proposed rule, the agencies noted the importance of
identifying the mechanisms by which vehicle manufacturers' compliance
with different CAFE and CO2 standards generated impacts on
manufacturers, owners of new and used vehicles, and the remainder of
the U.S. The agencies organized the analysis of alternative standards
using a framework that clarified the economic impacts on vehicle
producers, illustrated how costs were transmitted to buyers of new
vehicles, highlighted the collateral economic effects on owners of used
vehicles, and identified how these responses created various indirect
costs and benefits. Throughout the analysis, the agencies stressed the
distinction between the proposal's economic consequences for private
businesses and households, and its ``external'' economic impacts--those
ultimately borne by the rest of the U.S. economy.
To clarify the framework used in the proposal, the agencies used
Table VI-146 below (which is based on Tables II-25 to II-28 from the
NPRM) \1520\ to report costs and benefits and to trace how they pass
through the economy. As the table shows, the economic impacts of
standards initially fall on vehicle manufactures, but ultimately are
borne by consumers who purchase and drive new models. Smaller, indirect
economic effects of the proposal would be borne by owners of used cars
and light trucks (vehicles produced during model years prior to those
affected by the proposal, but still in use) as well as by the general
public and government agencies. On balance, the agencies projected that
most of the proposal's economic effects would fall on private
businesses and households, with the remainder of the U.S. economy
bearing much smaller impacts.
---------------------------------------------------------------------------
\1520\ See 83 FR at 43062-66.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24585]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.295
BILLING CODE 4910-59-C
More specifically, the agencies' analysis showed that the proposal
would initially have saved manufacturers the costs of adding the
technologies that would otherwise have been necessary to enable their
new cars and light trucks to comply with the baseline fuel economy and
CO2 emissions regulations, with the estimated dollar value
of those savings shown in line 1 of Table VI-146. The proposal also
enabled some manufacturers to make lower civil penalty payments for
failing to comply with the more demanding standards that were
supplanted (line 2), although these savings would have been exactly
offset by lower civil penalty revenue to the
[[Page 24586]]
Federal Government (line 16). The analysis assumed that manufacturers
would have the ability, in a competitive market, to pass their savings
in technology costs and any reduction in civil penalties paid on to
buyers, by charging lower prices for new vehicles. Although lower
prices reduced their revenues (line 3), on balance, their savings in
compliance costs, reduced civil penalty payments, and lower sales
revenue were assumed to leave manufacturers financially unaffected
(shown by the zero entry in line 4 of the table).
Under the proposal, the analysis showed that buyers of new cars and
light trucks benefited directly from those vehicles' lower purchase
prices and financing costs (line 5). They also avoided the increased
risk of crash-related injuries that would have resulted from reductions
in the weight of some new models, as manufacturers attempted to improve
fuel economy to comply with the baseline standards. The economic value
of this reduction in risk represented an additional benefit from the
proposal to reducing the stringency of the standards vis-[agrave]-vis
the baseline (line 6).
At the same time, however, the lower fuel economy that some new
cars and light trucks were expected to offer with less stringent
standards in place would have imposed various additional costs on their
buyers and users. Drivers experienced higher fuel costs as a
consequence of new vehicles' increased fuel consumption (line 7), as
well as the added time and inconvenience of having to make more
frequent refueling stops required by reduced driving range (line 8).
They also forfeited some mobility benefits as they drove newly-
purchased cars and light trucks less in response to their higher fuel
costs (line 9). On balance, the agencies' analysis of the proposal
showed that buyers of new cars and light trucks produced during the
model years it affected would experience significant economic benefits
(line 10).
A novel feature of the agencies' evaluation of the proposal showed
that lowering prices for new cars and light trucks, some owners of used
vehicles retired them from service earlier than they otherwise would
have done. In combination with increased sales of new models, this
transferred some driving that would have occurred with used cars and
light trucks to newer and safer models, thus reducing the total costs
of fatalities and injuries sustained in motor vehicle crashes.\1521\ In
the proposal, this reduction in injury risks provided benefits to
owners and drivers of older cars and light trucks that had not been
recognized or quantified in its analyses of previous CAFE and
CO2 standards (line 11).
---------------------------------------------------------------------------
\1521\ This improvement in safety resulted from the fact that
cars and light trucks have become progressively more protective in
crashes over time (and also slightly less prone to certain types of
crashes, such as rollovers). Thus, shifting some travel from older
to newer models reduced injuries and damages sustained by drivers
and passengers because they were traveling in inherently safer
vehicles, rather than because of changes to driver risk profiles.
---------------------------------------------------------------------------
Table VI-146 also showed that the changes in fuel consumption and
vehicle use resulting from the proposal would in turn generate both
benefits and costs to the remainder of the U.S. economy. The analysis
described these as ``external'' effects, in the sense that they were
by-products of households' choices among new vehicle models, decisions
about keeping older cars and light trucks in service, and allocations
of driving across the fleet that were experienced broadly throughout
the U.S. economy, rather than by the individuals making such decisions.
The largest of these was additional refining and consumption of
petroleum-based fuel and the associated increases in emissions of
carbon dioxide and other gases, which were projected to increase the
cost of economic damages inflicted on the U.S. economy by future
changes in the global climate (line 13). Added fuel production and use
under the proposal also led to higher emissions of localized air
pollutants, and the resulting increase in the U.S. population's
exposure and its adverse effects on health imposed additional external
costs (line 14).
Increased consumption of petroleum-derived fuel also imposed higher
external costs on the U.S. economy, in the form of potential losses in
economic output and costs to businesses and households for adjusting to
any sudden changes in energy prices (line 15 of the table). Reduced
driving by buyers of new cars and light trucks in response to their
higher operating costs also reduced the external costs from their
contributions to traffic delays and noise, benefits that were expected
to be experienced throughout the U.S. economy (line 17). Finally, some
of the higher fuel costs to buyers of new cars and light trucks will
consist of increased fuel taxes; this increase in revenue was projected
to enable Federal and State government agencies to improve upkeep of
roads and highways, fund increases in other services, or reduce other
tax burdens (line 18).\1522\
---------------------------------------------------------------------------
\1522\ In some States, levies on gasoline include both general
sales taxes as well as excise taxes, and not all proceeds are
dedicated to transportation purposes.
---------------------------------------------------------------------------
The net economic effect (line 22) of the proposal consisted of the
benefits and costs imposed directly on car and light truck
manufacturers, accompanying indirect effects on buyers of new vehicles
and owners of used ones, external costs driving decisions generated
throughout the U.S. economy, and changes in revenue to government
agencies. The agencies' organization was intended to convey the causal
connections among these impacts, by highlighting how the proposed
change in fuel economy standards faced by manufacturers would set in
motion the sequence of behavioral responses that determined its
economy-wide costs and benefits. This contrasted with the way benefits
and costs of previous proposals to establish CAFE and CO2
standards were analyzed and presented, which obscured their sequence
and causal connections.
In those previous analyses, most economic effects other than
manufacturers' costs to comply with proposed standards and anticipated
changes in fuel consumption were grouped together and reported as ``co-
benefits.'' This obscured how these various consequences arose from the
proposed standards, providing no information about who would ultimately
experience the costs of complying with the standards, or who would
experience their direct and indirect benefits. In contrast, the recent
analysis spelled out how each category of benefits and costs resulted
from the proposed change in standards, identified the mechanisms that
translated direct economic impacts into indirect costs and benefits,
and distinguished between those arising from changes in fuel
consumption, and safety consequences of changes in vehicle use. The
proposal's framework also clarified who would bear each category of
impacts, distinguishing between the proposal's economic impacts on
private actors--vehicle manufacturers, new car and light truck buyers,
and owners of used vehicles--and the external economic consequences for
the general public and government agencies that stem indirectly from
such private impacts.
2) Final Rule Framework
While the agencies received several comments about which economic
effects are included in the analysis, the agencies received no comments
about the specific structure of the framework. Substantive comments
about individual
[[Page 24587]]
effects are addressed over the next several sections.
The agencies have expanded the accounting framework for benefits
and costs shown in Table VI-146 above to include two additional
entries, as well as to distinguish financial impacts on government
agencies from externalities borne broadly across the remainder of the
U.S. economy. The revised accounting framework for costs and benefits
is shown in Table VI-147, below. Line 6 of the revised table reports
the change in consumer surplus experienced by buyers of new cars and
light trucks when prices and sales of those vehicles adjust in response
to changes in CAFE and CO2 standards. The gain in consumer
surplus that occurs when production costs and prices for vehicles fall
and sales increase in response represents a benefit to buyers, while
any loss in consumer surplus that occurs when more stringent standards
increase costs and prices and cause sales to decline appears as a loss
to new car and light truck buyers.
Line 7 of Table VI-147 reports the estimated value of changes to
attributes of new cars and light trucks other than fuel economy that
their manufacturers make to comply with changes in CAFE and
CO2 standards. In the case where standards are less
stringent, manufacturers are able to employ many of the same resources
they would have deployed to increase fuel economy for the alternative
purpose of improving other attributes of vehicles that their potential
buyers value more highly than the forgone improvements in fuel economy.
This response provides an additional benefit to purchasers of new cars
and light trucks that was not recognized in the agencies' analysis of
the proposal, but is included in the analysis of this final rule. Of
course, if CAFE and CO2 standards are made more stringent,
manufacturers employ those technologies to increase fuel economy, thus
sacrificing potential improvements in competing attributes--those that
entail tradeoffs with higher fuel economy--and the value of
improvements in those other attributes that is sacrificed or forgone
represents an opportunity cost to those buyers. This implicit
opportunity cost is analyzed in a sensitivity analysis and is not
included in the primary analysis.
Finally, the agencies revised the framework for reporting costs and
benefits of changes in CAFE and CO2 standards to identify
government agencies separately from the entry previously labeled ``Rest
of U.S Economy.'' This minor revision is intended to distinguish more
clearly between changes in external costs imposed by externalities that
result from fuel production and use, and the revenue effects on
government agencies from changes in tax and civil penalty payments.
While both effects ultimately result from manufacturers' compliance
with revised standards and the resulting changes in fuel consumption,
externalities represent real economic costs; in contrast, changes in
tax revenues received by government agencies are financial transfers,
whose offsetting effects on manufacturers and vehicle buyers are also
recognized elsewhere in the accounting framework.
BILLING CODE 4910-59-P
[[Page 24588]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.296
BILLING CODE 4910-59-C
b) Economic Assumptions
The agencies' analysis of CAFE and CO2 standards for the
model years covered by this final rule rely on a range of forecast
information, estimates of economic, safety, and environmental
variables, and input parameters. While the analysis accompanying the
proposal largely resembled previous CAFE and CO2 analyses,
the agencies updated many of the underlying inputs and assumptions--
based on the most up-to-date data--and expanded the central analysis to
account for changes in new vehicle sales and the retirement of older
vehicles.
EDF, UCS, CARB and others commented that the agencies acted
arbitrarily and capriciously by changing inputs and assumptions from
previous analyses, and argued that the agencies failed to provide
``good reasons'' for the changes.\1523\ In the following sections, the
agencies will respond directly to these comments. However, the agencies
note that it would be uncommon to retain inputs and assumptions from
prior analyses--which are typically informed by transitory empirical
observations--on the basis of precedent. The agencies are ``neither
required nor
[[Page 24589]]
supposed to regulate the present and the future within the inflexible
limits of yesterday.'' \1524\
---------------------------------------------------------------------------
\1523\ See, e.g,. IPI, Appendix, NHTSA-2018-0067-12213, at 99-
100.
\1524\ American Trucking Associations v. Atchison, 387 U.S. 397,
416 (1967).
---------------------------------------------------------------------------
The agencies also received a number of comments focused on the
agencies' attempt to incorporate the effects of changes in new vehicle
prices on new vehicle sales, retirement rates of used vehicles, and the
resulting ``turnover'' of the vehicle fleet. Some comments endorsed the
agencies' more comprehensive analysis, although many of those same
commenters later disagreed with aspects of the results. For example,
RFF noted that ``Incorporating sales and scrappage effects represents a
step in the right direction for modeling the effects of the
regulation.'' \1525\ Similarly, NRDC stated that ``it is reasonable and
appropriate to develop a mechanism for estimating future vehicle
populations, and the NPRM documents appropriately present considerable
discussion on the topic and the derivation of the utilized algorithm.''
\1526\ One commenter explicitly recognized that the narrower analysis
utilized in previous rules likely led to incorrectly estimating costs
and benefits, and endorsed the broader approach used by the proposal.
Specifically, American Fuel & Petrochemical Manufacturers stated that
the absence of scrappage in prior rules ``likely led to a significant
overestimation of the existing standard's benefits with respect to fuel
and air pollutant emission reductions and an underestimation of safety
risks and societal costs.'' FCA also expressed general support for the
agency's expanded analysis.\1527\
---------------------------------------------------------------------------
\1525\ Resources for the Future, NHTSA-2018-0067-11789, at 2.
\1526\ Meszler Engineering Services & Baum and Associates, on
behalf of Natural Resources Defense Council, NHTSA-2018-0067-11943-
43, NHTSA-2018-0067-11723.
\1527\ FCA, NHTSA-2018-0067-12078.
---------------------------------------------------------------------------
In contrast, some commenters objected to the inclusion of `new'
impacts, including the effect of fuel economy regulations on new
vehicle prices, the resulting changes in their sales, and retirement
rates for used cars. Workhorse Group, Inc. noted that the agencies
``made novel assumptions about the safety impacts of consumers delaying
vehicle purchases due to the increased costs of fuel economy
improvements that contradicts the analytical approach NHTSA has
followed in all prior safety and CAFE rulemakings.'' \1528\ Honda
agreed ``that significantly higher-priced new vehicles have the
potential to depress the new vehicle market and thus increase the fleet
of used vehicles, with concomitant increased safety risks associated
with driving greater numbers of older vehicles in lieu of newer ones,''
but found it ``premature and ill-advised'' to model the impact of fleet
turnover.\1529\ CBD et. al. argued that the sales and scrappage effects
were too uncertain to include in the analysis and cited EPA's 2016
proposed determination as stating, ``a reasonable qualitative
assessment is preferable to a quantitative estimate lacking sufficient
basis, or (due to uncertainties like those here) having such an
enormous range as to be without substantial value.'' \1530\
---------------------------------------------------------------------------
\1528\ Workhorse Group, Inc., NHTSA-2018-0067-12215.
\1529\ American Honda Motor Company, Inc., NHTSA-2018-0067-
11818.
\1530\ Environmental group coalition, Appendix A, NHTSA-2018-
0067-12000, at 174.
---------------------------------------------------------------------------
As was done repeatedly throughout the proposal, the agencies
acknowledge that dynamically modeling fleet turnover is new for this
rulemaking; however, the agencies disagree that the analysis relied on
`novel' assumptions or contradicted previous analyses. The agencies
have described the sales and scrappage responses similarly in prior
rulemakings,\1531\ and have expressed an interest in quantitatively
measuring them.\1532\ The agencies agree with commenters that--like
many of the effects included in today's analysis--there remains a
degree of uncertainty about the magnitude of the sales and scrappage
responses. However, CBD v. NHTSA stressed that a variable should not be
excluded from the analysis simply because it is uncertain when the
effect is quantifiable, ``certainly not zero,'' and the analysis
``monetize[s] other uncertain benefits.'' \1533\ As discussed in the
coming sections, the agencies are confident that (a) changes in new
vehicle prices impact the volume of new vehicle sales and rate of
retirement of older vehicle, (b) of the direction of those effects, and
(c) their ability to reasonably estimate the impacts. As such, the
agencies strongly believe that including the sales and scrappage
responses improves the thoroughness of the analysis, is consistent with
case law, and is necessary to comprehensively analyze the cost-benefits
of the rule.
---------------------------------------------------------------------------
\1531\ See, e.g., 76 FR 75153.
\1532\ See, e.g., 77 FR 61971.
\1533\ 538 F.3d 1172, 1200-02 (2008).
---------------------------------------------------------------------------
The following subsections briefly describes the sources of the
agencies' estimates of each of the economic, environmental, and safety
estimates. In reviewing these variables and the agencies' estimates of
their values for purposes of this final rule, NHTSA and EPA considered
comments received in response to the proposed rule and, in response,
made several changes to the economic assumptions used for the final
analysis.
1) Macroeconomic Assumptions That Affect the Agencies' Analysis
As the proposed rule noted, the more comprehensive economic impact
analysis of CAFE and CO2 included in this rulemaking
requires a more detailed and explicit explanation of the macroeconomic
context in which regulatory alternatives are evaluated. The agencies
continued to rely on projections of future fuel prices to evaluate
manufacturers' use of fuel-saving technologies, the resulting changes
in fuel consumption, and various other benefits. Furthermore, the
agencies expanded the scope of their analysis to include projecting
future sales of new cars and light trucks, as well as the retirement of
used vehicles under each regulatory alternative. In addition to
projections of future fuel prices, constructing these forecasts
requires explicit projections of macroeconomic variables, including
U.S. Gross Domestic Product (GDP), labor force participation (the
number of persons employed or actively seeking employment), and
bellwether interest rates, which are likely to vary according to
roughly the same pattern as interest rates on new car loans.
The analysis presented in the proposal as well as the accompanying
RIA and EIS employed forecasts of future fuel prices developed by the
agencies using the U.S. Energy Information Administration's (EIA's)
National Energy Model System (NEMS). An agency within the U.S.
Department of Energy (DOE), EIA collects, analyzes, and disseminates
independent and impartial energy information to promote sound
policymaking, efficient markets, and public understanding of energy and
its interaction with the economy and the environment. EIA uses NEMS to
produce its Annual Energy Outlook (AEO), which presents forecasts of
future fuel prices, among many other energy-related variables. AEO
projections of energy prices and other variables are not intended as
predictions of what will happen; rather, they are projections of the
likely course of these variables that reflect their past relationships,
specific assumptions about future developments in global energy
markets, and the forecasting methodologies incorporated in NEMS. Each
AEO includes a ``Reference'' case as well as a range of alternative
scenarios that each incorporate
[[Page 24590]]
somewhat different assumptions from those underlying the Reference
Case.
For the proposal, the agencies used the AEO2017 version of NEMS, as
this was the most current version of the model that was available at
the time. Using this version of NEMS, the agencies reevaluated the
``Reference,'' ``Low Oil Price,'' and ``High Oil Price'' cases
described in AEO2017, by setting aside their assumption that mandates
by California and other States to sell ``Zero Emission Vehicles''
(ZEVs) would be enforced. The agencies used the resulting modified
Reference case fuel prices as inputs to the proposal's central case
results, and used the modified ``Low Oil Price'' and ``High Oil Price''
case fuel prices, which were generated using NEMS, as inputs to several
of the sensitivity analysis cases that were presented in the proposal.
The sensitivity analysis also included a case that applied the
Reference case fuel prices from the then recently issued AEO2018, which
did not reflect the modification of EIA's forecasting model to set
aside state mandates for ZEV sales.\1534\
---------------------------------------------------------------------------
\1534\ The results of these and other sensitivity analyses were
reported in NHTSA and EPA, ``Notice of Proposed Rulemaking: The
Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Years
2021-2026 Passenger Cars and Light Trucks,'' Federal Register Vol.
83, No. 165, August 24, 2018, Tables Vii-90 to Vii-98, pp. 43353-69.
---------------------------------------------------------------------------
The analysis supporting the proposed rule simulated the economic
impacts of car and light truck manufacturers' compliance with
alternative CAFE and CO2 standards through model year 2032,
and in doing so estimated the number of vehicles originally produced
and sold in each model year that would remain in service during each
year of their useful lives (assumed to extend for a maximum of 40
years), as well as their usage, fuel consumption, and safety
performance. This required the forecasts of macroeconomic variables
that affect vehicle sales, use, and retirement rates, which include
U.S. Gross Domestic Product (GDP), the size of the domestic labor
force, and key interest rates, to extend well beyond calendar year
2050. One of the few sources that provides forecasts of these variables
spanning such a long time horizon was the 2017 OASDI Trustees Report
from the U.S. Social Security Administration, and the analysis
supporting the proposed rule relied on this source for forecasts of
these key macroeconomic measures.\1535\
---------------------------------------------------------------------------
\1535\ Social Security Administration, The 2017 Annual Report of
the Board of Trustees of the Federal Old-Age and Survivors Insurance
and Federal Disability Insurance Trust Funds, available at https://www.ssa.gov/OACT/TR/2017/.
---------------------------------------------------------------------------
(a) Comments on the Fuel Price Forecasts and Macroeconomic Assumptions
Used in the NPRM Analysis
The agencies received relatively few comments on the projections of
fuel prices and macroeconomic variables that were used in their
analysis supporting the proposed rule, virtually all of them focused on
the fuel price projections the agencies employed. While only one
comment questioned the agencies' use of price projections that rely on
EIA's methodology and assumptions, a few commenters called attention to
the unreliability of price projections reported in earlier editions of
AEO. Other comments noted the importance of updating projections used
to analyze the proposal to reflect more recent developments in energy
markets, without necessarily questioning the reliability of EIA's fuel
price projections. Several comments emphasized the implications for the
agencies' analysis of the wide variation in alternative fuel price
projections reported in both EIA's 2017 and 2018 Annual Energy
Outlooks, with most stressing the possibility that future prices might
be above even those projected in their High Oil Price cases. Only a
single comment identified a potential alternative source of fuel price
projections, but noted that it was within the range of projections the
agencies considered.
One commenter claimed that AEO's projections of fuel prices are
``inappropriate'' for the agencies to employ in analyzing the
consequences of CAFE and CO2 standards; because EIA ``does
not speculate on changes in international policy or geopolitics,''
which contribute to the uncertainty surrounding future prices.\1536\
However, this commenter did not identify an alternative source for fuel
price projections that reflect such considerations; and because
projections of fuel prices are a central element in the agencies'
evaluation of alternative future standards, the observation that EIA's
projections do not incorporate some sources of uncertainty is unhelpful
by itself.
---------------------------------------------------------------------------
\1536\ NHTSA-2018-0067-11837, Alliance to Save Energy, p. 2
(``EIA takes a transparently conservative approach in modeling
future oil prices, and does not speculate on changes in
international policy or geopolitics. As a result, their projections
are an inappropriate measure of future fuel prices.'').
---------------------------------------------------------------------------
Some commenters asserted that by relying on the AEO2017 Reference
Case projections of fuel prices in their central analysis of the
proposed rule while considering the significantly higher fuel prices
projection in the AEO High Oil Price scenario only in the accompanying
sensitivity analyses, the agencies inadequately considered the possible
effect of higher fuel prices on the estimated economic benefits from
alternatives that would have relaxed the augural standards, including
the preferred alternative.\1537\ Surprisingly, none of these comments
acknowledged that the fuel price projections reported in the High Oil
Price cases accompanying past editions of the Annual Energy Outlook
have so far proven to be significantly above actual prices, or that EIA
has consistently lowered its fuel price projections in more recent
editions of the AEO. In any case, supplemental material included in the
NPRM regulatory docket showed that the ranking of regulatory
alternatives by their estimated net economic benefits remained
unchanged from the central analysis in the sensitivity analysis that
substituted the AEO2017 High Oil Price case projection of fuel prices.
---------------------------------------------------------------------------
\1537\ See e.g., Securing America's Future Energy (SAFE), NHTSA-
2018-0067-11981, pp. 12 & 30 and Institute for Policy Integrity,
NHTSA-2018-0067-12213, p. 31.
---------------------------------------------------------------------------
None of the commenters who argued that the agencies inadequately
considered the possibility of higher fuel prices observed that the
agencies' analogous use of lower fuel price projections from the
AEO2017 Low Oil Price case only in their sensitivity analyses
inadequately considered the possibility that future fuel prices might
prove to be lower than projected in the AEO2017 Reference Case, and its
potential effect on the proposal's estimated benefits. Nor did any of
the commenters offer substantive guidance about how the agencies might
revise their analysis to accord greater emphasis to fuel price
projections above (or below) those from the AEO Reference Case.\1538\
---------------------------------------------------------------------------
\1538\ One commenter did refer to guidance to EPA contained in a
National Research Council report on incorporating and conveying
uncertainty about key inputs directly into that agency's estimates
of benefits from reducing air pollution, rather than simply
recognizing it in supplemental sensitivity analyses. This was
presumably intended as potential guidance to the agencies about how
they might do so in their evaluations of fuel economy and
CO2 standards, although that was not stated explicitly.
See American Fuel & Petrochemical Manufacturers, NHTSA-2018-0067-
12078, p. 19, citing National Research Council (2002), Estimating
the Public Health Benefits of Proposed Air Pollution Regulations,
2002, available at https://www.nap.edu/catalog/10511/estimating-the-public-health-benefits-of-proposed-air-pollution-regulations.
---------------------------------------------------------------------------
Other comments stressed the fact that EIA's current projections of
future fuel prices are significantly lower than those the agencies
relied on when they established CAFE standards through
[[Page 24591]]
model year 2021 and introduced the augural standards for subsequent
model years in the rulemaking they conducted in 2012, citing this as
support for the agencies' reconsideration of the augural standards in
the current rulemaking.\1539\
---------------------------------------------------------------------------
\1539\ For example, Fiat Chrysler Automobiles (FCA) pointed out
that the AEO 2017 Reference Case forecast of gasoline prices through
2025 is approximately 36% lower than that in the AEO 2012 Reference
Case, which the agencies relied on in the analysis supporting that
earlier rulemaking; see NHTSA-2018-0067-11943, p. 33.
---------------------------------------------------------------------------
One comment compared the range of fuel price projections spanned by
the High and Low Oil Price cases from AEO2017 and AEO2018 to the range
of future prices spanned by another widely-recognized and relied-upon
projection, concluding that the alternative scenarios included in
AEO2017 incorporated an even wider range of uncertainty about future
prices, and noted that the net economic benefits of the preferred
alternative were positive over this entire range of alternative future
fuel prices. This same commenter noted that by combining high and low
fuel price projections with alternative assumptions about other key
economic variables (such as GDP growth) and parameter assumptions
(principally payback period), the agencies' sensitivity analyses
captured potentially important interactions between uncertainty
regarding fuel prices and other key economic inputs.\1540\
---------------------------------------------------------------------------
\1540\ See Alliance of Automobile Manufacturers, NHTSA-2018-
0067-1207, p. 108.
---------------------------------------------------------------------------
(b) Macroeconomic Assumptions Used To Analyze Economic Consequences of
the Final Rule
After considering these comments, the agencies have concluded that
there is no convincing reason to rely on sources other than EIA's NEMS
model to project future energy prices, or to rely on alternatives to
the Reference Case scenario in the current edition of AEO as their
basis for using NEMS. The agencies agree that the resulting projections
will be uncertain, but note that EIA regularly publishes retrospective
analyses comparing past Reference case projections to subsequent market
price outcomes, thus enabling an assessment of this uncertainty.
Although EIA does not identify its Reference case as a ``most likely''
outcome, in the agencies' judgment that case's design--which assumes
future trends are consistent with historical and current market
behavior--makes it a reasonable and appropriate basis for projecting
fuel prices to use in the agencies' central analysis of alternative
CAFE and CO2 standards.
The agencies also conclude that the wide range of uncertainty about
future petroleum prices encompassed in EIA's ``Low Oil Price'' and
``High Oil Price'' cases means that including them in the accompanying
sensitivity analyses provides a meaningful basis for assessing the
potential economic consequences of future energy prices that prove to
be considerably lower or higher than those reflected in the Reference
case. Although these alternative cases do not incorporate unbridled
speculation regarding hypothetical changes in ``international policy or
geopolitics,'' the agencies believe that this restraint means that
relying on them produces a more, rather than less, meaningful test of
the effect of the inherent uncertainty surrounding projections of fuel
prices.
For today's final rule, the agencies have therefore used the
AEO2019 version of NEMS to develop projections of future prices for
transportation fuels, as this was the most current version available
when this analysis was conducted. Using this version of NEMS, the
agencies modified EIA's AEO2019 Reference case by (1) setting aside
presumed enforcement by California and other States of any mandates to
sell ``Zero Emission Vehicles'' (ZEVs), (2) setting aside post-2020
increases in the stringency of CAFE and CO2 standards, and
(3) modifying inputs regarding battery costs, in order to bring those
costs down to levels more consistent with battery cost estimates
applied in the CAFE model analysis.\1541\ All other NEMS inputs used to
develop the AEO2019 Reference case were left unchanged in this
analysis.
---------------------------------------------------------------------------
\1541\ These inputs are all contained in the ``trnldvx.xlsx''
NEMS input file. The input file utilized for today's analysis is
available in regulatory docket NHTSA-2018-0067, https://www.regulations.gov/docket?D=NHTSA-2018-0067 (see Supporting
Documents), as is the corresponding output file from which reference
case fuel and electricity prices were obtained to be used as inputs
to the CAFE model. The version of NEMS utilized for today's analysis
is available at https://www.eia.gov/outlooks/aeo/info_nems_archive.php.
---------------------------------------------------------------------------
Setting aside enforcement of state mandates to sell ZEVs makes the
supporting analysis consistent with the agencies' recent One National
Program Action,\1542\ under which EPA withdrew aspects of a Clean Air
Act Preemption waiver previously granted to California, and NHTSA
concluded that EPCA expressly and implied preempted State ZEV mandates.
Setting aside the post-2020 increase in the stringency of CAFE and
CO2 standards ensures that the fuel prices used in the
agencies' analysis are at least as high as those that would prevail
under the least stringent regulatory alternative considered, since that
alternative produces the highest level of fuel consumption and thus the
highest fuel prices.
---------------------------------------------------------------------------
\1542\ 84 FR 51310.
---------------------------------------------------------------------------
Figure VI-55 and Figure VI-56 below show the resulting modified
projections of BEV prices and sales, and compare them to the
projections reported in EIA's AEO2019 Reference case. As they
illustrate, the combination of these modifications led NEMS to project
significantly lower BEV prices and correspondingly higher BEV sales
volumes. Figure VI-57 and Figure VI-58 show the modified projections of
gasoline and electricity prices, and again compare these to the
projections reported in EIA's AEO2019 Reference case. As those figures
indicate, the agencies' modifications to NEMS did not significantly
affect its projections of future prices for transportation fuels.
BILLING CODE 4910-59-P
[[Page 24592]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.297
BILLING CODE 4910-59-C
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.298
BILLING CODE 4910-59-C
BILLING CODE 4910-59-P
[[Page 24593]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.299
BILLING CODE 4910-59-C
The agencies used the resulting Reference case fuel prices as
inputs to the rule's central analysis. The agencies also used the as-
published (by EIA) ``Low Oil Price'' and ``High Oil Price'' case fuel
prices as inputs to several of the cases included in the sensitivity
analysis presented in the accompanying RIA.
For the projections of macroeconomic variables used in the analysis
supporting this rule, the agencies elected to rely on different sources
from those that informed their analysis of the proposed rule.
Specifically, the agencies rely on projections of future growth in U.S.
GDP reported in AEO2019 to support their central analyses of the final
rule's impacts on new car and light truck sales and the retirement of
used vehicles. These incorporate underlying projections generated using
the IHS Markit Global Insight long-term macroeconomic model, as
modified via this model's interaction with NEMS' representation of
global energy markets and their future outcomes. The alternative
projections of future growth in GDP used in the agencies' accompanying
sensitivity analyses are drawn from the AEO2019 High Economic Growth
and Low Economic Growth cases. These reflect alternative future trends
in U.S. labor force and productivity growth, and are also consistent
with the energy market outcomes projected by NEMS under the resulting
future performance of the U.S. economy.
[GRAPHIC] [TIFF OMITTED] TR30AP20.300
[[Page 24594]]
For estimates of the number of U.S. households during future years,
which influence the projections of new car and light truck sales used
in the analysis, the agencies rely on projections of new household
formation developed the Harvard University Joint Center for Housing
Studies.\1543\ These are consistent with the most recent projections of
future growth in the nation's population prepared by the U.S. Bureau of
the Census.\1544\
---------------------------------------------------------------------------
\1543\ See Harvard University Joint Center for Housing Studies,
Updated Household Growth Projections: 2018-2028 and 2028-2038,
December 18, 2018, available at https://www.jchs.harvard.edu/sites/default/files/Harvard_JCHS_McCue_Household_Projections_Rev010319.pdf.
\1544\ Ibid., pp. 2-5.
---------------------------------------------------------------------------
(2) Approach To Estimating Sales Response Under Different Standards
Prior to the NPRM, all previous CAFE and CO2 rulemaking
analyses used static fleet forecasts that were based on a combination
of manufacturer compliance data, public data sources, and proprietary
forecasts (or product plans submitted by manufacturers). When
simulating compliance with regulatory alternatives, those analyses
projected identical sales across the alternatives, for each
manufacturer down to the make/model level--where the exact same number
of each model variant was assumed to be sold in a given model year
under both the least stringent alternative (typically the baseline) and
the most stringent alternative considered (intended to represent
``maximum technology'' scenarios in some cases). To the extent that an
alternative matched the assumptions made in the production of the
proprietary forecast, using a static fleet based upon those assumptions
may have been warranted. However, a sales forecast is unlikely to be
representative of a broad set of regulatory alternatives with
significant variation in the cost of new vehicles. A number of
commenters on previous regulatory actions encouraged consideration of
the potential impact of fuel efficiency standards on new vehicle prices
and sales, and the changes to compliance strategies that those shifts
could necessitate.\1545\ In particular, the continued growth of the
utility vehicle segment creates compliance challenges within some
manufacturers' fleets as sales volumes shift from one region of the
footprint curve to another, or as mass is added to increase the ride
height of a vehicle on a sedan platform to create a crossover utility
vehicle, which exists on the same place of the footprint curve as the
sedan upon which it might be based.
---------------------------------------------------------------------------
\1545\ See, e.g., Alliance of Automobile Manufacturers, Comment,
EPA-HQ-OAR-2015-0827-4089, at 115-16.
---------------------------------------------------------------------------
However, some NPRM commenters referenced the agencies' previous
omission of this effect as justification to continue ignoring this
issue in the current rulemaking. EDF commented,\1546\ ``use of a sales
response model constitutes an unexplained reversal in the agency's
position on the feasibility of doing so.'' To say that the agencies
never used a model is a misrepresentation. Assuming that sales never
change in any model year, even at the individual nameplate level,
regardless of the stringency of fuel economy regulations or the
technology costs required to comply with those regulations, is, itself,
a model. It is a model that implicitly asserts that, while fuel economy
regulation impacts vehicle prices, such regulations have no impact on
the quantity or mix of new vehicle sold, regardless of stringency. This
is an implicit argument that new vehicle demand is perfectly
inelastic--and that no change in vehicle prices can impact the number
of cars consumers will buy. Logically, however, there must exist a
level of stringency that would have a negative impact on new sales.
Picking an extreme example to prove the point, if the agencies set
standards at an extraordinarily stringent level that forced all
vehicles into battery electric propulsion systems next year, sales
would obviously be impacted. The increase in new vehicle price or
changes to other relevant attributes like range, refueling time, or
operating cost would surely affect the decisions of some buyers. But,
by arguing that the agencies should continue to model new vehicle sales
as if they are entirely unaffected by standards, commenters are
effectively asking the agencies to assume that the alternatives
considered in this rule are insufficiently stringent to affect the
market. By endorsing the approach from the 2012 final rule, which
assumed no impact on the new vehicle market from standards as stringent
as 7 percent increase, year-over-year, beginning in 2017, commenters
are suggesting that even those standards would have no impact on new
vehicle sales. Manufacturers have asserted in their comments that fuel
economy regulations change both the cost of producing new vehicles and
consumer demand for them. In the recent peer review of the NPRM release
of the CAFE model, all reviewers encouraged the inclusion of a sales
response to fuel economy regulations (albeit not necessarily the
version of the response model that appeared in the NPRM).\1547\ Based
on earlier comments and the agencies' own analysis, the agencies were
persuaded to include a sales response mechanism in the NPRM, and do so
again in this final rule.
---------------------------------------------------------------------------
\1546\ EDF, Appendix B, NHTSA-2018-0067-12108, at 37-38.
\1547\ CAFE Model Peer Review, DOT HS 812 590, Revised (July
2019), available at https://www.regulations.gov/contentStreamer?documentId=NHTSA-2018-0067-0055&attachmentNumber=2&contentType=pdf.
---------------------------------------------------------------------------
While several commenters (CARB, NCAT, CBD, Aluminum Association)
discouraged the agencies from attempting to account for the effect of
regulations on new vehicle sales, other commenters stated that the NPRM
analysis was improved by explicitly considering this effect (RFF,
Toyota, the Alliance of Automobile Manufacturers). CBD cited EPA's 2016
proposed determination, stating ``[a] reasonable qualitative assessment
is preferable to a quantitative estimate lacking sufficient basis, or
(due to uncertainties like those here) having such an enormous range as
to be without substantial value.'' \1548\ However, RFF supported the
inclusion of the effect (with caveats about the specific
implementation, for which they suggested alternative approaches),
stating ``[i]ncorporating sales and scrappage effects represents a step
in the right direction for modeling the effects of the
regulation.\1549\ It is reasonable to conclude that regulations as
transformative as fuel economy standards will impact the market for new
vehicles, and excluding the effect (as CBD and others suggested) is
equivalent to stating that it does not exist.
---------------------------------------------------------------------------
\1548\ Environmental group coalition, Appendix A, NHTSA-2018-
0067-12000, at 174.
\1549\ RFF, Comments, NHTSA-2018-0067-11789, at 3.
---------------------------------------------------------------------------
The NPRM version of the sales response relied on differences in the
average price of new vehicles to produce sales differences between
regulatory alternatives. Some commenters (ACEEE, IPI, CBD, UCS,
Aluminum Association, and Alliance to Save Energy) argued that new
vehicle prices do not increase with the addition of technology required
to comply with fuel economy regulations. Some argued that manufacturers
will choose not to ``pass through'' the full incremental cost of fuel
saving technologies to consumers, instead absorbing those costs into
their profit margin.\1550\ The question of cost pass-through is one
that academic and industry researchers have considered for decades--and
two of the
[[Page 24595]]
agencies' recent peer reviewers addressed this issue in their comments.
---------------------------------------------------------------------------
\1550\ E.g. IPI, Appendix, NHTSA-2018-0067-12213, 28-29; CBD et
al., Attachment 1, NHTSA-2018-0067-12123, at 23-24.
---------------------------------------------------------------------------
Dr. John D. Graham, one of the peer reviewers, argued that the
assumption of complete cost pass-through is defensible, and more likely
in the long-run than the short-run.\1551\ The reviewer also suggested
that changes to the CAFE (and subsequent CO2) program that
base a manufacturer's standard on the mix of vehicle footprints in each
fleet more equitably spreads the impact of the standards across the
industry, and that industry shifts toward increasingly competitive
market models (rather than the oligopolistic models that existed
earlier in the last century) both act to increase the likelihood that
manufacturers will pass regulatory costs through to consumers. In
particular, this reviewer stated: \1552\
---------------------------------------------------------------------------
\1551\ CAFE Model Peer Review, DOT HS 812 590, Revised (July
2019), pp. B31-B33, available at https://www.regulations.gov/contentStreamer?documentId=NHTSA-2018-0067-0055&attachmentNumber=2&contentType=pdf.
\1552\ Gron Anne, Swenson, Deborah L, Cost Pass-Through in the
US Automobile Market, Review of Economics and Statistics, Vol. 82(2)
(May 2000), at 3.
In a classic study, Gron and Swenson (2000) examined list prices
of automobiles at the model level in the U.S. from 1984 to 1994
coupled with data on production, vehicle characteristics, foreign
versus domestic firm ownership, wages of employees, exchange rates,
imported parts content, tariffs and other variables. Although their
work rejects the hypothesis of 100% pass through of cost to consumer
price, they find higher rates of pass through than previous studies,
and much of the incomplete pass through occurs when cost increases
impact only a few models or firms. Confirming earlier studies, they
show that U.S. auto manufacturers engage in more aggressive pass-
through pricing than Asian and European manufacturers (greater than
100% in some specifications), possibly due to the eagerness of
importers to enlarge market share in lieu of recovering regulatory
costs, at least in the short run (see Dinopolous and Kreinin, 1988;
\1553\ Froot, 1989 \1554\[hairsp]). This study helps explain why
pass-through pricing is a more viable hypothesis in the long run
than in the short run.
---------------------------------------------------------------------------
\1553\ Dinopoulos, Elias, Kreinin, Mordechai, Effects of U.S.-
Japan Auto VER on European Prices and on U.S. Welfare, The Review of
Economics and Statistics, Vol. 70(3) (1988), at 484-91.
\1554\ Froot, Kenneth A, Klemperer, Paul D, Exchange Rate Pass-
Through When Market Share Matters, American Economic Review, Vol.
79(4) (1989), at 637-54.
---------------------------------------------------------------------------
The original design of the CAFE program is a contrasting case
where pass-through pricing was difficult for some automakers. All
auto makers, regardless of their product mix, were subject to the
same fleet-wide average CAFE standard, such as 27.5 miles per gallon
for cars in 1990. In practice, those standards impacted only three
high-volume companies (General Motors, Ford and Chrysler) because
the Big Three produced a higher proportion of large and performance-
oriented vehicles than did Japanese companies. As a result,
manufacturers such as Toyota and Honda consistently surpassed the
federal fleet-wide standard for cars without any regulatory cost
(i.e., partly due to their smaller product mix). In the 1975-2007
period, the Big Three were not able to pass on all of their
compliance costs to consumers and thus experienced some declines in
profitability due to CAFE (Kleit, 1990; \1555\ Kleit, 2004; \1556\
Jacobsen, 2013\1557\[hairsp]).
---------------------------------------------------------------------------
\1555\ Kleit, Andrew N., The Effect of Annual Changes in
Automobile Fuel Economy Standards, Journal of Regulatory Economics,
Vol. 2. (1990,), at 151-72.
\1556\ Kleit, Andrew N, Impact of Long-Range Increases in the
Fuel Economy (CAFE) Standard, Economic Inquiry, Vol. 42(2) (2004),
at 279-94.
\1557\ Jacobsen, Mark R., Evaluating U.S. Fuel Economy Standards
in a Model with Producer and Household Heterogeneity, American
Economic Journal: Economic Policy, Vol. 5(2) (2013), at 148-87.
---------------------------------------------------------------------------
When the CAFE program was reformed for light trucks in 2008 (and
for cars in 2011) on the basis of vehicle size (the so-called
``footprint'' adjustments to CAFE stringency), the, the technology
costs of CAFE standards were spread more evenly among automakers,
although the overall societal efficiency of the regulation
diminished due to the removal of downsizing as a compliance
option.\1558\ Given that the size-based fuel economy programs are
not concentrating the costs of compliance on one or two automakers,
it is reasonable to predict a fairly high degree of pass-through
pricing for the 2021-2025 fuel economy standards. In related
literature on manufacturer pricing responses to a national carbon
tax, Bento and Jacobsen (2007) \1559\ and Bento (2013) \1560\ report
high rates of pass-through pricing (on the order of 85%). Carbon
taxes are more efficient than footprint-based CAFE standards, but
both instruments are likely to impact a wide range of companies in
the auto sector and result in a high degree of pass-through pricing
by impacted companies.
---------------------------------------------------------------------------
\1558\ See Ito, Koichiro, Sallee, James M., The Economics of
Attribute-Based Regulation: Theory and Evidence from Fuel-Economy
Standards, Review of Economics and Statistics, in press (2018).
\1559\ Bento, Antonio M., Jacobsen, Mark R, Environmental Policy
and the `double-dividend' hypothesis, Journal of Environmental
Economics and Management, Vol. 53(1) (January 2007) at 17-31.
\1560\ Bento, Antonio M. Equity Impacts of Environmental Policy,
Annual Review of Resource Economics, Vol. 5 (May 2013), at 181-96.
---------------------------------------------------------------------------
Also, it should be noted that the U.S. automotive industry is
much more competitive today than it was from 1970 to 2000. The
market share of General Motors, once the dominant, majority producer
in the U.S. market, has declined dramatically, and a variety of
Japanese and Korean companies have captured substantial market
share. Moreover, the rise of startups (e.g., Tesla and other
electric vehicle start-ups) and ride-sharing services (e.g., Uber)
are adding a new competitive dimension in the U.S. industry. As a
result, some of the most recent auto regulatory studies have given
more emphasis to analytic results based on competitive models than
oligopolistic models (see, e.g., Davis and Knittel (2016)
\1561\[hairsp]).
---------------------------------------------------------------------------
\1561\ Davis, Lucas, Knittel, Christopher R., Are Fuel Economy
Standards Regressive? Working Paper 22925, National Bureau of
Economic Research, Cambridge, MA (2016).
Another peer reviewer, Dr. James Sallee, suggested that costs would
pass through to new vehicle buyers to different degrees, depending upon
the stringency of the standards.\1562\ The reviewer argued that more
stringent standards, which result in larger increases to the cost of
production, are likely to induce greater degrees of pass-through than
less stringent standards, which automakers may, as some commenters have
suggested, be able to absorb in the form of lost profit. If the degree
of cost pass-through should vary by the stringency of the alternative,
the agencies are underestimating the difference in price between the
most and least stringent alternatives--which would favor alternatives
with higher stringency.
---------------------------------------------------------------------------
\1562\ CAFE Model Peer Review, DOT HS 812 590, Revised (July
2019), pp. B54-B75, available at https://www.regulations.gov/contentStreamer?documentId=NHTSA-2018-0067-0055&attachmentNumber=2&contentType=pdf.
---------------------------------------------------------------------------
Other commenters argued that manufacturers are able to compensate
fully for the costs of fuel economy standards by increasing the prices
of luxury vehicles--which would increase the average new vehicle price,
but leave large sections of the market unaffected by the increased cost
of producing fleets that comply with the standards. While it seems
likely that manufacturers employ pricing strategies that push
regulatory costs (as well as increases in costs like pension
obligations and health care costs for employees) into the prices of
models and segments with less elastic demand, the extent to which any
OEM is able to succeed at this is unknown by the agencies. At some
point, however, price increases on even luxury models will merely price
more and more purchasers out of the market, and make competition with
other manufacturers and market segments that much more difficult. And
the more that avoided price increases for lower ends of the vehicle
market are subsidized by luxury vehicles, the more either prices for
luxury models would need to be increased, or (if moderately increasing
prices) more of those luxury models would need to be sold. It is worth
noting that luxury vehicles tend to be more powerful and content-rich,
and often have fuel economy levels below (or CO2 levels
above) their targets on the curves--so that selling more of them to
compensate for lost profit elsewhere
[[Page 24596]]
further erodes the compliance levels of the fleets in which they
reside.
While manufacturers could conceivably push some small cost
increases into the prices of their vehicle segments that have less
elastic demand to cover accordingly small increases in stringency,
larger stringency increases would exhaust the ability of such segments
to absorb additional costs. In addition, the agencies do not attempt to
adjust the mix of vehicle models based on their own price elasticity of
demand; doing so would require a pricing model that takes the
compliance cost for each manufacturer (which the agencies' model
estimates dynamically) and apportions that cost to the prices of
individual nameplates and trim levels. The agencies have experimented
with pricing models (when integrating vehicle choice models, pricing
models are a necessity), but each manufacturer almost certainly has a
unique pricing strategy that is unknown to the agencies, and involves
both strategic decisions about competitive position within a segment
and the volumes needed fully to amortize fixed costs associated with
production. To the extent that the agencies assume all regulatory costs
are passed through and affect the average regulatory cost of each
vehicle instead of being priced in a fashion to minimize the impact on
aggregate sales, the agencies note that--more stringent alternatives
are provided an artificial analytical advantage because manufacturers
are better positioned to incorporate smaller price adjustments into
their current strategic pricing models. The agencies opted to take the
conservative approach instead of speculating on manufacturer's private
business models.
Finally, some commenters have argued that, even if regulations do
increase the cost of producing vehicles and those costs are passed on
to new vehicle buyers, it does not matter because sales have increased
in recent years under both rising standards and rising prices. EDF,
CARB, Aluminum Association, SAFE, CBD, and CA et al. and Oakland et
al., all make some version of this argument in their comments.\1563\
The commenters are confusing correlation with causation and failing to
consider the counterfactual case. Higher prices of new vehicles
certainly did not cause sales to increase since 2012. Sales increased
over that period, in large part, as a result of economic expansion
following the great recession.\1564\ The statistical model used in the
NPRM attempted to isolate the effect of average price on new vehicle
sales, independent of the overall health of the US economy which plays
an obviously important role. That model showed a negative relationship
between sales and price (albeit a modest one), and positive
relationships with GDP and employment. Even under the most stringent
alternative in the NPRM, sales increased over time. However, in other
alternatives, where the same macroeconomic conditions prevailed but
average new vehicle prices were lower, sales increased relative to the
baseline. That is the counterfactual case that is relevant for
regulatory analysis--it attempts to answer the question, ``would sales
have been even higher if average prices had been lower?''
---------------------------------------------------------------------------
\1563\ See, e.g. EDF, Appendix B, NHTSA-2018-0067-12108, at 37;
CARB, Detailed Comments, NHTSA-2018-0067-11873, at 198-204; Aluminum
Association, Comments, NHTSA-2018-0067-11952, at 19-21; SAFE,
Comments, NHTSA-2018-0067-11981 at 36; CBD et al., Attachment 1,
NHTSA-2018-0067-12123, at 20. States and Cities, Detailed Comments,
NHTSA-2018-0067-11735, at 87-89.
\1564\ Table VI-148 below shows a large and statistically
significant effect of GDP on sales.
---------------------------------------------------------------------------
As discussed below, identifying the independent contribution of
price to new vehicle sales is econometrically challenging. In the NPRM,
the agencies stated that the simultaneous nature of price and sales--
where transaction prices are higher in periods of higher demand,
because the market will bear them, and lower in periods of lower
demand, because the market will not, for an otherwise identical
vehicle--creates a form of reverse causality. As commenters suggested,
in recent years sales have increased along with average transaction
price increases--and transaction price increases will occur when
regulation forces manufacturers to add content, and their corresponding
costs, to the vehicles they sell. Thus, it is understandable that some
commenters could interpret the recent increase in new vehicle sales
following the recession as evidence that standards (and maybe prices)
have no impact on new sales. However, that view confuses correlation
for causation (or lack thereof, in this case).
In response to these comments, the agencies have modified their
approach to modeling the sales impacts of regulatory alternatives. In
order to isolate the impact of the standards, the agencies have broken
the sales response module into two discrete components. The first
captures the effects of broader economic forces such as GDP growth. The
second measures how changes in vehicle prices influence sales. As
elaborated in more detail in the following passages, the agencies
considered alternative approaches and specific changes suggested by
commenters, but concluded that the comments either lacked enough
information to implement a change, failed to remedy identified alleged
weaknesses of the NPRM model, or created new limitations for which
there were no practical solutions. Furthermore, the two-pronged
approach addresses many of the concerns raised by commenters better
than any specific modeling alteration. First, the structural changes to
the model address many of the econometric concerns raised by
commenters. Second, by modeling sales in the first step as a function
of macroeconomic conditions, and then applying an independent own-price
elasticity to estimate the change in sales across alternatives, the
agencies are able to more clearly distinguish between demand-side and
supply-side impacts on prices, the issue that appears to have tripped
up some of the commenters.
Comments on the Econometric Model Used in the NPRM
Any model of sales response must satisfy two requirements: It must
be appropriate for use in the CAFE model, and it must be based in both
sound economic theory and appropriate empirical analysis. The first of
these requirements implies that forecasts of any variable used in the
estimation of the econometric model must also be available as a
forecast throughout the duration of the years covered by the
simulations (this analysis explicitly simulates compliance through MY
2050). Some values the model calculates endogenously, making them
available in future years for sales estimation, but others must be
known in advance of the simulation. As the CAFE model simulates
compliance, it accumulates technology costs across the industry and
over time. By starting with the last known average transaction price
(associated with MY 2016, in this analysis) and adding accumulated
regulatory costs to that value, the model is able to represent an
estimated average selling price in each future model year, assuming
that manufacturers are able to pass their compliance costs on to buyers
of new vehicles. Other variables used in the estimation can be entered
into the model as inputs prior to the start of the compliance
simulation.
The NPRM analysis was based on an econometric model that attempted
to estimate the price elasticity of aggregate demand for new light-duty
vehicles based on exogenous factors, intended to represent (1)
macroeconomic forces that influence demand for new vehicles, and (2)
average new vehicle price, intended
[[Page 24597]]
to represent the impact of regulation. A number of commenters voiced
opposition to the approach. Some disagreed with the theoretical framing
of the issue--arguing that the model of sales response should have
acknowledged the relevance of other vehicle attributes, included
consumer valuation of fuel savings for new vehicles, based the response
on something other than price, and considered the effect at a lower
level of aggregation, rather than average price across the industry.
In the NPRM, the agencies relied upon an autoregressive distributed
lag (ARDL) statistical model to estimate the impact of price
differences between regulatory alternatives and to produce a time
series of total new vehicle sales in each year of the analysis. The
statistical model estimated new vehicle sales per year based on two
lagged variables of new sales (new sales in the previous period, and
the period before that), GDP and lagged GDP, and labor force
participation and lagged labor force participation. The model used
quarterly data and seasonally adjusted annual rates to increase the
number of observations over the sample period for which reliable sales
data existed (1978-2015). The ARDL model used in the NPRM was chosen to
address sales impacts at a high level of aggregation, namely the total
new vehicle market (across all vehicle brands and body styles), and to
resolve the econometric issues associated with the time series data
related to total new vehicle sales.
Stock et al. commented at length on the econometric specification
of the NPRM sales response model, identifying limitations and
suggesting alternative approaches.\1565\ In particular, they argued
that the length of the response to price shocks should dissipate faster
than the NPRM model allows--an artifact of using quarterly data and
seasonally adjusted annual rates to estimate the effect and
implementing it on an annual basis in the CAFE model. The agencies
agree that this was a flaw in the implementation of the NPRM model.
While this approach produced the correct units (i.e., annual sales) the
response to changes in price should have dissipated at a quarterly
rate, rather than an annual rate. As a result, a single price shock,
which appears in one year and disappears the next, was projected to
have a longer impact on sales in future years than was appropriate
given the specification. The sales response in the final rule corrects
for this objective error and takes a more conservative approach to
price shocks.
---------------------------------------------------------------------------
\1565\ EPA-HQ-OAR-2018-0283 and NHTSA-2018-0067.
---------------------------------------------------------------------------
Stock et al. commented that ``it is important to estimate the
dynamic effect on sales of a price increase, that is, the causal effect
on current and future demand of a price increase'' because ``it allows
the response to an intervention--here, a one-time price increase or
sequence of such increases--to evolve over time.'' \1566\ The comment
suggests that the agencies should include future responses in sales to
a one-time price increase that exists for a single period and then
disappears. In our analytical framework, this implies that a price
difference between any alternative and the baseline that causes a
difference in sales in that year should also produce a difference in
sales in the following year (and possibly subsequent years), though of
smaller magnitude, even if the price difference only exists for a
single period. The Stock et al. comment illustrates a quickly
diminishing response to a single price shock. The final rule assumes
(more conservatively) that each price shock lasts only for a single
year, and produces no future ``ripple'' effects in the new vehicle
market in subsequent years. Furthermore, the regulatory alternatives
considered in this analysis do not produce single period price shocks
(in the form of price differences between alternatives), but rather
persistent price differences between alternatives that result from
continued differences in stringency. The persistent nature of the price
differences resulting from fuel economy and CO2 regulations
further reduce the importance of capturing these multi-period effects
caused by single-period price shocks.
---------------------------------------------------------------------------
\1566\ Ibid.
---------------------------------------------------------------------------
Stock et al. also objected to the use of an ARDL model to estimate
the impact of price on new vehicle sales. In order for the estimation
of causality to be valid in a time series model, the current price
movements must be uncorrelated with unobserved demand shocks in the
past, present, and future; so-called strict exogeneity. The commenters
argue that the NPRM fails this test because actions taken in the market
(by both buyers and sellers) can influence the response to price
changes in the next period. They suggest the use of a vector
autoregression (VAR) model to address the relationship between past
demand disturbances and current prices to address the temporal
exogeneity issues they identify. However, an important caveat is that
this approach still does not resolve the largest econometric
challenge--that of contemporaneous endogeneity between price and sales
(in the same period). To address that challenge, one needs to employ
instrumental variable methods.
The agencies attempted several modifications to the statistical
model developed for the NPRM based on the Stock et al. comment. The
agencies reviewed the initial approach and attempted several
specifications that would explicitly address the temporal endogeneity
bias identified in the comment. In particular, the agencies addressed
data limitations that were raised by Stock et al. (and also by EDF),
who encouraged us to reconsider the quarterly specification and to use
quality-adjusted price data for new vehicles in order to ensure a more
consistent definition of the average vehicle over the time series, as
the ``average vehicle'' has consistently improved in a myriad of ways
over successive model years. The quarterly price series was
statistically interpolated in the NPRM to increase the number of
observations,\1567\ but represented a less-than-ideal solution. The
interpolating process may have impacted the underlying quarterly data
generating process, resulting in unreliable, or potentially biased,
regression results. This issue was remedied by sourcing both vehicle
sales and price data from IHS Markit, which provides these data at the
same base frequency (quarterly) and obviates the need for any
interpolation. In addition, the macroeconomic data used in the model
specification were also sourced from IHS, which provides consistency
between historical and forecast data (i.e., forecasts of sales, price,
personal income, etc., were all based on a consistent set of input
assumptions and modeling framework during testing).
---------------------------------------------------------------------------
\1567\ Interpolation is the practice of adding unobserved data
points based on observed trends to provide more observations to a
limited data set.
---------------------------------------------------------------------------
Historical quarterly series for new light vehicle average price and
total sales are presented in Figure VI-59 below. Due to the lack of
data availability for business investment in light vehicles, the
historical series for average vehicle price begins in 1987. Average
prices were transformed into quality adjusted real terms using the CPI
for new motor vehicles, and both series were seasonally adjusted.\1568\
Quality adjusted prices have risen overtime, while total sales have
remained relatively flat in recent years with the major exception being
the significant economic downturn of 2008-2009. The difference in these
trends suggests that the number of vehicles purchased per
[[Page 24598]]
household does not necessarily change, or grow, over time, as income
grows, but rather households adjust the ``amount'' of new vehicle they
are willing to purchase (i.e., switching from sedan to an SUV).\1569\
Moreover, while disposable income has steadily increased during this
period, sales have not seen the same type of upward trend, and instead
only returned to its pre-recession average of around 17 million annual
sales.
---------------------------------------------------------------------------
\1568\ Seasonal adjustment was made using X.12 in EViews.
\1569\ Aggregate light duty vehicle sales data does not allow
for observing the distribution of vehicles being sold, which will
have an effect on the average price.
[GRAPHIC] [TIFF OMITTED] TR30AP20.301
Even as real disposable income has risen since 2000, and outside of
the great recession, new vehicle sales have remained relatively steady.
This, in turn, suggests there are other economic, or behavioral,
factors beyond disposable income influencing the decision to purchase a
new vehicle. Given the significant cost to purchase a new vehicle, and
the long multiyear timeframe over which they are typically financed,
households' forward-looking view on the health of the economy likely
plays a role in their willingness to purchase a new vehicle. Put
differently, households may delay their purchasing decisions if their
view outlook on the economy sours, regardless of income level. These
observations are consistent with the framework of the NPRM model, and
Figure VI-60 presents the consumer sentiment index and total new sales,
with both series exhibiting similar trends over this period. Some
commenters advocated that consumer sentiment (also known as consumer
confidence) should be included in the sales forecast. For example, the
Aluminum Association indicated that prior sales models have shown
consumer behavior to be ``highly sensitive to macroeconomic conditions,
consumer confidence and employment levels.'' While consumer sentiment
was not included in the NPRM model, it was included in specifications
that the agencies tested and considered and is a component of the
forecasting model used in the final rule.\1570\
---------------------------------------------------------------------------
\1570\ Commenters mentioned consumer confidence as a predictor
of consumer behavior. For instance, the Aluminum Association
indicated that prior sales models have shown consumer behavior to be
``highly sensitive to macroeconomic conditions, consumer confidence
and employment levels.'' Comments, NHTSA-2018-0067-11952, at 14.
---------------------------------------------------------------------------
[[Page 24599]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.302
All macroeconomic data were sourced from IHS including real
disposable income, number of US households, and the University of
Michigan's consumer sentiment index. The summary statistics for all
series are presented below in Table VI-148.
[GRAPHIC] [TIFF OMITTED] TR30AP20.303
Each series was transformed into natural logarithms and tested for
stationarity using the modified Dicky-Fuller test.\1571\ Results
presented in Table VI-149 indicate each variable containing contained a
unit-root, while being differenced stationary (i.e., integrated of
order one).
---------------------------------------------------------------------------
\1571\ Using nonstationary variables would generate unreliable
estimates of their influence, as prior values of those variables are
correlated with their future values, and this violates the
assumption that values variables take on are independent over time.
---------------------------------------------------------------------------
[[Page 24600]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.304
Two separate variables lists were then tested for the existence of
one or more cointegrating relationships, with results from the Johansen
test presented in Table VI-150.\1572\ In each set of variables, both
total LDV sales and disposable income were converted to household units
as a means to control for the growth in US households and the possible
decision making process of buying/consuming a new unit of LDV. The
results show that 4 out of the 5 lag length selections for both
variable sets conclude there being one cointegrating relationship (rank
I(1)) among them.
---------------------------------------------------------------------------
\1572\ The number of lag lengths were also tested formally, with
general consensus between 2 and 6 lags as being optimal. Test
results are available upon request, however, the final lag length
selection was determined on the full set of VAR and VECM output that
includes satisfying time series conditions such as no presence of
autocorrelation and plausible interpretability of the estimated
output.
[GRAPHIC] [TIFF OMITTED] TR30AP20.305
Taken together, these tests confirm the need to address the time
series properties of each variable in any modeling framework. This will
become especially important when discussing the correct modeling
approach, as The pre-modeling tests provide evidence against running a
simple OLS regression or VAR in first differences, because doing so
would have the potential outcome of excluding important long-run
information.
Furthermore, the endogeneity between vehicle sales and price is
another element that needs to be considered for model specification.
The IHS historical series for average price of a new light duty vehicle
is defined as a
[[Page 24601]]
function of business and private residential spending on light vehicles
divided by total new light vehicle sales; from this identity, the
average price represents the nominal price per new unit of light duty
vehicle sold. This definition supports the existence of an endogenous
relationship between vehicle price and sales that needs to be accounted
for when developing an econometric estimation of the influence of new
vehicle price on sales. This is consistent with economic theory,
whereby vehicle sales and price are simultaneously determined in the
market, and therefore should be included together when specifying a
forecasting equation.\1573\ This restriction holds even if nominal
vehicle price is transformed into a quality adjusted real dollar
series, as some commenters (EDF, Stock et al) proposed.\1574\
---------------------------------------------------------------------------
\1573\ Endogeneity results in correlation between an independent
variable in a regression and the error term leading to biased
coefficient estimates.
\1574\ For reference on how the BLS measures quality adjustments
in vehicles: https://www.bls.gov/cpi/factsheets/new-vehicles.htm.
---------------------------------------------------------------------------
Models
Faced with the simultaneity problem associated with price and
sales, several specifications were reviewed to determine the best
method for addressing this issue. An Instrumental Variable (IV) method
was deemed the most direct approach, with the advantage of preserving
the initial model's autoregressive distributed lag structure. In order
to obtain consistent estimates of the price elasticity of demand, a
suitable instrument that is correlated with average LDV price but
uncorrelated with the error term is needed in the first stage. A
suitable instrument must also make economic sense and have a plausible
causal relationship. In theory, instruments that satisfy all three
conditions (exogeneity, causality, and non-weak correlation) should
exist. In practice, however, it is often prohibitively difficult to
find a viable instrument. Both Stock et al. and CARB suggested
instrumenting to resolve the endogeneity issue in the NPRM model, but
neither suggested specific candidates for instrumental variables.
For the purposes of modeling vehicle sales, candidate IVs would
reflect the price of inputs to production that are broad enough, so
that the underlying behavior of the variable is not deterministic of
LDV sales. Examples of candidate variables include producer price
indices (PPIs) of auto or other related manufacturing, cost of capital
required for production, labor market data, energy costs, technology
changes, and exogenous shocks to price, production, labor, or policy
changes.
The lack of data availability and quality concerns reduced the
primary list of candidate IVs to relatable PPIs such as for
manufacturing and automobile primary products. Even the most
``promising'' candidate IVs, however, proved to be poor instruments,
with counterintuitive signs, lack of statistical significance, and poor
overall first stage F-statistics (even by relatively lenient weak
instrument test standards).
The lack of reasonable results from the IV approach led to testing
vector autoregressive (VAR) and vector error correction (VECM) models.
Relaxing the strict exogeneity assumption needed under an ARDL
framework is the main advantage of modeling price, sales, and
macroeconomic variables as a system of equations where the feedback
from previous period shocks affect both price and sales.\1575\ In
addition, a VAR or VECM can also adequately handle the time series and
nonstationary properties discussed above. For both the VAR and VECM, a
parsimonious specification was preferred with either a three or four
variable system using the variables discussed above.
---------------------------------------------------------------------------
\1575\ Strict exogeneity requires there to be past,
contemporaneous, and future exogeneity between the variables of
interest.
---------------------------------------------------------------------------
We first estimated a simple VAR using a Wold causal ordering of
real disposable income per household, average price of new LDV, and new
total sales of LDVs per household.\1576\ The alternative specification
included the consumer sentiment variable in the ordering the consumer
sentiment variable after income and before price. This ordering assumes
that households' disposable income (and consumer sentiment) do not
respond to shocks to auto prices and sales within the same quarter. It
also assumes that prices are contemporaneously exogenous of sales
(demand), since the MSRPs are set in advance. Lastly, sales are able to
respond to unexpected changes in price in the same quarter. The
alternative ordering of placing sales before average price was deemed
unrealistic as it would presume sales responding independently to an
unexpected change in prices.
---------------------------------------------------------------------------
\1576\ The Wold causal ordering creates a lower triangular
matrix for our shocks, so by construction these shocks are
orthogonal to each other to allow for causal inference. This
recursive or Wold ordering technique should be predetermined and
based on economic theory as the causal interpretation of the impulse
responses are dependent on the correct/plausible ordering of
variables.
---------------------------------------------------------------------------
In the first specification, all variables were transformed to first
differences to ensure stationarity, while ignoring any possible long-
run information (for the moment). A combination of post-estimation
tests for autocorrelation and stability conditions were considered
along with impulse response functions to gauge the model performance.
The preferred model was estimated with five lags, and the impulse
response functions (IRF) of a 1 percent shock to price on sales for the
two specifications are presented in Figure VI-61.
[[Page 24602]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.306
Both figures show a similar trend of the response in sales
oscillating from negative to positive before ultimately returning to
zero 12 quarters out. The three variable VAR sees a positive response
in the first few periods, while the four variable VAR manages to dip
below zero briefly after 4 periods out. This behavior, which by
definition is short-run due to the differencing of the variables, could
be representing auto dealerships' attempts to pull sales back to its
equilibrium level after the price shock pushes sales negative, implying
some level of over compensation during this process. Nonetheless,
despite the model showing there is some evidence of an immediate and
negative price elasticity, the overly simplified VAR model is missing
key long run information (as identified in the cointegration tests),
creating some reservations about the results. It is also worth noting
that the lagged positive response in sales from an unexpected price
shock is persistent regardless of the lag length selection, and in many
cases even more pronounced.
A number of preliminary conclusions can be drawn from the IRF
results shown in Figure VI-61. First, at least at this level of
aggregation, any short-run and immediate effect of a price increase on
total LDV sales is relatively small in nature. This does not suggest,
however, that the price elasticity of demand is zero. Instead, what may
be the case is that when faced with an unexpected change in price,
consumers will choose to purchase a less expensive car with fewer
features as opposed to no car at all. In other words, the level of
aggregation being used, total car sales, removes important variation
between the type of vehicle being sold and consumer purchasing
decisions from the data; what is left is a clouded version of the true
relationship between price and sales. Second, this type of VAR ignores
and throws out any long run information that may exist, which would
create omitted variable bias if such a cointegrated relationship
exists.
Based on the conclusions from the Johansen cointegration test, the
next step involved estimating the system as a VECM. As with the VAR
models, the VECM employs either a three or four variable system with
five lag lengths and an unconstrained constant in the model (no trend
in either the first differenced or cointegrating equations). In each
model, the cointegrating vector is normalized around sales (i.e., the
sales' coefficient is set to 1), and the model results indicate strong
evidence of a cointegrating relationship between the variables.
Aside from general agreement on a cointegrating relationship, the
VECM performance was weak in nearly every specification attempted, with
implausible magnitudes for the long-run coefficient estimates and
insignificant short-run dynamics. Moreover, the adjustment coefficient
for the sales equation is particularly weak and insignificant.\1577\
The limitations of the VECM could be rooted in the system being
normalized around sales, which lacks significant variation,
correlation, or possibly true causation with the other variables.
---------------------------------------------------------------------------
\1577\ The lack of a statistically significant adjustment
variable could be an indication of weak exogeneity. In this case
that would not be plausible given the clear endogeneity between
price and sales, and is more likely an indication of poor data and
the absence of reliable modelling approaches.
---------------------------------------------------------------------------
As with the VAR analysis, a similar focus is placed on the IRFs
presented in Figure VI-62. Here a one percent shock in price on LDV
sales shows a similar response between the two specifications, with an
increase during the first several periods before returning to a
negative and permanent long-run effect. This response is erroneous in
two ways: First, the sharp positive response during the first 8 to 10
quarters defies economic logic as an increase in the price of a normal
good should not induce an increase in sales. Second, the permanent and
negative effect is equally as confounding because it rules out the
ability for dealerships or auto manufacturers to adjust prices or
supply.\1578\
---------------------------------------------------------------------------
\1578\ Note that error bounds cannot be generated for VECM IRFs
using most statistical packages, so determining statistical
significance is difficult. Given the change from positive to
negative and the low magnitude of the response, it is quite possible
that this effect is indistinguishable from zero.
---------------------------------------------------------------------------
The updated econometric models of light duty vehicle sales
(described above) thus did not provide clear, significant or robust
insight into the magnitude of the price elasticity of demand. While the
VAR model specification points to an immediate short-run negative price
elasticity of demand (i.e., sales fall in the face of an immediate
price shock), this relationship is relatively small. In addition, the
fact that this specification excludes the identified cointegration
between the variables suggests that it is not robust or unbiased. In
short, the VECM and IV approaches were unable to provide reasonable and
meaningful results.
These results strongly suggest that the relationship between sales
and price is
[[Page 24603]]
not adequately estimated with the macro-level data used in this
analysis. Recent peer reviewers of the CAFE model had similar concerns.
In particular, these data are insufficient to explain the individual
consumer (micro-) level decision making process of purchasing a new
LDV. Aggregating the sales response to the national level reduces the
useful variation in the decision making process to levels unsuitable
for estimation. Commenters generally agreed with this conclusion.
Even assuming a theoretically and econometrically correct model was
possible, this relationship is impossible to evaluate at the current
data aggregation level. Future research may focus on constructing an
aggregate price elasticity of demand from consumer level data utilizing
discrete choice modeling or something similar. However, constructing
such models and integrating them into the simulations of the final rule
are beyond the scope of this analysis.
[GRAPHIC] [TIFF OMITTED] TR30AP20.307
Many commenters suggested that the NPRM model was unable to find a
statistically significant influence of fuel economy on sales because
the model was too highly aggregated, as the agencies found with the
econometric experimentation to estimate a price response. EDF, CARB,
and CA et al. and Oakland et al. expressed concern that using industry
averages eliminated the variation needed to detect consumer valuation
of fuel economy in new vehicle purchases. The agencies noted a similar
concern in the NPRM, citing the level of aggregation as the most likely
reason that the average fuel economy of a new vehicle was not a
statistically significant explanatory variable in the ARDL model. The
approach for the final rule includes an average value of improved fuel
economy in the sales response, as commenters suggested it should.
(a) How Do Car and Light Truck Buyers Value Improved Fuel Economy?
Many commenters (CARB, CA et al. and Oakland et al., NRDC, EDF,
CBD, North Carolina Department of Environmental Quality, IPI, EPA
Science Advisory Board, Stock et al.) stated that the agencies should
explicitly consider fuel savings, and the value that consumers ascribe
to it, in addition to changes in price when estimating the response of
new vehicle sales to different regulatory alternatives. NRDC stated,
``The decision between new vehicle purchase alternatives must consider
both differential costs and differential benefits. The CAFE model sales
algorithm considers only differential costs and is, therefore,
flawed.'' \1579\ The agencies agree that the degree to which new
vehicle buyers value improvements in fuel economy is an important
consideration when estimating the response of new vehicle sales to
potential standards. The effect of vehicle prices on sales is difficult
to detect at the aggregate level because price movements are correlated
with the current strength of the economy, which can appear as a
positive price elasticity when modeling sales, and there are various
technical econometric difficulties in identifying the effect of price
on sales (simultaneity, cointegration, etc., addressed above). The
sales response model in the final rule accounts for fuel savings
realized by buyers of new vehicles.
---------------------------------------------------------------------------
\1579\ NRDC, Attachment 3, NHTSA-2018-0067-11723, at 4.
---------------------------------------------------------------------------
Some commenters and EPA's Science Advisory Board noted that the
sales response equation omitted any value of fuel savings to new
vehicle buyers, while other elements of the analysis--notably the
technology application algorithm--assumed that buyers would demand fuel
economy technologies that ``pay back'' within the first 2.5 years of
ownership (as a result of avoided fuel costs), and manufacturers would
supply fuel economy at those levels even in the absence of standards.
This observation was made in comments by CARB, CBD, and IPI--the last
of which stated that 2.5 year payback assumption ``clashes directly
with the contradictory assumption that the agencies rely on in the
model's sales module, where they implicitly assume that customers
entirely disregard fuel efficiency in their purchasing decisions.''
\1580\ The agencies agree that this represented an internal
inconsistency. The sales model used to analyze the final rule includes
the estimated value of fuel savings to vehicle buyers, and is
consistent with other assumptions throughout the analysis about the
``pay back'' period.
---------------------------------------------------------------------------
\1580\ IPI, Appendix, NHTSA-2018-0067-12213, at 16.
---------------------------------------------------------------------------
How potential buyers value improvements in the fuel economy of new
cars and light trucks is an important issue in assessing the benefits
and costs of government regulation. If buyers fully value the savings
in fuel costs that result from higher fuel
[[Page 24604]]
economy, manufacturers will presumably supply any improvements that
buyers demand, and vehicle prices will fully reflect future fuel cost
savings consumers would realize from owning--and potentially re-
selling--more fuel-efficient models. If consumers internalize fuel
savings this case, more stringent fuel economy standards will impose
net costs on vehicle owners and can only result in social benefits
through correcting externalities, because consumers would already fully
incorporate private savings into their purchase decisions, as discussed
further below. If instead consumers systematically undervalue some
market failure such as an information asymmetry leads to an
underinvestment in fuel-saving technology, the cost savings generated
by improvements in fuel economy when choosing among competing models,
more stringent fuel economy standards will also lead manufacturers to
adopt improvements in fuel economy that buyers might not choose despite
the cost savings they offer and improve consumer welfare.
The potential for car buyers voluntarily to forego improvements in
fuel economy that offer savings exceeding their initial costs is one
example of what is often termed the ``energy-efficiency gap.'' This
appearance of such a gap, between the level of energy efficiency that
would minimize consumers' overall expenses and what they actually
purchase, is typically based on engineering calculations that compare
the initial cost for providing higher energy efficiency to the
discounted present value of the resulting savings in future energy
costs.
There has long been an active debate about why such a gap might
arise and whether it actually exists. Economic theory predicts that
individuals will purchase more energy-efficient products only if the
savings in future energy costs they offer promise to offset their
higher initial costs. However, the additional up-front cost of a more
energy-efficient product includes more than just the cost of the
technology necessary to improve its efficiency; because consumers have
a scarcity of resources, it also includes the opportunity cost of any
other desirable features that consumers give up when they choose the
more efficient alternative. In the context of vehicles, whether the
expected fuel savings outweigh the opportunity cost of purchasing a
model offering higher fuel economy will depend, among other things, on
how much its buyer expects to drive, his or her expectations about
future fuel prices, the discount rate he or she uses to value future
expenses, the expected effect on resale value, and whether more
efficient models offer equivalent attributes such as performance,
carrying capacity, reliability, quality, or other characteristics.
Published literature has offered little consensus about consumers'
willingness-to-pay for greater fuel economy, and whether it implies
over- under- or full-valuation of the expected discounted fuel savings
from purchasing a model with higher fuel economy. Most studies have
relied on car buyers' purchasing behavior to estimate their
willingness-to-pay for future fuel savings; a typical approach has been
to use ``discrete choice'' models that relate individual buyers'
choices among competing vehicles to their purchase prices, fuel
economy, and other attributes (such as performance, carrying capacity,
and reliability), and to infer buyers' valuation of higher fuel economy
from the relative importance of purchase prices and fuel economy.\1581\
Empirical estimates using this approach span a wide range, extending
from substantial undervaluation of fuel savings to significant
overvaluation, thus making it difficult to draw solid conclusions about
the influence of fuel economy on vehicle buyers' choices.\1582\ Because
a vehicle's price is often correlated with its other attributes (both
measured and unobserved), analysts have often used instrumental
variables or other approaches to address endogeneity and other
resulting concerns.\1583\
---------------------------------------------------------------------------
\1581\ In a typical vehicle choice model, the ratio of estimated
coefficients on fuel economy--or more commonly, fuel cost per mile
driven--and purchase price is used to infer the dollar value buyers
attach to slightly higher fuel economy.
\1582\ See Helfand & Wolverton (2011) and Green (2010) for
detailed reviews of these cross-sectional studies.
\1583\ See, e.g., Barry, et al. (1995).
---------------------------------------------------------------------------
Despite these efforts, more recent research has criticized these
cross-sectional studies; some have questioned the effectiveness of the
instruments they use,\1584\ while others have observed that
coefficients estimated using non-linear statistical methods can be
sensitive to the optimization algorithm and starting values.\1585\
Collinearity (i.e., high correlations) among vehicle attributes--most
notably among fuel economy, performance or power, and vehicle size--and
between vehicles' measured and unobserved features also raises
questions about the reliability and interpretation of coefficients that
may conflate the value of fuel economy with other attributes (Sallee,
et al., 2016; Busse, et al., 2013; Allcott & Wozny, 2014; Allcott &
Greenstone, 2012; Helfand & Wolverton, 2011).
---------------------------------------------------------------------------
\1584\ See Allcott & Greenstone (2012).
\1585\ See Knittel & Metaxoglou (2014).
---------------------------------------------------------------------------
In an effort to overcome shortcomings of past analyses, three
studies published fairly recently rely on panel data from sales of
individual vehicle models to improve their reliability in identifying
the association between vehicles' prices and their fuel economy
(Sallee, et al. 2016; Allcott & Wozny, 2014; Busse, et al., 2013).
Although they differ in certain details, each of these analyses relates
changes over time in individual models' selling prices to fluctuations
in fuel prices, differences in their fuel economy, and increases in
their age and accumulated use, which affects their expected remaining
life, and thus their market value. Because a vehicle's future fuel
costs are a function of both its fuel economy and expected gasoline
prices, changes in fuel prices have different effects on the market
values of vehicles with different fuel economy; comparing these effects
over time and among vehicle models reveals the fraction of changes in
fuel costs that is reflected in changes in their selling prices
(Allcott & Wozny, 2014). Using very large samples of sales enables
these studies to define vehicle models at an extremely disaggregated
level, which enables their authors to isolate differences in their fuel
economy from the many other attributes, including those that are
difficult to observe or measure, that affect their sale prices.\1586\
---------------------------------------------------------------------------
\1586\ These studies rely on individual vehicle transaction data
from dealer sales and wholesale auctions, which includes actual sale
prices and allows their authors to define vehicle models at a highly
disaggregated level. For instance, Allcott & Wozny (2014)
differentiate vehicles by manufacturer, model or nameplate, trim
level, body type, fuel economy, engine displacement, number of
cylinders, and ``generation'' (a group of successive model years
during which a model's design remains largely unchanged). All three
studies include transactions only through mid-2008 to limit the
effect of the recession on vehicle prices. To ensure that the
vehicle choice set consists of true substitutes, Allcott & Wozny
(2014) define the choice set as all gasoline-fueled light-duty cars,
trucks, SUVs, and minivans that are less than 25 years old (i.e.,
they exclude vehicles where the substitution elasticity is expected
to be small). Sallee et al. (2016) exclude diesels, hybrids, and
used vehicles with less than 10,000 or more than 100,000 miles.
---------------------------------------------------------------------------
These studies point to a somewhat narrower range of estimates than
suggested by previous cross-sectional studies; more importantly, they
consistently suggest that buyers value a large proportion--and perhaps
even all--of the future savings that models with higher fuel economy
offer.\1587\
[[Page 24605]]
Because they rely on estimates of fuel costs over vehicles' expected
remaining lifetimes, these studies' estimates of how buyers value fuel
economy are sensitive to the strategies they use to isolate differences
among individual models' fuel economy, as well as to their assumptions
about buyers' discount rates and gasoline price expectations, among
others. Since Anderson et al. (2013) found evidence that consumers
expect future gasoline prices to resemble current prices, the agencies
use this assumption to compare the findings of the three studies and
examine how their findings vary with the discount rates buyers apply to
future fuel savings.\1588\
---------------------------------------------------------------------------
\1587\ Killian & Sims (2006) and Sawhill (2008) rely on similar
longitudinal approaches to examine consumer valuation of fuel
economy except that they use average values or list prices instead
of actual transaction prices. Since these studies remain
unpublished, their empirical results are subject to change, and they
are excluded from this discussion.
\1588\ Each of the studies makes slightly different assumptions
about appropriate discount rates. Sallee et al. (2016) use five
percent in their base specification, while Allcott & Wozny (2014)
rely on six percent. As some authors note, a five to six percent
discount rate is consistent with current interest rates on car
loans, but they also acknowledge that borrowing rates could be
higher in some cases, which could be used to justify higher discount
rates. Rather than assuming a specific discount rate, Busse et al.
(2013) directly estimate implicit discount rates at which future
fuel costs would be fully internalized; they find discount rates of
six to 21 percent for used cars and one to 13 percent for new cars
at assumed demand elasticities ranging from -2 to -3. Their
estimates can be translated into the percent of fuel costs
internalized by consumers, assuming a particular discount rate. To
make these results more directly comparable to the other two
studies, we assume a range of discount rates and uses the authors'
spreadsheet tool to translate their results into the percent of fuel
costs internalized into the purchase price at each rate. Because
Busse et al. (2013) estimate the effects of future fuel costs on
vehicle prices separately by fuel economy quartile, these results
depend on which quartiles of the fuel economy distribution are
compared; our summary shows results using the full range of quartile
comparisons.
---------------------------------------------------------------------------
As Table VI-148 indicates, Allcott & Wozny (2014) found that
consumers incorporate 55% percent of future fuel costs into vehicle
purchase decisions at a six percent discount rate, when their
expectations for future gasoline prices are assumed to reflect
prevailing prices at the time of their purchases. With the same
expectation about future fuel prices, the authors report that consumers
would fully value fuel costs only if they apply discount rates of 24
percent or higher. However, these authors' estimates are closer to full
valuation when using gasoline price forecasts that mirror oil futures
markets, because the petroleum market expected prices to fall during
this period (this outlook reduces the discounted value of a vehicle's
expected remaining lifetime fuel costs). With this expectation, Allcott
& Wozny (2014) find that buyers value 76 percent of future cost savings
(discounted at six percent) from choosing a model that offers higher
fuel economy, and that a discount rate of 15 percent would imply that
they fully value future cost savings. Sallee et al. (2016) begin with
the perspective that buyers fully internalize future fuel costs into
vehicles' purchase prices and cannot reliably reject that hypothesis;
their base specification suggests that changes in vehicle prices
incorporate slightly more than 100 percent of changes in future fuel
costs. For discount rates of five to six percent, the Busse et al.
(2013) results imply that vehicle prices reflect 60 to 100 percent of
future fuel costs. As Table VI-151 suggests, higher private discount
rates move all of the estimates closer to full valuation or to over-
valuation, while lower discount rates imply less complete valuation in
all three studies.
[GRAPHIC] [TIFF OMITTED] TR30AP20.308
The studies also explore the sensitivity of the results to other
parameters that could influence their results. Busse et al. (2013) and
Allcott & Wozny (2014) find that relying on data that suggest lower
annual vehicle use or survival probabilities, which imply that vehicles
will not last as long, moves their estimates closer to full valuation,
an unsurprising result because both reduce the changes in expected
future fuel costs caused by fuel
[[Page 24606]]
price fluctuations. Allcott & Wozny's (2014) base results rely on an
instrumental variables estimator that groups miles-per-gallon (MPG)
into two quantiles to mitigate potential attenuation bias due to
measurement error in fuel economy, but they find that greater
disaggregation of the MPG groups implies greater undervaluation (for
example, it reduces the 55 percent estimated reported in Table VI-148
to 49 percent). Busse et al. (2013) allow gasoline prices to vary
across local markets in their main specification; using national
average gasoline prices, an approach more directly comparable to the
other studies, results in estimates that are closer to or above full
valuation. Sallee et al. (2016) find modest undervaluation by vehicle
fleet operators or manufacturers making large-scale purchases, compared
to retail dealer sales (i.e., 70 to 86 percent).
Since they rely predominantly on changes in vehicles' prices
between repeat sales, most of the valuation estimates reported in these
studies apply most directly to buyers of used vehicles. Only Busse et
al. (2013) examine new vehicle sales; they find that consumers value
between 75 to 133 percent of future fuel costs for new vehicles, a
higher range than they estimate for used vehicles. Allcott & Wozny
(2014) examine how their estimates vary by vehicle age and find that
fluctuations in purchase prices of younger vehicles imply that buyers
whose fuel price expectations mirror the petroleum futures market value
a higher fraction of future fuel costs: 93 percent for one- to three-
year-old vehicles, compared to their estimate of 76 percent for all
used vehicles assuming the same price expectation.\1589\
---------------------------------------------------------------------------
\1589\ Allcott & Wozny (2014) and Sallee, et al. (2016) also
find that future fuel costs for older vehicles are substantially
undervalued (26-30%). The pattern of Allcott and Wozny's results for
different vehicle ages is similar when they use retail transaction
prices (adjusted for customer cash rebates and trade-in values)
instead of wholesale auction prices, although the degree of
valuation falls substantially in all age cohorts with the smaller,
retail price based sample.
---------------------------------------------------------------------------
Accounting for differences in their data and estimation procedures,
the three studies described here suggest that car buyers who use
discount rates of five to six percent value at least half--and perhaps
all--of the savings in future fuel costs they expect from choosing
models that offer higher fuel economy. Perhaps more important in
assessing the case for regulating fuel economy, one study (Busse et
al., 2013) suggests that buyers of new cars and light trucks value
three-quarters or more of the savings in future fuel costs they
anticipate from purchasing higher-mpg models, although this result is
based on more limited information.
In contrast, previous regulatory analyses of fuel economy standards
implicitly assumed that buyers undervalue even more of the benefits
they would experience from purchasing models with higher fuel economy,
so that, without increases in fuel economy standards, little
improvement would occur, and the entire value of fuel savings from
raising CAFE standards represented private benefits to car and light
truck buyers themselves. For instance, in the EPA analysis of the 2017-
2025 model year CO2 standards, fuel savings alone added up
to $475 billion (at three percent discount rate) over the lifetime of
the vehicles, far outweighing the compliance costs: $150 billion). The
assertion that buyers were unwilling to take voluntary advantage of
this opportunity implies that collectively, they must have valued less
than a third ($150 billion/$475 billion = 32 percent) of the fuel
savings that would have resulted from those standards. In fact, those
earlier analyses assumed that new car and light truck buyers attach
relatively little value to higher fuel economy, since their baseline
scenarios assumed that fuel economy levels would not increase in the
absence of progressively tighter standards, despite increasing fuel
prices. The evidence reviewed here makes that perspective extremely
difficult to justify and would call into question any analysis that
claims to show large private net benefits for vehicle buyers
attributable to increases in fuel economy standards.
What analysts assume about consumers' vehicle purchasing behavior,
particularly about potential buyers' perspectives on the value of
increased fuel economy, clearly matters a great deal in the context of
benefit-cost analysis for fuel economy regulation. In light of this
recent evidence on this question, warrants a more nuanced approach that
is more nuanced than merely assuming that buyers drastically undervalue
benefits from higher fuel economy, (and that, as a consequence, these
benefits are unlikely to be realized without stringent fuel economy
standards,) seems warranted. One possible approach would be to use a
baseline scenario where fuel economy levels of new cars and light
trucks reflected full (or nearly so) valuation of fuel savings by
potential buyers in order to reveal whether setting fuel economy
standards above market-determined levels could produce net social
benefits. Another might be to assume that, unlike in the agencies'
previous analyses, where buyers were assumed to greatly to undervalue
higher fuel economy under the baseline but to value it fully under the
proposed standards, buyers value improved fuel economy identically
under both the baseline scenario and with stricter CAFE standards in
place.
The agencies requested comment on the consumer valuation of fuel
economy and its use in the NPRM analysis. CBD and the North Carolina
Department of Environmental Quality took issue with the agencies'
characterization of the literature on the value of fuel economy, citing
EPA's previous determination that the estimates in the literature
represented too large a range, and the degree of uncertainty made
including a value of fuel economy challenging. This final rule analysis
accounts for the value of fuel economy in several places, though it
uses a more conservative value than is suggested by the literature
summarized above. Manufacturers have consistently told the agencies
that new vehicle buyers will pay for about 2 or 3 years' worth of fuel
savings before the price increase associated with providing those
improvements begins to impact affect sales. The agencies have assumed
the same valuation, 2.5 years, in all components of the analysis that
reflect consumer decisions regarding vehicle purchases and
retirements.\1590\ This analysis explicitly assumes that: (1) Consumers
are willing to pay for fuel economy improvements that pay back within
the first 2.5 years of vehicle ownership (at average usage rates); (2)
manufacturers know this and will provide these improvements even in the
absence of regulatory pressure; (3) potential buyers weigh these
savings against increases in new vehicle prices when deciding to retire
a vehicle; and (4) the amount of technology for which buyers will pay
rises (or falls) with rising (or falling) fuel prices.\1591\ Excluding
the value of fuel economy entirely from these calculations does not
remove it from the analysis; it merely imposes an implausibly low value
on the desired payback period of new
[[Page 24607]]
vehicle buyers and manufacturers--regardless of fuel prices or
technology costs. And while the agencies acknowledge the uncertainty
around the estimates in the literature, zero is far removed from the
lower bounds of any study.
---------------------------------------------------------------------------
\1590\ When accounting for social benefits and costs associated
with an alternative, the full lifetime value of fuel savings is
included.
\1591\ NADA, the Alliance of Automobile Manufacturers, and
American Fuel and Petrochemical Manufacturers argued that CAFE/
CO2 standards have already reached the point where the
price increases necessary to recoup manufacturers' increased costs
for providing further increases in fuel economy outweigh the value
of fuel savings, and requiring further increases in fuel economy
will reduce new vehicle sales. The sales response in the final rule
recognizes and incorporates the effect of fuel prices and fuel
economy on new vehicle purchases. See NADA, NHTSA-2018-0067-12064,
at 11; Auto Alliance, Full Comment Set, NHTSA-2018-0067-12073 at
163-64; AMFP, Comments, NHTSA-2018-0067-12078-29,at 3.
---------------------------------------------------------------------------
CARB asserted that the various market failures suggested by the
agencies in past rules (lack of information about fuel savings from
higher MPG, inability to calculate cost savings from higher MPG, loss
aversion, first-mover disadvantage), together with advertising that
only emphasizes fuel economy during periods of high fuel prices, leads
buyers to undervalue fuel economy.\1592\ In contrast, CARB (and
others--such as SCAQMD, Alliance to Save Energy, Save EPA, AAA,
Environmental group coalition, Consumers Union, EDF, and IPI) argues
elsewhere that new vehicle buyers do value fuel economy highly, and
nearly fully once fuel prices return to ``normal'' levels.\1593\ The
agencies' payback period assumption, and the matching adjustment it
makes to changes in new car prices to account for accompanying changes
in fuel economy, recognizes that on average potential car buyers value
a significant share of lifetime cost savings resulting from higher fuel
economy. The agencies considered longer payback periods along the lines
suggested by Consumer Federation of America (CFA),\1594\ but chose 2.5
years as a conservative approach. Our assumption is consistent with
survey evidence cited by the commenters, but at odds with their
assertions that this program is necessary to save buyers from their own
limited ability to make decisions in their best interest.
---------------------------------------------------------------------------
\1592\ See CARB, Detailed Comments, NHTSA-2018-0067-11873 at
212-16.
\1593\ E.g. id. at 190-91. See also, id. at 188-89. See also,
SCAQMD, Supplemental comments, NHTSA-2018-0067-11813, at 4-5;
Alliance to Save Energy, Comment, NHTSA-2018-0067-11837, at 2; Save
EPA, Comments, NHTSA-2018-0067-11930, at 6; AAA, Comments, NHTSA-
2018-0067-11979, at 2-3; Environmental group coalition, Appendix A,
NHTSA-2018-0067-12000, at 54-56; Consumers Union, Attachment A,
NHTSA-2018-0067-12068, 27-29; EDF, Appendix B, NHTSA-2018-0067-
12108, at 84-86; and IPI, Appendix, NHTSA-2018-0067-12213, at 40-47.
\1594\ CFA, Comments, NHTSA-2018-0067-12005, at 12.
---------------------------------------------------------------------------
More recently, the agencies have justified stricter CAFE and
CO2 emissions standards by asserting that buyers do not take
advantage of opportunities to improve their own well-being, by
purchasing models whose higher fuel economy would more than repay their
higher initial purchase prices via future savings in fuel costs. This
newer rationale is fundamentally different from asserting that some
externality--whereby buyers' choices cause economic harm to others--
exists to justify regulating fuel economy or CO2 emissions,
or adopting more demanding regulations. EPA and NHTSA have previously
labeled this behavior an example of the ``energy paradox,'' whereby
consumers voluntarily forego investments that conserve energy even when
those initial outlays appear likely to repay themselves--in the form of
savings in energy costs--over the relatively near term.\1595\
---------------------------------------------------------------------------
\1595\ See, e.g., EPA Regulatory Impact Analysis: Final
Rulemaking for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission
Standards and Corporate Average Fuel Economy Standards, available at
https://nepis.epa.gov/Exe/ZyPDF.cgi/P100EZI1.PDF?Dockey=P100EZI1.PDF.
---------------------------------------------------------------------------
However, recent research cast doubt on whether such an energy
paradox exists in the case of fuel economy--that is, on whether buyers
of new vehicles inadequately consider the value of future savings in
fuel costs they would experience from purchasing models that feature
higher fuel economy--and about how extensive it might be. Several
recent studies have estimated the fraction of appropriately discounted
lifetime fuel savings offered by models featuring higher fuel economy
that car shoppers appear to value or willing to pay for. These
estimates are typically drawn from one of three sources--(1) buyers'
choices among competing models with different purchase prices, fuel
economy levels, and other features; (2) statistically ``decomposing''
vehicle prices into the values buyers attach to their individual
features, one of which is fuel economy; or (3) analyzing how selling
prices for vehicles with different fuel economy levels respond to
variation in fuel prices and the changes it causes in their lifetime
fuel costs.
The estimates these studies report may partly reflect variation
among buyers' preferences for different vehicle features (such as fuel
economy, but also size or utility), the financial constraints they
face, how much they drive, or their expectations about future fuel
prices, so they should be interpreted cautiously. However, the most
careful recent studies suggest that on average buyers appear to
undervalue the savings from higher fuel economy at most modestly, and
perhaps not at all, after accounting for the influence of vehicles'
other attributes on prices and purchasing decisions.\1596\ This
research suggests that the energy paradox, sometimes described as
buyers' ``myopia'' in assessing the value of future fuel savings, is a
much weaker rationale for regulating fuel economy than the agencies had
previously asserted.
---------------------------------------------------------------------------
\1596\ For a review of these recent studies, see Table VI-120--
Percent of Future Fuels Costs Internalized in Used Vehicle Purchase
Price using Current Gasoline Prices to Reflect Expectations (for
Base Case Assumptions).
---------------------------------------------------------------------------
IPI commented that the agencies' obligation to consider market
failures in setting standards derives not just from Executive Order
12,866 but also from the agencies' respective statutes, and argued that
the agencies had defined market failures too narrowly in their
proposal.\1597\ Specifically, IPI stated that NHTSA's task under EPCA
is ``not so restricted to only protecting consumers from gas price
spikes,'' and argued that NHTSA must also consider ``externalities
relating to energy security, national security, positional goods,
global climate change, and air and water pollution associated with fuel
production and consumption; asymmetric information, attention costs,
and other information failures; internalities, including myopia; and
various supply-side market failures, including first-mover
disadvantage.'' \1598\
---------------------------------------------------------------------------
\1597\ IPI, Appendix, NHTSA-2018-0067-12213, at 9-10.
\1598\ Id.
---------------------------------------------------------------------------
For EPA's task under the CAA, IPI stated that, although while EPA
must ``protect the planet from unchecked climate change, [it] must not
ignore other related market failures that cause harm to public health
and welfare, including the issues and market failures [as described for
NHTSA above].'' \1599\ IPI argued that the proposal was arbitrary and
capricious for not ``consider[ing] important aspects of the problem set
before the agencies by Congress,'' and also for not considering the
market failures discussed in the 2012 final rule.\1600\ CBD, et al.,
asserted similarly that the agencies' respective statutes require their
actions to be more technology-forcing than what markets would otherwise
achieve, in effect asserting that innovations in technology confer
external benefits that vehicle manufacturers or buyers do not fully
consider.\1601\
---------------------------------------------------------------------------
\1599\ Id.
\1600\ Id.
\1601\ CBD, et al., NHTSA-2018-0067-12057, at 2 and 9.
---------------------------------------------------------------------------
With regard to the specific market failures CAFE and CO2
standards could potentially address, Global Automakers suggested that
climate effects are indeed an externality that more stringent standards
can address,\1602\ while CFA stated that regulating fuel economy and
CO2 emissions can address an extensive catalog of market
failures, including externalities, marketing, availability of
[[Page 24608]]
fuel-efficient models, transaction cost friction, information
asymmetry, behavioral issues, and access to capital, among
others.\1603\ CFA asserted that advances in economic theory had heavily
criticized the neoclassical model, and that ``a great deal of empirical
evidence supports [that the] standards are seen as an important and, in
many ways, preferred policy approach.'' \1604\ On this basis, CFA
stated that attribute-based standards that ``are set at a moderately
aggressive level'' and are ``consistent with the rate of improvement
that the auto industry achieved in the first decade of the fuel economy
standard setting program,'' among other things, would address the
market failure.\1605\
---------------------------------------------------------------------------
\1602\ Global Automakers, Attachment A, NHTSA-2018-0067-12032,
at A-22.
\1603\ CFA, Comments, NHTSA-2018-0067-12005, at 61-64.
\1604\ Id. at 63.
\1605\ Id. at 64.
---------------------------------------------------------------------------
IPI argued that regulation of fuel economy (presumably also
CO2 emissions) is necessary because ``many vehicle
attributes, like horsepower and size, are positional goods--that is,
they confer status on buyers of cars and light truck models that
feature them prominently, so regulation of fuel economy can help
correct the positional externality.'' \1606\ IPI also noted the
externality of health effects associated with refueling. IPI cited
Alcott and Sunstein (2015) to argue, like CFA, that fuel economy
standards can correct market failures like informational failure,
myopia, supply-side failures, positional externalities, etc., and by
doing so, can provide net private welfare gains--that is, improve the
utility of vehicle buyers themselves, not just that of other households
or businesses.\1607\
---------------------------------------------------------------------------
\1606\ IPI, Appendix, NHTSA-2018-0067-12213, at 33.
\1607\ Id. at 34. Note, however, that the reference cited does
not address the question of whether fuel economy standards can be
effective in correcting those market failures. Instead, it explores
the circumstances under which fuel economy standards can improve
welfare when vehicle buyers undervalue savings in fuel costs from
purchasing more fuel-efficient models. See generally, Allcott, Hunt,
and Cass R. Sunstein, ``Regulating Internalities,'' Working Paper
20087, National Bureau of Economic Research, May 2015, available at
https://www.nber.org/papers/w21187.pdf.
---------------------------------------------------------------------------
EDF and CARB both asserted that an energy paradox exists in the
case of fuel economy, with EDF arguing (like CFA) that information
asymmetry--that is, unequal access of vehicle manufacturers and
potential buyers to information about the cost savings likely to result
from owning higher-mpg models--coupled with limited availability of
fuel-efficient models, leads consumers to purchase vehicles with lower
fuel economy than they otherwise would.\1608\ CARB simply stated that
the NPRM analysis did not account for the energy paradox.\1609\
---------------------------------------------------------------------------
\1608\ EDF, Appendix B, NHTSA-2018-0067-12108, at 88-89.
\1609\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 188-
89.
---------------------------------------------------------------------------
The agencies agree with these commenters that the market failures
CAFE and CO2 standards can help address are likely to exist,
but note that little of the behavior in the broad catalog identified by
commenters actually represents market failures, and instead simply
reflects consumers' preferences for features other than fuel economy.
Even in the few cases of potential market failures that commenters
identify related to the hypothetical energy paradox, the agencies
question whether more stringent CAFE and CO2 standards are
necessary to address the phenomena, or are even likely to be effective
in doing so. In the agencies' view, neither the logical arguments nor
the limited empirical evidence that commenters presented convincingly
demonstrate the capacity of more stringent CAFE and CO2
standards to resolve, or even mitigate, most of the various phenomena
they describe as market failures.
For example, the idea that regulating fuel economy and
CO2 emissions can mitigate the consequences of inadequate
access to information by placing decisions that depend on access to
complete information in the hands of regulators rather than buyers has
superficial appeal. Yet commenters do not establish that such a drastic
step is necessary to overcome any inadequacy of information, or that
requiring manufacturers to supply higher fuel economy will be more
effective than less intrusive approaches such as expanding the range of
information available to buyers. As OMB Circular A-4 notes, ``Because
information, like other goods, is costly to produce and disseminate,
your evaluation will need to do more than demonstrate the possible
existence of incomplete or asymmetric information.'' \1610\
---------------------------------------------------------------------------
\1610\ Circular A-4, at 5.
---------------------------------------------------------------------------
In the few cases where commenters cited empirical evidence to
support their arguments that stricter fuel economy and CO2
regulations are an appropriate response to market failures, that
evidence is limited and unpersuasive. As one illustration, the frequent
assertion that buyers' widespread aversion to the prospect of financial
losses makes them hesitant to purchase higher-mpg models appears to be
traceable to findings from classroom experiments on small numbers of
university students, rather than to large-scale empirical evidence
drawn from buyers' observed behavior.\1611\ Commenters' repeated
emphasis on loss aversion as a critical source of buyers' unwillingness
to choose levels of fuel economy that appear to be in their own
financial interest also ignores recent research questioning whether
loss aversion is a plausible motivation for such systematic or
universal behavior by consumers.\1612\
---------------------------------------------------------------------------
\1611\ CFA, Comments, NHTSA-2018-0067-12005, at 16 et seq;
Consumers Union, Attachment 4, NHTSA-2018-0067-12068, at 12;
Attachment 3, NHTSA-2018-0067-11741, at 5-6, CARB at 214, and States
at 87 each assert that loss aversion is an important source of car
buyers' hesitance to purchase higher-mpg models, variously citing
Greene, David L., John German, and Mark A. Delucchi, ``Fuel Economy:
The Case for Market Failure,'' Reducing Climate Impacts in the
Transportation Sector, Springerin James S. Cannon and Daniel
Sperling, eds., Springer, 2009, at pp. 181-205; (2009); Greene,
David L. (2010). How consumers value fuel economy: A literature
review (No. EPA-420-R-10-008); Greene, David L., ``Uncertainty, Loss
Aversion and Markets for Energy Efficiency,'' Energy Economics, vol.
33, at pp. 608-616, (2011) and Greene, David L., ``Consumers'
Willingness to Pay for Fuel Economy: Implications for Sales of New
Vehicles and Scrappage of Used Vehicles,'' attachment to comments by
CARB, Oct. 10, 2018. However, none of these sources presents
empirical evidence on how the frequency of actual common loss
aversion actually is among real world vehicle buyers, instead simply
asserting (or implicitly assuming) that loss aversion it is likely
to be widespread. Further, their (identical) estimates of the degree
of loss aversion are difficult to trace, and appear to be drawn from
classroom exercises administered to limited numbers of university
students, not from empirical research involving real world vehicle
buyers. One source cited for their repeated assertion that losses of
a given dollar amount are valued twice as highly as gains of the
same amount is Gal, David, ``A psychological law of inertia and the
illusion of loss aversion,'' Judgment and Decision Making, Vol. 1,
No. 1, at pp. 23-32 (July 2006,), pp. 23-32, but this reference does
not report such a value. Another source repeatedly cited by Greene
and co-authors, Benartzi, Shlomo, and Richard H. Thaler, ``Myopic
Loss Aversion and the Equity Premium Puzzle,'' Quarterly Journal of
Economics, Vol. 110, No. 1, at pp. 73-92 (February 1995), pp. 73-92,
does report this value (at p. 74), although only in passing, and
cites other references as its original source. The original sources
of the claim that losses are values twice as highly as equivalent
gains appear to be Kahneman, Daniel, Jack L. Knetsch, and Richard H.
Thaler, ``Experimental Tests of the Endowment Effect and the Coase
Theorem,'' Journal of Political Economy, Vol. 98, No. 6, pp. 1325-
48. (Dec., 1990) (pp. 1325-1348, specifically Section II), pp. 1329-
1336; and Tversky, Amos, and Daniel Kahneman, ``Loss Aversion in
Riskless Choice: A Reference-Dependent Model,'' Quarterly Journal of
Economics, Vol. 106, No. 4, at pp. 1039-61 (Nov., 1991) (pp. 1039-
1061, specifically pp. 1053-1054). Neither of these references,
however, makes any claim about the generality of the estimate or its
applicability to non-experimental settings for consumer behavior.
\1612\ See Gal, David, ``A psychological law of inertia and the
illusion of loss aversion,'' Judgment and Decision Making, Vol. 1,
No. 1, pp. 23-32 (July 2006,) pp. 23-32,; Erev, I., E. Ert, and E.
Yechiam, ``Loss aversion, diminishing sensitivity, and the effect of
experience on repeated decisions.'', Journal of Behavioral Decision
Making, Vol. 21 (2008), pp. 575-97; (2008); Ert, E., and I. Erev,
``On the descriptive value of loss aversion in decisions under risk:
Six clarifications,'' Judgment and Decision Making, Vol. 8 (2013),
at pp. 214-35; (2013); Gal, David and Rucker, Derek, ``The Loss of
Loss Aversion: Will It Loom Larger Than Its Gain?'' Journal of
Consumer Psychology, Vol. 28 No. 3, (July 2018), at pp. 497-516
(July 2018) available at (https://onlinelibrary.wiley.com/doi/abs/10.1002/jcpy.1047); and Gal, David, ``Why the Most Important Idea in
Behavioral Decision-Making Is a Fallacy,'' Scientific American,
Observations, (July 31, 2018), available at (https://blogs.scientificamerican.com/observations/why-the-most-important-idea-in-behavioral-decision-making-is-a-fallacy/).
---------------------------------------------------------------------------
[[Page 24609]]
Another example is commenters' repeated citation of the study of
households' difficulties in analyzing the financial value of purchasing
vehicles with higher fuel economy conducted by Turrentine and Kurani,
which relies on interviews with a limited number of subjects (57
California households) to conclude that consumers are systematically
unable to perform the calculations necessary to estimate the value of
fuel savings.\1613\ These same commenters consistently ignore the
wealth of detailed, publicly-available information on the fuel economy
of new vehicle models, and shoppers' ready access to user-friendly
tools to estimate the savings they are likely to realize from
purchasing higher-mpg models. These tools include the label that
prominently displays how much a vehicles' fuel economy will save, or
conversely, cost a purchaser in fuel costs over 5 years of use in color
and large type (see Figure VI-63), which is legally required to be
prominently displayed on all new cars vehicles offered for sale.\1614\
Separately, new car dealers are also required to prominently display
the Federal Fuel Economy Guide for each model year of new vehicles
offered for sale, which provides fuel economy information for all
vehicles from that model year.\1615\
---------------------------------------------------------------------------
\1613\ ICCT at p. 4 and Consumers Union at p. 12 (among others),
citing Turrentine, T.S., & Kurani, K.S., ``Car buyers and fuel
economy?,'' Energy policy, Vol. 35 No. 2 (2007), at 1213-1223,
available at https://www.sciencedirect.com/science/article/pii/S0301421506001200, as evidence that most or all new-car shoppers are
incapable of calculating the savings they would realize from
purchasing a higher-mpg model, and further misinterpret the study as
evidence that buyers invariably underestimate the value of increased
fuel economy. Yet this widely relied-upon analysis included only 57
households, all located in California. As an illustration, citing
Turrentine and Kurani, ICCT asserts ``There is substantial
circumstantial evidence that most consumers in the U.S. place a low
value on fuel economy.'' See ICCT at 4 (emphasis added). Similarly,
Consumers Union simply asserts that ``Households do not track
gasoline prices over time and cannot accurately estimate future gas
prices or cost savings.'' See Consumers Union at 12, again citing
Turrentine and Kurani as authority).
\1614\ See 15 U.S.C. 1531, et seq., and 49 CFR 575.401.
\1615\ 40 CFR 600.405-08 and 600.407-08.
[GRAPHIC] [TIFF OMITTED] TR30AP20.309
Similarly, no commenters offered empirical evidence to support
their repeated assertions that buyers or the public actually view
features such as styling, size, or performance as ``positional goods''
to which other potential buyers might aspire, or considered the
possibility that high fuel economy or advanced technology (such as
hybrid or electric propulsion) might themselves represent such
positional attributes.\1616\ Nor do commenters
[[Page 24610]]
provide any empirical evidence that the various aspects of behavior
they allege lead buyers to underinvest in fuel economy--ranging from
unwillingness to spend time or effort estimating likely fuel savings,
to inattentiveness to the economic and social importance of improved
fuel economy, inability to obtain information about the savings it
offers them, and incorrect ``framing'' of the choice among models with
different levels of fuel economy--are widespread, empirically
significant, or systematically likely to lead buyers to under- rather
than over-invest in fuel economy.
---------------------------------------------------------------------------
\1616\ For evidence that prestige appears to be a motivation for
purchasing advanced-technology vehicles, see Hidrue, Michael K., et
al., ``Willingness to pay for electric vehicles and their
attributes,'' Resource and Energy Economics, Vol. 33, Issue 3
(September 2011), at pp. 686-705; Chua, Wan Ying, Lee, Alvin and
Sadeque, Saalem 2010, ``Why do people buy hybrid cars?,''
Proceedings of Social Marketing Forum, University of Western
Australia, Perth, Western Australia, Edith Cowan University,
Churchlands, W.A., at pp. 1-13; Liu, Yizao, ``Household demand and
willingness to pay for hybrid vehicles,'' Energy Economics, Volume
44, 2014, at pp. 191-197; Hur, Won-Moo, Jeong Woo, and Yeonshim Kim,
``The Role of Consumer Values and Socio-Demographics in Green
Product Satisfaction: The Case of Hybrid Cars,'' Psychological
Reports, Volume 117, issue 2, October 2015, at pp. 406-427. A useful
summary of many studies appears in Table 1 (p. 196) of Makoto
Tanaka, Takanori Ida, Kayo Murakami, Lee Friedman, ``Consumers'
willingness to pay for alternative fuel vehicles: A comparative
discrete choice analysis between the US and Japan,'' Transportation
Research Part A: Policy and Practice, Volume 70, 2014, at pp. 194-
209 (Table 1 at p. 196). Some of these studies find that buyers are
apparently willing to pay significant price premiums for the
prestige or status value of hybrids or battery-electric vehicles--
which their authors speculate may derive from their ``greenness''--
because their purchases cannot be explained on the basis of economic
or financial considerations. Others find that average or typical
shoppers' willingness to pay advanced-technology vehicles is below
the price premiums they command, suggesting that their purchasers
must derive some status or prestige value from owning and driving
them.
---------------------------------------------------------------------------
The most frequent argument that an energy paradox or energy
efficiency ``gap'' exists in the case of fuel economy is the
observation that many U.S. vehicle buyers seem unwilling to pay higher
prices for models whose increased fuel economy would appear to repay
their additional investment within a relatively brief ownership period.
However, this argument is unpersuasive for at least three reasons: Most
obviously, it does not acknowledge the possibility that engineering
studies systematically underestimate costs to produce vehicles with
higher fuel economy, and thus the prices that buyers would be asked to
pay for models with improved fuel economy. Nor does it account for
potential sacrifices in other vehicle attributes that manufacturers may
make in order to achieve higher fuel economy without increasing
vehicles' purchase prices beyond consumers' willingness to pay.
Finally, claims that consumers are acting irrationally by refusing to
purchase higher-mpg models usually reach this conclusion by comparing
rates at which they implicitly discount future fuel costs--and thus
evaluate savings from purchasing more fuel-efficient models--to
interest rates in financial markets that incorporate time horizons or
risk profiles that may be very different from those of consumers.
Even putting these concerns aside, comparing future fuel savings to
the costs of purchasing more expensive models that offer higher fuel
economy demonstrates only that buyers are not behaving as analysts
expect them to and believe they should behave. These comparisons do not
demonstrate that consumers are necessarily acting irrationally, and
cannot diagnose the nature of information shortcomings buyers face,
reasons that they might interpret such information incorrectly, or
identify behavioral inconsistencies they may exhibit. In short,
conjectures about why buyers might undervalue potential savings from
investing in higher-efficiency vehicle models do not represent evidence
that they actually do so, and as discussed above, recent research seems
to show that such behavior is not widespread, if it exists at all.
Past joint rulemaking efforts by NHTSA and EPA have repeatedly
sought to identify a plausible explanation for car buyers' perceived
undervaluation of improved fuel economy. The agencies have occasionally
relied on explanations such as consumers' insufficient appreciation of
the importance of fuel economy, the difficulty of obtaining adequate
information about the fuel economy of competing models or of converting
competing models' fuel economy ratings to future fuel costs and
savings, or consumers' misunderstanding or mistrust of such information
when it is provided to them. At other times, the agencies have pointed
to consumers' ``myopia'' about the future--asserting that for some
reason, they appear to underestimate future fuel costs and savings--or
argued that shoppers are insufficiently attentive to fuel costs when
comparing competing models, that the value of improved fuel economy is
obscured (``shrouded'') by vehicles' other, more visible attributes, or
that uncertainty about the savings in fuel costs owners will actually
realize causes them to undervalue those savings when comparing the
upfront costs of models with different fuel economy.
Despite the frequency with which the agencies have cited these
hypotheses, clear support for any of them remains elusive. Consumers
have long had ready access to detailed information about individual
models' fuel economy, which appears prominently on the labels displayed
by new cars,\1617\ and is published online and in printed outlets that
shoppers use routinely rely widely on to compare models.\1618\ In
addition, the fuel economy actually experienced by previous buyers of
individual models is increasingly reported in readily accessible on-
line databases.\1619\
---------------------------------------------------------------------------
\1617\ Fuel economy labels have been displayed on the window
sticker of all new light duty cars and trucks since the mid-1970s,
as required by the Energy Policy and Conservation Act. See https://www.epa.gov/fueleconomy/history-fuel-economy-labeling. Among the
information currently required to be posted on the fuel economy
label is both an estimated annual fuel cost for the vehicle, as well
as an estimate of how that cost compares to the fuel cost over five
years for an average new vehicle, so it is unclear what information
consumers lack that prevents them from making an informed decision
in this regard.
\1618\ See, e.g., http://www.fueleconomy.gov, where consumers
can find and compare the fuel economy (and greenhouse gas
CO2 and smog emissions) of different vehicle models
across model years, as well as upload information about their own
real-world fuel economy and compare it to other drivers.
\1619\ See id.
---------------------------------------------------------------------------
Similarly, consumers appear to be well aware of the prices they pay
for gasoline and how those vary among retail outlets, and are reminded
clearly and frequently of the financial consequences of their fuel
economy choices each time they purchase fuel. Increasingly, consumers
also have ready online access to comparisons of fuel prices at
competing locations near their homes or along routes they travel.\1620\
There is also considerable evidence that drivers' forecasts of future
fuel prices are more accurate than those issued by government agencies
or private forecasting services.\1621\ Evidence exists
[[Page 24611]]
that car buyers and owners anticipate extreme volatility in fuel
prices, recognize that there is considerable uncertainty about future
fuel prices and potential savings from driving a higher-mpg model, and
respond cautiously to these uncertainties when evaluating competing
vehicle models,\1622\ none of which suggests a market failure as much
as it suggests that consumers balance multiple, often competing
objectives, and make choices based on the outcome of such balancing.
---------------------------------------------------------------------------
\1620\ See, e.g., Gas Buddy, available at www.gasbuddy.com.
\1621\ Anderson et al. report evidence that consumers believe
fuel prices are likely to remain constant in inflation-adjusted
terms.; see Anderson, Soren T., Ryan Kellogg, and James M. Sallee,
``What do consumers believe about future gasoline prices?'' Journal
of Environmental Economics and Management, vol. 66 no. 3 (2013), at
pp. 383-403. (2013). Other evidence generally supporting this view
is reported by Allcott, Hunt, ``Consumers' Perceptions and
Misperceptions of Energy Costs,'' American Economic Review: Papers &
Proceedings, Vol. 101 No. 3 (2011), at pp. 98-104, (2011), although
Allcott finds that some fraction of consumers consistently believes
that gasoline prices will rise in the future. In related research,
Anderson et al. demonstrate that consumers' expectations that
gasoline prices will return to their current levels, even after
sudden and significant variation, is generally accurate; see
Anderson, Soren T., Ryan Kellogg, James M. Sallee, and Richard T.
Curtin, ``Forecasting Gasoline Prices Using Consumer Surveys.''
American Economic Review: Papers & Proceedings, Vol. 101 No. 3
(2011), at pp. 110-14. (2011). In contrast to many consumers'
expectation that fuel prices may vary over the future but will
generally return to current levels, the U.S. Energy Information
Administration predicted that gasoline prices would rise
significantly over the future at the time the two previous rules
establishing CAF[Eacute]E standards for model years 2012-16 and
2017-21 were adopted, in 2010 and 2012; see Energy Information
Administration (EIA), Annual Energy Outlook 2010), Table A12, p.
131, available at https://www.eia.gov/outlooks/archive/aeo10/pdf/0383(2010).pdf, Table A12, p. 131; and Annual Energy Outlook 2012,
Appendix A, Table A12, at p. 155, available at https://www.eia.gov/outlooks/archive/aeo12/pdf/appa.pdf, Table A12, p. 155. As of those
same dates, forecasts of future petroleum prices issued by other
government agencies and most private forecasting services (with the
notable exception of HIS-Global Insight, which projected little or
no increase in future prices) agreed closely with EIA's forecasts
that prices would increase significantly over both the near- and
longer-term futures; see EIA, Annual Energy Outlook 2010, Table 10,
at p. 86; and Annual Energy Outlook 2012, Table 23, available at
https://www.eia.gov/outlooks/archive/aeo12/table_23.php. Expressed
in constant-dollar terms, U.S. gasoline prices in 2019 are
essentially unchanged from those in 2010, although prices have
varied significantly above and below that level during the
intervening period. See https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=emm_epm0_pte_nus_dpg&f=m.
\1622\ For such evidence, see Allcott, Hunt, ``Consumers'
Perceptions and Misperceptions of Energy Costs,'' American Economic
Review: Papers & Proceedings, Vol. 101 No. 3 (2011), at pp. 98-104;
(2011); Greene, David L., (2010). ``How consumers value fuel
economy: A literature review'' No. EPA-420-R-10-008 (2010) (No. EPA-
420-R-10-008); Brownstone, David, David Bunch, and Kenneth Train,
``Joint Mixed Logit Models of Stated and Revealed Preferences for
Alternative-Fuel Vehicles,'' Transportation Research Part B, Vol. 34
(2000), at pp. 315-338, (2000), among many other sources.
---------------------------------------------------------------------------
In past rulemakings, the agencies have also hypothesized that
consumers may ``satisfice''--that is, select some minimum acceptable
level of fuel economy, and then evaluate models that achieve that
minimum on the basis of their other attributes. This explanation for
buyers' reluctance to purchase more fuel-efficient vehicles ignores the
possibility that they do account fully for the value of higher fuel
economy in their decision-making, but simply value differences in
vehicles' other attributes more highly than they do fuel economy, which
would not reveal irrational or myopic behavior.
A related argument has been that calculating future savings
attributable to fuel economy is complicated, so car shoppers resort to
simplified decision rules to choose among models with different fuel
economies, and relying on these rules-of-thumb causes them to choose
models with lower fuel economy.\1623\ However, it is unclear why
buyers' reliance on simplified procedures or approximations for
estimating the value of fuel savings would necessarily lead them to
systematically choose models with lower fuel economies rather than
leading some to underinvest in fuel economy while others overinvest.
---------------------------------------------------------------------------
\1623\ See, e.g., 77 FR at 63115 (Oct. 15, 2012).
---------------------------------------------------------------------------
The agencies have also frequently described consumers as ``loss
averse,'' making them reluctant to pay the upfront and certain higher
prices for models offering better fuel economy when the future savings
they expect to realize are more distant and less certain.\1624\ The
agencies' past assumption that loss aversion is universal (and equally
strong) among new-car shoppers appears to be a simplification that is
largely unsupported by empirical evidence, and in any case has been
challenged both as a widespread feature of consumer behavior and more
specifically as an explanation for vehicle shoppers' reluctance to
purchase more costly models that offer higher fuel economy.\1625\
Further, the extremely wide variety of competing models among which car
buyers can choose enables many of those searching for a model with
better fuel economy at a comparable price to do so simply by choosing a
version with fewer other features, which might partly offset the effect
of their aversion to the prospect of losses from paying a higher
purchase price. Lastly, the agencies note that both increased fuel
costs and increased upfront car prices will appear as ``losses,'' so it
is not obvious why potential buyers would react to the prospects of
these different forms of losses in different ways.
---------------------------------------------------------------------------
\1624\ Id. at 63114-15; see also 74 FR at 25511, 25653 (May 7,
2010).
\1625\ See supra notes 1611 and 1612.
---------------------------------------------------------------------------
OMB Circular A-4 does acknowledge that ``[e]ven when adequate
information is available, people can make mistakes by processing it
poorly.'' It goes on to say that people may rely on ``mental rules-of-
thumb'' that produce errors, or cognitive ``availability'' may lead to
consumers overstating the likelihood of an event. However, Circular A-4
also cautions that ``the mere possibility of poor information
processing is not enough to justify regulation,'' and that potential
problems with information processing ``should be carefully
documented.'' Some of the above examples of potential market failures
may fall into this category, but lack evidentiary support. As with
claims of asymmetric information, it is very difficult to distinguish
between information processing errors and behavior consistent with
consumer preferences for time and other vehicle attributes that differ
from what government agency analysts believe they should be.
Similarly, the agencies have occasionally noted (and seemingly been
critical of) some consumers' apparent preferences for vehicle
attributes that convey social status, such as size or styling, and
suggested that they may give inadequate attention to fuel economy
because it does not provide similar status. The agencies have also
suggested that consumers may be reluctant to purchase more fuel-
efficient models because they associate higher fuel economy with
inexpensive, less well-designed vehicles. These might be plausible
explanations, were they not contradicted by concurrent arguments that
potential buyers are inattentive to or uninformed about fuel economy,
or have difficulty isolating it from vehicles' other attributes.
Moreover, the market currently offers a wide range of highly fuel
efficient (and advanced technology) vehicles at many different price
points, including in the luxury and performance segments, which belies
the assumption that fuel economy is inconsistent with positional
attributes. In any case, consumers' hesitance to choose models offering
higher fuel economy because they are reluctant to sacrifice
improvements in other vehicle attributes on which they place higher
values cannot reasonably be characterized as a market failure.
Although past rulemakings have raised the possibility that car
buyers' apparent tendency to underinvest in fuel economy could
plausibly be explained by their use of discount rates exceeding those
the agencies employ to assess the present value of fuel savings, the
agencies have generally dismissed that possibility. In combination with
factors such as their valuation of vehicles' attributes other than fuel
economy, differences in driving habits that affect fuel economy and in
how much they expect to drive newly- purchased cars, and variation in
their expectations about future fuel prices, differing attitudes about
the importance of future costs relative to more immediate ones could
readily explain buyers' apparent reluctance to purchase models offering
fuel economy levels that the agencies interpret as privately
``optimal.''
As with consumption of any good or service, the agencies believe
consumers'
[[Page 24612]]
choice in vehicles represents what economists call ``constrained
optimization.'' That is, consumers select a bundle of vehicle
features--within their budget constraint--that optimizes the value to
them. The agencies also believe, as is the case in every constrained
consumer choice, that each of these attributes provide what economists
call diminishing marginal returns (or value) to consumers. For
instance, the agencies believe that consumers value vehicle size,
comfort, performance, trim-level, appearance, etc. As such, fuel-saving
technologies that increase the cost of the car are just one of many
vehicle attributes that consumers balance against each other. And
instead of using their entire budget on a single vehicle attribute,
consumers tend to sacrifice some degree of many or all attributes in a
degree that varies according to their preferences so that they can
consume some degree of most or all attributes they value. This means
that many consumers may not maximize fuel-saving technologies in their
vehicle selection, but instead may choose some other bundle of
attributes. The agencies' use of a 30 month pay-back period in this
analysis--as opposed to fuel-savings over the life of the vehicle--is
consistent with the constrained optimization consumers perform when
selecting a vehicle. It is a reasonable representation of consumers'
valuation of fuel-saving technologies, given the diminishing marginal
returns of additional fuel economy. If the agencies had used the entire
undiscounted fuel-savings over the entire life of the vehicle, the
agencies would be effectively modeling a scenario where consumers
maximize fuel economy to the detriment of all other vehicle
attributes--an assumption that is evidently wrong. As such, it is not
necessary that purchasers do not value lifetime fuel savings--and, in
all likelihood, purchasers would prefer vehicles with better fuel
efficiency and all of their preferred attributes--but rather consumers
are forced to choose between fuel economy and other vehicle attributes
while weighing how much each attribute contributes to the total cost of
the vehicle.
Finally, the agencies have also previously speculated that vehicle
producers may be reluctant to offer models featuring the higher levels
of fuel economy that buyers are willing to pay for, and that buyers'
apparent underinvestment in fuel economy reflects this lack of choice.
The agencies have speculated that such behavior by manufacturers could
arise from their collective underestimation of the value that buyers
attach to fuel economy, or failing this, from limitations on
competition among them to supply improved fuel economy, whether
voluntarily or as a consequence of the industry's structure.\1626\ The
agencies have also raised the seemingly contradictory argument that
producers have more complete knowledge about fuel economy than
potential buyers (``asymmetric information'') causing them to provide
lower levels than buyers demand, and speculated that deliberate
decisions by manufacturers may limit the range of fuel economy they
offer in particular market segments.\1627\
---------------------------------------------------------------------------
\1626\ See 75 FR at 25653-64 (May 7, 2010); and 77 FR at 63115
(Oct. 15, 2012).
\1627\ See, e.g. 75 FR 25510-13; 76 FR 57315-19; 77 FR 62914.
---------------------------------------------------------------------------
The overarching theme of these arguments seems to be that vehicle
manufacturers cannot identify--or can, but voluntarily forego--
opportunities to increase sales and profits at the expense of their
rivals by offering models that feature higher fuel economy. The
agencies have sometimes ascribed this behavior to the risk that
producers might incur large investments to produce the more fuel-
efficient models that would enable them to seize these opportunities,
but subsequently lose sales and profits to competitors who simply
followed suit after their rivals were successful. This explanation is
at odds with the customary view that innovative producers can be
rewarded--substantially, even if only temporarily--with commensurate
profits that justify taking such risks, when they correctly assess
consumer demand for innovative features or products.
In any case, behavior on the part of individual businesses that
leaves obvious opportunities to increase profits unexploited by an
entire industry seems extremely implausible, particularly in light of
the fact that auto manufacturers are profit-seeking businesses whose
ownership shares are publicly traded and subject to regular market
valuation. This notion also seems to ignore the range of choices
already available in the current automobile market, where
extraordinarily efficient models are available in nearly every vehicle
class or market segment, including plug-in hybrid and fully electric
versions of a rapidly increasing number of models. Automobile
manufacturers can, and in fact are, competing on the basis of fuel
economy.
The central analysis presented in this final regulatory impact
analysis does not account for the possibility that imposing stricter
standards may require manufacturers to make sacrifices in other vehicle
features that compete with fuel economy, and that some buyers may value
more highly. If this proved to be the case, more stringent alternatives
could impose offsetting losses on buyers well beyond the increases in
vehicle prices that are necessary for manufacturers to recover their
outlays for adding new technology (or changing design features) to
improve fuel economy. By doing so, it could significantly reduce the
estimates of total and net benefits the agencies report. To further
illustrate this issue, the agencies have conducted a sensitivity
analysis that incorporates a conservative estimate of consumers'
valuation of other vehicle attributes, as further discussed in Chapter
VII of the FRIA accompanying today's notice.\1628\ The agencies also
recognize that buyers may have time preferences that cause them to
discount the future at higher rates than the agencies are directed to
consider in their regulatory evaluations.
---------------------------------------------------------------------------
\1628\ This sensitivity analysis assumes that consumer's value
of other vehicle attributes is at least as great as a portion of the
fuel savings that consumers supposedly ``leave on the table.'' In
this analysis, the private net benefits of the final rule are a
positive $15 billion using a 7% discount rate--which is consistent
with the theory that providing consumers with greater choices will
enhance their private welfare. The net external benefits are
identical to the primary analysis, or $34 billion, so the
sensitivity results show the final rule improves net social benefits
by $49 billion.
---------------------------------------------------------------------------
If either case is true--that the analysis is incomplete regarding
consumer valuation of other vehicle attributes or discount rates used
in regulatory analysis inaccurately represent consumers' time
preferences--no market failure would exist to support the hypothesis of
a fuel efficiency gap. In either case, the agencies' central analysis
would overstate both the net private and social benefits from adopting
more stringent fuel economy and CO2 emissions standards. For
instance, Table VII-93 (Combined LDV Societal Net Benefits for MYs
1975-2029, CAFE Program, 7% Discount Rate) shows that the CAFE final
rule would generate $16.1 billion in total social net benefits using a
7% discount rate, but without the large net private loss of $26.1
billion, the net social benefits would equal the external net benefits,
or $42.2 billion. Because government action cannot improve net social
benefits in the absence of a market failure, if no market failure
exists to motivate the $26.1 billion in private losses to consumers,
the net benefits of these final standards would be $42.2 billion.
In sum, the agencies do not take a position in this rule on whether
a fuel
[[Page 24613]]
efficiency gap exists or constitutes a failure of private markets.
Accordingly, the final regulatory impact analysis is not constrained in
any manner that ensures the private net benefits of more stringent
standards will necessarily be either positive or negative. In fact,
however, the analysis supporting this final rule does present a
situation where adopting more stringent CAFE and CO2
emission standards aligns consumers' decisions with a simplified
representation of their own economic interests, and by doing so
improves their well-being from what they would experience under less
stringent standards. In other words, our final modelling results
reflect the case where some fuel efficiency gap persists (albeit of
smaller magnitude than the agencies found in previous analyses),
despite our expressed reservations about its likelihood.
(b) Representing Sales Responses in CAFE/CO2 Analysis
The approach used in the NPRM relied on a single model to produce
the total number of new vehicle sales in each calendar year for a given
regulatory scenario. Many commenters expressed reservations about the
predictive capabilities of the model (CARB, North Carolina Department
of Environmental Quality, EDF, Aluminum Association). As the Aluminum
Association commented, ``[D]eveloping a model to predict consumer
reaction to changes in prices is complicated and highly sensitive to
macroeconomic conditions, consumer confidence and employment levels.''
\1629\ As discussed above, the agencies agree that development of such
a model is complicated, and the agencies have elected to simplify the
approach for the final rule. For the purposes of regulatory evaluation,
the relevant sales metric is the difference between alternatives rather
than the absolute number of sales in any of the alternatives. As such
and in response to these comments and others previously addressed, the
agencies divided the sales response model for the final rule into two
parts: A nominal forecast that provides the level of sales in the
baseline (based primarily upon macroeconomic inputs), and a price
elasticity that creates sales differences relative to that baseline in
each year. The nominal forecast does not include price, and is merely a
(continuous) function of several macroeconomic variables that are
provided to the model as inputs. While the statistical model used in
the NPRM attempted to account for the influence of these other factors
in estimating the price elasticity, the forecast in this analysis
separates the two completely (as described further below). The price
elasticity is also specified as an input, but this analysis assumes a
unit elastic response of 1.0--meaning that a one percent increase in
the average price of a new vehicle produces a one percent decrease in
total sales.\1630\
---------------------------------------------------------------------------
\1629\ NHTSA-2018-0067-11952-4.
\1630\ The ``price increase'' in this case represents the new
vehicle price net of a portion of fuel savings, described further in
this section.
---------------------------------------------------------------------------
The revised sales model features three broad changes: (1) It uses
the change in average vehicle price net of fuel costs instead of
vehicle prices on their own, (2) it uses macroeconomic factors to
project baseline sales without considering vehicle prices, and (3) it
assesses the change in sales across the various regulatory alternatives
considered using an own-price elasticity from the literature. These
changes were made in response to comments that consumers are willing to
pay for some level of fuel economy and vehicle prices and sales are
simultaneously and jointly determined (e.g. endogenous). This section
discusses these three broad changes, as well as other more technical
and minor changes.
The first component of the new sales response model is the nominal
forecast, which is a function (with a small set of inputs) that
determines the size of the new vehicle market in each calendar year in
the analysis for the baseline. It leverages some of the same structure
of the statistical model used in the NPRM, though the dependent
variable and some of the explanatory variables have changed. It is of
some relevance that this statistical model is intended only as a means
to project a baseline sales series. Some commenters raised econometric
objections about the NPRM specification's ability to isolate the causal
effect of new vehicle prices on new vehicle sales. The agencies note
that the nominal forecast model does not include prices and is not
intended for statistical inference.
The forecast is derived from a statistical model that accounts for
a similar set of exogenous factors related to new light-duty vehicle
sales. In particular, the model accounts for the number of households
in the U.S., recent number of new vehicles sold, GDP, and consumer
confidence. The structure of the forecast model is similar to the NPRM
model, which also used a ARDL specification, but even the variables
that are common between the two models have different structural forms
in the final rule version. In particular, the dependent variable has
been transformed to reflect the fact that, as some commenters
suggested, households are an important component of demand for new
vehicles. As such, the dependent variable is defined as new vehicles
sold per household.\1631\ While this variable still exhibits the cyclic
behavior that new vehicle sales exhibit over time, the trend shows the
number of new vehicles sold per household declining since the 1970's,
as shown in Figure VI-64, where the dotted line is the trend over time.
As this time series is non-stationary,\1632\ a lagged variable (the
value in the previous year) is included on the right-hand side of the
regression equation. In addition, the model includes a lagged variable
that represents the three-year running sum of new vehicle sales,
divided by the number of households in the previous year. This variable
represents the saturation effect, where the existing number of
households can only buy so many new vehicles before a significant
number of households already have one (and do not need to buy another).
As vehicle durability and cost has increased over time, and average
length of initial ownership has increased similarly, this variable acts
to put downward pressure on sales after successive years of high sales
(particularly during extrapolation).
---------------------------------------------------------------------------
\1631\ Number of U.S. households is taken from Federal Reserve
Economic data, https://fred.stlouisfed.org/series/TTLHH.
\1632\ Stationary refers to whether a time series statistical
properties are constant over time. Since car sales are increasing
over time, the time series non-stationary.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24614]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.310
BILLING CODE 4910-59-C
Similar to the NPRM model, the forecast model includes real U.S.
GDP,\1633\ but in natural logarithm form (as some commenters suggested
was more appropriate).\1634\ The final variable is consumer sentiment,
as measured by the University of Michigan survey of consumers.\1635\ As
both of these series are non-stationary (determined by applying
augmented Dickey-Fuller unit root tests to the time series), lagged
versions of the variables are included to ensure stationarity in the
residuals. The functional form appears below in Equation 2.
---------------------------------------------------------------------------
\1633\ Federal Reserve Economic Data, available at https://fred.stlouisfed.org/series/GDPC1#0.
\1634\ EPA-HQ-OAR-2018-0283-6220-1.
\1635\ http://www.sca.isr.umich.edu/tables.html.
---------------------------------------------------------------------------
Equation 2--Statistical Model Used to Generate Nominal Forecast
The model fit is described in Table VI-152. The included lag term
of the dependent variable and both GDP variables are statistically
significant at nearly zero, while both the lagged three year sum term
and consumer sentiment are both marginally significant. Being a time
series model, the agencies also computed the Durbin-Watson test
statistic for autocorrelation (1.77) and the Breusch-Godfrey test for
serial correlation (0.65) at order 1. The signs of the coefficients are
all correct, in the sense that they are consistent with our
expectations.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.311
[[Page 24615]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.312
BILLING CODE 4910-59-C
Because the dependent variable is the number of new vehicles sold
per household, it is necessary to multiply by the number of households
to produce an estimate of new vehicle sales. This model is used to
produce a forecast of new vehicle sales out to 2050, so it is necessary
to have projections of each variable used in Equation 2 through
calendar year 2050. In an effort to be consistent with other inputs to
the analysis, the projection of U.S. GDP is taken from the 2019 AEO.
The forecast of households in this analysis comes from the Harvard
Joint Center for Housing Studies 2018 Household projections.\1636\ The
consumer confidence forecast is taken directly from the University of
Michigan index for 2017 and 2018, and from the Global Insight forecast
of consumer confidence for all subsequent years.
---------------------------------------------------------------------------
\1636\ https://www.jchs.harvard.edu/research-areas/working-papers/updated-household-growth-projections-2018-2028-and-2028-2038.
---------------------------------------------------------------------------
While the analysis could have relied on a forecast of new vehicle
sales taken from a published source (the 2019 AEO, for example), using
a function is an attractive option because it allows the CAFE Model
dynamically to adjust the forecast in response to input changes. If a
sensitivity case requires a forecast that is consistent with a set of
specific, possibly unlikely, assumptions, a forecast of new vehicle
sales that is consistent with those assumptions may not exist in the
public domain, for example low GDP growth sensitivity cases. As
implemented in this rulemaking, using a functional form allows the user
to vary some of the assumptions to the analysis without creating
inconsistencies with other elements of the analysis. However, it is
incumbent upon the analyst to ensure that any set of assumptions that
deviate from the central analysis are logically consistent.
This function, and the set of assumptions contained in the central
analysis, produces a projection that is comparable in magnitude to the
forecast in the 2019 AEO reference case, though there are differences.
The two forecasts, and the percentage difference relative to the AEO
2019, appear in Table VI-153, as does a recent forecast published by
the Center for Automotive Research.\1637\ The reader will notice that
even 2017 shows a discrepancy of nearly 7 percent between the final
rule forecast and the Annual Energy Outlook, one of the larger
differences between annual forecasts. However, the final rule analysis
is based upon the certified production volumes of MY2017, which exceed
17 million units. So, while the difference may seem significant, the
final rule volumes in 2017 represent the ground truth for model year
production.\1638\ The CAR forecast, while shorter in length, is
consistently higher than both the AEO and final rule forecasts--though
likely also includes class 2b (and possibly class 3) pickup trucks in
its light vehicle forecast. Finding a public forecast that explicitly
excludes light-duty vehicles exempt from these regulations is
challenging. However, all three forecasts exhibit similar trends--
decreases in sales starting in 2019 that last for a few years before
ticking up again slowly. As commenters observed, all forecasts are
almost guaranteed to have some errors, and projections out to 2050
should be taken as potential future projections limited by our
knowledge at the time, rather than an ironclad prediction of the
future.
---------------------------------------------------------------------------
\1637\ https://www.cargroup.org/u-s-light-vehicle-sales-expected-to-take-a-dip-in-2019/, last accessed 11.21.2019.
\1638\ See CAFE Public Information Center, https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24616]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.313
BILLING CODE 4910-59-C
Although the forecast produces the total number of new vehicle
sales in the baseline, an elasticity is imposed on price differences to
produce sales changes between alternatives. The NPRM version of the
model considered only differences in average new vehicle prices between
alternatives, and the agencies received a number of comments (from CBD,
IPI, EDF, CARB, CA et al., and Oakland et al., as well as recent peer
reviewers) encouraging the agencies to account for some component of
fuel savings associated with those price changes. In their comment,
California et al. and Oakland et al. stated the model failed ``to
consider
[[Page 24617]]
how consumers will respond to the reduced cost of operating the vehicle
from better gas mileage and therefore inaccurately predicts a decline
in vehicle sales under the existing standards.'' \1639\ The agencies
agree that price is not the only consideration, and that the value of
fuel savings to new vehicle buyers is also relevant to the purchase
decision.
---------------------------------------------------------------------------
\1639\ States and Cities, Attachment 1, NHTSA-2018-0067-11735,
at 86.
---------------------------------------------------------------------------
In previous rules, while the agencies produced analyses that
qualitatively considered sales and employment impacts, the agencies
acknowledged that fuel economy and CO2 standards were likely
to increase vehicle prices, while simultaneously reducing operating
costs, and that estimating how consumers would choose to balance those
two factors in the new vehicle market was challenging.\1640\
Furthermore, the agencies recognized that there is a broad consensus in
the economic literature that the price elasticity of demand for
automobiles is approximately -1.0.\1641\ The agencies feel that a unit
elasticity of -1.0 is still a reasonable estimate.\1642\
---------------------------------------------------------------------------
\1640\ Final Regulatory Impact Analysis, Corporate Average Fuel
Economy for MY 2017-MY 2025 Passenger Cars and Light Trucks, August
2012, at 821.
\1641\ See, e.g., Kleit, A.N., ``The Effect of Annual Changes in
Automobile Fuel Economy Standards,'' Journal of Regulatory
Economics, Vol. 2 (1990), at pp 151-72; Bordley, R., ``An
Overlapping Choice Set Model of Automotive Price Elasticities,''
Transportation Research B, Vol. 28B no. 6 (1994), at pp 401-408; and
McCarthy, P.S. ``Market Price and Income Elasticities of New Vehicle
Demands,'' The Review of Economics and Statistics, Vol. LXXVII no. 3
(1996), at pp. 543-547.
\1642\ For example, a recent review of 12 studies examining
vehicle price elasticities conducted by the Center of Automotive
Research (``CAR'') found an ``average short-run elasticity of -
1.09'' and focusing ``only those models which also employ time
series methods, the average short-run own-price elasticity is higher
yet, at -1.25.'' CAR's own analysis found a -.79 short-run
elasticity. Appendix II of the CAR report shows that the long-run
elasticities ranged from -.46 and -1.2 with an average of -.72. In
sum, a -1.0 elasticity is well-aligned with the totality of
research. McAlinden Ph.D., Sean P., Chen, Yen, Schultz, Michael,
Andrea, David J., The Potential Effects of the 2017-2025 EPA/NHTSA
GHG/Fuel Economy Mandates of the US Economy, Center for Automotive
Research, Ann Arbor, MI (Sept. 2016), available at https://www.cargroup.org/wp-content/uploads/2017/02/The-Potential-Effects-of-the-2017_2025-EPANHTSA-GHGFuel-Economy-Mandates-on-the-US-Economy.pdf.
---------------------------------------------------------------------------
Because the elasticity assumes no perceived change in the quality
of the product, and the vehicles produced under different regulatory
scenarios have inherently different operating costs, the price metric
must account for this difference. As commenters suggested is
appropriate, the price to which the unit elasticity is applied in this
analysis represents the residual price change between scenarios after
accounting for 2.5 years' worth of fuel savings to the new vehicle
buyer. This approach is consistent with the 2012 FRIA analysis of sales
impacts, that which considered several payback periods over which the
value of fuel savings was subtracted from the change in average new
vehicle price.
Similar to the NPRM, the price elasticity is applied to the
percentage change in average price (in each year). However, the average
price to which the elasticity is applied is calculated differently in
the final rule in response to comments. As discussed below the price
change does not represent an increase/decrease over the last observed
year, but rather the percentage change relative to the baseline. In the
baseline, the average price is defined as the observed new vehicle
price in 2017 plus the average regulatory cost associated with the
alternative. In the case of CO2 standards, the regulatory
cost is equivalent to the retail equivalent price of technology
improvements. In the case of CAFE standards, the regulatory cost
includes both technology costs and civil penalties paid for non-
compliance in a model year. So the change in sales for alternative a in
year y is:
[GRAPHIC] [TIFF OMITTED] TR30AP20.314
[Delta]RegCost is the difference in average regulatory cost between
alternative a and the baseline scenario in year y to make a vehicle
compliant with the standards, $34,449 is the average transaction price
of a new vehicle in 2016, NominalSales is the forecasted sales (in the
baseline) in year y, [Delta]FuelCosts is the change in average fuel
costs over 2.5 years relative to the baseline in year y and
PriceElasticity is -1.0:
[GRAPHIC] [TIFF OMITTED] TR30AP20.315
Where 35,000 miles is assumed to be equivalent to 2.5 years of
vehicle usage.\1643\ The agencies assume that consumers behave as if
the fuel price faced at the time of purchase is the fuel price that
they will face over the first 2.5 years of ownership and usage.
Essentially, they behave as if fuel prices follow a random walk,
where the best prediction of (near) future prices is the price
today. Scrappage rates in the first few years of ownership are close
to zero, so buyers can reasonably expect to travel the full annual
mileage in each of the first three years of ownership. Total sales
in each alternative (that is not the baseline) will equal
NominalSalesy + [Delta]Salesa,y for
alternative a in year y.
---------------------------------------------------------------------------
\1643\ Based on odometer data, 35,000 miles is a good
representation of typical new vehicle usage in the first 2.5 years
of ownership and use--though the distribution of usage is large.
This implementation produces a range of differences in total sales,
both between alternatives and over time. Table VI-154 shows the range
of differences in the final rule at the industry level for
CO2, and Table VI-155 shows the sales changes under CAFE.
While cost decreases between the baseline and alternatives differ by
program, one can see that removing the value of fuel savings from the
price limits the sales increases in the alternatives to under 300,000
units in a single year under the preferred alternative, and about one
percent of total sales between 2017 and 2050.
BILLING CODE 4910-59-P
[[Page 24618]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.316
[GRAPHIC] [TIFF OMITTED] TR30AP20.317
BILLING CODE 4910-59-C
Table VI-154 and Table VI-155 show sales under the baseline
(augural standards), and differences under the proposal (0 percent
increase in stringency) and final rule (1.5 percent increase in
stringency) of MYs 2017-2050.
c) Dynamic Fleet Share (DFS)
The first module described above (the forecast function and applied
elasticity)
[[Page 24619]]
determine the total industry sales in each model year from 2018 (in
this analysis, 2017 is based on certified compliance data) to 2050. A
second module, the dynamic fleet share, acts to distribute the total
industry sales across two different body-types: ``cars'' and ``light
trucks.'' While there are specific definitions of ``passenger cars''
and ``light trucks'' that determine a vehicle's regulatory class, the
distinction used in this phase of the analysis is more simplistic. All
body-styles that are obviously cars--sedans, coupes, convertibles,
hatchbacks, and station wagons--are defined as ``cars'' for the purpose
of determining fleet share. Everything else--SUVs, smaller SUVs
(crossovers), vans, and pickup trucks--are defined as ``light
trucks''--even though they may not be treated as such for compliance
purposes. In the case of SUVs, in particular, many models may have
sales volumes that reside in both the passenger car and light fleets
for regulatory purposes, but the dynamic fleet share does not make this
distinction. The fleet share model was applied at the same level in the
NPRM--namely, at the level of body-style rather than regulatory class.
EDF expressed concern that any simulated increase in the light truck
share represented consumers shifting from sedans to either 4WD drive
crossovers, SUVs or pickup trucks.\1644\ However, this was not the
case. All crossovers are considered light trucks for the purposes of
fleet share, even though they may be 2WD crossovers treated as
passenger cars for compliance purposes. So, while the number may
increase overall for a given scenario, the proportion of crossovers
sold as 4WD, rather than 2WD, does not.
---------------------------------------------------------------------------
\1644\ EDF, Appendix B, NHTSA-2018-0067-12108, at 40-41.
---------------------------------------------------------------------------
EDF was also concerned that the sales implementation in the NPRM,
which relied on the absolute average price to determine differences
between alternatives, was unduly influenced by fleet share--as
differences in the share of light-trucks had the potential to skew
differences in average price because light-trucks are generally more
expensive than sedans and hatchbacks. The final rule implementation,
which starts from an observed average transaction price and evolves the
average price in the alternatives based on average regulatory cost, is
less vulnerable to this potential distortion. Even if the fleet share
model (described in greater detail below) increases the share of light
trucks (for example), the inherent price difference between passenger
cars and light trucks does not pass through to the average price--only
the relative difference in compliance costs associated with the vehicle
types. Despite the fact that light trucks have generally higher
transaction prices than passenger cars, there is no guarantee that
regulatory costs will be higher for light-trucks than for cars (which
depend upon the mix of footprints, their distance from the relevant
curve, and the technology cost needed to bring each fleet into
compliance). Thus, the average price differences used in the sales
calculations are relatively unaffected by the fleet share model.
As in the NPRM, the dynamic fleet share represents two difference
equations that independently estimate the share of passenger cars and
light trucks, respectively, given average new market attributes (fuel
economy, horsepower, and curb weight) for each group and current fuel
prices, as well as the prior year's market share and prior year's
attributes. The two independently estimated shares are then normalized
to ensure that they sum to one. As with the Sales Response model, the
DFS utilizes values from one and two years preceding the analysis year
when estimating the share of the fleet during the model year being
evaluated. For the horsepower, curb weight, and fuel economy values
occurring in the model years before the start of analysis, the DFS
model uses the observed values from prior model years. After the first
model year is evaluated, the DFS model relies on values calculated
during analysis by the CAFE model. The DFS model begins by calculating
the natural log of the new shares during each model year, independently
for each vehicle class, as specified by the following equation:
[[Page 24620]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.318
HPVC,MY 1: The average horsepower of all vehicle models belonging to
vehicle class VC, in the year immediately preceding model year MY,
---------------------------------------------------------------------------
\1645\ As discussed elsewhere in this final rule, model year and
calendar year are assumed to be equivalent in the simulation--as
they always have been in all prior rulemaking analyses.
---------------------------------------------------------------------------
HPVC,MY 2: The average horsepower of all vehicle models belonging to
vehicle class VC, in the year preceding model year MY by two years,
CWVC,MY 1: The average curb weight of all vehicle models belonging
to vehicle class VC, in the year immediately preceding model year
MY,
CWVC,MY 2: The average curb weight of all vehicle models belonging
to vehicle class VC, in the year preceding model year MY by two
years,
FEVC,MY 1: The average on-road fuel economy rating of all vehicle
models (excluding credits, adjustments, and petroleum equivalency
factors) belonging to vehicle class VC, in the year immediately
preceding model year MY,
FEVC,MY 2: The average on-road fuel economy rating of all vehicle
models (excluding credits, adjustments, and petroleum equivalency
factors) belonging to vehicle class VC, in the year preceding model
year MY by two years,
0.423453: a dummy coefficient, and
1n(ShareVC,MY): The natural log of the calculated share of the total
industry fleet classified as vehicle class VC, in model year MY.
In the equation above, the beta coefficients, [beta]C through
[beta]Dummy, are provided in the following table. The beta coefficients
differ depending on the vehicle class for which the fleet share is
being calculated.
[[Page 24621]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.319
Once the initial car and light truck fleet shares are calculated
(as a natural log), obtaining the final shares for a specific vehicle
class is simply a matter of taking the exponent of the initial value,
and normalizing the result at one (or 100%). This calculation is
demonstrated by the following:
[GRAPHIC] [TIFF OMITTED] TR30AP20.320
These shares are applied to the total industry sales derived in the
first stage of the sales response. This produces total industry volumes
of car and light truck body styles. Individual model sales are then
determined from there based on the following sequence: (1) individual
manufacturer shares of each body style (either car or light truck)
times the total industry sales of that body style, then (2) each
vehicle within a manufacturer's volume of that body-style is given the
same percentage of sales as appear in the 2017 fleet. This implicitly
assumes that consumer preferences for particular styles of vehicles are
determined in the aggregate (at the industry level), but that
manufacturers' sales shares of those body styles are consistent with
MY2017 sales. Within a given body style, a manufacturer's sales shares
of individual models are also assumed to be constant over time. The
agencies assume that manufacturers are currently pricing individual
vehicle models within market segments in a way that maximizes their
profit. Without more information about each OEM's true cost of
production and operation, fixed and variables costs, and both desired
and achievable profit margins on individual vehicle models, the
agencies have no reason to assume that strategic shifts within a
manufacturer's portfolio will occur in response to standards.
The Global Automakers noted in their comments that the market share
of SUVs continues to grow, while conventional passenger car body-styles
continue to lose market share.\1646\ The agencies are aware of this,
and include the DFS model in an attempt to address these market
realities. In the 2012 final rule, the agencies projected fleet shares
based on the continuation of the baseline standards (MY2012-2016) and a
fuel price forecast that was much higher than the realized prices since
that time. As a result, that analysis showed passenger car body-styles
comprising
[[Page 24622]]
about 70 percent of the new vehicle market by 2025. The reality, as
Global Automakers note, has been quite different.
---------------------------------------------------------------------------
\1646\ Global Automakers, Attachment A, NHTSA-2018-0067-12032,
at 13.
---------------------------------------------------------------------------
The coefficients of the DFS model show passenger car styles gaining
share with higher fuel prices and losing them when prices are lower.
Similarly, as fuel economy increases in light truck models, which offer
consumers other desirable attributes beyond fuel economy (ride height
or interior volume, for example) their relative share increases. NRDC,
in particular, found this counterintuitive.\1647\ However, this
approach does not suggest that consumers dislike fuel economy in
passenger cars, but merely recognizes the fact that fuel economy has
diminishing returns. As the fuel economy of light trucks increases, the
tradeoff between passenger car and light truck purchases increasingly
involves a consideration of other attributes. Similarly, the
coefficients show a relatively stronger preference for power
improvements in cars than light trucks because that is an attribute
where trucks have outperformed cars, like cars have outperformed trucks
for fuel economy.
---------------------------------------------------------------------------
\1647\ NRDC, Attachment 3, NHTSA-2018-0067-11723, at 5.
---------------------------------------------------------------------------
Rather than estimate new functions to determine relative market
shares of cars and light trucks, the agencies applied existing
functions from the transportation module of the National Energy
Modeling System (NEMS) that was used to produce the 2017 Annual Energy
Outlook. The functions above appear in the ``tran.f'' input file to
that version of NEMS, and were embedded (in their entirety) in the CAFE
model in the NPRM (and this final rule). NEMS uses the functions to
estimate the percent of total light vehicles less 8,500 GVW that are
cars/trucks. While NRDC asserted that the agencies must demonstrate the
propriety of the fleet share model before relying on its
estimates,\1648\ they ignore the fact that, by using the AEO to develop
a static fleet in prior rulemakings, the agencies have always relied on
NEMS estimates. The primary difference between those analyses and the
NPRM (and this final rule), is that prior analyses applied the fleet
share that was simulated for the baseline to all regulatory scenarios
considered. Based on the fleet share functions in NEMS, NPRM corrected
this internal inconsistency found in previous analyses. This approach
also enables consistent sensitivity cases--where higher fuel prices
produce fleets with more transitional passenger car body styles, for
example--and ensures that the starting point (MY 2017) evolves in
response to both fuel economy improvements and fuel prices in a way
that is internally consistent.
---------------------------------------------------------------------------
\1648\ Id.
---------------------------------------------------------------------------
The agencies are making one change to the DFS function, which is
the level of application. While NEMS intended the fleet shares to be
defined by regulatory classes, vehicles are defined much more coarsely
in NEMS than in the CAFE model, and manufacturers are not
differentiated at all. In order to produce well-behaved fleet share
projections with this model, the agencies applied the share functions
to body-styles rather than regulatory classes. For many years, there
was little overlap between nameplates in a manufacturer's passenger car
regulatory class and its light truck regulatory class. However, with
the recent emergence of smaller FWD SUVs and crossovers, it is
increasingly common to have nameplates with model variants in both the
passenger car and light truck regulatory classes, and it is also common
for there to be only minor differences (like the presence of 4WD or
AWD) between versions regulated as cars and versions regulated as light
trucks. The agencies have modified the application of the fleet share
equations to focus on body-style, rather than regulatory class, in
recognition of the increased ambiguity between the regulatory class
distinction for popular models like the Honda CR-V and Toyota RAV4,
that sell more than 100K units in each regulatory class (typically
using the same powertrain configuration). The Nissan Rogue sold more
than 400K units in MY2017, and almost exactly half of them were in the
light truck (LT) regulatory class. Applying the fleet share at the
body-style level preserves the existing regulatory class splits for
nameplates that straddle the class definitions. It also serves to
minimize the deviation from the observed MY2017 regulatory class shares
over time. Had the agencies applied the share equations at the
regulatory class level, as some commenters incorrectly claimed the
agencies were doing in the proposal, the passenger car regulatory class
would have eroded much faster than we've seen in the real world and
ceased to resemble the composition of the MY2017 fleet. Our
implementation allows the passenger car (PC) regulatory class to
continue evolving toward crossover-type cars, if that is what economic
and policy conditions favor.\1649\
---------------------------------------------------------------------------
\1649\ The ``passenger car'' fleet for CAFE represents the
combination of both imported passenger cars (IC) and domestic cars
(DC). While Table VI-157 illustrates shares for the CAFE program,
resulting shares under the tailpipe CO2 emissions
standards are comparable.
---------------------------------------------------------------------------
[[Page 24623]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.321
Table VI-157 shows the regulatory class shares under the baseline
(augural standards), proposal (0 percent increase in stringency), and
final rule (1.5 percent increase in stringency) between 2017 and 2030.
The shares move relatively little between the classes in the baseline,
with larger (but still small) deviations occurring in the least
stringent alternative (0 percent increase) and the final rule. As the
sensitivity cases show, the changes in shares (both over time and
between regulatory classes) respond to the fuel price case, but remain
internally consistent due to the inclusion of the DFS.
Some commenters encouraged the agencies to consider vehicle
attributes beyond price and fuel economy when estimating a sales
response to fuel economy/CO2 standards, and suggested that a
more detailed representation of the new vehicle market would allow the
agencies to simulate strategic mix shifting responses from
manufacturers and diverse attribute preferences among consumers. Doing
so would have required a discrete choice model (at some level), and
below the reasons why the agencies have not chosen to employ that
approach in this final rule.
d) Using Vehicle Choice Models in Rulemaking Analysis
Some commenters argued that the NPRM's statistical model used to
estimate changes in sales between alternatives was too highly
aggregated and missed consumers' valuation of other vehicle attributes.
CARB, Cities and States, and EDF all made some version of the argument
that the sales model in the NPRM operated at too high a level of
aggregation to estimate the real sales response, which primarily occurs
at the model level where consumers are making decisions based on the
comprehensive set of attributes and body styles available in the
market. They also argued that a model must operate at the same level,
such as a discrete choice model, in order to capture consumer response
accurately. EPA's Science Advisory Board, Bento, Toyota, Automobile
Alliance, RFF, and Bunch (writing on behalf of CARB) insisted that the
best approach to estimating the change in sales across alternatives is
to use a discrete choice model and embed it in the simulation.
Other commenters expressed different views on the importance of a
consumer choice model. For example, while the Aluminum Association
supported a consumer choice model, they suggested that total new
vehicle sales may not change due to increases in price, but rather the
attributes of new vehicles would shift, as consumers would likely shift
their purchases toward lower content vehicles (in terms of safety,
luxury, or other option content) when faced with generally higher
prices. Other commenters, including UCS and CBD, strongly encouraged
the agencies to avoid using consumer choice models; commenters asserted
that consumer choice models have historically lacked reliability and
predictive power.\1650\
---------------------------------------------------------------------------
\1650\ UCS, Technical Appendix, NHTSA-2018-0067-12039 at 50.
---------------------------------------------------------------------------
In general, these various comments present the agencies with
considerably different suggestions on how to address these issues, and
certain suggestions are in direct opposition to each other. That is,
while some commenters argue that only micro-level consumer responses
are relevant to the analysis, and that a consumer choice model is
required to estimate these responses, others argue that it is
inappropriate to use a discrete choice model--the method by which those
responses are econometrically estimated--in a regulatory analysis.
Adding to the confusion, some of the same commenters who argued against
a consumer choice model,\1651\ also argued that it was necessary for
the analysis to account for the influence of other vehicle attributes
in purchasing decisions, which would require incorporating a discrete
choice model.
---------------------------------------------------------------------------
\1651\ For example, see EDF, NRDC, RFF, NCAT, and CBD comments.
---------------------------------------------------------------------------
CARB argued that ``accurately capturing the relative impact of
sales shifts versus no-buy decisions would require a more detailed
consumer choice model, as recommended by the CAFE Model peer reviewers.
The current new vehicle sales model has no
[[Page 24624]]
way of capturing these types of effects.'' \1652\
---------------------------------------------------------------------------
\1652\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 192.
---------------------------------------------------------------------------
David Bunch, writing for CARB, said, ``In fact, in previous
versions of the CAFE model there were no attempts to directly simulate
consumer response from within the CAFE model at all. Instead, NHTSA
relied on fixed projections of future vehicle market behavior from
multiple sources for the purpose of performing the required economic
cost and benefit calculations. While this might possibly be less than
ideal, this approach is only a problem if, in the real world, there
[are] notable differences in future market behavior [that] occur under
different regulation scenarios, and, moreover, that these differences
would be large enough to compromise the validity of the net benefit
comparisons.'' Bunch essentially argues that the old approach,
asserting that standards can have no impact on sales, even at the
individual model level, is more appropriate than trying to capture the
general idea that when all new vehicles get more expensive, consumers
are likely to buy fewer of them, all else being equal. The agencies
disagree with that perspective.
There are a number of practical challenges to using estimates of
consumer attribute preferences to simulate market responses. Discrete
choice models typically rely on fixed effects (or alternative-specific
constant terms) to account for the unobserved characteristics of a
given model that influence purchasing decisions, such as styling,\1653\
but are not captured by independent variables that represent specific
vehicle attributes (horsepower, interior volume, or safety rating, for
example). Ideally, these constant terms would contribute relatively
little to the fit and performance of the model, assuming that the most
salient characteristics are accounted for explicitly. In practice, this
is seldom the case. While the fixed effects at the model level are
statistically sound estimates of consumer preferences for the
unobserved vehicle characteristics of the individual models, the
estimates are inherently historical--based on observed versions of the
specific vehicle models to which they belong. However, once the
simulation starts, and new technologies are added to each
manufacturer's product portfolio over successive generations, it is no
longer obvious that those constant terms would still be valid in the
context of those changes.
---------------------------------------------------------------------------
\1653\ Aesthetics such as styling are difficult, if it not
impossible, to define in a manner that allows meaningful comparison
between choices.
---------------------------------------------------------------------------
Another complication is that discrete choice models are highly
dependent on their inputs and are unable to account for future market
changes. For example, the Draft TAR relied on a MY 2014 market (for
EPA's analysis) and a MY 2015 market (for NHTSA's analysis), while the
NPRM used a MY 2016 fleet, and this final rule has updated the market
characterization to a MY 2017 fleet. A discrete choice model estimated
on any of those model years would probably produce different fixed
effects estimates for each model variant in the fleet. Even assuming
that no new variants of a given model are offered over time, new
nameplates emerge as others are retired--and for those new nameplates
and all of their model variants, no constant terms would exist. They
would have to be imputed (either from comparable vehicles in the
market, some combination of their attributes, or both). Some studies
have attempted to estimate fixed effects for a single new entrant to
the market,\1654\ but none have attempted to do so at the scale
required to migrate a discrete choice model fit on an earlier model
year to a newer model year for simulation.
---------------------------------------------------------------------------
\1654\ Berry, Steven, James Levinsohn, and Ariel Pakes (2004).
Differentiated products demand systems from a combination of micro
and macro data: The new car market. Journal of Political Economy
112(1): 68-105.
---------------------------------------------------------------------------
Figure VI-65 shows the cumulative percentage of nameplates in the
2017 new vehicle market by year of introduction. About ten percent of
nameplates in 2017 have been around since the 1970s, but another ten
percent have only existed since about 2010. This fact illustrates the
likely necessity of constructing vehicle model fixed effects for the
inevitable new entrants between the estimating fleet and the rulemaking
fleet. But it also suggests another challenge. New model entrants are
driven by the dynamics of the market, where some vehicle models succeed
and others fail, but a simulated market with a discrete choice model
can only simulate failure--where consumer demand for specific
nameplates erode to the point that the nameplate volumes trend toward
zero. It has no mechanism to generate new nameplates to replace those
nameplates whose sales it estimates will erode beyond some minimal
practical level of production.
Consumer choice models are typically fit on a single year of data
(a cross-section of vehicles and buyers), but this approach misses
relevant trends that build over time, such as rising GDP or shifting
consumer sentiment toward emerging technologies. If such a model is
used to estimate total sales, but lacks trends in GDP growth or
employment, etc., it will have the wrong set (likely a smaller set) of
new vehicle buyers and exaggerate price responses and attribute
preferences. Consumer preferences change over time in response to any
number of factors--given manufacturers' recent investments in electric
powertrains, they are counting on this fact. But a choice model
estimated on observed consumer preferences for EVs--or other vehicle
attributes with comparatively little experience in the market--would
necessarily disadvantage a technology that is currently (or only
recently) unpopular, but gaining popularity. While these are problems
that may not matter in the estimation process, where a researcher is
attempting to measure revealed consumer preference for given attributes
at a single point in time, they become material once that model is
integrated into the simulation and dynamically carried forward for
three decades. The agencies note that models that examine aggregate
trends, such as the one utilized in this analysis, are able to side-
step this issue by not placing a value on unique vehicle attributes.
[[Page 24625]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.322
The agencies' compliance simulation model estimates the additional
cost of technology required to achieve compliance, or to satisfy market
demand for additional fuel economy. While it necessarily calculates
these costs on a per-vehicle basis, estimating the cost of additional
technologies as they are applied to each specific model in order to
bring an entire fleet into compliance, it is agnostic about how these
costs are distributed to buyers. Manufacturers have strategic, complex
pricing models that rely on extensive market research and reflect each
company's strategic interests in each segment. Automobile companies
attempt to maximize profit from the sale of their vehicles, rather than
solely focusing on minimizing the cost of compliance, as this
rulemaking simulates. Lacking reliable data for each manufacturer on
production costs and profit margins for each vehicle model in their
portfolios, the most reasonable course of action is to simulate
compliance as if OEMs are attempting to minimize costs, and, worth
noting, this approach is also the one NHTSA takes in its rulemakings
related to the FMVSS. However, it is obvious that some market segments
and individual models are much less elastic than others.\1655\ As
reflected in the prices of those models, consumers are able to bear a
greater share of the total cost of compliance before negatively
affecting sales and manufacturer profits.
---------------------------------------------------------------------------
\1655\ See, for example, Kleit, A.N. (2004), Impacts of Long-
Range Increases in the Fuel Economy (CAFE) Standard. Economic
Inquiry, 42: 279-294. doi:10.1093/ei/cbh060.
---------------------------------------------------------------------------
Several commenters (CARB, CBD, IPI, and Bento et al.) suggested
that the agencies should employ a pricing model that allows
manufacturers to vary prices in response to heterogeneous consumer
preferences and different levels of willingness to pay for fuel
economy, and other attributes, in the new vehicle market.
Fundamentally, this would require the agencies to model strategic
pricing for each manufacturer individually--no single pricing model
would be appropriate for every manufacturer. Bento et al. stated that
the agencies should simulate the market by allowing manufacturers to
dynamically adjust vehicle prices to ensure compliance with the
standards.\1656\ There is no reasonable expectation that the agencies
could embed and utilize each manufacturer's pricing strategy, as this
is an essential feature of competitive corporate behavior and that
automakers closely hold pricing strategy information and the agencies
have insufficient information to model manufacturer pricing strategies.
Furthermore, models in the academic literature that commenters have
suggested are superior because they allow prices to adjust, merely
demonstrate that the mechanics of those adjustments work; they do not
imply that the resulting prices are reasonable or realistic. Given the
burden to estimate each manufacturer's standard under the attribute-
based system, where the mix of vehicles sold defines not only the
achieved fuel economy of each fleet but also the standard to which it
is compared, the agencies are understandably reluctant to implement
models that might shift a manufacturer's mix of vehicles sold within a
market segment.
---------------------------------------------------------------------------
\1656\ NHTSA-2018-0067-12326 at 10.
---------------------------------------------------------------------------
Bunch suggested the agencies use a joint model of household vehicle
holdings and sales that encompasses decisions to purchase new vehicles,
retain existing ones, or reduce or augment current holdings of vehicles
of all types and vintages in each period. Manufacturers would modify
either new vehicle content, prices, or both to produce a supply of new
vehicles that allowed them each to comply with standards. And,
subsequently, households and manufacturers would iteratively interact
until the market reached equilibrium. The model described by Bunch
would face many of the same issues outlined above. There are
significant econometric challenges associated with estimating a
household's decision to buy a new vehicle instead of a used vehicle (of
some vintage), or to maintain its current set. And integrating such a
model would
[[Page 24626]]
require the agencies to simulate the dynamics of the used vehicle
market--hundreds of unique nameplates for each of dozens of vintages--
in order to provide the correct choice set in each simulated year. Such
a model is beyond the scope of the current analysis.
While the agencies believe that these challenges provide a
reasonable basis for not employing a discrete choice model in today's
final rule analysis, the agencies also believe they are not
insurmountable, and that some suitable variant of such models may yet
be developed for use in future fuel economy and CO2
emissions rulemakings. The agencies have not abandoned the idea and
plan to continue experimenting with econometric specifications that
address heterogeneous consumer preferences in the new vehicle market as
they further refine the analytical tools used for regulatory analysis.
Operating at the level of individual auto and light truck model
variants--the same level at which compliance is, necessarily,
simulated--may not be tractable for rulemaking analyses. However,
market shares for brands and manufacturers within market segments are
more stable over time--even if the volumes of segments across the
industry fluctuate. In the 2012 final rule, the agencies' analysis
showed a new vehicle market where the share of passenger car body
styles--sedans, coupes, hatchbacks--reached almost 70 percent of the
new vehicle market by 2025, while light trucks, including many
crossovers, accounted for the remaining 30 percent. Those results were
consistent with the assumptions made in 2012, but the combination of
low fuel prices and decreasing differences in fuel consumption between
body styles has instead reduced the market share of those body styles
significantly (only 40% in the MY 2017 fleet), and, thus eroded the
value of the 2012 analysis to inform current decisions. Including a
choice model that operated on existing market shares, albeit at a
higher level of aggregation than specific nameplates, such as brand/
segment/powertrain, may be able to improve internal consistency with
the interaction of assumptions about fuel prices and regulatory
alternatives. The agencies will continue to engage with the academic
community and other stakeholders to ensure that future work on this
question improves our analysis of regulatory alternatives.
3) Scrappage
a) The Impacts of New Vehicle Fuel Economy Standards on Fleet Turnover
Economic literature and theory indicate that the retirement (or
scrappage) rates of existing vehicles slows when new vehicle fuel
economy standards increase and cause new vehicle price increases.
Slower retirement rates result in an older distribution of the on-road
fleet. Today's on-road fleet is the oldest it has ever been,
approaching an average of 12 years old.\1657\ Since older vehicles are,
on average, less safe and less fuel efficient, modeling this reduction
in the scrappage rates of existing vehicles has important implications.
As mentioned in the sales section above, past quantitative analyses of
CO2 and CAFE standards excluded the scrappage effect (though
the agencies discussed the scrappage effect qualitatively), which could
have resulted in an overestimate of the benefits of increasing
standards.
---------------------------------------------------------------------------
\1657\ Bureau of Transportation Statistics (BTS). ``Average Age
of Automobiles and Trucks in Operation in the United States.''
Available at https://www.bts.gov/content/average-age-automobiles-and-trucks-operation-united-states.
---------------------------------------------------------------------------
For the NPRM, the agencies chose for the first time to model the
change in existing vehicle retirement rates across regulatory
alternatives. The agencies used a logistic function to estimate the
instantaneous scrappage rate for vehicles of different body styles and
model year vintages using registration data from Polk, the estimated
durability of specific model year vintages, the prices of new vehicles,
a measure of the cost of travel for the model year cohort versus new
vehicles in any given calendar year, and other cyclical macroeconomic
indicators.\1658\
---------------------------------------------------------------------------
\1658\ For a more detailed explanation of the NPRM model, see
PRIA Chapter 8.10.
---------------------------------------------------------------------------
The agencies received many comments about the NPRM's scrappage
model. While some commenters objected to the inclusion of a scrappage
model, most commenters supported the inclusion of a dynamic scrappage
model as an improvement in the agencies' analysis; these comments are
discussed in Section VI.C.1.b)(3)(a)(ii). Other commenters raised
concerns about the specific scrappage models used in the NPRM analysis;
these are discussed in Section VI.C.1.b)(3)(b). Specifically,
commenters raised concerns about overfitting in the models, the
identification strategy, the modeling of new and used vehicle fuel
economy in general, the exclusion of certain variables, about how the
agencies captured macroeconomic effects, and about the lack of
integration with the sales model.
The agencies contemplated all of the comments and suggestions made
by commenters and, in response, have made several changes to final
rule's model. First, the agencies changed the time-series strategy used
in the model, as discussed in Section VI.C.1.b)(3)(c)(iii)(a). This
change allows the agencies to simplify the models significantly,
addressing commenters' concerns about potential overfitting of the
model and difficulty of interpreting individual coefficient values
(discussed in Section CI.C.1.b)(3)(b)(i)). Second, the agencies changed
the modeling of the durability effect as discussed in Section
VI.C.1.b)(3)(c)(iii)(c); this change reduces the reliance on the decay
function and has the added benefit of addressing concerns about
overfitting and out-of-sample projections discussed in Section
VI.C.1.b)(3)(b)(i). Third, a portion of anticipated fuel savings from
increased fuel economy are netted from new vehicle prices--meaning
consumers are now assumed to value fuel economy at the time of purchase
to a certain extent--as discussed in Section VI.C.1.b). This change is
in response to comments discussed in Section VI.C.1.b)(3)(c)(iii)(d)
and addresses inconsistent treatment of consumer valuation within the
NPRM's analysis. Finally, the agencies consider the inclusion of
additional or alternative variables in the scrappage model in response
to comments discussed in Section VI.C.1.b)(3)(b)(ii). After extensive
testing, the agencies concluded that these additional variables do not
improve the model fits or would introduce autocorrelation in the error
structures (see Sections VI.C.1.b)(3)(c)(iii)(e) and
VI.C.1.b)(3)(c)(iii)(f) for further discussion). As such, the agencies
rejected the additional terms suggested by commenters. Input from
commenters was used to simplify the scrappage model, make it more
consistent with modeling of new vehicle prices elsewhere in the
analysis, and improve its predictions for the instantaneous scrappage
rates of vehicles beyond age 20.
i) Basis for `The Gruenspecht Effect'
Gruenspecht (1981) and (1982) recognized that since fuel economy
standards affect only new vehicles, any increase in price (net of the
portion of reduced fuel savings valued by consumers) will increase the
expected life of used vehicles and reduce the number of new vehicles
entering the fleet. The effects of differentiated regulation in the
context of fuel
[[Page 24627]]
economy is often deemed the Gruenspecht Effect.\1659\ Jacobsen and van
Bentham (2015) first quantified the Gruenspecht Effect, or the share of
new vehicle fuel savings lost to the used vehicle fleet due to delayed
scrappage, to be between 13 and 16 percent.\1660\
---------------------------------------------------------------------------
\1659\ Gruenspecht, H. ``Differentiated Regulation: The Case of
Auto Emissions Standards.'' American Economic Review, Vol. 72(2),
pp. 328-331 (1982).
\1660\ M. Jacobsen and A. van Benthem, ``Vehicle Scrappage and
Gasoline Policy,'' American Economic Review, Vol. 105, pp. 1312-38
(2015).
---------------------------------------------------------------------------
As discussed in the write up of the sales model, fuel economy
standards increase the cost of acquiring new vehicles, but also improve
the quality of those vehicles by increasing their fuel economy. The
CAFE analysis assumes that consumers value 30 months of fuel savings,
so that the quality-adjusted change in new vehicle prices is the
increase in regulatory costs less 30 months of fuel savings. As long as
the quality-adjusted price is positive,\1661\ it becomes more expensive
for manufacturers to produce vehicles and, as a result, prices of new
vehicles increase. From a supply and demand perspective, this equates
to the supply curve for new vehicles moving inwards or to the left and
a corresponding increase in the equilibrium price and decrease in the
equilibrium quantity of new vehicles purchased.
---------------------------------------------------------------------------
\1661\ The quality adjusted price is positive when regulatory
compliance costs exceed 30 months of fuel savings.
---------------------------------------------------------------------------
New and used vehicles are substitutes. When the price of a good's
substitute increases, the demand curve for that good shifts upwards and
the equilibrium price and quantity supplied also increases. Thus,
increasing the quality-adjusted price of new vehicles will result in an
increase in equilibrium price and quantity of used vehicles. Since, by
definition, used vehicles are not being ``produced'' but rather
``supplied'' from the existing fleet, the increase in quantity must
come via a reduction in their scrappage rates. Practically, when new
vehicles become more expensive, demand for used vehicles increases (and
they become more expensive). Because used vehicles are more valuable in
such circumstances, they are scrapped at a lower rate, and just as
rising new vehicle prices push marginal prospective buyers into the
used vehicle market, rising used vehicle prices force marginal
prospective buyers of used vehicles to acquire older vehicles or
vehicles with fewer desired attributes.
ii) Commenter Response to the Inclusion of the Gruenspecht Effect
(a) Many Commenters Support the Inclusion of the Effect
Academic researchers and automakers widely agree with the existence
and direction of the Gruenspecht Effect. For example, RFF commented,
``There's good evidence supporting the scrappage effect.'' \1662\ The
Auto Alliance stated that the agencies ``made significant strides
toward improving their modeling of consumer behavior by adding new
modules to estimate new vehicle sales and in-use vehicle scrappage in
response to changes to new vehicle prices.'' \1663\ FCA agreed ``that
an outcome of the current augural stringency of the CAFE/
[CO2] emission regulations may be a decreasing trend in
vehicle scrappage rates as consumers delay purchases [. . .] forc[ing]
consumers to hold their current vehicles for additional time.'' \1664\
---------------------------------------------------------------------------
\1662\ RFF, Comments EPA NHTSA, NHTSA-2018-0067-11789, at 4.
\1663\ Auto Alliance, Full Comment Set, NHTSA-2018-0067-12073,
at 47.
\1664\ FCA, Comments for CAFE-GHG NPRM Final Public Version,
NHTSA-2018-0067-11943, at 22.
---------------------------------------------------------------------------
Other commenters agreed with the existence of the effect, but took
issue with the implications of the combination of the sales and
scrappage models. Mark Jacobsen stated ``while we agree that the
scrappage effects we study will mitigate changes in the used fleet, we
do not believe they could be strong enough to reverse completely the
direction of change in the used fleet.'' \1665\ Jacobsen's contention
was echoed by many commenters; the main point was that they believed
that the prices of both new and used vehicles should be less expensive
in the NPRM's preferred alternative than the augural standards, and
that this should, if anything, result in a larger fleet in the NPRM's
preferred alternative. This issue is further discussed in Section
(b)(iv) with other comments about integrating the sales and scrappage
models and the incremental fleet size across alternatives. Here it is
important to note that this concern does not suggest that a scrappage
model should not exist, but takes issue with the specific modeling of
scrappage and/or sales implemented in the NPRM analysis.
---------------------------------------------------------------------------
\1665\ Mark Jacobsen and Arthur van Benthem, Letter Describing
Scrappage Effects, NHTSA-2018-0067-7788, at 2.
---------------------------------------------------------------------------
b) Some Commenters Worry About the Shift in Agency Perspective
Some commenters argued that the agencies modeling of sales and
scrappage in the NPRM analysis contradicted previous positions that
these effects were too uncertain to model. For example, the Center for
Biological Diversity (CBD) commented:
In the 2012 rulemaking for fuel economy and [CO2]
standards, both NHTSA and EPA stated that analysis of the standards'
impact on new vehicles sales and on the ``scrappage'' of used
vehicles was too uncertain to be used in the rulemaking. The
agencies reiterated this position in their 2016 technical assessment
of the standards.\1666\
---------------------------------------------------------------------------
\1666\ CBD, Appendix A, NHTSA-2018-0067-12000, at 171.
---------------------------------------------------------------------------
They further stated:
The agencies have not provided a meaningful rationale or
justification for the change in position regarding their ability to
present quantified estimates of the impact of the standards on new
vehicle sales and the scrappage of used vehicles.\1667\
---------------------------------------------------------------------------
\1667\ CBD, Appendix A, NHTSA-2018-0067-12000, at 178.
To respond to these comments, it is useful to look at the reasons
the agencies gave for not considering fleet turnover effects on pages
---------------------------------------------------------------------------
845-46 of the 2012 rulemaking:
If the value of fuel savings resulting from improved fuel
efficiency to the typical potential buyer of a new vehicle outweighs
the average increase in new models' prices, sales of new vehicles
will rise, while scrappage rates of used vehicles will increase
slightly. This will cause the ``turnover'' of the vehicle fleet--
that is, the retirement of used vehicles and their replacement by
new models--to accelerate slightly, thus accentuating the
anticipated effect of the rule on fleet-wide fuel consumption and
CO2 emissions. However, if potential buyers value future
fuel savings resulting from the increased fuel efficiency of new
models at less than the increase in their average selling price,
sales of new vehicles will decline, as will the rate at which used
vehicles are retired from service. This effect will slow the
replacement of used vehicles by new models, and thus partly offset
the anticipated effects of the final rules on fuel use and
emissions.
Because the agencies are uncertain about how the value of
projected fuel savings from the final rules to potential buyers will
compare to their estimates of increases in new vehicle prices, we
have not attempted to estimate explicitly the effects of the rule on
scrappage of older vehicles and the turnover of the vehicle
fleet.\1668\
---------------------------------------------------------------------------
\1668\ 77 FR 62,623, 63,112-13 (emphasis added).
The agencies' reason for not modeling the fleet turnover effects in
prior rulemakings was not uncertainty about the direction or impact of
vehicle prices on sales or scrappage rates, but rather uncertainty
about how consumers value fuel savings. The agencies now have
sufficient knowledge regarding the amount of fuel savings consumers are
assumed to value at the time they purchase new vehicles and make these
[[Page 24628]]
assumptions in the technology application simulation. With this
assumption, it becomes possible to model the fleet turnover effects,
including the scrappage effect.
c) Some Commenters Think the Effects Are Uncertain
Other commenters argue that the sales and scrappage effects are too
uncertain to include in a rulemaking analysis. For example, CBD argued
that ``the models are attempting to evaluate the small and uncertain
effects of changes in vehicle standards on certain dynamics--vehicle
sales, scrappage rates, and vehicle usage--which are largely determined
by much stronger forces, such as the state of the economy.'' \1669\
---------------------------------------------------------------------------
\1669\ CBD, Appendix A, NHTSA-2018-0067-12000, at 177.
---------------------------------------------------------------------------
The agencies agree that there is uncertainty around the magnitude
of the sales and scrappage response, but do not agree that sign of
either effect is uncertain. Importantly, excluding modeling of the
sales and scrappage effects would only make sense if there was a
legitimate existential concern--the sales and scrappage effects are
founded in very basic economic theory, as noted above, in Section
VI.C.1.b)(3)(a)(i). Furthermore, the agencies believe that assessing
the magnitudes of the sales and scrappage effects is a tractable task
for researchers and sufficient data exists to quantify these effects.
Thus, excluding these effects would be a serious omission that limits
accurate accounting of the costs and benefits of fuel economy
standards. Other stakeholders commented that the NPRM analysis did not
thoroughly consider the uncertainty around the magnitudes of the sales
and scrappage responses. These comments and the agencies response is
discussed in Section VI.C.1.b)(3)(b)(i), below. The agencies believe it
is better to consider a range of the scrappage and sales response to
address concerns about uncertainty, and that excluding them would be
inappropriate.\1670\ The agencies did just that with the proposal
through sensitivity analyses--including seeking comment and having the
scrappage modeling peer reviewed--and continue to do so for the final
rule.
---------------------------------------------------------------------------
\1670\ See, e.g. Ctr. for Biological Diversity v. Nat'l Highway
Traffic Safety Admin., 538 F.3d 1172, 1203 (9th Cir. 2008), (finding
that NHTSA inappropriately assigned no value to reducing carbon
emissions when the value for doing so was ``certainly not zero.'').
---------------------------------------------------------------------------
b) Summary of Notice, Request for Comments, and the Agencies' Response
The comments related to the scrappage model are summarized here
into five major categories: Overfitting and identification strategies,
modeling fuel economy and new vehicle prices, consideration of other
additional variables, integration with sales or VMT, and evaluations of
associated costs and benefits due to changes in scrappage rates within
the CAFE model. Specific modeling decisions the agencies have made or
considered in response to the public comments summarized in this
section are discussed in Sections VI.C.1.b)(3)(c)(ii)(d) and
VI.C.1.b)(3)(c)(iii).
i) Overfitting and Identification Strategy
Several commenters argued that the NPRM scrappage model did not
have a clear identification strategy, or that the model over-fit the
data. These commenters suggest that the NPRM model may not capture a
causal relationship, but picks up other correlation or noise within the
data. This section outlines the specific claims made by commenters.
a) Overfitting and the Use of Lagged and Interactions Terms
Several commenters argued that the results presented in the NPRM
could be driven by the specific structure of the price effect used in
the scrappage models that were implemented into the CAFE Model. IPI,
California States et. al., CARB, and other commenters suggested that
the NPRM model is over-fit. CARB outlined its argument that the
agencies overfit the data in the following passage:
[T]he model appears to be significantly overfit and to suffer
from multicollinearity. An overfit model means that the model is
able to precisely replicate past trends, but only through the use of
too many variables. An overfit model fits the data too well, fitting
the noise or errors in the data in addition to the underlying
relationships between the variables of interest. Because an overfit
model also fits the noise and errors of the data, the out-of-sample
predictions are unreliable. Comments from Jeremy Michalek and Katie
Whitefoot suggest that choice of specification of the scrappage
model could result in substantially different predictions, and that
the agencies should make only those claims that are robust to
reasonable variations in the model specifications.\1671\
---------------------------------------------------------------------------
\1671\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 245.
The agencies agree that it is important that the scrappage model
results are robust across those specifications that meet a set of
econometric criteria (these criteria are discussed further in Section
VI.C.1.b)(3)(c)(iii)). However, the agencies acknowledge that the NPRM
could have provided further evidence that the specification did not
drive the results. In the analysis for the final rule the agencies have
presented more than one specification of the price effect as evidence
that the specification chosen here does not drive the results of the
analysis. Further, claims that the specification of the scrappage
response in the NPRM is inconsistent with economic theory are false.
Theoretically, changes in average new prices may have longer-term
trends that can be picked up by including lagged terms, and/or be non-
linear with age, so that vehicles of different ages have different
elasticities of scrappage (relative to changes in average new vehicle
prices). Further, sometimes the effect of one independent variable on
the dependent variable depends on the magnitude of another independent
variable--this is called an interaction effect. Regression analysis can
capture these interaction effects by defining a new variable using some
combination of independent variables.\1672\ It is necessary to retain
such interaction terms when doing so.\1673\ For example, it is not
obvious that the elasticities of scrappage rates to changes in new
vehicle prices should be constant for all vehicle ages, or put another
way, the older a vehicle is, the higher likelihood the vehicle will be
scrapped instead of being retained or resold.
---------------------------------------------------------------------------
\1672\ Davis, J. B., Statistics using SAS enterprise guide.
Cary, NC: SAS Institute, pp. 411-415 (2012).
\1673\ As explained in more detail in Section
I.A.1.a)(1)(a)(ii)(a), below, the agencies perform several
sensitivity analyses to ensure the model captures the correct impact
of interactive effects.
---------------------------------------------------------------------------
Michalek and Whitefoot, Honda, and other commenters, argued that
the fact that some of the interaction terms were not statistically
significant was evidence that the response measured is uncertain. CBD
in particular claimed that the ``scrappage model is poorly constructed,
and its results are not statistically significant.''
In response to such comments, it is important to note that when
interaction terms are included, the significance of the overall effect
of a variable should be tested by performing a restricted F-test, which
simultaneously tests that all coefficients of the variable of interest
are jointly indistinguishable from zero. The insignificance of one term
of the interaction does not imply that the effect is indistinguishable
from zero.\1674\
---------------------------------------------------------------------------
\1674\ Davis, J. B., Statistics using SAS enterprise guide.
Cary, NC: SAS Institute, pp. 411-415 (2012).
---------------------------------------------------------------------------
Commenters also noted the lagged terms and age interactions make
the new vehicle price effect difficult to interpret. IPI argued that
``[t]he inclusion of interaction variables make it very difficult to
evaluate the results of the regression for an individual variable
[[Page 24629]]
of interest.'' Michalek and Whitefoot suggested ``using a Monte Carlo
analysis to understand the distribution of scrappage outcomes implied
by uncertainty of the value of the coefficients in the model regression
and reporting 95% confidence intervals.''
We agree that the inclusion of lags and age interactions of new
vehicle prices can make interpreting the sign and magnitude of the
price effect difficult. It also makes it difficult to use the
confidence intervals on the coefficients as a way to capture
uncertainty, since the interaction variables are jointly estimated.
Thus, for the NPRM analysis, the agencies could not independently
sample each coefficient from the confidence intervals and perform a
Monte Carlo analysis.
While the agencies think that the inclusion of lags and interaction
terms is theoretically plausible, in response to commenter and peer
reviewer concerns about overfitting and the difficulty of interpreting
coefficients, the agencies reconsidered the time series approach. The
agencies found that new vehicle prices are integrated to order one and
that the dependent variable is stationary (as discussed in Section
VI.C.1.b)(3)(c)(iii)(a)). It is therefore sufficient to fit the first
difference of new vehicle prices within the models. Thus, the agencies
have simplified the central model of the response of scrappage rates to
changes in new vehicle prices to exclude lags of the effect. The
agencies further simplified the central scrappage models to exclude
interaction of new vehicle prices and vehicle age; this allows the
agencies to take the 95 percent confidence intervals as a low and high
range for the magnitude of the price effect for the sensitivity
analysis. The agencies also include a sensitivity analysis which
includes interaction terms between new vehicle price and vehicle age to
allow the elasticity of scrappage to changes in new vehicle price to
vary by vehicle age.
Commenters also noted that the model did not perform well for
vehicles beyond age 20. The agencies noted in the PRIA that the Polk
dataset for older vehicles was limited and likely led to the inability
to estimate the scrappage rates for older ages.\1675\
---------------------------------------------------------------------------
\1675\ FR, Vol 83, No. 165, August 24, 2018, p.43097.
---------------------------------------------------------------------------
The final rule dataset includes almost 30 percent more data for
vehicles fifteen years or older than the NPRM, which improves estimates
of the scrappage rate of vehicles aged 20 to 30 (see Table VI-158). The
agencies are still unable to capture the scrappage trends for vehicles
over 30, as the dataset is still limited for the oldest ages of
vehicles, and still rely on the decay function used in the NPRM for
vehicles over the age of 30. The limited data explains the inability to
predict the scrappage rates for older vehicles. However, including
model year fixed effects and including the share of the initial cohort
remaining does improve predictions of the final share remaining in the
final rule models. These changes are discussed in Section
VI.D.1.b)(c)(i)(c).
b) Reduced Form and Endogenous Prices
California States et. al., CARB, EDF, IPI and academic commenters
expressed concerns that the NPRM analysis fit a reduced form of the
scrappage model, rather than a structural model. In other words,
instead of explicitly modeling new and used vehicle prices in
equilibrium under different regulatory alternatives and applying a
measurement of the elasticity of scrappage to the resulting used
vehicle prices, the agencies modeled the elasticity of scrappage from
changes to new vehicle prices. For example, California States et. al.,
argued that the model ``does not link the new and used vehicle markets
as required by economic theory, nor does it attempt to measure used
vehicle prices, which form the basis of scrappage theory.''
While the agencies recognize that there are certain advantages to a
structural model, they disagree that the sales of new and used vehicles
must be modeled simultaneously. The agencies do link the new and used
car markets by including new vehicle prices as an independent variable
in scrappage regression equation. However, it would be inappropriate to
include used vehicle prices in this equation due to endogeneity
concerns. A change in used vehicle prices may change scrappage rates,
but also an exogenous shock to scrappage rates may cause used car
prices to vary.
Furthermore, the agencies are unaware of a viable structural model
for the scrappage effect. The agencies performed an extensive review of
economic of literature, both before creating the scrappage model for
the proposal and revising it for the final rule, but were unable to
find such a model or any insights on how to construct one. The agencies
note that commenters did not suggest a structural model that the
agencies should use or give any indication of whether such a model
exists.
In order to understand why such a model is difficult to construct,
it is important to understand what a structural model of the sales and
scrappage responses would entail. A hypothetical structural model for
the new vehicle market can be represented by the following simultaneous
demand and supply equations:
DNew = [beta]0 + [beta]1 * PNew +
[beta]2 * PUsed + [beta]3 * PTransit +
[beta]4 * Income + [beta]5 * Households
SNew = [beta]6 + [beta]7 * PNew +
[beta]8 * Production CostNew
The demand equation for new vehicles in a given year is determined by
the annual price of owning and operating new vehicles, the annual price
of owning and operating used vehicles, the annual price of other
substitutes, average household income, and the number of households.
The supply equation is made up of the average price of new vehicles and
the average cost to produce them.
As noted in the sales model write up, reducing required fuel
economy stringency reduces the cost of producing new vehicles, and
shifts the supply curve to the right. This results in an increase in
the quantity supplied of new vehicles.
The structural model for the used vehicle market can be represented
by the following simultaneous demand and supply equations:
DUsed = [gamma]0 + [gamma]1 * PUsed +
[gamma]2 * PNew + [gamma]3 * PTransit +
[gamma]4 * Income + [gamma]5 * Households
SUsed = [gamma]6 + [gamma]7 * PUsed +
[gamma]8 * Maint RepairUsed + [gamma]9 * Scrap
ValueUsed
The aggregate demand equation for used vehicles is determined by
the price of owning and operating used vehicles, the price of owning
and operating new vehicles, the price of other transit substitutes,
average income, and the number of households. The supply curve equation
for used vehicles is determined by the price of used vehicles, the cost
to repair and maintain them in service, and the opportunity cost of the
scrappage value of doing so. Relaxing new vehicle standards reduces new
vehicle prices and shifts the demand curve for used vehicles downward,
which reduces demand for used vehicles and the equilibrium price and
quantity of used vehicles, and increases the annual scrappage rate.
Modeling the structural equations would require that the agencies
predict new and used vehicle prices in equilibrium, allowing prices of
new and used vehicles be determined simultaneously from estimates of
the supply and demand curves for each market. As CARB stated in the
following comment, new and used vehicle prices are endogenous--the
equilibrium prices of each good are simultaneous:
[[Page 24630]]
Because both scrappage rates and new vehicle prices may
influence one another, the Agencies would need to utilize different
statistical techniques to credibly identify the impact of new
vehicle prices on scrappage rates. For example, the Agencies would
need to identify an instrumental variable that impacts new vehicle
price but that does not impact the scrappage rate. Models that
suffer from endogeneity problems will have biased estimates. In
other words, the estimates from these models cannot be used to
inform policy, because they do not actually tell us how new vehicle
prices impact scrappage.
CARB suggested a way to correct for endogeneity: Using an
instrumental variable in a two-stage least squares methodology where
the instrumental variable is correlated with new vehicle prices, but
not scrappage rates.\1676\ The agencies could also address the
potential for endogeneity in two steps: First, they could model the
impacts of exogenous changes in new vehicle prices on used vehicle
prices, and second, they could model the impacts of exogenous changes
in used prices on scrappage rates. Implementing the first step would
require using an instrumental variable to isolate exogenous shifts to
the new vehicle supply curve, and then using the predicted values of
new vehicle prices to model changes in prices for used vehicles of all
ages. Because prices and scrappage rates are jointly determined in the
market for used vehicles, predicting the elasticity of scrappage with
respect to price variation also requires isolating exogenous changes in
used vehicle price via the use of an instrumental variable.
---------------------------------------------------------------------------
\1676\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 244.
---------------------------------------------------------------------------
There is one literature example that approaches the structural
model that some commenters would like the agencies to implement.
Jacobsen and van Bentham \1677\ developed a structural model that
simultaneously solves for prices that clear new and used vehicle
supplies, and then applies an elasticity of scrappage measure that
corrects for potential endogeneity of used vehicle values and scrappage
rates using an instrumental variable methodology. Specifically, they
use changes in fuel prices as an instrumental variable; changes in fuel
prices shift the demand for different vehicle models, but not the cost
of supplying them. This should capture exogenous changes in value, so
that an exogenous measure of the scrappage elasticity can be isolated
in the second stage of the two-staged least squares method.
---------------------------------------------------------------------------
\1677\ M. Jacobsen and A. van Benthem, ``Vehicle Scrappage and
Gasoline Policy,'' American Economic Review, Vol. 105, pp. pp. 1312-
38 (2015).
---------------------------------------------------------------------------
While Jacobsen and van Bentham are able to correct for potential
endogeneity between used vehicle values and their scrappage rates,
their structural model to set new and used vehicle values
simultaneously makes some presumptions that the agencies are not
comfortable making. First, they calibrate their constant elasticity of
substitution (CES) utility function using 1999 data from GM's internal
model. This type of model would estimate elasticities of specific
vehicle models and require a pricing strategy other than allotting all
additional technology costs to the vehicle models to which they are
applied. The agencies have avoided a pricing strategy for the reasons
cited in the sales model write up. Second, by relying on GM's internal
model, Jacobsen and van Bentham used elasticities calculated using only
1999 data of the GM fleet. The agencies do not expect that elasticities
estimated from 20-year old data from a single OEM's portfolio of
vehicles would translate to the entirety of the current vehicle
fleet.\1678\ Finally, Jacobsen and van Bentham represent total vehicle
demand of a representative consumer from a composite vehicle. This
approach precludes the realistic consideration that a household may
prefer two used vehicles over one new vehicle, which is accounted for
in the agencies' functional equations.
---------------------------------------------------------------------------
\1678\ Kleit, Andrew N., 2004. ``Impacts of Long-Range Increases
in the Corporate Average Fuel Economy (CAFE) Standard.'' Economic
Inquiry 42:279-94.
---------------------------------------------------------------------------
Jacobsen's and A. van Benthem's model is not a household level
choice model, and is not meant to determine fleet size, as noted in
their comment:
In summary, while the Jacobsen and van Benthem (2015) paper
cannot inform by how much the total vehicle fleet would expand under
a CAFE rollback (since we do not estimate by how much it shrinks
under CAFE), all the evidence and economic logic points to a larger
total vehicle fleet under a rollback, at odds with NHTSA's fleet
turnover model.\1679\
---------------------------------------------------------------------------
\1679\ Mark Jacobsen and Arthur van Benthem, Letter Describing
Scrappage Effects, NHTSA-2018-0067-7788, at 2.
The agencies agree that the long-term fleet should be smaller in
the augural case, as fewer new vehicles flow into the used car market
(because of lower sales), but do think it is plausible that in the
short term the fleet size could increase under augural standards if in
some cases consumers substitute two used vehicles for one new one or
choose to retain an additional vehicle on the margin because the higher
value makes doing so a more reasonable investment (at the annual
level). This sort of outcome is not possible with the Jacobsen and van
Bentham 2015 model, because the overall demand for vehicles is set by
the annual rent prices of a composite vehicle. The updates to the
scrappage model for the final rule are consistent with this view, but
do show a smaller fleet size under the augural standards relative to
the proposal. This is discussed further in Section
VI.C.1.b)(3)(b)(iv)(b).
Fitting the reduced form equation requires that endogenous
variables are excluded from the model to avoid biased coefficients. As
a result, used vehicle prices were omitted by design, because used
vehicle prices and scrappage rates are endogenous.\1680\ Some
commenters argue that new vehicle prices and scrappage rates are also
endogenous; CARB argued that ``the model tries to rely solely on new
vehicle prices to predict scrappage rates without realizing or
controlling for the fact that scrappage rates may also affect new
vehicle prices.'' \1681\
---------------------------------------------------------------------------
\1680\ Hill, R. C., Griffiths, W. E., & Lim, G. C. Chapter 11:
Simultaneous Equation Models. In Principles of Econometrics (3rd
ed., pp. 303-24). Hoboken, NJ: John Wiley & Sons, Inc. (2008).
\1681\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 244.
---------------------------------------------------------------------------
Commenters provided neither evidence nor an explanation as to why
there may be some degree of ``reverse causality'' or endogeneity
between new vehicle prices and scrappage rates. Two potential
econometric explanations for such endogeneity could be that: (1) These
variables are jointly or simultaneously determined, so each one
influences the other; or (2) the model omitted a variable that causes
covariance between new vehicle prices and scrappage rates. The agencies
believe the first source of potential endogeneity can be dismissed, as
any causal relationship between scrappage rates and new vehicle prices
would necessarily flow through the used car market, which are
substitute products for new vehicles, and specifically through the
mechanism of used car prices. For example, an exogenous shock to
scrappage rates might cause the supply curve in the market for the
lowest-price used vehicles to shift, and the resulting change in their
price might cause price responses in higher-price segments of the used
vehicle market, which in turn might eventually filter up to the new
vehicle market and affect the prices for new vehicles. This chain of
events suggests omitted variable bias might be a concern, rather than
simultaneity.
The agencies believe that supply and demand for used vehicles (or
some measure of their interaction, such as
[[Page 24631]]
used vehicle prices) are the most likely sources of any potential
omitted variable bias. If an omitted variable is causing bias in the
estimates, then the bias is observable. Whether endogeneity--through an
omitted variable--is causing bias is an empirical question, which can
be answered by conducting common empirical test--the Durbin-Wu-Hausman
test. The Durbin-Wu-Hausman test requires identifying a suitable
instrument(s)--a variable--that is correlated with new vehicle prices
but not with scrappage rates, so any effect exerted on scrappage rates
by the instrument will occur through their association with prices for
new vehicles.\1682\ The agencies tested a few alternative approaches,
which included using the change in new vehicle prices during the
preceding time period and the level of prices during the current period
as instrumental variables for the change in prices during the current
period, and another test using the current-period growth rate in GDP as
an instrument for the change in new vehicle prices during the current
period. Each of these tests fails to reject the null hypothesis that no
endogeneity is present at the 0.05 level of significance.
---------------------------------------------------------------------------
\1682\ For a conceptual overview of this test, see https://www.statisticshowto.datasciencecentral.com/hausman-test/. For a more
detailed description of the logic underlying the test and how to
interpret its results, see http://personal.rhul.ac.uk/uhte/006/ec2203/Lecture%2015_IVestimation.pdf.
---------------------------------------------------------------------------
For both theoretical and empirical reasons, the agencies are
therefore skeptical about both the likelihood that scrappage rates will
affect prices for new vehicles, and the extent to which they might do
so. The agencies find the theoretical underpinnings for endogeneity to
be tenuous, and believe the empirical evidence suggests such
endogeneity is not an issue for today's analysis.
The agencies chose not to fit a model predicting used vehicle
prices directly from new vehicle prices for the proposal because
currently-available time-series data on the prices of used vehicles of
a given vintage going back to 1975 is limited. EDF cited the lack of
available data as the reason not to fit the structural model:
In the absence of any data or analysis, NHTSA did not describe
the extent to which changes in new vehicle prices affect used
vehicle prices of varying age, condition, etc. \1683\
---------------------------------------------------------------------------
\1683\ EDF, Appendix B, NHTSA-2018-0067-12108, at 56.
The agencies note that acquisition, assembly, and cleaning of a
nationally representative database for calendar years 1974 to 2017 on
used vehicle prices by vintage from Kelly Blue Book (or a similar
source) would take months to years, and would push the final rule
beyond the necessary April 2020 lead time requirement to set MY 2022
standards. Kelly Blue Book data is readily searchable for current
prices, but without a time series of used vehicle prices the data
cannot be used to answer the causal relationship of changes in used
vehicle prices over time on vehicle retirement rates. Even assembling a
nationally representative sample of used vehicle prices by vintage
would be a major undertaking. This is not to suggest that doing so is
out of scope for future analyses; the agencies plan to consider further
the possibility of conducting additional analysis on the relationship
between new and used vehicle prices.
The agencies considered use of the Consumer Expenditure Survey
(CEX), which has reported vehicle transaction data annually since
1984.\1684\ However, the sample of used vehicle purchase prices aged
twenty and older is severely limited. For vehicles purchased between
1996 and 2017, the average number of transaction prices reported for
vehicles aged 20 is 58, and for vehicles aged 25 is 18. Any computation
of average used vehicle prices from such a small sample would not be
reliable, and in fact, would be quite noisy. The agencies do not think
that estimates of a structural model based on such limited sampling
would improve the prediction of the scrappage effects over use of the
reduced form equation.
---------------------------------------------------------------------------
\1684\ U.S. Bureau of Labor Statistics. (2016). Consumer
Expenditures and Income: Collections & Data Sources. Retrieved from
https://www.bls.gov/opub/hom/cex/data.htm.
---------------------------------------------------------------------------
EDF argued that modeling the impact of changes in new vehicle
prices directly on used vehicle scrappage may not capture the fact that
changes in used vehicle prices impact vintages differently. Further,
they argue that if new and used vehicle prices change by the same
proportion, the effect will have a very small impact on the prices of
the oldest used vehicles. They argue that these small changes are not
enough to change the scrappage decisions:
Given that vehicles can sell for as little as a couple of
hundred dollars and new vehicle prices average over $30,000, used
vehicle prices can be as little as 1% of that of a new vehicle.
Given that the largest increase in new vehicle prices projected by
NHTSA in the NPRM is less than $3000, and assuming that its effect
on used vehicle prices is likely to be roughly proportional to
current relative prices, this might mean that the value of a very
old vehicle or one in poor condition might only increase by $30
(decline by $30 under the proposal). It is difficult to see how such
a change in value would have a measurable impact on scrappage. Of
course, the impact of an increase in new vehicle prices on used
vehicle prices might be more or less than proportional to their
current relative values. However, NHTSA has done nothing to show
which might be the case. The probability of any realistic change in
used vehicle prices to induce the scrappage of used vehicles is
still a complete mystery.\1685\
---------------------------------------------------------------------------
\1685\ EDF, Appendix B, NHTSA-2018-0067-12108, at 52.
However, the age interaction on the new vehicle price effect allows
that the elasticity of scrappage to changes in new vehicle prices may
not be constant for all ages. Allowing the scrappage elasticity to new
vehicle prices to vary by age incorporates the fact that the elasticity
of scrappage of used vehicles and the cross-price elasticity of used
vehicle demand to new vehicle prices may not be constant with age. At
some point, the thirty-dollar increase EDF cited could be the
difference in keeping a marginally used vehicle on the road; it would
be a 10 percent increase in the price of a used vehicle, and may cover
State registration fees on a marginally scrapped vehicle.
(c) Time Series
The scrappage model utilizes panel data. Panel data observes
multiple individuals or cohorts over time. The data employed by the
scrappage model observes the scrappage rates of individual model year
cohorts between successive calendar years. The model allows for the
isolation of trends over time and across individuals.\1686\ Since the
scrappage model uses aggregate model year cohorts to estimate scrappage
rates by age and time-dependent variables (new vehicle prices, fuel
prices, GDP growth rate, etc.) panel data is necessary to estimate the
model. A major challenge to using panel data is that the data structure
requires consideration of potential violations of econometric
assumptions necessary for consistent and unbiased estimates of
coefficients both across the cross-section and along the time
dimension. The cross-section of the scrappage data introduces potential
heterogeneity bias--where model year cohorts may have cohort-specific
scrappage patterns. \1687\ Another way to put this is that each model
year may have its own inherent durability. The NPRM captured this
potential bias by including model year as a continuous variable, but
the model amended for the final rule includes the more traditional
[[Page 24632]]
individual fixed effects. This is discussed in Section
VI.C.1.b)(3)(c)(iii)(a). The time dimension of a panel introduces a set
of potential econometric concerns present in time series analysis. The
agencies considered potential autocorrelation in the error structures
and included lags of the dependent and specific independent variables
to correct for it; this is not an uncommon practice in dynamic panel
models.\1688\ Some commenters argued that time series approaches were
not appropriate in the scrappage model at all. CARB stated the
following:
---------------------------------------------------------------------------
\1686\ Cambridge University Press. (1989). Analysis of Panel
Data. New York, NY.
\1687\ Cambridge University Press. (1989). Analysis of Panel
Data. New York, NY.
\1688\ Bun, M. J. G., & Sarafidis, V. (2015). Dynamic Panel Data
Models. In The Oxford Handbook of Panel Data (pp. 76-110). New York,
NY: Oxford University Press.
Time-series analysis for modeling scrappage is also
inappropriate for the same reasons as it was for the new vehicle
sales model--particularly because time-series analysis does not
capture structural changes, which the scrappage model seeks to
illustrate.\1689\
---------------------------------------------------------------------------
\1689\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 243.
The agencies disagree with CARB's assessment. The potential
scrappage effect can only be measured with a time series dimension; the
agencies are interested in how changes in new vehicle prices over time
impact the retirement rate of the on-road fleet over time. In order to
isolate this effect, the agencies need multi-period data on the
scrappage rates of used vehicles and prices of new vehicles.
The literature on vehicle scrappage rates utilizes panel data, but
most research has ignored potential autocorrelation issues caused by
the structural properties of independent variables that vary along the
time dimension. With the NPRM analysis, the agencies found evidence of
auto-correlated errors, which were corrected by including three lagged
terms of the dependent variable.\1690\ While in a pure time series
analysis, this can be an appropriate methodology to account for
autocorrelation in the error structure; estimates of the coefficients
of the lagged dependent variable are biased downwards when applied in
fixed or random effects panel models. The reason for this is that the
constant individual specific terms are correlated with the lagged
dependent variable (by definition, since the individual specific terms
are constant for all time periods, including the previous period),
creating a bias in the estimate of the coefficient on the lagged
dependent variable, and potentially other measures.\1691\ The eponymous
bias was first discussed in a paper written by Nickell in 1982.\1692\
There is an increasing body of work developing estimators built
specifically for dynamic panel data (DPD), or panel data where there is
an autoregressive component to the data-generating process. In other
words, the previous value of the dependent variable impacts the current
value.
---------------------------------------------------------------------------
\1690\ FR, Vol 83, No. 165, August 24, 2018, p.43097.
\1691\ Allison, P., Don't Put Lagged Dependent Variables in
Mixed Models, (2015, June 2). Retrieved June 1, 2019, from https://statisticalhorizons.com/lagged-dependent-variables.
\1692\ Nickell, Stephen. ``Biases in Dynamic Models with Fixed
Effects.'' Econometrica, vol. 49, no. 6, 1981, pp. 1417-26. JSTOR,
www.jstor.org/stable/1911408.
---------------------------------------------------------------------------
Further research into this literature (discussed above), comments
on the NPRM, and peer review comments prompted the agencies to
reconsider the approach developed for the NPRM. The NPRM analysis did
not use fixed effects for specific model years, but instead imposed a
parametric logarithmic relationship of successive model years. This
parametric model year term will still result in biased estimates of the
lagged dependent variable because it also does not vary over time for
the same model year, and is therefore correlated with the
autoregressive term. Since the autoregressive term carries through
effects from the previous period (the new vehicle price effect), this
will also bias the predicted Gruenspecht effect in the NPRM model.
Updates to the model used for the final rule correct this issue by more
deliberately considering the time series properties of both the
dependent and independent variables.
In reconsidering the appropriate way to address the time series
properties of the scrappage model, the agencies first consider the
stationarity of dependent and independent variables. This was suggested
in James Sallee's peer review:
In contrast to the new vehicle sales regression reported in the
PRIA's section 8.6, the discussion of the scrappage regressions does
not include any discussion of the time series properties of the
estimators. It is important to test for non-stationarity, for
example.\1693\
---------------------------------------------------------------------------
\1693\ CAFE Model Peer Review (Report No. DOT HS 812 590).
Washington, DC--National Highway Traffic Safety Administration, B-
64.
Importantly, the agencies find that the instantaneous scrappage rate is
stationary, so that there is no longer term information in the
scrappage rates to recover with an autoregressive term. This means that
a DPD model is not necessary to correct for potential autocorrelation
in the model. This also implies that the autocorrelation in the errors
is a result of non-stationarity in some or all of the regressors, and
not the independent variable. The solution to this problem is to
identify the order of integration of each regressor and difference
until each is non-stationary. Table VI-160 in Section
VI.C.1.b)(3)(c)(iii)(a) shows the order of integration of variables
considered in the scrappage modelling.
(ii) Modeling Fuel Economy
(a) Counterintuitive Signs
In the NPRM analysis, the agencies controlled for the changes in
the relative fuel economy of new and used vehicles by including the
cost per mile of travel in the current period and the previous period
for both new vehicles and the model year cohort whose scrappage is
being predicted. This allowed fuel prices to alter the scrappage rates
of existing vehicles, meaning model year cohorts with lower-than-
average fuel economies were impacted by increases to fuel prices to a
greater extent than cohorts with higher-than-average average fuel
economies. It also allowed increases in the fuel economy of new
vehicles to impact the scrappage rates of existing vehicles; the idea
is that when new vehicles have a higher average fuel economy, holding
price constant, the demand for new vehicles should increase relative to
used vehicles, and scrappage rates should increase. While this was a
plausible way of controlling for changes in the relative fuel cost per
mile of usage of new and used vehicles, the agencies noted in the NPRM
that some of the signs on new vehicle cost per mile were
counterintuitive, so that increases in the average new vehicle fuel
economy of certain body styles actually increased the scrappage rates
of existing vehicles.
IPI, CARB, CBD, Natural Resources Defense Council (NRDC), and other
commenters argued that these results were driven more by modeling
decisions than by actual relationships within the data. NRDC suggested
that the conclusions from the NPRM model should be treated with
suspicion until validated by further research:
[A]n increase in fuel price for a given level of fuel economy
results in longer vehicle retention even though operational costs
per mile increase. While it is not possible to rationalize this
response without significant additional research, it is indicative
of the fact that the algorithm response functions may not be
properly defined.\1694\
---------------------------------------------------------------------------
\1694\ NRDC, Attachment 3: CAFE Model Activity Review, NHTSA-
2018-0067-11723, at 20.
The agencies agree that the results were counter-intuitive--having
identified this issue in the NPRM and
[[Page 24633]]
specifically seeking comment on the matter--and considered multiple
alternative methods of capturing the fuel economy improvements of new
vehicles within the scrappage model in response to comments. Among the
changes considered were alternate forms of modeling the form of new
---------------------------------------------------------------------------
vehicle fuel economy, as suggested by IPI:
A paper by Shanjun Li et al., provides a useful example of how
the agencies could include fuel efficiency in their regression
without raising the econometric concerns that may be leading to
their nonsensical results. Li et al. include fuel price and vehicle
fuel efficiency (gallons per mile) of used vehicles as well as a
variable that captures the interaction of fuel efficiency of used
vehicles and fuel price in their regression as explanatory
variables. Unlike the agencies' model, the regression analysis used
in the Li et al. paper found results that are consistent with
economic theory: A decrease in overall demand for vehicles and an
increase in demand for more fuel-efficient cars.\1695\
---------------------------------------------------------------------------
\1695\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 72.
The NPRM included changes in new vehicle cost-per-mile, but did not
include separate variables for fuel prices or fuel economy. This could
potentially have conflated changes in the cost-per-mile of new vehicles
from changes in fuel prices and changes in new vehicle fuel economy.
The agencies considered including changes in fuel prices and new
vehicle fuel economy as separate measures, as suggested in IPI's
comment above, but opted for a different method of addressing the
concern of how to include changes to new vehicle fuel economy in the
scrappage model. However, specifications considering this approach are
shown in Section VI.C.1.b)(3)(c)(iii)(d).
(b) New Vehicle Prices Net of Fuel Savings
UCS, CBD, NRDF, EDF, and other commenters expressed concern that
quality adjustments were not included in the price series used to fit
the NPRM model. In particular, commenters suggested that the valuation
of fuel savings at the time of purchase should be deducted from the new
vehicle price increases. For example, CBD argued:
. . . [T]he agencies rely heavily on work by Howard Gruenspecht
regarding the scrappage effect, and the NPRM acknowledges that
Gruenspecht considered the effect of an increase in price ``net of
the portion of reduced fuel savings valued by consumers.'' Yet
consumer valuation of fuel savings is excluded from the scrappage
model, as well.\1696\
---------------------------------------------------------------------------
\1696\ CBD, Appendix A, NHTSA-2018-0067-12000, at 177.
The scrappage model cannot include both independent variables on
the fuel economy and cost-per-mile of new vehicles, and adjust the new
vehicle prices by the value of fuel savings considered at the time of
purchase, which would account for the improvement of the fuel economy
of new vehicles twice. Thus, the agencies must choose between these
methods to capture the value improvement of new vehicles when their
fuel economy increases. The agencies show both methods in Section
VI.C.1.b)(3)(c)(iii)(d). However, additional comments give reason to
prefer a methodology that does not model the fuel economy or cost per
mile of new model year cohorts directly, but instead adjusts the new
vehicle price series by the amount of fuel savings valued at the time
of purchase.
IPI expressed concern that the cost-per-mile measure was included
in the scrappage model, but not in the sales model:
[T]he CPM results in the scrappage model are inconsistent with
the agencies' sale model. In the sales module, the agencies have
chosen to ignore consumer demand for fuel economy and significantly
boosted the price impact of the baseline standards as a result. But
in the scrappage model, the agencies have incongruously allowed
consumer valuation of fuel economy to drive a significant portion of
the estimated fatalities.\1697\
---------------------------------------------------------------------------
\1697\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 79.
The agencies note that the fuel economy of new vehicles was not
included in the sales model because the signs were statistically
insignificant when it was included, and the fit of the overall model
was not improved. It was not excluded because the agencies do not think
that new vehicle fuel economy does not affect their sales. One way to
consider the value of increased fuel economy in both the sales and the
scrappage model (in the same way) is to adjust the price of new
vehicles by the amount of fuel savings consumers value at the time of
purchase in both models. This is also consistent with how the CAFE
model applies technology in the absence of CAFE standards, or when a
manufacturer is already in compliance with existing standards. In
response to comments about the counterintuitive signs of the change in
new vehicle cost per mile for some body styles, and about the
disconnect in how the fuel economy of new vehicles is modelled in the
sales and scrappage models, the agencies have adjusted the new vehicle
price series in both models by the amount of fuel savings consumers are
assumed to value at the time of purchase (30 months of fuel savings).
As noted in Section VI.C.1.b)(3)(b)(ii)(a), alternatives to this
solution are presented in Section VI.C.1.b)(3)(c)(iii)(d). The agencies
also discuss consideration of other quality improvements over
successive model years in Section VI.C.1.b)(3)(b)(iii)(d).
(iii) Consideration of Other Additional Variables
Some commenters expressed concern that the scrappage model
implemented in the NPRM analysis omitted several theoretically
important variables in predicting the scrappage rates of the existing
vehicle fleet. To understand these comments more fully it is useful to
recall that existing vehicle owners can be private households/
individuals, businesses, or dealerships. They supply the used vehicle
(in the sense of making it available for use) to the market either by
reselling them, or continuing to own the vehicle for their own use.
Theoretically an existing owner will supply a used vehicle for
additional use if the value of the vehicle (net of the opportunity cost
of its value as scrap metal and used parts) exceeds the cost of
maintenance, repair, insurance, and registration fees for the vehicle.
If a seller does not perform necessary repair or maintenance services
on the vehicle prior to sale, the value of the vehicle should be offset
by the cost of those services. Accordingly, the scrappage threshold for
a vehicle should remain the same regardless of whether the seller or
buyer pays for any necessary maintenance or repair services on the
vehicle.
Under this framework, commenters have argued that the agencies
should include maintenance and repair costs, the value of the used
vehicle when scrapped, and other costs to purchase the vehicle, all of
which were excluded in the NPRM version of the scrappage models. IPI
stated the following:
The agencies should include the variables that Gruenspecht and
others have traditionally included in their scrappage analysis,
including price of vehicles indexed by maintenance and repair costs,
the price of scrap metal, and interest rates.\1698\
---------------------------------------------------------------------------
\1698\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 91.
The agencies agree that these variables are relevant to determining the
scrappage rates of existing vehicles, but have concerns that the level
of aggregation of available series related to each of these factors may
obscure the ability of a statistical model to capture their impact on
vehicle scrappage rates.
[[Page 24634]]
Below, the agencies discuss commenter concerns about the omission of
maintenance and repair costs, scrap steel prices, and interest rates,
in turn. This rulemaking then outline the agencies' further
consideration of each factor in this final rule analysis, and why each
chose whether to consider each factor in the analysis for the final
rule. Empirical results of models considering these factors are shown
in Sections VI.C.1.b)(3)(c)(iii)(e) and VI.C.1.b)(3)(c)(iii)(f); the
decision to exclude them from the primary analysis is further explained
in these sections.
(a) Maintenance and Repair Costs
EDF, IPI, California States et. Al., CARB, CBD, and other
commenters suggest that the omission of maintenance and repair costs by
the agencies was not justified, and that the measure should be included
in future models. CARB claimed that:
parameters for repair costs and used vehicle prices towards the end
of life should likely be included in a scrappage model. However,
neither of these variables appear in the Agencies' model.\1699\
---------------------------------------------------------------------------
\1699\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 244.
The agencies agree that the theoretically ideal model of scrappage
would include maintenance and repair costs. For this reason, the
agencies explored several methods for explicitly incorporating
maintenance and repair costs. Section VI.C.1.b)(3)(c)(iii)(f) reports
model results both with and without a maintenance and repair variable.
Since the variable is integrated of order one, (see Table VI-158), the
models including it take the first difference; in this form, increases
in maintenance and repair costs result in an increase in the scrappage
rate of existing vehicles, as expected. The sign is also statistically
significant. While the agencies would prefer a maintenance and repair
price series that varies by calendar year and vintage, such a series is
not currently available. The agencies hope to continue to improve this
variable in future work on the scrappage model, but respond to comments
by including the first difference of the maintenance and repair series
in some of the models considered for the model used for the final rule.
Commenters were apparently confused about the agencies' discussion
of the impact of fuel economy standards on durability. The agencies
discussed a finding from the Greenspan and Cohen (1996) paper that
suggested that higher EPA emission standards actually decreased the
durability of certain model years. The discussion from the PRIA
follows:
In addition to allowing new vehicle prices to affect cyclical
vehicle scrappage [agrave] la the Gruenspecht effect, Greenspan &
Cohen also note that engineering scrappage seems to increase where
EPA emission standards also increase; as more costs goes towards
compliance technologies, it becomes more expensive to maintain and
repair more complicated parts, and scrappage increases. In this way,
Greenspan and Cohen identify two ways that fuel economy standards
could affect vehicle scrappage--(1) through increasing new vehicle
prices, thereby increasing used vehicle prices, and finally,
reducing on-road vehicle scrappage, and (2) by shifting resources
towards fuel-saving technologies--potentially reducing the
durability of new vehicles by making them more complex.\1700\
---------------------------------------------------------------------------
\1700\ PRIA at 1000.
EDF and IPI misinterpret the agencies' discussion of findings from
Greenspan and Cohen's work to imply that the fuel efficiency variable
is meant to control for changes in maintenance and repair costs. The
---------------------------------------------------------------------------
following quote from IPI exemplifies their confusion:
In addition, the agencies have explicitly excluded several
theoretically important explanatory variables (e.g., the cost of
maintenance and repair), which are potentially correlated with fuel
efficiency. [Footnote 405: Id. at 1000 (indirectly making this point
with respect to fuel efficiency and maintenance and repair costs
when emphasizing that `Greenspan & Cohen also note that engineering
scrappage seems to increase where EPA emission standards also
increase; as more costs goes towards compliance technologies, it
becomes more expensive to maintain and repair more complicated
parts, and scrappage increases'). In other words, maintenance and
repair costs are correlated with respect to fuel efficiency and
scrappage rates.]\1701\
---------------------------------------------------------------------------
\1701\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 78.
The agencies did not mean to imply that including some measure of the
fuel economy of a model year cohort (cost per mile, in the NPRM model)
would control for variation in maintenance and repair costs over time.
The discussion of Greenspan and Cohen's results was intended only to
demonstrate that durability and standards that increase technological
complexity may be correlated, so that durability increases may not be
independent of CAFE/CO2 standards.
Maintenance and repair costs for a given model year cohort likely
are correlated with the fuel saving technologies applied to that
cohort, but there is also a dimension of maintenance and repair costs
that are correlated with other macroeconomic factors (i.e., wages,
materials, etc.). Controlling for fuel economy would not capture
calendar-year-specific changes to maintenance and repair costs that are
caused by factors other than fuel economy. It also does not seem likely
that variation in maintenance and repair costs from different fuel
savings technology would be linearly related to fuel consumption, so
that even model year variation in maintenance and repair costs could
not be captured by including some measure of fuel economy or fuel
consumption. As noted above, the agencies agree that maintenance and
repair prices exist in the theoretically ideal scrappage model, and
consider the variable in some of the models presented in Section
VI.C.1.b)(3)(c)(iii)(f).
(b) Scrap Values
In the NPRM model, the agencies considered inclusion of the BLS
scrap steel CPI series. The agencies gave the following reasons for
excluding the measure in the final NPRM models in the PRIA:
As noted by Parks (1977), the value of a scrapped vehicle can be
derived either from the value of recoverable scrap metal or from the
value of sellable used parts. There are several issues with using
the BLS scrap steel CPI. First, as in Park's work, the coefficient
on scrap steel is statistically insignificant--model results
including the CPI of scrap steel are not shown, as there were other
theoretical problems with the measure. The material composition and
mass of vehicles has changed over time so that the absolute amount
of recoverable scrap steel is not constant over the series. The
average weight of recoverable steel by vintage would have to be
known, and this measure would still be missing any other recoverable
metals and other materials. Further, projecting the future value of
the recoverable scrap metal would involve computing the amount of
recoverable steel under all scenarios of fuel economy standards,
where mass and material composition are assumed to vary across all
alternatives. This value is not calculated explicitly in the current
model, which is another reason some estimate of the value of
recoverable metal is not included in the preferred model
specification.\1702\
---------------------------------------------------------------------------
\1702\ PRIA at 1012.
The concerns the agencies raised in the NPRM continue to be present for
the model used for the final rule. The BLS scrap steel CPI will not
have the same effect on the opportunity cost (the scrap value) of
keeping an existing vehicle on the road as opposed to scrapping it for
successive model year cohorts. The average weight of vehicles has
changed over successive model years, as has the average steel
composition.
Even considering the limitation of using the BLS scrap steel price
series, commenters expressed concern about the exclusion of a variable
to capture changes in the value of a vehicle as
[[Page 24635]]
scrapped metal and/or used vehicle parts. As noted in Section
VI.C.1.b)(3)(b)(iii)(a), IPI suggested that ``the price of scrap
metal'' should be included, while CARB suggested the model include
``used vehicle prices towards the end of life.'' The agencies made
several further attempts to capture this component of vehicle
scrappage, and address commenters' concerns, in the scrappage models
used in the final rule. The agencies continue to consider models which
include the BLS iron and scrap steel CPI series; results of these
considerations are shown in Section VI.C.1.b)(3)(c)(iii)(f).
(c) Interest Rates
IPI and EDF expressed concerns that changes in the real interest
rates of vehicle loans had not been included in the final NPRM
scrappage model. EDF commented the following:
NHTSA's model also does not include interest rates or the cost
of financing a vehicle, another variable which NHTSA acknowledges
affects scrappage. NHTSA itself states that ``[a]s the real interest
rate increases so does the cost of borrowing and the opportunity
cost of not investing. For this reason, it is expected that as real
interest rates increase that vehicle scrappage should decline.
Consumers delay purchasing new vehicles because the cost of
financing increases. Conversely, as real interest rates decrease,
vehicle scrappage should increase . . . . Yet, NHTSA chooses not to
include interest rates in its model since inclusion of interest
rates yields results that are opposite to what is expected--``as
real interest rates increase, so does the scrappage rate'' in
NHTSA's model. As discussed above, this is yet another indication
that the model is flawed and cannot be relied upon.\1703\
---------------------------------------------------------------------------
\1703\ EDF, Appendix A, NHTSA-2018-0067-12108, at 41.
The agencies considered real interest rates in the NPRM analysis.
Increasing the cost of purchasing a vehicle should increase the
incentive for households to hold onto existing vehicles (as opposed to
purchasing one) and scrappage rates should decline. The agencies
excluded real interest rates from the final NPRM model for the reasons
---------------------------------------------------------------------------
stated in the PRIA:
Table 8-14, Table 8-15, and Table 8-16 include interest rates
and maintenance and repair CPI for cars, vans/SUVs, and pickups,
respectively. For cars, as shown in Table 8-8, real interest rate is
of the opposite sign than expected; as real interest rates increase,
so does the scrappage rate--this model is also a worse fit by
measures of AIC and BIC relative to the preferred model.\1704\
---------------------------------------------------------------------------
\1704\ PRIA at 1028.
In response to commenters' concerns, the agencies continue to
consider interest rates in the model used for the final rule, as shown
in Section VI.C.1.b)(3)(c)(iii)(e). However, interest rates only affect
scrappage rates where a household might be unable to finance the
purchase of a new or used vehicle and instead decides to maintain an
existing vehicle that would have otherwise been scrapped. The most
likely substitute for a marginal scrapped vehicle would not be a
vehicle that could be financed. Accordingly, the relationship between
interest rates and scrappage rates may be weaker than that between new
vehicle prices and scrappage rates. The most likely substitutes for new
vehicles are vehicles just off lease, and the resulting increase in
residual values will affect slightly older vehicles. Eventually, the
price of the most likely substitutes for marginally scrapped vehicles
will also increase, so that scrappage rates will also be affected.
(d) Other Vehicle Quality Adjustments
CARB and other commenters expressed concerns that the NADA series
used by the agencies in development of the NPRM scrappage model did not
make quality adjustments. CARB made the following specific comment:
By only including new vehicle prices and no other controls for
vehicle quality, the Agencies' scrappage model omits variables that
are important predictors of scrappage rates and of vehicle prices.
Prior work that has relied on new vehicle prices to estimate
scrappage rates have also included some aspects of quality
improvements, meaning considering that the vehicle is improving in
some way. For example, Greenspan and Cohen (1996) include both the
Bureau of Labor Statistics (BLS) new vehicle price index and the BLS
cost of repair index.\1705\
---------------------------------------------------------------------------
\1705\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 244.
The NADA average new vehicle transaction price does not control for
other average characteristics that may change over successive model
years. The agencies considered controlling for average body style and
model year characteristics in the scrappage model as an alternative to
including fixed effects in the model. The considered characteristics
included: Horsepower to weight, zero to sixty acceleration time, and
average curb weight. However, performing the pFtest implementation of
an F-test of goodness-of-fit, from the ``plm'' R package, suggested
that fixed effects are necessary to control for heterogeneity across
model years.\1706\ For this reason, average characteristics that are
constant over calendar years for a given model year cohort cannot be
included in the model. The agencies do present specifications that
include the ratio of new to used vehicle performance (since this has
calendar year level variation and can be included with model year fixed
effects) in Section VI.C.1.b)(3)(c)(iii)(f).
---------------------------------------------------------------------------
\1706\ Croissant, Y., Millo, G., & Tappe, K. (2019, September
7). Package `plm.' Retrieved from https://cran.r-project.org/web/packages/plm/plm.pdf.
---------------------------------------------------------------------------
(iv) Integration of Sales and/or VMT, Total Fleet Size, and Total VMT
Some commenters believe the ideal model of how CAFE/CO2
standards affect sales, scrappage, and usage would be a joint household
choice model. RFF makes the following comment:
The agencies can fix those problems by making two changes.
First, they can jointly model VMT and vehicle holdings (i.e.,
scrappage and new-vehicle purchases). The literature provides many
examples of such modeling for guidance (see citations above).
Jointly modeling these choices will make the analysis internally
consistent and will account for the fact that households do not make
scrappage and vehicle use decisions in isolation. If the model
predicts that weaker standards cause more scrappage, it will
simultaneously estimate any increase in VMT for the remaining
vehicles.\1707\
---------------------------------------------------------------------------
\1707\ RFF, Comments EPA NHTSA, NHTSA-2018-0067-11789, at 14.
The advantage of such a model is that sales, scrappage, and usage
would be jointly determined so that the impacts on scrappage is
conditional on how increased new vehicle prices affect sales and
vehicle prices, and usage is dependent on both effects. The agencies
agree that this type of model would better capture the joint nature of
the choices of which vehicles to buy, which to sell or scrap, and how
much to use each than modelling each effect separately. However, the
agencies are not aware of any national dataset that would allow sales,
scrappage and usage to be jointly predicted, nor are they confident of
such a model's ability to predict better than carrying current market
shares forward.
The papers cited in the RFF comment, Linn and X. Dou, 2018; \1708\
Berry, Levinsohn, and Pakes, 1995; \1709\ and Jacobsen and van Bentham,
2015,\1710\ either use the CEX or the NADA transaction price series
merged with the Polk registration counts. The CEX is a relatively small
sample of households (about 160,000), their vehicle holdings,
[[Page 24636]]
vehicle purchases, and usage. However, it does not report retirement
rates, but only when a vehicle exits a household's fleet (most often it
is sold or traded in). Thus, at best, the CEX could be used to build a
household consumer vehicle holdings and usage model, but the vehicles
that are scrapped would be implied; scrappage would not be modeled
directly, nor would it be attached to the number of miles on a vehicle.
The NADA and Polk datasets used by Jacobsen and van Bentham links
vehicles prices and scrappage rates, but does not track individual
household decisions. The Jacobsen and van Bentham paper relies instead
on a model of the new and used vehicle market which takes cross-price
elasticities as an assumption derived from the outputs of a 1997 GM
consumer choice model.1728 1711 The agencies will continue
investigating whether a consumer/household choice model can serve as an
alternative to aggregate estimates of sales and scrappage, but are
skeptical about the ability of such models to predict future model
shares accurately.
---------------------------------------------------------------------------
\1708\ J. Linn and X. Dou, ``How Do US Passenger Vehicle Fuel
Economy Standards Affect Purchases of New and Used Vehicles?''
(Washington, DC: Resources for the Future, 2018).
\1709\ Berry, S., J. Levinsohn, and A. Pakes, ``Differentiated
Product Demand Systems from a Combination of Micro and Macro Data:
The New Car Market,'' Journal of Political Economy 112(1) (2004):
68-105.
\1710\ M. Jacobsen and A. van Benthem, ``Vehicle Scrappage and
Gasoline Policy,'' American Economic Review 105 (2015): 1312-38.
\1711\ Kleit, Andrew N., 2004. ``Impacts of Long-Range Increases
in the Corporate Average Fuel Economy (CAFE) Standard.'' Economic
Inquiry 42:279-94.
---------------------------------------------------------------------------
As was the case with the 2012 final rule and the 2016 TAR, the
agencies again note there is no credible consumer choice model which
can be implemented in the CAFE model. Literature comparing the
performance of consumer choice models to holding manufacturers constant
suggest that the latter predicts future market shares better than the
former. NCAT raises this point in their comment below:
Academic and other researchers have developed a number of
vehicle demand (consumer choice) models for the new and/or used
vehicle markets to look at effects on sales and fleet mix. Rarely
has there been any effort to validate these models, either for
consistency across models, or for ability to predict out of sample.
Recent academic research, as well as work by EPA, has found that
these models commonly perform worse, especially in the short run,
than simply holding market shares constant.\1712\
---------------------------------------------------------------------------
\1712\ NCAT, NCAT Comments, NHTSA-2018-0067-11969, at 11.
For these reasons, the agencies have not used a consumer choice model
to capture the sales and/or scrappage impacts, but have built reduced
form equations from aggregate data instead.
NCAT and CBD also refer to EPA attempts to develop a consumer
choice model in conjunction with Oak Ridge National Labs, and note that
the agencies did not use this model for the NPRM analysis. This
specific choice model, as referenced in the excerpted NCAT comment
above, has not predicted future market shares as well as projecting
current shares forward. For this reason the model was not deemed fit to
include in the policy analysis. NHTSA also worked to develop a consumer
choice model, but when implemented, the model predicted that some OEM's
would have unrealistic declines in total sales. The limitations of the
consumer choice models the agencies have considered is overlooked in
the following comments from CBD:
The sales model the agencies use is not the consumer-choice
model that EPA has been developing and refining for almost a decade.
Rather, both it and the scrappage model appear to have been
developed by NHTSA in just the last two years. Neither model has
been peer-reviewed, nor even released publicly until the publication
of this NPRM.\1713\
---------------------------------------------------------------------------
\1713\ CBD, Appendix A, NHTSA-2018-0067-12000, at 175.
The agencies did not use the consumer choice models either agency
developed because the predictions are not reliable--which has
disappointed not only the commenters mentioned above, but the agencies
and researchers who have spent significant resources attempting to
develop models for these purposes. Instead, the agencies have modelled
the effects from reduced form equations from aggregate data.
(a) Integration With Sales Model
The NPRM models did not include any direct linkage between the
sales, scrappage, and usage functions, as noted by the agencies. Here,
the agencies consider comments from stakeholders about the lack of
integration of the scrappage model with sales (and the effect on total
fleet size), and the lack of integration with the vehicle usage
schedules (and the effects on total VMT).
NCAT, EDF, CBD, CARB, and other commenters argued that the sales
and scrappage models should be directly linked, and that their
independence predicts the higher fleet size and total VMT under the
augural standards. CBD makes the following statement:
The agencies now, irrationally, decouple those two effects, such
that the number of new vehicles sold (or left unsold) has no effect
on the number of vehicles scrapped. Relying on the deeply flawed
scrappage model, the agencies have predicted a massive ballooning of
fleet size under the existing standards that leads, automatically
under their model, to a massive increase in VMT. \1714\
---------------------------------------------------------------------------
\1714\ CBD, Appendix A, NHTSA-2018-0067-12000, at 185.
The agencies note that the structural model presented in Section
VI.C.1.b)(3)(b)(i)(b) demonstrates that both the equilibrium quantity
and the price of new vehicles sold are changed when the production cost
of new vehicles changes under different regulatory alternatives.
Specifically, under relaxed standards, the equilibrium price is lower
and equilibrium sales are higher than the counterfactual augural
standards. Controlling for other variables that might shift the new
vehicle supply or demand curves, either new vehicle prices or sales
could enter the used vehicle demand equation (as in the structural
model, there is a functional relationship between the two, again,
controlling for factors that shift the supply and demand curves for new
vehicles). Thus, the agencies could use either new vehicle sales or
prices to control for changes in the new vehicle equilibrium solution
in the scrappage equation. It is important to control for factors that
affect the demand for vehicles overall (business cycle conditions,
etc.). The agencies present the preferred models using either new
vehicle prices or new vehicles sales in Section
VI.C.1.b)(3)(c)(iii)(d). Since there should be a collinearity between
the two, it would be inappropriate to include both variables
simultaneously.
(b) Total Fleet Size
NCAT, EDF, CBD, CARB, UCS, IPI, California et. al., academic
commenters, and other stakeholders argue that the fleet size should not
change much with new vehicle prices. Some commenters go further to
argue that higher vehicle prices under the augural standards should
result in a smaller fleet size in the augural case relative to the
proposal. The agencies agree that the long-term impact of higher new
vehicle prices should be a slight reduction in fleet size, but do not
agree that the short-term impacts of the standards on fleet size are
obvious.
Many examples from the literature make assumptions that ensure that
the fleet size under different regulatory alternatives remain constant.
UCS cites this assumption in the original Gruenspecht works (their
emphasis):
Though the agencies cite the Gruenspecht effect for its basis
for the scrappage model, they ignore a central constraint of
Gruenspecht's work--namely, his assumption that FLEET SIZE AND TOTAL
VMT ARE INSENSITIVE TO PRICE.\1715\
---------------------------------------------------------------------------
\1715\ UCS, UCS MY2021-2026 NPRM: Technical Appendix, NHTSA-
2018-0067-12039, at 60.
Other works ensure the same conclusion with different assumptions.
Within the
[[Page 24637]]
Jacobsen and van Bentham, 2015 and Goulder et. al., 2012 framework, a
household first chooses the number of vehicles to own based on the
average price of all vehicles subject to a budget constraint. After
choosing the number of vehicles to hold, the household chooses the
specific type and age of vehicles to hold. However, for some households
the choice of how many and which vehicles to hold is not disjoint, so
that a household may choose to hold two used vehicles as a second
choice to one new vehicle. When new vehicle prices increase, under the
same budget constraint, they may choose to hold two vehicles instead of
one. If enough households make this choice, the fleet size could
slightly increase.
IPI gives a literature example of a model that does not ensure this
outcome with initial assumptions. This model directly predicted fleet
size, and not sales and scrappage. The fleet size in the CAFE model is
the result of the sales and scrappage models, and not the result of a
single of the models. Small and Van Dender, 2007 finds that higher new
vehicle prices are associated with lower total vehicle stock, as IPI
states in the quote below: \1716\
---------------------------------------------------------------------------
\1716\ Auto Alliance, Attachment 1: NERA Evaluation, NHTSA-2018-
0067-1207, at D-3.
In their 2007 study estimating the rebound effect caused by
changes in fuel efficiency, Kenneth Small and Kurt Van Dender
derived estimates of the relationship between vehicle price and
fleet size. By simultaneously estimating a system of equations for
VMT per capita, fleet size, and fuel efficiency for the United
States from 1966 to 2001, Small and Van Dender also found that an
increase in new vehicle price has a negative, statistically
significant effect on total vehicle stock.\1717\
---------------------------------------------------------------------------
\1717\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 70.
However, it is worth noting that Hymel, Small, and Van Dender in 2010
published a study finding a statistically insignificant result of the
opposite sign.\1718\ The general framework of the two papers are very
similar, so that the updated results show that the fleet size impact is
ambiguous.
---------------------------------------------------------------------------
\1718\ Hymel, Kent M. & Small, Kenneth A. & Dender, Kurt Van,
2010. ``Induced demand and rebound effects in road transport,''
Transportation Research Part B: Methodological, Elsevier, vol.
44(10), pages 1220-1241.
---------------------------------------------------------------------------
Toyota and the Automobile Alliance mentioned that NERA built sales
and scrappage models, and requested that the agencies ``review the NERA
econometric study's methodologies for adoption or to refine their own
models.'' The agencies considered the NERA scrappage model, but note
that the model merges the data for all vehicle types, so that the
scrappage relationship by age for pickups is adjusted by the same
constant for all ages. However, the agencies note that each body style
has a unique functional form with age--as evidenced in Section
VI.C.1.b)(3)(c)(iii)(c))--so that it does not seem appropriate to merge
them. Further, it does not seem likely that the elasticity of scrappage
is the same for all vehicle types.
While the agencies think there are reasons not to adopt the NERA
scrappage model as is, this suggested general approach does support
simplifying the model as further suggested in Section
VI.C.1.b)(3)(b)(i). Also, this research supports the notion that the
relative fleet size of the proposed and augural standards is not a
given. NERA's comments about their model provided:
The separate changes in new vehicle sales and changes in
scrappage rates would lead to differences in the overall fleet size
for the CAFE standard alternatives. The net effects of these two
changes did not have a substantial effect on the overall fleet
population under any of the three CAFE alternatives (never more than
0.25% change in fleet size compared to the augural standards).\1719\
---------------------------------------------------------------------------
\1719\ Auto Alliance, Attachment 1: NERA Evaluation, NHTSA-2018-
0067-1207, at D-3.HONDA.
The NERA model shows the same directional fleet impacts as the NPRM
sales and scrappage model. This lends some further support to the
notion that the fleet impacts are not as certain as some commenters
suggest.
Another empirical model predicts a larger total fleet size under
the augural standards than under the proposed standards. Comments by
David Bunch offer an extended comparison of the sales, fleet size, and
retirement rate results of the Department of Energy's National Energy
Modeling System (NEMS) model under the proposed and augural standards.
NEMS predicts fleet size from input assumptions about the size of the
on-road fleet, endogenous new vehicle sales estimates, and exogenous
assumptions about scrappage.\1720\ However, in his comments Bunch said:
---------------------------------------------------------------------------
\1720\ From page 109 of 2016 NEMS documentation ``exogenously
estimated vehicle scrappage and fleet transfer rates.'' https://www.eia.gov/outlooks/aeo/nems/documentation/archive/pdf/m070(2016).pdf.
Scrappage is an implied behavior determined by projecting total
fleet size and new vehicle sales. Through this mechanism, all else
equal, an increase in new vehicle sales would yield an increase in
scrappage.\1721\
---------------------------------------------------------------------------
\1721\ David Bunch, Bunch-UC Davis: Consumer Behavior Modeling,
at 77.
NEMS does not project total fleet size endogenously in their model
as Bunch assumes. Nor is scrappage an implied behavior determined by
fleet size and new sales projections. Instead, total fleet size is
implied from an endogenous sales model, and constant age- and body-
style-specific scrappage rates. The difference between the CAFE Model
and NEMS is that the CAFE model has both endogenous new vehicles sales
and scrappage rates--scrappage rates are not assumed to be constant for
all regulatory alternatives. Fleet size is the implied variable in both
models.
Bunch finds that the NEMS model also predicts a larger fleet size
under the augural standards than the proposed standards. Specifically,
he finds the following:
The differences are initially about 100K, increasing linearly
from 2031 from 200K to 1.8M in 2050. Because even the Existing
standards remain at the same level after 2025, this would seem to
represent a very different effect from what might be going on in the
CAFE model results.\1722\
---------------------------------------------------------------------------
\1722\ David Bunch, Bunch-UC Davis: Consumer Behavior Modeling,
at 69.
Bunch goes on to discuss the relationship between sales, scrappage
---------------------------------------------------------------------------
and fleet size in NEMS in the following passage:
New vehicle sales generally are growing in both scenarios, so
economic theory suggests that fleet sizes should also be growing
(they are). Specifically, although the Gruenspecht effect logic
suggests that increasing new vehicle sales should lead to increased
used vehicle scrap rates, the total ``value'' of the fleet is
increasing, so this would suggest an increase in the fleet size.
Moreover, new vehicle sales are higher under Existing, so the fleet
size should be also.\1723\
---------------------------------------------------------------------------
\1723\ David Bunch, Bunch-UC Davis: Consumer Behavior Modeling,
at 71.
Bunch makes several claims that are not consistent with available
data and the agencies' understanding of how the NEMS model. First, he
states that because sales are growing fleet size should also be
growing. However, change in fleet size is the result of new vehicle
sales less the number of existing vehicles scrapped; if new vehicle
sales and used vehicle scrappage rates both increase, the fleet size is
not necessarily increasing. Second, he states that the `Gruenspecht
effect logic' suggests that increasing new vehicle sales results in
increasing scrappage rates. However the NEMS model does not change
vintage-specific scrappage rates endogenously, but takes them as an
exogenous input. Thus, the NEMS model does not capture the Gruenspecht
effect, and its fleet size projections can only vary from changes in
new vehicle sales. Any differences in the projected total fleet
scrappage rates Bunch considers later are due to
[[Page 24638]]
different initial sales of each body style, and therefore a different
weighting of the constant body-style- and vintage-specific scrappage
rates. This makes the comparison of the fleet size and scrappage rates
of the two models not particularly meaningful. However, the difference
in the projected sales impacts are worth a second glance. NEMS predicts
prices that are at most about $1,000 higher in the Augural than the
proposed standards, while the CAFE model predicts prices that are up to
approximately $2,500 higher. The difference in the projected costs to
meet the CAFE standards is likely the main reason for the difference in
the sales outcomes--if the average fuel savings exceed the average
incremental cost of the augural standards (relative to the proposal) in
the NEMS model, the expected outcome is that sales should be higher in
the augural case, as shown.
It is also worth noting Bunch's discussion of the empirical results
of the CAFE scrappage model. Bunch purports to calculate the scrappage
elasticity relative to new vehicle price increases, but his point of
comparison does not hold constant other factors that might impact used
vehicle scrappage rates. Instead, Bunch calculates the inter-annual
percentage change in the scrappage rates for each regulatory
alternative, then calculates the inter-annual change in new vehicle
prices for each regulatory alternative, and finally takes the quotient.
However, for inter-annual changes in scrappage rates, different
projected GDP growth rates and fuel prices will have also played a
critical role in the scrappage rates. The better point of comparison
would be the incremental percentage decrease in scrappage rates for the
augural standard relative to the proposal, over the incremental
percentage increase in new vehicle price in the augural standard
relative to the proposal for each calendar year. This ensures that the
point of comparison holds constant all other factors that determine
scrappage, as the regulatory alternatives use the same GDP growth rate
and fuel price projections. When computing the implied scrappage
elasticity in this way, the implied elasticities vary between
approximates -0.1 and -1.1, with the average being approximately -0.5--
which is more in line with what Bunch determines reasonable for his
incorrect calculations of the NEMS model scrappage elasticities, as
cited below:
Finally, the average values are -0.90 and -0.88 for the Existing
and Rollback scenarios, respectively. On one hand, these are
reasonably close to the Jacobsen and van Benthem (2015) estimate for
scrap elasticity with respect to used vehicle prices. On the other
hand, the Bento et al. (2018) estimate was -0.4, and one might
expect the elasticity with respect to new vehicle price to be
smaller. In any case, these results are not unreasonable.\1724\
---------------------------------------------------------------------------
\1724\ David Bunch, Bunch-UC Davis: Consumer Behavior Modeling,
at 79.
The implied elasticities from the NEMS model are approximately zero,
which is not a surprise since these are merely the result of different
new vehicle sales affecting the relative weighting of NEMS' constant
age-specific scrappage rates. Figure VI-66, below, shows a comparison
of fleet sizes under the baseline, preferred alternative, and AEO 2019.
The agencies see that, as commenters believed likely, the fleet size
under the preferred alternative (where sales are larger in many years
and scrappage rates higher) is eventually larger than in the baseline.
However, those differences are minimal in the early years of the
simulation where policy differences produce only small differences in
sales and scrappage. Furthermore, the agencies see that the magnitudes
of the fleet sizes in today's rule are generally similar to those
produced by the AEO 2019 model. NEMS tends to produce growth that is
more linear, leading to slightly smaller fleet sizes than those
simulated by the CAFE Model through the 2030's and slightly larger
fleet sizes through the 2040's. However, these differences are at most
three percent of fleet size, and typically closer to one or two
percent.
[[Page 24639]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.323
As discussed above, commenters offered NERA's model and NEMS as
points of comparison for NHTSA's sales and scrappage models and their
combined implied fleet size. However, since NEMS does not model the
scrappage effect, but takes static scrappage rates, it is not a fair
point of comparison. NERA's model shows a larger fleet under the
Augural standards, providing evidence that the impacts of the sales and
scrappage models are ambiguous.
(c) Integration With VMT
In the NPRM the agencies noted that the average VMT by age is
constant regardless of instantaneous or cumulative scrappage rates. The
agencies noted that this was a limitation of the model, and sought
comment on ways to integrate the two effects:
[O]ur scrappage model assumes that the average VMT for a vehicle
of a particular vintage is fixed--that is, aside from rebound
effects, vehicles of a particular vintage drive the same amount
annually, regardless of changes to the average expected lifetimes.
The agencies seek comment on ways to further integrate the survival
and mileage accumulation schedules.\1725\
---------------------------------------------------------------------------
\1725\ EDF, Appendix B, NHTSA-2018-0067-12108, at 51.
Several commenters suggest that the lack of integration between VMT and
scrappage rates is not justified. Some commenters suggested that the
VMT should be determined from a household holdings model, while others
suggested merely that delayed scrappage under higher standards should
increase average mileage accumulation, which will have some feedback
for the next year's scrappage rates.
Joshua Linn and other commenters suggest that VMT is determined at
the household level and should thus be modelled as such. EDF makes the
following comment, which seems to reflect a fundamental
misunderstanding of the type of model used to predict the scrappage
effect:
When describing the process whereby a potential new vehicle
purchaser chooses to forego buying a new vehicle and continues to
drive their existing vehicle, NHTSA's scrappage model ignores the
fact that this action shifts VMT from a new vehicle with a higher
average mileage per year to a used vehicle with a lower average
mileage. Either the driver of this vehicle will drive their older
vehicle less, causing overall VMT to decline, or the average mileage
of the used vehicle will increase without any need to affect
scrappage. By focusing solely on scrappage, and focusing the change
in scrappage on those vehicles with the worst fuel economy (i.e.,
the oldest vehicles), NHTSA essentially shifts new vehicle VMT to
the oldest vehicles. According to NHTSA's own rationale, much of the
lost VMT from new vehicles will be replaced by vehicles only a few
years old. The VMT of these relatively new used vehicles which is
then replaced by VMT from older used vehicles, and so on.\1726\
---------------------------------------------------------------------------
\1726\ EDF, Appendix B, NHTSA-2018-0067-12108, at 51.
The agencies' scrappage model does not capture household choices,
but uses aggregate data to predict new vehicle sales and age-specific
scrappage rates in response to changes in new vehicle prices. In
addition, the scrappage rates of all ages change in response to
increases in new vehicle prices, not just the oldest vehicles. Further,
the household that does not buy a new vehicle but holds onto an
existing vehicle instead, in EDF's example, results in one fewer used
vehicle supplied to the used market--this will result in an increased
price for used vehicles and potentially lead to some used vehicles not
being scrapped. Because the VMT schedules the agencies use in modelling
show usage declining with age, the agencies' model does assume that
younger vehicles that are not scrapped are driven more than older
vehicles that are not scrapped.
[[Page 24640]]
EDF, IPI, and Honda further argue that mileage accumulation should
not be constant under all scrappage rates. Specifically, they suggest
that the assumption that average VMT accumulation by age is constant
even when scrappage rates decline, results in an overestimate of VMT.
IPI suggests that the marginally unscrapped vehicles should drag down
the average VMT accumulation under higher standards in the following
comment:
Because those schedules assume each vehicle of a certain age and
type in the fleet drives a set amount of miles without any
adjustment for the increase in total fleet size or vehicle quality
(i.e., wear and tear and durability), the finding that the standards
cause the fleet size to increase results in a significant increase
in total VMT.\1727\
---------------------------------------------------------------------------
\1727\ IPI, Policy Integrity Comments: NHTSA Final--Appendix,
NHTSA-2018-0067-12213, at 61.
The agencies note that mileage accumulation and scrappage are not
disjoint. A vehicle that is driven more miles is more likely to be
scrapped. However, since the National Vehicle Population Profile (NVPP)
data does not track individual vehicles, there is no obvious way to
merge individual vehicle odometer readings with those that are
scrapped. The agencies explored different data sources that could be
used to capture the joint relationship of the two effects, but
unfortunately were unable to identify a workable dataset. Furthermore,
the agencies note that while commenters could be correct about the
relationship between mileage accumulation and scrappage, they did not
provide the agencies with any empirical evidence supporting their
assertions.\1728\ In the meantime, the agencies have adjusted the final
rule analysis to conservatively assume that total demand for VMT, not
including the rebound effect, should be constant for all regulatory
alternatives, as discussed in Section VI.C.1.b)(3)(b)(iv)(d), below.
This requires that the VMT schedules are no longer constant for all
fleet sizes.
---------------------------------------------------------------------------
\1728\ EDF, Appendix B, NHTSA-2018-0067-12108, at 54.
---------------------------------------------------------------------------
(d) Total VMT
Many commenters think that total VMT, not considering rebound
miles, should be constant, regardless of the number of new vehicles
sold and used vehicles scrapped. NCAT, Global, Auto Alliance, CBD, EDF,
IPI, CARB, and Honda all make this argument. CARB makes the following
statement suggesting that even a larger fleet size should not increase
aggregate demand for VMT (again, not including rebound miles):
A change in the overall fleet size due to the Augural standards
might not in and of itself be problematic, as long as the VMT
schedules are adjusted to account for overall travel activity that
is distributed over a larger number of vehicles. However, the As-
Received version of the [scrappage] model does not adjust VMT
schedules, with the result that the additional unscrapped vehicles
inflate total VMT proportionally.\1729\
---------------------------------------------------------------------------
\1729\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 238.
The agencies agree that the aggregate demand for VMT should be roughly
constant across alternatives, and stated this in the NPRM, where the
differences in non-rebound VMT were on the order of 0.4%.
NERA's modelling efforts found similar small decreases in VMT in
regulatory alternatives where the standards are relaxed. The Alliance
stated:
Under all three scenarios, vehicle miles traveled (``VMT'')
decreases relative to the augural standards. This is due primarily
to rebound effects. Because NERA was only examining vehicles through
MY 2029, the difference in VMT between the alternatives and the
augural standards decreases over time, since fewer of the MY 2029
and earlier vehicles are on the road in those later years.\1730\
---------------------------------------------------------------------------
\1730\ Auto Alliance, Full Comment Set, NHTSA-2018-0067-12073,
at 11.
NERA's model used similar assumptions as the NPRM analysis and,
like the NPRM results, the NERA model results suggest that it is
plausible that total VMT could decline under less stringent standards.
A key assumption common to NERA's model and the NPRM analysis is that
the VMT schedules are constant under all scrappage rates. However, as
discussed in Section VI.C.1.b)(3)(b)(iv)(c), this can potentially
overestimate total VMT in the augural case, where vehicles that were
marginally scrapped in the proposal are kept on the road.
Presumably, vehicles that are scrapped in the proposal, but not in
the augural, are in more disrepair than others in the same age cohort.
As a result, these vehicles would on average be driven less, bringing
down the average usage of the entire age cohort. This effect could
alter the relative size of total VMT under the regulatory alternatives,
as Honda notes in the following comment:
According to our calculations, if the impact of lowering the
average cohort's utility is even 0.2% the augural standards would
become safer than the preferred alternative. We believe that the
agencies should consider VMT behavior change as part of an effort to
mature and refine the scrappage model.\1731\
---------------------------------------------------------------------------
\1731\ Honda, Honda Comment, NHTSA-2018-0067-11818, at 18.
As Honda suggests, a relatively small reduction in the average VMT
schedules for the more stringent regulatory alternatives could result
in a change in the direction of the safety impact. This shows the
importance of investigating the linkage between usage and scrappage
rates, but also shows that small changes to the total VMT assumptions
can have meaningful impacts on the predicted effects of the analysis.
Other commenters make similar points.
As noted above, the difference in total non-rebound VMT in the NPRM
analysis was only 0.4%. However, CBD notes that this relatively small
change in VMT across the alternatives in a single year can result in a
large number of cumulative additional miles in more stringent
regulatory alternatives:
While 0.4% sounds small, when the scrappage model's effect it is
multiplied by all the VMT that NHTSA includes in its analysis,
spanning decades, it becomes highly significant--at least 692
billion additional VMT under the CAFE standards and 894 billion
under the CO2 program, both relative to the preferred
alternative.\1732\
---------------------------------------------------------------------------
\1732\ CBD, Appendix A, NHTSA-2018-0067-12000, at 180.
Since VMT is related to many of the costs and benefits of the program,
differences in cumulative VMT of this magnitude can have meaningful
impacts on the incremental net benefit analysis. This point was implied
by comments from CBD, EDF, NCAT, EAO, and in a paper published by
academics after the issuance of the NPRM.\1733\ For this reason, the
agencies have opted to constrain total non-rebound VMT across
regulatory alternatives.
---------------------------------------------------------------------------
\1733\ Bento, Antonio M., et al. ``Flawed Analyses of U.S. Auto
Fuel Economy Standards.'' Science, vol. 362, no. 6419, 2018, pp.
1119-21., doi:10.1126/science.aav1458.
---------------------------------------------------------------------------
Such a constraint was suggested by EDF, IPI and other commenters.
EDF states the following:
A sophisticated model is not needed to correct this problem. One
only needs to adjust the VMT added by the ``scrappage model'' so
that it matches the VMT lost by the sales response model. Put
another way, used vehicles would be used to the same extent as new
vehicles since they meet the identical demand (possibly minus a
rebound effect). \1734\
---------------------------------------------------------------------------
\1734\ EDF, Appendix B, NHTSA-2018-0067-12108, at 49.
EDF goes on to suggest some potential issues with implementing this
---------------------------------------------------------------------------
constraint:
Even this adjustment would still be in favor of the proposal, as
it assumes that all the VMT lost from fewer new vehicle sales would
be replaced by used vehicle VMT. This assumes that travel is
inelastic. This is
[[Page 24641]]
clearly not the case given NHTSA's position on the rebound effect.
NHTSA must first justify the used vehicle response to any change in
new vehicle sales. Then, in the unlikely event that this can be
done, NHTSA must link the scrappage model to the sales response
model to ensure that the combination of the two models does not
increase VMT in any calendar year (and probably show a decrease, as
the overall cost of driving will have increased).\1735\
---------------------------------------------------------------------------
\1735\ EDF, Appendix B, NHTSA-2018-0067-12108, at 49.
The agencies disagree that lost new vehicle sales would impact the VMT
of the new vehicles that are sold. The agencies do, however, as EDF
notes, adjust the VMT of new vehicles to consider changes in the cost
per mile of travel. In fact, when fuel prices increase, the agencies
assume that owners of all existing vehicles drive less; the reduction
will be greater when the vehicles on the road are less efficient, which
seems consistent with what EDF suggests in the last sentence above. The
agencies have justified the scrappage effect throughout this
discussion, above.
EDF identifies another reason the agencies think a constraint on
total VMT is reasonable for purpose of the final rule analysis. The
scrappage, sales, and VMT models each have a certain amount of
uncertainty associated with it (the uncertainty of the scrappage model
is discussed in Section VI.C.1.b)(3)(b)(i)(a)), so that when the three
models are combined, the uncertainty is compounded. EDF characterizes
these results as being inconsistent with economic theory in the comment
below:
We are not aware of any economic arguments which would support
such an increase. All that can be said is that NHTSA put data from a
variety of sources through a statistical regression and never
bothered to see if the results were reasonable or consistent with
its own economic theory. \1736\
---------------------------------------------------------------------------
\1736\ EDF, Appendix B, NHTSA-2018-0067-12108, at 57.
The NPRM analysis discussed total fleet size and VMT at length; the
agencies noted that the fleet was 1.5% bigger for the augural standard
than the proposal, resulting in 0.4% additional non-rebound VMT in
CY2050.\1737\ However, given the amount of uncertainty around each of
the models, and considering that differences in total VMT can have
meaningful impacts on the cost benefit analysis, the agencies are
conservatively assuming for the final rule analysis that non-rebound
VMT is constant, to constrain the outputs derived from the combination
of the three models.
---------------------------------------------------------------------------
\1737\ FR, Vol 83, No. 165, August 24, 2018, p.43099.
---------------------------------------------------------------------------
(v) Comments on the Evaluation of Associated Costs and Benefits
(a) Presentation and Valuation of Non-Rebound Miles
IPI and EDF argued that it was inconsistent to exclude the costs
and benefits of additional rebound driving but include them for the
sales and scrappage effect. For example, EDF stated:
[W]henever a vehicle is driven an additional mile, there is
value associated with that travel. NHTSA completely ignores the
value of any additional travel which occurs due to reduced
scrappage. Including this value would not be an adequate surrogate
for the additional repair costs required to keep older vehicles on
the road. Just as NHTSA is now recognizing that rebound VMT is due
to drivers' express decision to drive more, any driving of older
vehicles in lieu of new vehicles is due to the same choice. To treat
these identical choices in 180 degree different manners is of course
manifestly arbitrary. \1738\
---------------------------------------------------------------------------
\1738\ EDF, Appendix B, NHTSA-2018-0067-12108, at 58.
The agencies agree that there is value associated with additional
miles driven. The NPRM did not directly attribute costs for the loss of
additional miles in the scrappage analysis when the fleet size shrank.
The final rule analysis addresses this issue by holding non-rebound
total VMT constant across regulatory alternatives. However, contrary to
what EDF suggests above, the cost of additional maintenance and repair
for otherwise-scrapped vehicles are not directly related to the
additional miles. The cost of additional maintenance and repair is
incurred because the value of used vehicles has increased. The increase
in value of the used vehicles should at least offset the maintenance
and repair costs.
Holding aggregate non-rebound VMT constant across alternatives
addresses IPI's and EDF's concerns that additional miles due to a
larger fleet size were not adequately valued. However, on average newer
vehicles tend to be safer, more efficient, more powerful, and more
spacious than used vehicles. Because of this, driving a newer vehicle
will be more enjoyable, and provide more utility per mile, than driving
a used vehicle. Even disregarding trends in vehicle quality, the
utility of a mile driven in a newer vehicle is on average higher than
that driven in an older vehicle because the average newer vehicles in
better condition. The regulation is responsible for the shift in the
distribution of miles driven at each vehicle age. Including the
additional safety risks and fuel costs accrued from more miles being
driven by older vehicles accounts for part of the reduction in the
utility of the average mile under more stringent standards. Quantifying
the remaining change in utility of more miles being driven by older
vehicles is currently beyond the scope of this rulemaking analysis and
will require extensive future research. The agencies do not think
excluding other sources of changes in the utility of driving
(performance, comfort, etc.) will significant change the outcome of the
analysis.
(b) Increase in Maintenance and Repair Costs and Used Vehicle Values
EDF and others also commented that the agencies should include the
value of additional maintenance and repair costs and the increase in
value for used vehicles explicitly in the cost and benefit analysis.
They state the following:
``It is important to note that NHTSA fails to account for three
large economic impacts occurring during this process.
1. The increase in value of the entire used vehicle fleet from
2017-2050. This is a windfall gain for all current vehicle owners
that is completely ignored;
2. The cost of repairing and maintaining the older vehicles
which are no longer scrapped;
3. The value of the additional driving that these vehicles
provide.
NHTSA only counts the costs related to the additional driving
performed by the non-scrapped vehicles. Again, NHTSA's decision to
only include this cost maximizes monetary costs related to the
current standards and minimizes those related to the proposal.''
\1739\
---------------------------------------------------------------------------
\1739\ EDF, Appendix B, NHTSA-2018-0067-12108, at 50.
As discussed above, in Section VI.D.1.b)(3)(a)(a), the agencies
hold the non-rebound fleetwide VMT constant to an exogenous projection
of aggregate VMT. This addresses EDF's third concern, above. Without a
model of the used vehicle market it is impossible for the agencies to
estimate the value increase of used vehicles due to a substitution
towards used vehicles when new vehicle prices increase. However, the
maintenance and repair costs should be less than or equal to the
increase in vehicle value (or the current owner would not pay to
maintain the vehicle). Not including the additional maintenance and
repair costs should at least partially offset not including the
increase in the value of used vehicles. The remaining increase in
vehicle value should be a transfer between the seller and buyer of a
used vehicle so that it should be both a cost and benefit exactly
offsetting. Thus, the total costs and benefits are understated by the
same amount, and including them
[[Page 24642]]
should not affect the reported net benefits of the rule.
(c) Scrappage Effects From MY2030 and Beyond
The NPRM analysis considered cost per mile as a continuous
variable, and new vehicle prices in discrete levels. This means that
persistently higher new vehicle prices in more stringent standards
would continue to suppress the scrappage rate of existing vehicles. It
also means that higher fuel economies in more stringent scenarios would
continue to affect the scrappage rates as well. EDF noted that the cost
and benefit accounting that considered the costs and benefits accruing
to the remaining lifetimes of MYs 1977-2029 included some of the costs
of the scrappage effect due to the higher prices of MYs beyond 2030,
but did not include the benefits of the reduced fuel economy for these
MYs. EDF proposed that the agencies consider a CY analysis instead of
the model year presented in the NPRM:
[A] 2017-50 CY analysis would include the operation of 2017-2029
MY vehicles through CY 2050. This would include the any scrappage
effects on these vehicles through 2050, consistent with the
inclusion of new 2050 MY vehicles in the analysis. Some of the
operation of all the 2017-2029 MY vehicles would be excluded from
the analysis, as these vehicles are not assumed to be scrapped in
the Volpe Model until CY 2052-2068. Such an analysis would include
the benefits over the clear majority of the operation of 2017-2029
MY vehicles compared to both the shorter calendar year analysis and
NHTSA's 1977-2029 MY analysis. It would also include the scrappage
effects caused by 2017-2050 MY vehicles through CY 2050. Any
scrappage effects would be applied to 2030-2050 MY vehicles, as well
as 2017-2029 MY vehicles.\1740\
---------------------------------------------------------------------------
\1740\ EDF, Appendix B, NHTSA-2018-0067-12108, at 22.
However, as the commenter also notes, a CY analysis would exclude some
of the lifetime costs and benefits of improving the fuel economy of MYs
impacted by the rule (MYs 2017-2029). For this reason, the agencies do
not think that a CY analysis should supplant the MY perspective shown
in the NPRM.
EDF presents an alternative to switching to a CY analysis which
would exclude the scrappage effects due to differences in the prices
and fuel efficiencies of MYs not included in the cost benefit analysis
(MY 2030 and beyond):
An alternative that keeps the model year structure of NHTSA's
1977-2029 MY analysis would be to modify it by removing any
scrappage effects occurring in 2030 CY and beyond. This analysis
would still have the disadvantage of barely including any vehicles
which reflect full compliance with the current and proposed
standards in 2025. However, it would at least remove the primary
problem with NHTSA's current MY analysis. The impact of including
the scrappage effects caused by 2030 and later MY vehicles simply
and straightforwardly increases the VMT of used vehicles under the
current standards.\1741\
---------------------------------------------------------------------------
\1741\ EDF, Appendix B, NHTSA-2018-0067-12108, at 23.
The agencies note that previous analyses have not considered the costs
and benefits of MYs beyond those which could be a response to the
change in the considered set of standards. Part of the reason for this
was that future standards are unknown, and without existing standards
in place, manufacturers may choose to shift application of fuel saving
technologies to increases in vehicle performance or safety. The CAFE
model does not currently simulate such actions, so that including MYs
too far into the future may overstate the costs and benefits of the
rule.
While the agencies disagree that excluding cost and benefits of MYs
beyond 2030 is an issue for the cost benefit analysis, the agencies
agree that allowing persistently higher prices and fuel economies of
future MYs to impact the scrappage of the on-road fleet but not
considering the costs and benefits of those MYs is inconsistent.
However, changes to the scrappage model mitigate this issue. As noted
in Section VI.C.1.b)(3)(b)(i)(c) and VI.C.1.b)(3)(b)(ii), updates to
the time series strategy and the way that new vehicle fuel economy is
modelled in the FRM scrappage model change the form of how new vehicle
prices and fuel economy enter the equation. First, addressing the
autocorrelation by taking the first difference of variables with first
order integration instead of including lags of the dependent variables
means that cost per mile variables and new vehicle prices are captured
as changes rather than in levels. This means that constant, but higher,
new vehicle prices in the augural standards will not continue to impact
the scrappage rates of existing vehicles. More specifically, higher
prices of MYs 2030 and beyond in the augural case will no longer result
in lower scrappage rates for prior MYs. Further, since new vehicle cost
per mile is no longer explicitly included, but rather the amount of
fuel savings consumers of new vehicles value at the time of purchase is
excluded from the new vehicle prices series, differences in new vehicle
fuel economies for MYs beyond 2029 will no longer impact the scrappage
rates of earlier MYs. This naturally takes care of the concern raised
by several commenters that the accounting for costs and benefits due to
changes in MYs 2030 and beyond was inconsistent due to the scrappage
model.
(c) Estimation of the FRM Scrappage Models
(i) Framing Dynamic Scrappage Models in the Literature
(a) How Fuel Economy Standards Impact Vehicle Scrappage
As noted above, any increase in price (net of the portion of
reduced fuel savings valued by consumers) will increase the expected
life of used vehicles and reduce the number of new vehicles entering
the fleet (the Gruenspecht effect). In this way, increased fuel economy
standards slow the turnover of the fleet and the entrance of any
regulated attributes tied only to new vehicles. Gruenspecht tested his
hypothesis in his 1981 dissertation using new vehicle price and other
determinants of used car prices as a reduced form to approximate used
car scrappage in response to increasing fuel economy standards.
Greenspan and Cohen (1996) offer additional foundations from which
to think about vehicle stock and scrappage. Their work identifies two
types of scrappage: Engineering scrappage and cyclical scrappage.
Engineering scrappage represents the physical wear on vehicles which
results in their being scrapped. Cyclical scrappage represents the
effects of macroeconomic conditions on the relative value of new and
used vehicles--under economic growth the demand for new vehicles
increases and the value of used vehicles declines, resulting in
increased scrappage. In addition to allowing new vehicle prices to
affect cyclical vehicle scrappage [agrave] la the Gruenspecht effect,
Greenspan and Cohen also note that engineering scrappage seemed to
increase where EPA vehicular-criteria pollutant emissions standards
also increased; as more costs went towards compliance technologies,
scrappage increased. In this way, Greenspan and Cohen identify two ways
that fuel economy standards could affect vehicle scrappage: (1) Through
increasing new vehicle prices, thereby increasing used vehicle prices,
and finally, reducing on-road vehicle scrappage, and (2) by shifting
resources towards fuel-saving technologies--potentially reducing the
durability of new vehicles.
[[Page 24643]]
(b) Aggregate vs. Atomic Data Sources in the Literature
One important distinction in literature on vehicle scrappage is
between those that use atomic vehicle data (data following specific
individual vehicles), and those that use some level of aggregated data
(data that counts the total number of vehicles of a given type). The
decision to scrap a vehicle is made on an individual vehicle basis, and
relates to the cost of maintaining a vehicle, and the value of the
vehicle both on the used car market, and as scrap metal. Generally, a
used car owner will decide to scrap a vehicle when the value of the
vehicle is less than the value of the vehicle as scrap metal, plus the
cost to maintain or repair the vehicle. In other words, the owner gets
more value from scrapping the vehicle than continuing to drive it, or
from selling it.
Recent work is able to model scrappage as an atomic decision due to
the availability of a large database of used vehicle transactions. Work
by authors including Busse, Knittel, and Zettelmeyer (2013), Sallee,
West, and Fan (2010), Alcott and Wozny (2013), and Li, Timmins, and von
Haefen (2009) consider the impact of changes in gasoline prices on used
vehicle values and scrappage rates. In turn, they consider the impact
of an increase in used vehicle values on the scrappage rate of those
vehicles. They find that increases in gasoline prices result in a
reduction in the scrappage rate of the most fuel efficient vehicles and
an increase in the scrappage rate of the least fuel efficient vehicles.
This has important implications for the validity of the average fuel
economy values linked to model years, and assumed to be constant over
the life of that model year fleet within this study. Future iterations
of such studies could further investigate the relationship between fuel
economy, vehicle usage, and scrappage, as noted in other places in this
discussion.
While the decision to scrap a vehicle is made atomically, the data
available to NHTSA on scrappage rates and variables that influence
these scrappage rates are aggregate measures. This influences the best
available methods to measure the impacts of new vehicle prices on
existing vehicle scrappage. The result is that this study models
aggregate trends in vehicle scrappage, and not the atomic decisions
that make up these trends. Many other works within the literature use
the same data source and general scrappage construct, including those
by Walker (1968), Park (1977), Greene and Chen (1981), Gruenspecht
(1981), Gruenspecht (1982), Feeney and Cardebring (1988), Greenspan and
Cohen (1996), Jacobsen and van Bentham (2015), and Bento, Roth, and
Zhuo (2016.). These works all use aggregate vehicle registration data
as the source to compute vehicle scrappage.
Walker (1968) and Bento, Roth and Zhuo (2016) use aggregate data
directly to compute the elasticity of scrappage from measures of used
vehicle prices. Walker (1968) uses the ratio of used vehicle Consumer
Price Index (CPI) to repair and maintenance CPI. Bento, Roth, and Zhuo
(2016) use used vehicle prices directly. While the direct measurement
of the elasticity of scrappage is preferable in a theoretical sense,
the CAFE model does not predict future values of used vehicles, only
future prices of new vehicles. For this reason, any model compatible
with the current CAFE model must estimate a reduced form similar to
Park (1977), Gruenspecht (1981), and Greenspan and Cohen (1996), who
use some form of new vehicle prices or the ratio of new vehicle prices
to maintenance and repair prices to impute some measure of the effect
of new vehicle prices on vehicle scrappage.
(c) Historical Trends in Vehicle Durability
Waker (1968), Park (1977), Feeney and Cardebring (1988), Hamilton
and Macauley (1999), and Bento, Ruth, and Zhuo (2016) all note that
vehicles change in durability over time. Walker (1968) simply notes a
significant distinction in expected vehicle lifetimes pre- and post-
World War I. Park (1977) discusses a `durability factor' set by the
producer for each year, so that different vintages and makes will have
varying expected lifecycles. Feeney and Cardebring (1988) show that
durability of vehicles appears to have generally increased over time
both in the U.S. and Swedish fleets using registration data from each
country. They also note that the changes in median lifetime between the
Swedish and U.S. fleet track well, with a 1.5-year lag in the U.S.
fleet. This lag is likely due to variation in how the data is
collected--the Swedish vehicle registration requires a title to
unregister a vehicle, and therefore gets immediate responses, where the
U.S. vehicle registration requires re-registration which creates a lag
in reporting further discussed in Section VI.C.1.b)(3)(c)(ii)(b).
Hamilton and Macauley (1999) argue for a clear distinction between
embodied versus disembodied impacts on vehicle longevity. They define
embodied impacts as inherent durability similar to Park's producer
supplied `durability factor' and Greenspan's `engineering scrappage'
and disembodied effects as those which are environmental, not unlike
Greenspan and Cohen's `cyclical scrappage.' They use calendar year and
vintage dummy variables to isolate the effects--concluding that the
environmental factors are greater than any pre-defined `durability
factor.' Some of their results could be due to some inflexibility of
assuming model year coefficients are constant over the life of a
vehicle, and also some correlation between the observed life of the
later model years of their sample and the `stagflation' \1742\ of the
1970's. Bento, Ruth, and Zhuo (2016) find that the average vehicle
lifetime has increased 27 percent from 1969 to 2014 by sub-setting
their data into three model year cohorts. To implement these findings
in the scrappage model incorporated into the CAFE model, this study
takes pains to estimate the effect of durability changes in such a way
that the historical durability trend can be projected into the future;
for this reason, the agencies include a continuous `durability' factor
as a function of model year vintage.
---------------------------------------------------------------------------
\1742\ Continued high inflation combined with high unemployment
and slow economic growth.
---------------------------------------------------------------------------
(ii) Polk/IHS Registration Data
As in the NPRM, NHTSA uses proprietary data on the registered
vehicle population from IHS/Polk for the scrappage models. IHS/Polk has
annual snapshots of registered vehicles counts beginning in calendar
year (CY) 1975 and continuing until CY2017. Notably, the data
collection procedure changed in CY2002, which requires some special
consideration (discussed below). The data includes the following
regulatory classes as defined by NHTSA: Passenger cars, light trucks
(classes 1 and 2a), and medium and heavy-duty trucks (classes 2b and
3). Polk separates these vehicles into another classification scheme:
cars and trucks. Under their schema, pickups, vans, and SUVs are
treated as trucks, and all other body styles are included as cars. In
order to build scrappage models to support the model year (MY) 2021-
2026 light duty vehicle (LDV) standards, it was important to separate
these vehicle types in a way compatible with the existing CAFE model.
(a) Choice of Aggregation Level: Body Style
Two compatible methods existed by which the agencies could
aggregate scrappage rates: By regulatory class or by body style. Since,
for CAFE
[[Page 24644]]
purposes, vans/SUVs are sometimes classified as passenger cars and
sometimes as light trucks (depending upon vehicle-specific attributes)
and there was no simple way to reclassify some SUVs as passenger cars
within the Polk dataset, the agencies chose to aggregate survival
schedules by body style. This approach is also preferable because it is
consistent with the level of aggregation of the VMT schedules. Since
usage and scrappage rates are not independent of each other, if average
usage rates are meaningfully different at the level of body style, it
is likely that scrappage rates are as well.
Once stratified into body style level buckets, the data can be
aggregated into population counts by vintage and age. These counts
represent the population of vehicles of a given body style and vintage
in each calendar year. The difference between the counts of a given
vintage and vehicle type from one calendar year to the next is assumed
to represent the number of vehicles of that vintage and type scrapped
in each year.
(b) Greenspan and Cohen Correction
One issue with using snapshots of registration databases as the
basis for computing scrappage rates is that vehicles are not removed
from registration databases until the last valid registration expires--
for example, if registrations are valid for a year, vehicles will still
appear to be registered in the calendar year in which they are
scrapped. To correct for the scrappage that occurs during a calendar
year, a similar correction as that in Greenspan and Cohen (1996) is
applied to the Polk dataset. It is assumed that the real on-road count
of vehicles of a given MY registered in a given CY is best represented
by the Polk count of the vehicles of that model year in the succeeding
calendar year (PolkCY+1). For example, the vehicles scrapped
between CY2000 and CY2001 will still remain in the Polk snapshot from
CY2000 (PolkCY2000), as they will have been registered at
some point in that calendar year, and therefore exist in the database.
Using a simplifying assumption that all States have annual registration
requirements,\1743\ vehicles scrapped between July 1st, 1999 and July
1st, 2000 will not have renewed registration between July 1st, 2000 and
July 1st, 2001, and will not show up in PolkCY2001. The
vehicles scrapped during CY2000 are therefore represented by the
difference in count from the CY2000 and CY2001 Polk datasets:
PolkCY2001-PolkCY2000.
---------------------------------------------------------------------------
\1743\ In future analysis, it may be possible to work with
State-level information and incorporate State-specific registration
requirements in the calculation of scrappage, but this correction is
beyond the initial scope of this rulemaking analysis. Such an
approach would be extraordinarily complicated as States can have
very different registration schemes, and, further, the approach
would also require estimates of the interstate and international
migration of registered vehicles.
---------------------------------------------------------------------------
For new vehicles (vehicles where MY is greater than or equal to
CY), the count of vehicles will be smaller than the count in the
following year--not all of the model year cohort will have been sold
and registered. For these new model years, Greenspan and Cohen assume
that the Polk counts will capture all vehicles which were present in
the given calendar year and that approximately one percent of those
vehicles will be scrapped during the year. Importantly, this analysis
begins modeling the scrappage of a given model year cohort in: CY =
MY+2,\1744\ so that the adjustment to new vehicles is not relevant in
the modeling because it only considers scrappage after the point where
the on-road count of a given MY vintage has reached its maximum.
---------------------------------------------------------------------------
\1744\ Calculating scrappage could begin at CY=MY+1, as for most
model year the vast majority of the fleet will have been sold by
July 1st of the succeeding CY, but for some exceptional model years,
the maximum count of vehicles for a vintage in the Polk data set
occurs at age 2.
---------------------------------------------------------------------------
(c) Polk Data Collection Changes
Prior to calendar year 2002, Polk vehicle registration data was
collected as a single snapshot on July 1st of every calendar year. All
vehicles that are in the registration database at that date are
included in the dataset. For calendar years 2002 and later, Polk
changed the timing of the data collection process to December 31st of
the calendar year. In addition to changing the timing of the data
collection, Polk updated the process to a rolling sample. That is, they
consider information from other data sources to remove vehicles from
the database that have been totaled in crashes before December 31st,
but may still be active in State registration records.
The switch to a partially rolling dataset will mean that some of
the vehicles scrapped in a calendar year will not appear in the dataset
and their scrappage will wrongly be attributed to the year prior to
when the vehicle is scrapped. While this is less than ideal, these
records represent only some of the vehicles scrapped during crashes and
scrappage rates due to crashes should be relatively constant over the
2001 to 2002-time period. For these reasons, the agencies expect the
potential bias from the switch to a partially rolling dataset to be
limited. Thus, the Greenspan and Cohen adjustment applied does not
change for the dataset complied from Polk's new collection procedures.
As indicated in Figure VI-67, the scrappage counts computed from the
old Polk snapshot series represent vehicles scrapped between July 1st
of a given calendar year and the succeeding July 1st, and is computed
for CY1976-2000. The new Polk snapshot series represents vehicles
scrapped between December 31st of a given calendar year and the
succeeding calendar year, and is computed for CY2002-2016.
[[Page 24645]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.324
There is a discontinuity between the old and new methods so that
the computed scrappage for calendar year 2001 represents the difference
between the vehicle count reported in PolkCY2002 and
PolkCY2001. PolkCY2001 represents all vehicles on
the road as of July 1st, 2000, and PolkCY2002 represents all vehicles
on the road as of December 31, 2001. For this one timespan, the
scrappage will represent vehicles scrapped over a 17-month time period,
rather than a year. For this reason, the CY2001 scrappage data point is
dropped, and because of the difference in the time period of vehicles
scrapped under the old and new collection schemes, an indicator for
scrappage measured before and after CY2001 was considered; however,
this indicator is not statistically significant, and is dropped from
the preferred model.
(d) Updated FRM Dataset
As noted in section II.A.1, some commenters expressed concern about
the inability of the scrappage model to predict the scrappage rates of
vehicles over age 20. The inability was in large part due to the
limited data on the scrappage rates of older vehicles. NHTSA has worked
with Polk/IHS to construct some of the historical registration
databases using the new methodology for the purposes of other research.
As a result, the agency has registration data using both Polk
collection methods for CY's 2001-2012. Importantly, the old Polk
dataset censored data on older vehicles, with CY's 1975-1993 including
vehicles ages 0-15 and each successive CY past 1993 adding one
additional age to the dataset--so that by 2000 ages 0-22 are included.
The new datasets do not censor data on older vehicles, giving these
datasets an advantage over the old datasets--for this reason, NHTSA
uses as many years of the new data as is available.
The NPRM analysis also used all of the available data using the new
methodology at the time of publication (CY's 2005-2015). Since the NPRM
was published, NHTSA has gained access to registration data using
Polk's new methodology for CY's 2002-2005 and CY's 2016-2017. Table VI-
158 shows the calendars years of data in the NPRM and the final rule
datasets by age, as well as the total number of data points for each
age. There are a total of 330 and 420 data points for ages over 15 in
the NPRM and final rule datasets, respectively. That represents almost
a 30 percent increase in the number of data points for vehicles over
15, and a 50 percent increase in the number of data points for the
oldest vehicles considered in the dataset (ages 27-39). This additional
data on older vehicles allows the new scrappage models to better
predict the survival rates of older vehicles than the NPRM models.
[[Page 24646]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.325
(e) Models of the Gruenspecht Effect Used in Other Policy
Considerations
This is not the first estimation of the `Gruenspecht Effect' for
rulemaking policy considerations. In their Technical Support Document
(TSD) for its 2004 proposal to reduce emissions from motor vehicles,
CARB outlined how they utilized the CARBITS vehicle transaction choice
model in an attempt to capture the effect of increasing new vehicle
prices on vehicle replacement rates. They considered data from the
National Personal Transportation Survey (NPTS) as a source of revealed
preferences and a University of California (UC) study as a source of
stated preferences for the purchase and sale of household fleets under
different prices and attributes (including fuel economy) of new
vehicles.
The transaction choice model represents the addition and deletion
of a vehicle from a household fleet within a short period of time as a
``replacement'' of a vehicle, rather than as two separate actions.
CARB's final data set consists of 790 vehicle replacements, 292
additions, and 213 deletions; they do not include the deletions, but
assume any vehicle over 19 years old that is sold is scrapped. This
allowed CARB to capture a slowing of vehicle replacement under higher
new vehicle prices. That said, because their model does not include
deletions, it does not explicitly model vehicle scrappage, but assumes
all vehicles aged 20 and older are scrapped rather than resold. CARB
calibrated the model so that the overall fleet size is benchmarked to
Emissions FACtors (EMFAC) fleet predictions for the starting year; the
simulation then produced estimates that match the EMFAC predictions
without further calibration.
The CARB study captures the effect on new vehicle prices on the
fleet replacement rates, and offers some precedence for including an
estimate of the Gruenspecht Effect. However, because vehicles that
exited the fleet without replacement were excluded, the agencies do not
learn the effect of new vehicle prices on scrappage rates where the
scrapped vehicle is not replaced. New and used vehicles are
substitutes, and therefore the agencies expect used vehicle prices to
increase with new vehicle prices. And because higher used vehicle
prices will lower the number of vehicles whose cost of maintenance is
higher than their value, the agencies expect the replacements of used
vehicles to slow, but the agencies also expect that some vehicles that
would have been scrapped without replacement under lower new vehicle
prices will now remain on the road because their value will have
increased. The agencies' aggregate measures of the Gruenspecht effect
includes changes to scrappage rates both from slower replacement rates,
and from slower non-replacement scrappage rates.
(f) Car Allowance Rebate System (`Cash for Clunkers')
On June 14, 2009, the Car Allowance Rebate System (CARS) became
law, with the intent to stimulate the economy through automobile sales
and accelerate the retirement of older, less fuel efficient and less
safe vehicles. The program offered a $3,500 to $4,500 rebate for
vehicles traded-in for the purchase of a new vehicle. Vehicles were
subject to several program eligibility criteria: First, the vehicle had
to be drivable and continuously registered and insured by the same
owner for at least one year; second, the vehicle had to be less than 25
years old; third, the MSRP had to be less than $45,000; and finally,
the new vehicle purchased had to be more efficient than the trade-in
vehicle by a specified margin. The fuel economy improvement
requirements by body style for specific rebates are presented in Table
VI-159.
[[Page 24647]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.326
The program was originally budgeted for $1 billion dollars and to
end on November 1, 2009, but that amount was spent far more quickly
than expected and the program received an additional $1.85 billion in
funding. Even with that additional funding, the program only lasted
through August 25, 2009, expending $2.85 billion on 678,359 eligible
transactions. To ensure that the replaced vehicles did not remain on
the road, the vehicles were scrapped at the point of trade-in by
destroying the engine. While the program resulted in the replacement of
more vehicles and at a faster rate than expected, critics have argued
that many of the trade-ins would have happened even if the program had
not been in place, so that any economic stimulus to the automobile
industry during the crisis cannot be attributable to the CARS program.
Further, others have argued that forcing the scrappage of vehicles that
could still remain on the road has negative environmental impacts that
could outweigh any environmental benefits of the reduced fuel
consumption from the accelerated retirement of these less efficient
vehicles.
Li, Linn, and Spiller (2010) use Canada as a counterfactual example
to identify the portion of CARS trade-ins attributable to the policy,
i.e., trade-ins that would not have happened anywhere if the program
were not in place. They argue that the Canadian market is largely
similar to the U.S. market, in part based upon the fact that 13 to 14
percent of households purchased new vehicles one year pre-recession in
both countries. They also argue that the economic crisis affected the
Canadian economy in a similar manner as it affected the U.S. economy.
While they note that Canada offered a small rebate of $300 to vehicles
traded in during January, 2009, hey further note that only 60,000
vehicles were traded in under that program. Using those assumptions,
Li, et al., applied a difference-in-difference methodology to isolate
the effect of the CARS program on the scrappage of eligible vehicles.
Li, et al., found a significant increase in the scrappage only for
eligible U.S. vehicles, suggesting they isolated the effect of the
policy. They conclude that of the 678,359 trade-ins made under the
program, 370,000 of those would not have happened during July and
August 2009. They conclude that the CARS program reduced gasoline
consumption by 0.9-2.9 billion gallons, at $0.89-$2.80 per gallon
saved.
The agencies find the evidence from Li, et al., persuasive toward
the inclusion of a control for the CARS program during calendar year
2009. The importance is discussed further both in the data section,
Section VI.C.1.b)(3)(c)(ii), which provides more evidence for the
effect of the CARS program, and in the model specifications Section
VI.C.1.b)(3)(c)(iii), which describes the control used for the effect
of the program. This ensures that the measurements of other determining
factors are not biased by the exceptional scrappage observed in
calendar year 2009.
(iii) Updated Final Rule Modeling
The agencies contemplated all of the comments and suggestions made
by commenters and, in response, have made several changes to final
rule's model. First, the agencies changed the time-series strategy used
in the model, as discussed in Section VI.C.1.b)(3)(c)(iii)(a). This
change allows the agencies to simplify the models significantly,
addressing commenters' concerns about potential overfitting of the
model and difficulty of interpreting individual coefficient values
(discussed in Section VI.C.1.b)(3)(b)(i)). Second, the agencies changed
the modeling of the durability effect as discussed in Section
VI.C.1.b)(3)(c)(iii)(c); this change reduces the reliance on the decay
function and has the added benefit of addressing concerns about
overfitting and out-of-sample projections discussed in Section
VI.C.1.b)(3)(b)(i). Third, a portion of anticipated fuel savings from
increased fuel economy are netted from new vehicle prices--meaning
consumers are now assumed to value fuel economy at the time of purchase
to a certain extent--as discussed in Section VI.C.1.b)(3)(c)(iii)(d).
This change is in response to comments discussed in Section
VI.C.1.b)(3)(b)(ii) and addresses inconsistent treatment of consumer
valuation within the NPRM's analysis. Finally, the agencies consider
the inclusion of additional or alternative variables in the scrappage
model in response to comments discussed in Section
VI.C.1.b)(3)(b)(iii). After extensive testing, the agencies concluded
that these additional variables do not improve the model fits or would
introduce autocorrelation in the error structures (see Sections
VI.C.1.b)(3)(c)(iii)(e) and VI.C.1.b)(3)(c)(iii)(f) for further
discussion). As such, the agencies rejected the additional terms
suggested by commenters. Input from commenters was used to simplify the
scrappage model, make it more consistent with modeling of new vehicle
prices elsewhere in the analysis, and improve its predictions for the
instantaneous scrappage rates of vehicles beyond age 20.
(a) Changes to the Time Series Strategy
As discussed in Section VI.D.1.b)(3)(b)(i)(c), the agencies
reconsidered the time series strategy for the final rule in response to
comments. The first step in doing so is to test the time series
properties of the dependent and independent variables. The agencies use
the Augmented Dickey-Fuller (ADF) unit root test implemented in the
`CADFtest' R package to test for stationarity.\1745\ The agencies find
that the logistic scrappage rate is I(0), or stationary in levels.
Since the dependent variable is stationary, there is no long-term trend
in scrappage rates to capture. Lags of dependent variables need not be
included, but their stationary forms should be used in the regressions.
The following table summarizes the order of integration of each of the
considered regressions; the
[[Page 24648]]
regression forms represent the form of the variable that is included in
the considered models.\1746\ All the variables considered are either
I(0) or I(1), meaning that they should be run in either levels or first
differences, respectively. This significantly simplifies the
regressions. Two unintended, positive outcomes of this change in time
series strategy are that the coefficients on variables are easier to
interpret and the models are less likely to be overfit. In this way,
the shift to address concerns about the time series strategy (discussed
in Section VI.D.1.b)(3)(b)(i)(c)) also addresses commenter concerns
outlined in Section VI.D.1.b)(3)(b)(i)(a).
---------------------------------------------------------------------------
\1745\ Lupi, Claudio (2019, September 7). Package `CAFtest.'
Retrieved from https://cran.r-project.org/web/packages/CADFtest/CADFtest.pdf.
\1746\ Note: Some of these variables were considered or added in
response to comments presented in Sections I.A.1.a)(1)(b)(ii),
I.A.1.a)(1)(b)(iii), and I.A.1.a)(1)(b)(iv), and may not be present
in the NPRM.
[GRAPHIC] [TIFF OMITTED] TR30AP20.327
(b) Final Rule Preferred and Sensitivity Specifications
After consideration of comments on, and subsequent peer review of,
the NPRM analysis, the agencies updated the scrappage model
specifications for the final rule. Section VI.C.1.b)(3)(c)(iii)(a)
through VI.C.1.b)(3)(c)(iii)(f) discuss other considered specifications
and variables. The equation below represents the final
[[Page 24649]]
form of the scrappage equation included in the central and sensitivity
analysis:
[GRAPHIC] [TIFF OMITTED] TR30AP20.328
Here, ``S'' represents the instantaneous scrappage rate in a
period, so that the dependent variable is the logit form of the
scrappage rates. Logit models ensure that predicted values are
bounded--in this case between zero and one. It is not possible to scrap
more than all the remaining vehicles, nor fewer than zero percent of
them, which is illustrated in the graph below:
[GRAPHIC] [TIFF OMITTED] TR30AP20.329
[GRAPHIC] [TIFF OMITTED] TR30AP20.330
Solving for instantaneous scrappage yields the following:
[[Page 24650]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.331
In the equation above, [Sigma][beta]iXi represents the right-hand
side of the above model specification. Within the right-hand side of
the equation, Age represents the age of the model year cohort in a
specific calendar year, defined by the Greenspan and Cohen adjustment
discussed in Section VI.C.1.b)(3)(c)(ii)(b). The coefficient on the
cubic age term is assumed to be zero for the van/SUV and pickup
specifications as this term is not necessary to capture the general
scrappage trend for these body styles. Share Remaining represents the
share of the original cohort remaining at the start of the period.
These two components represent the engineering portion of scrappage--
the inherent durability of a model year and the natural life cycle of
how vehicles scrap out of a model year cohort as the cohort increases
with age. The determination of these specific forms is discussed in
detail in Section VI.C.1.b)(3)(c)(iii)(g).
New Price--FS represents the average price of new vehicles minus 30
months of fuel savings for all body styles. The central analysis
assumes the coefficient on the age interactions for this term are zero
for all body styles, but a sensitivity case allows the elasticity of
scrappage to vary with age. Fuel Price represents the real fuel prices,
weighted by fuel share of the model year cohort being scrapped. CP100M
represents the cost per 100 miles of travel for the specific body style
of the model year cohort being scrapped under the current period fuel
prices and using fuel shares for that model year cohort. These measures
capture the response of scrappage rates to new vehicle prices, fuel
savings, and to changes in fuel prices that make the used model year
cohort more or less expensive to operate. Because these measures are
all I(1), as discussed above in 0, the first difference of all of these
variables is used in modelling. The other specific modelling
considerations that resulted in this form of modelling the new and used
vehicles markets are discussed in Section VI.C.1.b)(3)(c)(iii)(d).
GDP Growth represents the GDP growth rate for the current period.
This captures the cyclical components of the macro-economy. Section
VI.C.1.b)(3)(c)(iii)(e) discusses how this specific measure was chosen,
and what other measures were considered as alternative or additional
independent variables.
CY2009 and CY2010 represent calendar year dummies for 2009 and 2010
when the CARS program was in effect; this controls for the impact of
the program. [Age = 25] represents an indicator for vehicles
25 years and older. The interaction of the calendar year dummies with
this indicator allows for the effect of the CARS program to be
different for vehicles under 25 versus vehicles 25 and older. Since
only vehicles under 25 were eligible for the program (see the
discussion of the program in Section VI.C.1.b)(3)(c)(ii)(f)), this
flexibility is important to correctly control for the program.
Finally, FE represents a set of model year fixed effects used to
control for heterogeneity across different model years. This is related
to the durability and engineering scrappage. The NPRM model did not
include fixed effects because it fit a parametric relationship to model
year as a continuous variable as a way to capture durability. This
change in how the durability effect is modelled is discussed further in
Section. Further, Section VI.C.1.b)(3)(c)(iii)(g) discusses trends in
the fixed effects and how these are projected forward within the CAFE
model.
(c) Modeling Durability Trends Over Time
As noted in the NPRM, the durability of successive model years
generally increases over time. However, this trend is not constant with
vehicle age--the instantaneous scrappage rate of vehicles is generally
lower for later vintages up to a certain age, but increases thereafter
so that the final share of vehicles remaining converges to a similar
share remaining for historically observed vintages. The NPRM
parameterized this trend by using the natural log of the model year as
a continuous variable interacted with a polynomial form of the age
variable--this predicted an increasing but diminishing trend in vehicle
durability for younger ages. The analysis for the final rule makes a
change that allows more flexibility in durability trends. Below, the
agencies consider the survival and scrappage patterns by body style.
Figure VI-69 to Figure VI-71 shows the survival and scrappage
patterns of different vintages with vehicle age for cars, SUVs/vans and
pickups, respectively. Cars have the most pronounced durability
pattern. Figure VI-69 shows that newer vintages scrap slower at first,
but that scrap more heavily so that the final share remaining of cars
is more or less constant by age 25 for all vintages.
[[Page 24651]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.332
SUVs/vans have a less pronounced durability pattern. Model year
1980 actually lives longer than model years 1985 and 1990. This is
likely due to a switch of SUVs/vans to be based on car chassis rather
than pickup chasses over time. However, through the later model years,
the durability trend is more like that of cars. The lack of a
continuous trend in durability of SUVs/vans make how this trend is
captured particularly important. Below the agencies discuss a change in
how the durability trend is modelled for the final rule, which is more
flexible than the NPRM model.
[GRAPHIC] [TIFF OMITTED] TR30AP20.333
There is no clear trend in durability for pickups. Like SUVs/vans,
this makes parameterizing by using a form of vintage as a continuous
variable problematic. Such a parametric form does not allow for each
model year to have its own durability pattern.
[[Page 24652]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.334
As noted above, the NPRM model used the natural log of model year
as a continuous variable interacted with age to capture an increasing
but diminishing trend of vehicle durability for the younger ages.
However, enforcing a parametric form on a continuous model year
excluded the possibility of including model year specific fixed effects
and required that durability have a parametric trend with successive
vintages. As seen above, SUVs/vans and pickups certainly do not follow
such a trend, so that this constraint was too restrictive, at least for
these body styles. The final rule analysis makes an adjustment that
allows for an initial increase in the durability of a model year to
persist, while including fixed effects and relaxing the parametric
assumption.
Instead of regressing the natural log of the vintage share in the
remaining models, shown in Table VI-161 through Table VI-163, the
agencies use the share remaining in the previous period as an
independent variable. Since the logistic instantaneous scrappage rate
is stationary (it is independent of the previous periods' logistic
instantaneous scrappage rate), the share remaining should not be
endogenous. The share remaining models for the final rule include model
year specific fixed effects and project a linear trend in durability by
fitting a regression through the fixed effects. This latter part still
requires a parametric assumption about durability (discussed in Section
VI.C.1.b)(3)(c)(iii)(g)), but not while jointly estimating other
coefficients. In this way, the other coefficients should not be biased
by projecting the durability trend forwards in the implementation of
the scrappage regressions within the CAFE model.
BILLING CODE 4910-59-P
[[Page 24653]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.335
[[Page 24654]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.336
[[Page 24655]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.337
BILLING CODE 4910-59-C
As Table VI-161 shows, the NPRM specification and both the constant
and the quadratic forms of the age interaction with the share remaining
variable to capture the durability effect show evidence of
autocorrelation. The linear form of the interaction of age and share
remaining does not show evidence of autocorrelation and also has the
lowest AIC and highest adjusted R-squared. For these reasons, this is
the preferred specification of the durability effect. Since the share
remaining coefficient is negative and larger than the positive
coefficient on the share remaining interacted with age, a cohort that
has a higher share remaining at an early age will have a lower
instantaneous scrappage rate in this period until a certain age and
then a higher scrappage rate after that age. To find the age where the
sign of the share remaining coefficient will switch from predicting a
lower instantaneous scrappage rate to a higher one, the agencies must
take the ratio of the coefficient on the share remaining variable to
the share remaining interacted with age--this suggests that at age 19,
the sign of the share remaining variable flips. That is, the
instantaneous scrappage rate of cars is predicted to be lower if the
share remaining is higher until age 18, after which a higher share
remaining predicts a higher instantaneous scrappage rate.
As Table VI-162 shows, the linear interaction of age and share
remaining is the only specification of the durability effect for SUVs/
vans that do not show autocorrelation in the error structure. The
linear interaction of age and share remaining has the lowest AIC and
highest R-squared; for this reason, this is the preferred specification
of the durability effect for SUVs/vans. The signs for share remaining
and share
[[Page 24656]]
remaining interacted with age show a similar trend as that to cars.
Taking the ratio again of the share remaining to the share remaining
interacted with age, for ages 0 to 18 a higher share remaining predicts
lower instantaneous scrappage, and for ages beyond 18 it predicts a
higher instantaneous scrappage rate.
As Table VI-163 shows, all but the NPRM specification of the
durability effect for pickups do not show autocorrelation in the error
structures. However, similar to cars and SUVs/vans, the linear
interaction of age and share remaining has the lowest AIC and highest
adjusted R-squared. For this reason, this is the preferred
specification for all body styles. Taking the ratio of the coefficient
on share remaining to share remaining interacted with age shows that a
higher share remaining will predict a lower instantaneous scrappage
rate in the next period for ages 0 through 14, but a higher
instantaneous scrappage rate for ages 15 and older.
Using the preferred forms of the engineering scrappage rates for
each body style as the reference point, Section VI.C.1.b)(3)(c)(iii)(d)
considers different forms to predict the Gruenspecht effect for each
body style. Section VI.C.1.b)(3)(c)(iii)(e) uses the preferred
engineering and Gruenspecht forms to consider alternative macroeconomic
variables to predict the effects of the business cycle. Finally,
Section VI.C.1.b)(3)(c)(iii)(f) uses the preferred engineering,
Gruenspecht and business cycle forms to consider the inclusion of other
additional independent variables.
(d) Modeling Impacts of New Vehicle Market on Used Scrappage Rates
Table VI-164 through Table VI-166 show the relationship between
car, SUV/van, and pickup scrappage rates and changes in new vehicle
price and fuel economies. The agencies consider two methods in response
to comments outlined in Section VI.C.1.b)(3)(b)(ii). (1) changes in
average new vehicle prices net of 30 months of fuel savings (consistent
with the technology selection and sales model) and (2) change in
average new vehicle prices, change in average fuel prices, changes in
new vehicle cost per mile and changes in new vehicle fuel consumption.
The agencies allow the elasticity of average new vehicle prices net of
30 months of fuel savings to vary by age by including interaction
terms.
BILLING CODE 4910-59-P
[[Page 24657]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.338
[[Page 24658]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.339
[[Page 24659]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.340
For all body styles, the specification of the Gruenspecht effect as
the change in new vehicle prices net of fuel savings does not show
signs of auto-correlated errors. However, for cars and vans/SUVs, the
specification which separates the effect of new vehicle prices and fuel
economy does show evidence of autocorrelation. For this reason, the
changes in new vehicle fuel prices net of fuel savings is the preferred
specification of the Gruenspecht effect.
The agencies consider the interaction of the change in average new
vehicle prices with vehicle age. This relaxes an assumption that the
elasticity of scrappage rates to change in new vehicle prices is
constant. For cars and
[[Page 24660]]
vans/SUVs the linear interaction of change to new vehicle prices net of
fuel savings show evidence of autocorrelation. The quadratic
interaction of age with change in new vehicle prices shows
autocorrelation with cars. For this reason, the agencies consider the
constant elasticity of scrappage rates to changes in new vehicle prices
to be the preferred specification (as the only specification that does
not show evidence of autocorrelation for all body styles). However, the
agencies do consider the quadratic form of the elasticity with age as a
sensitivity case (even though there is evidence of autocorrelation (but
only in the car specification)). This allows the agencies to test the
impact of relaxing the assumption around constant elasticity on CAFE
model outcomes.
(e) Considering Alternative/Additional Macroeconomic Indicators
Table VI-167 through Table VI-169 show alternative macroeconomic
indicators for cars, vans/SUVs and pickups, respectively. The agencies
consider unemployment rate and per capita personal disposable income as
alternatives to GDP growth rate to capture the cyclical component of
the macro economy. The unemployment rate and the per capita personal
disposable income are both I(1), so that the first difference of each
is the form included. For the car and van/SUV specifications, the
specifications replacing GDP growth rate show evidence of
autocorrelation in the error structures. For this reason, the GDP
growth rate is the preferred specification for the cyclical components
of instantaneous scrappage rates, as in the NPRM models.
As discussed in Section VI.D.1.b)(3)(b)(iii)(c), some commenters
were concerned with the exclusion of interest rates. In response, the
agencies considered including the change in interest rates for the
otherwise preferred specification. For vans/SUVs the model has a higher
AIC and shows evidence of autocorrelation in the error structures. For
pickups, the sign changes on the change in cost per mile when the
interest rate is included, which would be an implausible result.
Finally, the AIC for cars is nearly identical regardless as to whether
the interest rate is included. For these reasons, the agencies continue
to exclude the interest rate from the preferred specification.
[[Page 24661]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.341
[[Page 24662]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.342
[[Page 24663]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.343
[[Page 24664]]
(f) Considering Other Additional Variables
Table VI-170 through Table VI-172 show specifications that consider
additional variables not included in the preferred specifications. As
discussed in Section VI.D.1.b)(3)(b)(iii)(a), some commenters
criticized the fact that maintenance and repair costs were excluded
from the scrappage models. In response to comments, and since the
maintenance and repair costs are I(1), the agencies considered
including the difference in maintenance and repair costs. When
included, changes in maintenance and repair costs show the expected
sign--when maintenance and repair costs are higher, instantaneous
scrappage rates are predicted to be higher (as used vehicles are more
expensive to maintain). When included, the AIC is higher for the car
and van/SUV specifications. That is, including the change in
maintenance and repair costs does not improve the fit of the models.
Because of this, and because there is no obvious way to predict future
change to maintenance and repair costs (as discussed in the NPRM), the
preferred specification continues to exclude maintenance and repair
costs.
As discussed in Section VI.D.1.b)(3)(b)(iii)(b), some commenters
criticized the exclusion of steel and iron scrap prices from the
scrappage models. In response to comments, and since this variable is
also I(1), the agencies considered including the change in steel and
iron scrap prices. When included, the AIC of cars and vans/SUVs is
higher. Further, the car specification includes evidence of
autocorrelation in the error structures. In addition, there is no known
projection of steel and iron scrappage prices, so that the agencies
would have to make projections to include this variable in the
scrappage models. Accordingly, the central case continues to exclude
steel and iron scrap prices.
As discussed in Section VI.D.1.b)(3)(b)(iii)(d), some commenters
and peer reviewers suggested that controlling for aggregate measures of
model year cohorts, such as performance, might correct some unexpected
signs. The preferred specification already addresses these concerns.
Further, because fixed effects are included for model years, the
agencies cannot include aggregate model year specific attributes that
are constant over the lifetime of the cohort. The agencies do consider
the ratio of the average horsepower to weight of a model year cohort to
the new vehicle cohort, as this will change along with changes to the
horsepower to weight ratio over successive calendar years. Including
this variable results in a higher AIC for cars and vans/SUVs and shows
evidence of autocorrelation in the errors for these two body styles.
For this reason, the preferred specification excludes this metric.
The agencies also considered including new vehicles sales directly
as a predictor of instantaneous scrappage rates. Since new vehicle
sales are I(1), the difference in new vehicle sales is the included
form. Including the change in new vehicle sales results in a higher AIC
for cars and vans/SUVs. It also introduces evidence of autocorrelation
in the error structure for the car model, and reduces the effect of the
change in fuel prices by two orders of magnitude for vans/SUVs. It
seems unlikely that the magnitude of the effect of fuel prices would so
drastically vary between body styles. For these reasons, the preferred
specifications exclude the change in new vehicles sales. The agencies
also considered including changes in vehicle stock, but this similarly
did not improve the fit of the scrappage models--and doing so limited
the ability to link the sales and scrappage models as some commenters
suggested (see Sections (b)(iv)(a) and (b)(iv)(b)).
[[Page 24665]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.344
[[Page 24666]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.345
[[Page 24667]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.346
[GRAPHIC] [TIFF OMITTED] TR30AP20.347
BILLING CODE 4910-59-C
(g) Projecting Durability in the CAFE Model
The left graphs in Figure VI-72 through Figure VI-74 show the fixed
effects for the preferred scrappage specifications for cars, vans/SUVs,
and pickups, respectively. For all body styles there is a general
downward trend in the fixed effects. This suggests an increase in the
durability of successive model years. However, since the panel datasets
are not balanced, there is likely potential bias for the fixed effects
that include only certain ages. This makes projecting the durability
increase from the fixed effects a little more complicated than merely
fitting to all fixed effects. First, the agencies must determine what
part of this trend is likely due to increases in vehicle durability
(and should be projected forward) and which part of the trend may
conflate other factors.
The right graphs in Figure VI-72 through Figure VI-74 show the
average observed logistic scrappage rates by model year for all ages
where data exists. As can be seen, the average observed scrappage rates
decline dramatically for model years after 1996 for all body styles.
There are two reasons this trend exists. First, as Figure VI-72 through
Figure VI-74 show, the instantaneous scrappage rate generally follows
an inverted u-shape with respect to vehicle age. The instantaneous
scrappage rates generally peak between ages 15 and 20 for all body
styles. Model year 1996 is the first model year which will be at least
age 20 at the last date of available data (calendar year 2016). This
means that all model years newer than 1996 have likely not yet reached
the age where the instantaneous scrappage rate will be the highest for
the cohort. Accordingly, the fixed effects could be biased downwards
(consistent with the sharper downward slope in the fixed effects for
most body styles for model years beyond 1996) because of the unbalanced
nature of the panel, and not because of an actual increase in inherent
vehicle durability for those model years.
[[Page 24668]]
The second reason the average logistic scrappage rates for model
years before 1996 is more stable is because each data point in the
average has increasingly less effect on the average as more data
exists. For model years 1996 and older there are at least 18 data
points (we start the scrappage at age 2, by which point effectively all
of a model year has been sold), and each will have a smaller effect on
the average than for newer model years with fewer observations. For
these reasons, the average observed logistic scrappage rate is more
constant for model years before 1996. As a result, the agencies do not
consider the trend in fixed effects after model year 1996 to rely on
enough historical data to represent a trend in vehicle durability, as
opposed to a trend in the scrappage rate with vehicle age.
In considering which model year fixed effects should be considered
in projecting durability trends forward, another important factor is
whether there are discrete shifts in the types of vehicles that are in
the market or category of each body style over time. For cars, an
increasing market share of Japanese automakers which tend to be more
durable over time might result in fixed effects for earlier model years
being higher. This trend is shown in the fixed effects in Figure VI-72,
which follow a steeper trend before model year 1980.
[GRAPHIC] [TIFF OMITTED] TR30AP20.348
For vans/SUVs, earlier model years are more likely to be built on
truck chassis (body-on-frame construction) instead of car chassis
(unibody construction). Since pickups tend to be more durable, the
earlier fixed effects are likely to be lower for vans/SUVs for earlier
model years. The 1984 Jeep Cherokee was the first unibody construction
SUV.\1747\ As Figure VI-73 shows, the fixed effects before 1986 show
inconsistent trends; these are likely due to changes in what was
considered a van/SUV over time. For this reason, the agencies build the
trend of fixed effects from model years 1986 to 1996.
---------------------------------------------------------------------------
\1747\ https://www.autoguide.com/auto-news/2018/01/10-interesting-facts-from-the-history-of-the-jeep-cherokee.html.
---------------------------------------------------------------------------
[[Page 24669]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.349
[GRAPHIC] [TIFF OMITTED] TR30AP20.350
While the trend for pickups and cars could be extrapolated before
1986, the agencies opt to keep the fixed effects included constant for
all body styles. Thus, the projections are built from model year 1986
to model year 1996 fixed effects. Table VI-173 below, shows the linear
regressions shown as the line on the left side of Figure VI-70 through
Figure VI-72. The durability cap represents the last model year where
the durability trend is assumed to persist. The agencies cap the
durability impacts at model year 2000, as data beyond this point does
not exist for enough ages to determine if durability has continued to
increase since this point. The implication of this cap, is that model
years after 2000 are assumed to have the same initial durability as
model year 2000 vehicles. Since there is a limit to the potential
durability of vehicles, this acts as a bound on this portion of the
scrappage model.
[[Page 24670]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.351
The durability projections enter the scrappage equation in the CAFE
modelling in accordance to the following equation:
[GRAPHIC] [TIFF OMITTED] TR30AP20.352
The intercept enters as a constant added to the predicted logistic
of the instantaneous scrappage rate. The model year slope enters as the
model year for all model years older than 2000 and enters as 2000 for
all model years 2000 and newer.
Once the predicted logistic scrappage rate is calculated in the
CAFE model (including the projections of the fixed effect portion of
the equation), the future population of model year cohorts can be
predicted. The instantaneous scrappage can be calculated directly from
S. It identifies the share of remaining vehicles in each calendar year
that are scrapped in the next year. The population of vehicles in the
next calendar year can be calculated as follows:
Populations MY,CY +1 = Population MY,CY *(1 -SMY,CY).
This process is iteratively calculated at the end of the CAFE model
simulation to determine the projected population of each model year in
each future calendar year. This allows the calculation of vehicle miles
travelled, fuel usage, pollutant and CO2 emissions, and
associated costs and benefits. The CAFE model documentation released
with this final rule further details how the scrappage model is
projected within the simulations.
(d) Updates to the Decay Function
The scrappage models described above fit the historical data of car
and truck scrappage well, but when used to project the scrappage of
future model years they over-predict the remaining cars and trucks for
ages greater than 30 in an unrealistic manner. Nearly six percent of
the MY2015 van/SUV fleet and eight percent of the pickup fleet is
projected to persist until age 40. This is unrealistic, and likely due
to the fact that the agencies do not observe enough model years for
those ages and over-predict the impact of durability increases for
those ages. For this reason, the agencies are using the curves with an
accelerated decay function to predict instantaneous scrappage beyond
age 30 for pickups and SUVs/vans. The implementation and parameter
stricture of the decay function have not changed since the NPRM model.
Table VI-174, below, shows the inputs used for the final rule analysis.
[[Page 24671]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.353
The final survival rate has not changed since the NPRM, but the
input Decay age has changed. In the NPRM, the decay function was
specified to begin after age 20, while the decay function begins after
age 30 in the final rule analysis. This input change was possible
because the scrappage model for the final rule predicts shares
remaining in line with observed historical trends through age 30,
rather than through age 20. This improvement in the model fits for
older ages is driven both by the shift of the modelling of the
durability effect discussed in Section VI.D.1.b)(3)(a)(g) and the
increase in available data on the scrappage rates of older vehicles
discussed in Section VI.C.1.b)(3)(c)(ii)(d). Overall, this outcome
suggests that the final rule model predicts the scrappage rates of
older vehicle better than the NPRM model.
As in the NPRM, the decay function is implemented in the model
using the following conditions:
[GRAPHIC] [TIFF OMITTED] TR30AP20.354
Where:
t = (age + 1 - b15
And:
[GRAPHIC] [TIFF OMITTED] TR30AP20.355
Here, the population for ages beyond the start age of the decay
function depends on the population of the cohort at that start age and
the final share expected for that body style at age 40. The rate of
decay necessary to make the final population count equal that observed
in the historical data is applied.
(4) The Rebound Effect in the NPRM
The fuel economy rebound effect--a specific example of the well-
documented energy efficiency rebound effect for energy-consuming
capital goods--refers to the tendency of motor vehicles' use (as
measured by vehicle-miles traveled, or VMT) to increase when their fuel
economy is improved and, as a result, the cost per mile (CPM) of
driving declines. Amending and establishing CAFE and CO2
standards at a lower degree of stringency than the baseline level will
lead to comparatively lower fuel economy for new cars and light trucks,
thus increasing the amount of fuel consumed to travel each mile. The
resulting increase in CPM will lead to a reduction in VMT over the
lifetime of new vehicles, an example of the rebound effect working in
reverse. In the NPRM, the agencies assumed a fuel rebound effect of 20
percent, meaning that a 5 percent decrease in fuel economy would result
in a one percent decrease in the annual number of miles driven at each
age over a vehicle's lifetime.
Many of the comments received on different components of the CAFE
model can be traced back to the agencies' rebound selection. The
agencies recognize that the value selected for the rebound effect
influences overall costs and benefits associated with the regulatory
alternatives under consideration as well as the estimates of lives
saved under various regulatory alternatives, and that the rebound
estimate, along with fuel prices, technology costs, and other
analytical inputs, is part of the body of information that agency
decision-makers have considered in determining the final levels of the
CAFE and CO2 standards. The agencies also note that the
rebound effect diminishes the economic and environmental benefits
associated with increased fuel efficiency.
For the analysis supporting the NPRM, the agencies conducted a
thorough re-examination of the basis for the estimate of the fuel
economy rebound effect used to analyze the impacts of CAFE and
CO2 emission standards for model years 2012-16 and 2017-21.
This was prompted by three developments. First, more recent updates of
the 2007 study by Small and Van Dender that had provided the basis for
assuming the 10 percent rebound effect used in those previous analyses
reported larger values. Second, projected growth in the income measure
used in those authors' 2007 study, which was anticipated to reduce the
magnitude of the rebound effect over the future period spanned by those
analyses, did not occur during the decade following the 2007 study's
publication. Finally, extensive new research on the rebound effect had
become available since those previous
[[Page 24672]]
analyses were conducted, and while its findings were mixed, many of
those more recent studies reported values significantly above the
agencies' previous 10 percent estimate.
In the NPRM, the agencies first summarized estimates of the fuel
economy rebound effect for light-duty vehicles in the U.S. from studies
conducted through 2011, when the agencies originally surveyed research
on this subject. As the accompanying discussion in the proposal
indicated, the research available through 2011 collectively suggested
that the rebound effect was likely to fall in the range from 20 percent
to 25 percent, although the then-recent study by Small and Van Dender
(2007) pointed to smaller values, particularly for future years. The
agencies then identified 16 additional studies of the rebound effect
that had been conducted since their original survey, and the NPRM
discussed the various approaches they used to measure the magnitude of
the rebound effect, their data sources and estimation procedures,
reported findings, and strengths and weaknesses of each study.
Based on this re-examination, the agencies concluded that currently
available evidence did not appear to support the 10 percent estimate
relied upon in previous rules, and identified a value of 20 percent as
more representative of the totality of evidence, including both the
research covered by the earlier and more recent studies examined in the
NPRM. While acknowledging the wide range of estimates reported in more
recent research--which extended from zero to more than 80 percent--the
agencies noted that the central tendency of recent estimates appeared
to lie in the same 20-25 percent range suggested by their extensive
review of earlier research. The agencies also recognized that a 20
percent estimate differed markedly from the 10 percent estimate used in
the regulatory analyses for the 2010 and 2012 final rules, but noted
that it represented a return to the value NHTSA originally used to
analyze the impacts of CAFE standards for model years prior to 2011.
(a) Comments on the Rebound Effect Used in the NPRM
The agencies received numerous comments on the decision to revise
their previous estimate of the rebound effect, virtually all of which
echoed a few common arguments. First, commenters generally agreed that
the most appropriate measure for the agencies to rely on is the current
long-run fuel economy rebound effect for U.S., although a few suggested
that using an estimate of its short-run value might be
preferable.\1748\ However, many commenters argued that some of the more
recent studies the agencies relied upon to support the revised 20
percent estimate may have limited relevance to the appropriate measure
for analyzing the current rule, and that the agencies should place more
emphasis on those that commenters asserted were more appropriate to
rely upon.
---------------------------------------------------------------------------
\1748\ See, e.g., RFF, Comments, NHTSA-2018-0067-11789, at 30.
For an thorough example of the arguments made for a short- to
medium-term rebound effect, see generally IPI, Appendix, NHTSA-2018-
0067-12213, at 61.
---------------------------------------------------------------------------
To identify the most relevant research, some commenters proposed
applying various selection criteria to choose which studies were most
appropriate to rely on when estimating the value of the rebound effect
to use in this analysis. While commenters proposed using certain
criteria as ``filters''--that is, to eliminate any studies that did not
meet those criteria--they also suggested applying other criteria to
emphasize studies with particular features they argued made them more
relevant to identifying the current value of the rebound effect for the
U.S.\1749\ Among these suggested criteria were the following:
---------------------------------------------------------------------------
\1749\ See, e.g., IPI, Appendix, NHTSA-2018-0067-12213, at 58-
64; EDF, Analysis of the Value and Application of the Rebound
Effect, NHTSA-2017-0069-0574, at 16-19; California Office of the
Attorney General et al., Attachment 1, NHTSA-2017-0069-0625, at 8;
States and Cities, Attachment 1, Docket No. NHTSA-2018-0067-11735,
at 78; RFF, Comment, NHTSA-2018-0067-11789, at 3; CARB, Detailed
Comments, NHTSA-2018-0067-11873, at 120; Aluminum Association,
Comments, NHTSA-2018-0067-11952, at 5; NCAT, Appendix A, NHTSA-2018-
0067-11969, at 34; and North Carolina Department of Environmental
Quality, Comments, NHTSA-2018-0067-12025, at 12; among others. EPA's
Science Advisory Board shared similar policy opinions. SAB at 26-27.
---------------------------------------------------------------------------
Exclude estimates based upon data from outside the U.S.;
Include only estimates based upon ``more recent'' data,
usually taken to mean those published within approximately the last
decade;
View estimates based on the U.S. 2009 National Household
Travel Survey skeptically, or exclude them from consideration
completely;
Emphasize estimates derived from vehicle use and fuel
economy data spanning multiple years (such as aggregate time-series or
panel data), while according less weight to those based on a single-
year cross section (such as most household survey data);
Emphasize estimates of the rebound effect that measure the
response of vehicle use to variations in fuel efficiency, rather than
in fuel cost per mile driven or fuel price per gallon;
Emphasize estimates that rely on identification strategies
that account for potential endogeneity in fuel economy (as would
result, for example, if households with high levels of demand for
travel purchase vehicles with higher fuel economy);
Emphasize estimates based on measures of vehicle use
obtained from odometer readings; and
Emphasize estimates that explicitly control for purchase
prices of new vehicles in order to account for changes in new vehicle
prices due to CAFE standards.
A few commenters illustrated how applying these criteria could
reduce the large number of published studies of the rebound effect to a
limited subset that suggested a smaller value than 20 percent.\1750\
Using multiple criteria to exclude or de-emphasize studies that did not
meet all of those applied, these commenters argued that the most
appropriate value for this analysis was closer to (or possibly even
below) the 10-percent estimate the agencies used for the previous
rulemaking.\1751\ However, one commenter noted that applying these
criteria individually to exclude any estimates not meeting them had
almost no effect on formal measures of the central tendency (the mean
and median values) of the remaining estimates.\1752\ This commenter
suggested that only by applying two or more of these criteria jointly
and excluding any studies that did not meet all of those applied could
the universe of research on the rebound effect be reduced to a subset
supporting a lower value than the 20 percent figure the agencies used
to analyze the NPRM.
---------------------------------------------------------------------------
\1750\ See, e.g., Gillingham, Nera-Trinity Responses, NHTSA-
2018-0067-12403, at 16-30.
\1751\ See supra note 1749.
\1752\ Alliance of Automobile Manufacturers, Attachment 3,
NHTSA-2018-0067-12386, at 15-17.
---------------------------------------------------------------------------
Commenters also identified several additional recent studies that
were not included in the agencies' review of recent evidence for the
NPRM, and suggested revised interpretations of the empirical estimates
reported in two studies that had been included (the agencies also
clarified a third). Commenters represented these additional studies as
generally supporting lower values than the agencies' revised 20 percent
estimate, although this appeared to be a selective interpretation of
some of the results they reported.\1753\ Other commenters asserted
[[Page 24673]]
that the two most commonly-demonstrated features of the rebound effect
are that it varies directly with fuel prices and declines in response
to rising income over time, and argued that the latter suggests that a
declining value is likely to be more appropriate for analyzing the
longer-term impacts of this final rule.\1754\
---------------------------------------------------------------------------
\1753\ For example, some commenters (e.g., Gillingham, Nera-
Trinity Responses, NHTSA-2018-0067-12403, Table 2, at 24)
represented the recent analysis of vehicle use data from Texas by
Wenzel and Fujita as reporting a rebound effect of 8-15 percent,
which appears to be based on those authors' estimates of the
response of vehicle use to changes over time in fuel prices alone.
This range appears to ignore those same authors' estimates of the
sensitivity of vehicle use to variation in fuel costs per mile,
which provides a more direct measure of the fuel economy rebound
effect because it incorporates fuel economy as well as fuel prices.
Those estimates range from 7-40 percent, with most falling in the
interval from 15-25 percent; see generally, Wenzel and Fujita
(2018), Table 4-12, at 38.
\1754\ See particularly Small, NHTSA-2018-0067-7789, at 3.
---------------------------------------------------------------------------
Some commenters suggested that the rebound effect is asymmetrical,
meaning that drivers are more responsive to price increases than price
decreases. These commenters asserted that the asymmetrical nature of
the rebound effect favors a lower estimate.\1755\ Similarly, other
commenters suggested that the rebound effect had to be lower than 20
percent because congestion would limit additional driving.\1756\
---------------------------------------------------------------------------
\1755\ EDF, Analysis of the Value and Application of the Rebound
Effect, NHTSA-2017-0069-0574, Comment, 37-38.
\1756\ For example, the South Coast Air Quality Management
District argued that, logistically, rebound cannot exist in Southern
California because ``any rebound effect will only worsen congestion
in Southern California, such a result cannot be predicted.'' NHTSA-
2018-0067-11813 at 45.
---------------------------------------------------------------------------
(b) Agencies' Response to Comments on the NPRM
In response to commenters who argued that the agencies' estimate of
the rebound effect should be reduced, because research that
incorporates the effects of congestion or allows asymmetrical responses
to price changes suggests lower values, the agencies note that, for the
final rule's analysis, those factors would be difficult and perhaps
even inappropriate to incorporate in their analysis. In the case of
congestion, the agencies note that their estimate of the rebound
effect--like research on the rebound effect in general--represents a
change in aggregate VMT, and has no clear implication about how that
change in travel is likely to be distributed over times of the day or
geographic locations.\1757\
---------------------------------------------------------------------------
\1757\ The agencies' estimate of increased congestion costs
associated with additional driving due to the rebound effect
implicitly assumes that increased driving will be distributed
according to current travel patterns, producing similar proportional
increases at various hours of the day and geographic locations. Such
an assumption is made out of necessity to model congestion and
noise; the agencies acknowledge that the rebound effect is unlikely
to affect vehicle use in such a uniform fashion.
---------------------------------------------------------------------------
As for possible asymmetry in the response of vehicle use to changes
in driving costs, the CAFE model applies a single estimate of the
rebound effect for all changes in cost-per-mile, and cannot accommodate
a rebound effect that varies with the magnitude or direction of changes
in driving costs, which would be necessary to capture asymmetrical or
non-linear responses to cost changes. The agencies also remind
commenters that this rule will result in an increase in driving costs,
for which the research they cite generally suggests a larger value of
the rebound effect is appropriate. In any case, using a different
estimate of the rebound effect to analyze impacts of raising and
lowering standards would not promote consistency or replicability, both
desirable characteristics of regulatory analysis.
The agencies decided to include the previously omitted studies
raised by commenters in their rebound analysis supporting the final
rule, but do not feel that they suggest a value different from that
used to analyze the proposal. Adding these studies to the list of
recent research discussed in the NPRM, deleting one unpublished
analysis, and revising the entries for selected studies to reflect more
accurately the values reported by their authors produces a more
extensive catalog of recent research, which is summarized in Table VI-
175 below.
[[Page 24674]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.356
As evidenced in Table VI-175, studies continue to have a wide range
of estimates, but collectively the research looks remarkably similar to
the historical estimates. The newer studies suggest that a plausible
range for the rebound effect is 10-50 percent. The central tendency of
this range appears to be roughly 30 percent.
In response to comments proposing the application of specific
criteria to eliminate or reduce the consideration accorded to studies
without certain features thought to increase the relevance of their
findings, the agencies note that measuring the rebound effect is both
conceptually and technically challenging, and that analysts have used
many different approaches in an attempt to surmount these challenges.
The agencies' view is that each of the studies included in its previous
survey and in Table VI-175 above provides some useful evidence on the
likely value of the rebound effect, and while all have some conceptual
or theoretical weaknesses, each nevertheless provides some useful
insights into the appropriate magnitude of the rebound effect for the
current analysis.
As a general approach to estimating parameters that are uncertain,
the agencies prefer to rely on the totality of empirical evidence,
rather than restricting the available evidence by categorically
excluding or according less weight to that do not meet selection
criteria that may not be widely agreed upon. From this perspective,
analyses that rely on different measurement approaches, data sources,
and estimation procedures all have the potential to provide valuable
information for choosing the most representative value. The agencies
also view sound measurement strategies and careful empirical analysis
using reliable data as equally important features when compared to a
study's vintage or geographic scope. Examining the widest possible
range of research also enables useful comparisons and ``cross-checks''
on the estimates that individual studies report.
Notwithstanding this more inclusive perspective, the agencies
endorse certain of the characteristics preferred by commenters,
although the agencies view them as indicators of a strong study, rather
than a bright-line test of whether to accord it any weight rather than
discarding it from consideration. Specifically, the agencies agree with
many commenters that both the extended time span encompassed by their
analysis of the impacts of CAFE and CO2 standards and the
long expected lifetimes of vehicles subject to this final rule means
that estimates of the long-run rebound effect are most relevant for
purposes of the final rule
[[Page 24675]]
analysis.\1758\ The agencies also agree with commenters that estimates
based upon more recent data are generally preferable, but nevertheless
note that older studies that combine careful analysis with unusually
reliable or novel data can offer evidence that remains useful.\1759\
The agencies also concur with some commenters' argument that estimates
of the rebound effect that are derived from the relationship of vehicle
use to fuel efficiency, rather than to fuel cost per mile or gasoline
prices, are likely to provide more direct measures of the fuel economy
rebound effect itself, which is the desired parameter for the purposes
of this analysis. Finally, the agencies generally view identification
strategies and econometric methods that account or control for
potential endogeneity in fuel economy as likely to provide more
reliable estimates.
---------------------------------------------------------------------------
\1758\ Most of the vehicles affected by today's standards will
remain on the roads for at least a decade, with a significant
fraction surviving considerably longer. As such, long-run estimates
are more likely to reflect the lifetime mileage accumulation of the
new fleet than either short-run or medium-run estimates.
Furthermore, a long-run rebound estimate better reflects the
cumulative impact of successive CAFE and CO2 standards
such as those adopted by the agencies beginning as early as 2010.
\1759\ One example is the study by Greene et al. (1999), which
used advanced econometric analysis of unusually detailed and
reliable data on household demographic and economic characteristics,
household members' use of individual vehicles, and fuel purchases to
estimate the response of households' use of individual vehicles to
their actual on-road fuel economy, and its implications for total
household driving.
---------------------------------------------------------------------------
In contrast, the agencies view other criteria proposed by
commenters as unnecessarily restrictive, particularly when they are
used to disqualify otherwise informative research from consideration.
For instance, categorically excluding from consideration non-U.S.
studies--which the agencies agree should be treated cautiously--seems
likely to exclude useful evidence, particularly recognizing some of
those studies' access to unusually reliable data on vehicle use and
fuel economy and use of sophisticated econometric analysis. In
addition, many foreign studies have been conducted in nations with
income levels comparable to the U.S., and in some cases levels of auto
ownership that are beginning to approach U.S. levels. Furthermore,
driving habits throughout the U.S. are not homogenous. In fact, some
regions in the U.S. may exhibit driving habits that more closely
resemble those in some foreign nations than driving patterns in other
regions of the U.S.\1760\
---------------------------------------------------------------------------
\1760\ For example, drivers in Manhattan, Kansas likely respond
to changes in fuel prices and fuel economy differently than drivers
in Manhattan, New York.
---------------------------------------------------------------------------
In response to some commenters' recommendation that the agencies
more heavily weigh studies using data spanning multiple years than
those relying on data for a single year, the agencies note that
household surveys, the most common form of data for a single year,
provide cross-sectional variation in vehicle use and other
characteristics that is helpful for identifying the desired long-run
measure of the rebound effect. Household surveys are also an important
source of information that enable analysts to measure the response of
individual vehicles' use to variation in their fuel economy, while also
controlling adequately for household characteristics that affect travel
patterns and vehicle use. Household survey data can also enable
analysts to identify the vehicle substitution patterns within multiple-
vehicle households that are increasingly responsible for producing the
rebound effect, while even modest-scale household surveys include many
more observations than are typically available in aggregate time-series
or panel data.
These strengths of course need to be balanced against the potential
drawbacks of relying on a one-time snapshot of households' behavior
during a single time period. Surveys also frequently rely on owner-
reported estimates of vehicle use and usually require analysts to
impute vehicles' fuel economy ratings from limited and sometimes
incomplete information on the specific vehicle models and vintages that
households report owning. One result is that estimates of the rebound
effect derived from household survey data may be based on inaccurate
estimates of vehicles' use and fuel economy. Assuming the errors in
measuring these variables are random, the errors would increase the
uncertainty surrounding the estimates of the rebound effect, but would
not bias the estimate.
In contrast, studies using nationwide aggregate or average measures
of vehicle use and fuel economy or fuel cost rarely provide adequate
independent variation to support reliable estimates of the response of
vehicle use to variation in fuel economy, even where extended time
series are available, while State-level measures of these variables are
subject to potentially extreme measurement error that can compromise
estimates of these relationships.\1761\ Moreover, controlling for the
many other demographic and economic factors likely to affect vehicle
use using national or even State-level aggregate data presents
difficult challenges.
---------------------------------------------------------------------------
\1761\ For example, State-level estimates of travel by
individual vehicle classes such as cars and light-duty trucks often
exhibit implausible year-to-year variability due to the measurement
procedures states employ and the difficulty of distinguishing among
different types of vehicles. At the same time, the potential
geographic ``mismatch'' between State-level vehicle use and fuel
sales complicates any effort to measure fuel efficiency or fuel
costs at the State level.
---------------------------------------------------------------------------
Finally, the agencies note that no single selection criterion
proposed by commenters noticeably reduces the central tendency
displayed by the universe of estimates of the rebound effect, and
multiple criteria must be applied simultaneously to restrict the
universe to a subset of studies that points toward a significantly
lower value than the 20 percent estimate the agencies used to analyze
the proposal. Applying multiple criteria drastically reduces the number
of studies that remain available to guide the agencies, while at the
same time discarding potentially valuable information provided by
research those criteria exclude from consideration.\1762\ Doing so
would thereby necessarily reduce the confidence that the agencies can
have in the resulting estimate.
---------------------------------------------------------------------------
\1762\ As an illustration, excluding non-U.S. studies reduces
the number of recent analyses surveyed in the proposal from 15 to 8,
while eliminating those that rely on the 2009 National Household
Travel Survey (NHTS) discards another 5, leaving only 3.
---------------------------------------------------------------------------
Regarding some commenters' assertion that the rebound effect is
known to decline in response to rising income, and that this
observation warrants using a lower value for long-term future
evaluation of the standards' effects, the agencies note that some
evidence based on household and vehicle use surveys suggests that the
rebound effect increases with the level of household vehicle ownership,
which is itself highly correlated with income. Together with forecasts
of limited future growth in most measures of U.S. household income,
this finding casts some doubt on whether the rebound effect is likely
to decline over the time period spanned by the agencies'
analysis.\1763\
---------------------------------------------------------------------------
\1763\ For example, the widely cited IHS Markit Long-Term
Macroeconomic Outlook for Spring 2019 projects that per Capita
disposable personal income in the U.S. will grow at 1.6 percent
annually over the next 30 years; see Federal Highway Administration,
Forecasts of Vehicle Miles Traveled (VMT): Spring 2019, Table 2,
available at https://www.fhwa.dot.gov/policyinformation/tables/vmt/vmt_forecast_sum.cfm.
---------------------------------------------------------------------------
The agencies also note that one of the studies cited in Table VI-
175 above (DeBorger et al., 2016) finds that the decline in the fuel
economy rebound effect with income reported in the
[[Page 24676]]
earlier analysis by Small and Van Dender (2007)--on which the agencies
relied in reducing their original estimate of the rebound effect to 10
percent--results entirely from a reduction in drivers' sensitivity to
fuel prices as their incomes rise, rather than from any effect of
rising income on the sensitivity of vehicle use to fuel economy.\1764\
This latter measure--which DeBorger et al. find is quite small and has
not changed significantly as incomes have risen over time--is the most
direct measure of the fuel economy rebound effect, so their analysis
calls into question its widely-assumed sensitivity to income.
---------------------------------------------------------------------------
\1764\ DeBorger, B., Mulalic, I., and Rouwendal, J., ``Measuring
the rebound effect with micro data: A first difference approach.''
Journal of Environmental Economics and Management, 79 (2016), at 1-
17.
---------------------------------------------------------------------------
Finally, because there is not a clear consensus around a single
rebound estimate within the literature, the agencies believe it is
important to benchmark their analysis with other large scale surveys of
the literature published by neutral observers. In one early survey,
Greening, Greene, and Difiglio (2000) reviewed studies that estimated
the rebound effect for light-duty vehicles in the U.S., concluding that
those relying on aggregate time-series data found it was likely to
range from 10-30 percent, while those using cross-sectional analysis of
household vehicle use suggested a larger rebound effect, in the range
of 25-50 percent.\1765\ Sorrell et al. (2009) found that the magnitude
of the rebound effect for personal automobile travel is likely to fall
in the 10-30 percent range, with some evidence suggesting that the
lower end of that range might be most appropriate.\1766\
---------------------------------------------------------------------------
\1765\ Greening, L.A., Greene, D.L. and Difiglio, C., ``Energy
efficiency and consumption--the rebound effect--a survey.'' Energy
Policy, Vol. 28 (2000), at 389-401.
\1766\ Sorrell, Steve, John Dimitropoulos, and Matt Sommerville,
``Empirical Estimates of the Direct Rebound Effect: A Review,''
Energy Policy 37(2009), at 1356-71.
---------------------------------------------------------------------------
Most recently, a meta-analysis of 74 published studies of the
rebound effect conducted by Dimitropoulos et al. (2018) estimated that
the long-run rebound effect ranges from 22-29 percent when measured by
the response of vehicle use to variation in fuel efficiency (the
authors' preferred measure), from 21-41 percent when it is measured
using the variation fuel cost per unit distance, and from 25-39 percent
using fuel price per gallon.\1767\ The authors concluded that ``the
magnitude of the rebound effect in road transport can be considered to
be, on average, in the area of 20%,'' but noted that the long-run
estimate was about 32 percent.\1768\ A subsequent published study by
these same authors (Dimitropoulos et al. (2018)) concludes that the
most likely estimate of the long-run rebound effect is in the range of
26-29 percent, but could range from as low as 15 percent to as high as
49 percent at income levels, development densities, and fuel prices
that are currently representative of the U.S.\1769\
---------------------------------------------------------------------------
\1767\ Dimitropoulos, Alexandros, Walid Oueslati, and Christina
Sintek, ``The rebound effect in road transport: a meta-analysis of
empirical studies,'' Paris, OECD Environment Working Papers, No.
113; see esat Table 5, at 25 (and accompanying discussion).
\1768\ Id. at 28.
\1769\ Dimitropoulos, Alexandros, Walid Oueslati, and Christina
Sintek, ``The Rebound Effect in Road Transport: A Meta-Analysis of
Empirical Studies,'' Energy Economics 75 (2018), at 163-79; see esat
Table 4, at 170, Table 5, at 172 (and accompanying discussion), and
Appendix B, Table B.V., at 177.
---------------------------------------------------------------------------
(c) Selecting a Value of the Rebound Effect for Evaluating the Impacts
of This Rule
After reviewing the evidence on the rebound effect previously
summarized in the NPRM, comments the agencies received, other recent
studies of the rebound effect that were not summarized in the NPRM but
suggested by commenters, and published surveys of literature, a
reasonable case can be made to support values of the rebound effect at
least as high as 30 percent. The totality of evidence, without
categorically excluding studies on grounds that they fail to meet
certain criteria, and evaluating individual studies based on their
particular strengths, suggests that a plausible range for the rebound
effect is 10-50 percent. The central tendency of this range appears to
be at or slightly above its midpoint, which is 30 percent. Considering
only those studies that the agencies believe are derived from unusually
reliable data, employ identification strategies that are likely to
prove effective at isolating the rebound effect, and apply rigorous
estimation methods suggests a range of approximately 10-45 percent,
with most of their estimates falling in the 15-30 percent range.\1770\
---------------------------------------------------------------------------
\1770\ As indicated previously, these are the selection criteria
proposed by commenters with which the agencies concur. In
chronological order, the studies the agencies feel best meet those
criteria include Greene et al. (1997), Small and Van Dender (2007)
and subsequent updates by Hymel, Small, and Van Dender (2010, 2015),
Linn (2016), Anjovic and Haas (2012), Gillingham (2014), and
DeBorger et al. (2016). Other studies the agencies believe warrant
serious consideration because they offer some or most of these same
advantages include those by Liu et al. (2014), Knittel and Sandler
(2018), and Wenzel and Fujita (2018).
---------------------------------------------------------------------------
At the same time, the agencies conclude that a reasonable case can
also be made to support values of the rebound effect falling in the 5-
15 percent range. This argument relies on using the criteria proposed
by commenters to restrict the studies considered to include recently
published analyses using U.S. data, and to accord the most weight to
research that relies on measures of vehicle use derived from odometer
readings, controls for the potential endogeneity of fuel economy, and
estimates the response of vehicle use to variation in fuel economy
itself, rather than to fuel cost per distance driven or fuel prices.
This approach suggests that the rebound effect is likely in the range
from 5-15 percent, and is more likely to lie toward the lower end of
that range. The agencies note that estimates of very low or no rebound
effect cited by some commenters are either misinterpretations of the
findings reported by their authors, or do not represent measures of the
fuel economy rebound effect.\1771\
---------------------------------------------------------------------------
\1771\ For example, some commenters misinterpret Greene's (2012)
inability to identify a statistically significant estimate of the
response of vehicle use to variation in fuel economy as evidence
that its true value is zero. Similarly, some commenters misinterpret
the result reported by West et al. (2017) that buyers of more fuel-
efficient vehicles did not increase their driving as evidence that
fuel economy itself has no effect on vehicle use, when--as the
study's authors and some commenters acknowledge--it reveals instead
that buyers regarded those vehicles as providing inferior
transportation service and drove them less as a consequence. Because
the agencies repeatedly insist that vehicle attributes other than
fuel economy will not change as a consequence of this rule, those
authors' finding is of limited or no relevance to the analysis
supporting this rule.
---------------------------------------------------------------------------
Finally, the agencies note that surveys of evidence on the rebound
effect have consistently found that the most appropriate estimate falls
in the range of 10-40 percent. These findings have remained
surprisingly consistent over time, despite a rapidly expanding universe
of empirical evidence that includes estimates drawn from more diverse
settings, and reflects continuing improvements in the data they rely
upon, an expanding range of strategies for identifying the rebound
effect and distinguishing it from other influences on vehicle use, and
advances in the econometric procedures analysts use to estimate its
magnitude.
For the aforementioned reasons, the agencies have elected to retain
the 20 percent rebound effect used to analyze the effects of the NPRM
on vehicle use and fuel consumption for analyzing the comparable
effects of this final rule. As explained above and in the NPRM, older
research suggests a rebound of 20 to 25 percent. The new research in
Table VI-175 supports a similar--or even larger--range. Extensive
survey studies support
[[Page 24677]]
a rebound at or above 20 percent. As such, the agencies feel 20 percent
is a reasonable--and probably even conservative--estimate of the
totality of the evidence. While a lower estimate may be reasonable
under certain circumstances, the agencies are uncomfortable making the
requisite assumptions regarding which specific criteria should be used
to identify relevant studies and relying on a subset of the literature
for the central analysis. However, recognizing the uncertainty
surrounding the rebound value, the agencies also examine the
sensitivity of those estimated impacts to values of the rebound ranging
from 10 percent to 30 percent, both in isolation and in conjunction
with plausible variation in other key parameters.
(5) Vehicle Miles Traveled (VMT)
VMT directly influences many of the various effects of fuel economy
and CO2 standards that decision-makers consider in
determining what levels of standards to set. For example, fuel savings
is a function of a vehicle's efficiency, miles driven, and fuel price.
Similarly, factors like criteria pollutant emissions and fatalities are
direct functions of VMT. In the CAFE model, VMT is the product of
average usage per vehicle in the fleet and fleet composition, which is
itself a function of new vehicle sales and vehicle retirement
decisions, otherwise known as scrappage. These three components--
average vehicle usage, new vehicle sales, and older vehicle scrappage--
jointly determine total VMT projections for each alternative.
As the following discussion explains, today's VMT analysis provides
aggregate results comparable to other well-regarded VMT estimates.
However, because the agencies' analysis looks at the incremental costs
and benefits across alternatives (see Section VII), it is more
important that the analysis capture the variation of VMT across
alternatives than accurately to predict total VMT within a scenario. As
such, the agencies note that today's VMT estimates are logical,
consistent, and precise across alternatives. Furthermore, as will be
described in further detail below, while the agencies, in response to
comments, have decided to modify their approach to calculating VMT and
to use different VMT estimates than those used in the NPRM, the general
trends between alternatives are comparable.
Commenters addressed a number of topics related to the total amount
of estimated VMT, the incremental differences in estimated VMT between
regulatory alternatives, and per-vehicle VMT estimates in the NPRM
analysis. In general, commenters felt that the NPRM's VMT numbers were
inaccurate and should not be relied on for the analysis.\1772\ Some
commenters were more specific and argued that the total amount of
estimated VMT projected in the NPRM started at too low a level, and
increased too much over the years simulated. Similarly, some commenters
argued that the agencies' estimates were too different from other
recognized estimates and suggested that the agencies benchmark VMT
projections to other sources to ensure both a consistent starting point
and comparable VMT throughout the calendar years analyzed.
---------------------------------------------------------------------------
\1772\ See, e.g., Securing America's Energy Future, NHTSA-2018-
0067-11981 at 37-38.
---------------------------------------------------------------------------
A few commenters objected to the underlying mileage accumulation
schedules, which form the basis for per-vehicle VMT estimates in CAFE
Model simulations. Such commenters speculated that revisions to these
schedules undertaken in 2016 might be the reason for discrepancies in
total VMT. Other commenters were less concerned about how VMT was
computed within each scenario but were apprehensive about differences
in VMT estimates across regulatory alternatives. For instance, Honda
argued that, ``[a]ssuming all other parameters are held constant--and
excluding the rebound effect--it is not obvious why one scenario should
have different total VMT than another.'' \1773\ While commenters
generally provided few specific recommendations about the level to
which VMT estimates should be constrained among alternatives, several
commenters argued that VMT projections would benefit from consideration
of travel demand modeling.
---------------------------------------------------------------------------
\1773\ Honda, NHTSA-2018-0067-11818, at 17.
---------------------------------------------------------------------------
Additionally, some commenters (RFF, IPI, NRDC) argued that a
superior, and perhaps even necessary, approach would be to incorporate
a model that considers jointly the decision to buy, use, and retire
vehicles at the household level. As RFF posited ``a household makes
decisions about its vehicle ownership and use jointly: people don't buy
new vehicles or get rid of existing ones without considering how these
actions will affect the use of their vehicles.'' \1774\ IPI further
argued that ``[i]n sum, VMT is influenced by vehicle choice and vehicle
choice is influenced by VMT. And a `unified model of vehicle choice and
usage' is necessary.'' \1775\ While the agencies agree that a joint
household consumer choice model--if one could be developed adequately
and reliably to capture the myriad circumstances under which families
and individuals make decisions relating to vehicle purchase, use and
disposal--would reflect decisions that are made at the household level,
the agencies do not agree that it is necessary, or necessarily
appropriate, to model the national program at that scale in order to
produce meaningful results that can be used to inform policy decisions.
The most useful information for policymakers relates to national
impacts of potential policy choices. No other element of this analysis
occurs at the household level, and the error associated with allocating
specific vehicles to specific households over the course of three
decades would easily dwarf any error associated with the estimation of
these effects in aggregate. The agencies have attempted to incorporate
estimates of changes to the new and used vehicle markets at the highest
practical levels of aggregation, and worked to ensure that these
effects produce fleetwide VMT estimates that are consistent with the
best, current projections given our economic assumptions. While future
work will always continue to explore approaches to improve the realism
of CAFE/CO2 simulation, there are important differences
between small-scale econometric studies and the kind of flexibility
that is required to assess the impacts of a broad range of regulatory
alternatives over multiple decades. The agencies have read and
evaluated the comments on the NPRM, incorporating many suggestions that
improve the fidelity of this analysis--taking particular care to be
conservative with the analysis. The modifications the agencies have
made in response to these comments are described below (and in the
RIA).
---------------------------------------------------------------------------
\1774\ RFF, NHTSA-2018-0067-11789, at 5.
\1775\ IPI, Appendix, NHTSA-2018-0067-12213, at 80 (internal
citation omitted).
---------------------------------------------------------------------------
The agencies carefully assessed all comments. To address them, the
agencies have revised their calculation of estimated VMT in two,
significant respects. First, in response to comments regarding the
mileage accumulation schedules, the agencies have revised the schedules
using panel data. Second, to deal with commenters' concerns with the
fluctuation of estimated ``non-rebound'' VMT across regulatory
alternatives, the agencies have created a method that constrains ``non-
rebound'' VMT across regulatory alternatives. The agencies believe
these two changes collectively resolve the substantive issues raised by
commenters. The total VMT for the final rulemaking (FRM) analysis now
aligns with estimates of the Federal Highway Administration
[[Page 24678]]
(FHWA) and the only differences in VMT between alternatives is
attributable to changes in the fleet's fuel economy. The following
sections discuss these changes in detail.
(a) Mileage Accumulation Schedule
To account properly for the average value of consumer and societal
costs and benefits associated with vehicle usage under various CAFE and
CO2 alternatives, it is necessary to estimate the portion of
these costs and benefits that will occur each calendar year for each
model year cohort. Doing so requires some estimate of how many miles
the average vehicle of each body type is expected to drive at each age.
The agencies call these ``mileage accumulation schedules.'' For this
final rule, the agencies are modifying the mileage accumulation
schedules, largely in response to comments.
(i) Data
As mentioned in previous sections, NHTSA purchased a data set
containing 70 million vehicle odometer readings from Polk in part to
create the vehicle mileage accumulation schedules used in the NPRM. In
the proposal, the agencies explained that Polk data was newer and
believed to be qualitatively superior to the 2001 and 2009 National
Household Travel Survey (NHTS) data used in prior rules.\1776\
Consistent with previous analyses,\1777\ the agencies used a cross-
sectional sample of the Polk data for the NPRM. Cross-sectional data is
like a ``snapshot'' in time. Rather than tracking vehicles over a
period, the sample contained a single odometer reading from each
vehicle sampled. In other words, the sample contained observations of
the total lifetime accumulation of miles (represented by its odometer
reading) through CY2015 of all MYs still present on the road. The
cross-sectional sample was limited in the number of vintages included
in the sample. While the sample was suitable to capture the heaviest
usage ages (age zero to 15 years), it contained no observations for
vehicles older than 16 years. This required the agencies to rely on
mileage accumulation schedules developed from other data sources to
produce annual VMT rates for older vehicles. Furthermore, in order to
develop a schedule of mileage accumulation by age, it was necessary to
assume that each vehicle traveled the same number of miles each year to
reach its odometer reading, e.g. if a MY 2007 vehicle had an odometer
reading of 88,000 in CY2015, the analysis assumed the vehicle drove
11,000 miles each year from CY2007 to CY2015.
---------------------------------------------------------------------------
\1776\ See, e.g., 83 FR at 43089-90 (Aug. 24, 2018).
\1777\ Previous rules were based on odometer data from the 2001
National Household Travel Survey (NHTS). S. Lu, ``Vehicle
Survivability and Travel Mileage Schedules,'' Report Number: DOT HS
809 952 (January 2006).
---------------------------------------------------------------------------
The agencies acknowledged that this approach missed some of the
nuances of car ownership.\1778\ For example, vehicles are commonly part
of multi-vehicle household fleets and their usage changes over time as
households buy new vehicles and replace older ones. Similarly, most
vehicles belong to multiple owners over the course of their useful
lives, each of whom may have different patterns of usage. The most
significant limitation of using cross-sectional data is the presence of
an attrition bias. As a cohort ages, vehicles that have been used more
heavily are more likely to be retired at each age than vehicles that
are driven less. As the most heavily-driven vehicles drop out of the
fleet, the remaining vehicles, which likely have been driven less at
each age throughout their lives, will have lower odometer readings.
Making the common, but necessary assumption that each vehicle is driven
uniformly at each age results in lower miles-per-age estimates because
of this attrition bias. In the schedules used for the NPRM, the effect
of this bias occurred during the ages where each model year cohort
typically scraps at the highest rates--9 to 15 years. These limitations
led to lower estimates, which led commenters such as EDF to state
``[g]iven that the Volpe Model VMT falls far short of confident
measurements of gasoline consumption, these mileage accumulation
schedules need to be increased.'' \1779\ The agencies note that many of
these data limitations were present in previous CAFE and CO2
analyses.\1780\
---------------------------------------------------------------------------
\1778\ See 83 FR at 43092 (Aug. 24, 2018).
\1779\ EDF, Appendix B (Rykowski comments), NHTSA-2018-0067-
12108, at 46.
\1780\ See, e.g., NHTSA Final Regulatory Impact Analysis:
Corporate Average Fuel Economy for MY 2012-MY 2016 Passenger Cars
and Light Trucks, NHTSA-2010-0131, at 372-79.
---------------------------------------------------------------------------
Several commenters noted the agencies' reliance on cross-sectional
data, and urged the use of panel data instead to develop mileage
accumulation schedules. For example, API argued that cross-sectional
data cannot accurately capture mileage accrual and suggested ``the
agencies re-consider the use of the [Polk] data for developing revised
mileage accumulation schedules unless the data can capture mileage
accumulation rates versus age on an individual-vehicle basis.'' \1781\
The NPRM discussed the possible use of panel data in the future and the
benefits that doing so could provide.\1782\
---------------------------------------------------------------------------
\1781\ API, EPA-HQ-OAR-2018-0283-4548, at 10.
\1782\ See 83 FR at 43092 (Aug. 24, 2018).
---------------------------------------------------------------------------
In response to these comments, the agencies created new mileage
accumulation schedules based on panel data for this final rule. Unlike
cross-sectional data, panel data includes a temporal element, which
resolves the limitations imposed by cross-sectional data. The data
source used for the final rule contains sequential readings of
individual vehicles over time, and the vehicles are tracked at the VIN
level. Polk accumulates readings about individual vehicles through
state inspection programs, title changes, and maintenance events, among
other sources. The Polk data includes observations of a specific
vehicle's odometer readings over the course of many years, capturing
the accumulated lifetime mileage at multiple ages. By using the
observation date and accumulated miles (represented by the odometer
reading), the agencies can compute the rate of driving (miles per year,
or month) between observations for each vehicle. This is a superior
method to assuming that the rate of accumulation, over all ages, is
simply the ratio of odometer to age, as commenters noted. In
particular, calculating the rates of mileage accumulation using
successive observations of the same vehicle explicitly resolves the
attrition bias and matches the approach to estimating driving rates
with panel data in other studies.\1783\
---------------------------------------------------------------------------
\1783\ See, e.g., Kenneth Gillingham, Alan Jenn, and In[ecirc]s
M.L. Azevedo (2015), ``Heterogeneity in the Response to Gasoline
Prices: Evidence from Pennsylvania and Implications for the Rebound
Effect, Energy Economics,'' Volume 52, Supplement 1, 2015, Pages
S41-S52, ISSN 0140-9883, available at https://doi.org/10.1016/j.eneco.2015.08.011.
---------------------------------------------------------------------------
The panel dataset has another advantage over other sources: Because
it tracks individual vehicles over time, the agencies have more precise
information about each vehicle's useful age. In previous analyses, the
agencies were forced to assume that ``age'' was simply equal to the
calendar year minus the model year in which the vehicle was produced.
For example, a MY2010 vehicle was assumed to be five years old in 2015.
This created, as API stated, a ``discontinuity in the values between
year 1 and year 2'' within the schedules.\1784\ It is common for
vehicles produced in a given model year to be sold and registered over
the course of multiple calendar years. Thus, a MY2010 vehicle assumed
to be five years old in 2015, could have been
[[Page 24679]]
registered for the first time in CY2012 and might have a real driving
age of three years, rather than five, simply because it sat on a
dealership lot for a couple of years before being purchased. The Polk
data allows us to identify the first registration date of each vehicle
in the sample and compute its true driving age at each point in time.
This not only improves the precision of the mileage accumulation rate
in the first year, but in subsequent years as well. The odometer data
used in the NPRM had another limitation: Odometer readings were grouped
into cohorts by nameplate, for which only distributional information
was available. It was necessary to use the mean odometer reading for
each cohort at each age, but in cases where the distribution was
skewed, the mean could be misleading. Making the same assumption about
registration date, as each cohort contained information about the
average registration date, further compounded the potential for
distortion.
---------------------------------------------------------------------------
\1784\ API, EPA-HQ-OAR-2018-0283-5458, at 9-10.
---------------------------------------------------------------------------
To the extent that commenters objected to the NPRM's use of Polk
data on the basis of it being proprietary, the agencies note that using
proprietary data is common in rulemakings, and, specifically, Polk data
has been used for CAFE and CO2 analyses on multiple
occasions previously. For the 2016 final medium- and heavy-duty rule
and Draft TAR, the agencies used Polk odometer data to develop the
vehicle mileage accumulation schedules.\1785\ Further, the specific
data set was cited and is available for acquisition through Polk.
---------------------------------------------------------------------------
\1785\ See, e.g., 81 FR 73478, 73746 (Oct. 25, 2016); see also
81 FR 49217 (Jul. 27, 2016).
---------------------------------------------------------------------------
Recently, the 2017 National Household Travel Survey has become
available as a possible data source to develop mileage accumulation
schedules. While attractive from the standpoint of transparency, it
suffers from the same flaws as data sources used to develop previous
schedules. In particular, it represents a cross section of odometer
readings at a single point in time, requiring the assumption that the
rate of usage is simply reported odometer divided by vehicle/age, or an
extrapolation of respondents' daily travel behavior into representative
annual schedules, which commenters suggested was a poor assumption.
Additionally, all of the odometers in the newest NHTS are self-
reported, leading to questionable reliability of the individual data
points (and notably round numbers in many cases). Finally, the NHTS is
intended to be a representative sample of households, but not a
representative sample of vehicles. Research has found that creating a
representative sample of households can represent a significant
challenge, as past iterations of the NHTS have systematically
oversampled high income households. The nature of the sample also
explicitly excludes vehicles used for commercial purposes, which
nonetheless compose a meaningful portion of the new vehicle market,
accumulate miles of travel, and consume fuel. The data set on which the
mileage accumulation schedule used for this final rule is based
contains at least two readings (and frequently several) for over 70
percent of the registered light duty vehicle population in 2016.
(ii) Methodology
The data used to construct the schedules initially included between
two and fifty odometer readings from each of over 251 million unique
vehicles. While most of the readings had plausible reading dates,
odometer counts, and implied usage rates, some of the readings appeared
unrealistic and received additional scrutiny. The agencies used a set
of criteria to identify and remove readings that were likely record
errors. For example, odometer readings predating the commercial release
of the vehicle, showing negative VMT accumulation over time, or taken
too closely together to provide meaningful insight into annual vehicle
usage were removed from the analysis.\1786\ Such sanitization of real
datasets is typically necessary, and each step in the process was
recorded and described in conformity with standard econometric
practice.\1787\
---------------------------------------------------------------------------
\1786\ Refer to Section VI.D.1.(5).(b) of the FRIA for a full
accounting of the process used to clean the Polk odometer data.
\1787\ See, e.g., Osborne, Jason W., Best Practices in Data
Cleaning, SAGE Publications, Inc., January 2012.
---------------------------------------------------------------------------
Similar to the NPRM, the remaining readings were sorted into five
categories: Cars, SUV's/vans, pickups, MDHD pickups/vans, and chassis.
The car, SUVs/vans and pickup categories match the definitions used to
build the VMT schedules used in the NPRM, as well as those used to
build the scrappage model. Table VI-176 shows the number of VINs,
reading pairs, and average readings per VIN by body style.
[GRAPHIC] [TIFF OMITTED] TR30AP20.357
*Not used in this final rule analysis, in part in response to
comments.
Once the dataset was cleaned, the agencies created a sample of one
million reading pairs, where each pair represented an initial odometer/
date reading and a subsequent odometer/date reading from the same
vehicle. Analysis
[[Page 24680]]
of the entire dataset was too computationally demanding and
statistically unnecessary. Two conditions were created for sampling.
The first controlled for Polk's censoring in the odometer readings
recorded in the dataset (described below), and the second ensured the
usage data was not biased by survival and that it represented usage
rates over a relatively short period of time compatible with the
beginning of the FRM analysis. Further analysis suggests that shorter
periods between readings is still correlated with higher usage rates so
that further filtering of the data sample was considered in the
regression analysis. Once these filters were applied, the agencies
considered several polynomial fits to the average odometer readings.
These fits inform the final usage rates by age and body style used in
this FRM analysis. The details are further described below.
One element of the usage data (mentioned above as the first
condition control) required the agencies to filter the dataset. The
odometer readings recorded are censored at 250k miles.\1788\ For this
reason, the agencies exclude readings recorded exactly as 250k miles.
The censoring could bias estimates of usage rates if odometer readings
and future usage rates are correlated, which they likely are. While the
agencies hope to reconcile this limitation of the dataset in future
work, the benefits of observing actual usage data through 30 years
(rather than average odometer readings by model through 15 years) far
outweigh the limitation. Still, the agencies filtered out these
censored data points, since the actual odometer readings for such
vehicles are likely higher than reported.
---------------------------------------------------------------------------
\1788\ Polk codes any vehicle whose odometer exceeds 250K miles
as 250K miles exactly, regardless of the actual odometer reading.
---------------------------------------------------------------------------
The Polk dataset is conditional on survival so that it represents
the usage of vehicles on the road at the time of the sample (the end of
the first quarter of 2017). In this way, it captures the actual
observed usage rates of vehicles surviving to their current age in the
dataset. An issue with this is that all readings of a vehicle are
included in the sample. If usage rates from earlier ages and survival
are correlated, which they likely are, then including the readings for
a 30-year-old vehicle when it was 10 years old will bias the estimated
usage rates of 10-year-old vehicles downward because vehicles that
survive to advanced ages tend to be used less than vehicles that are
retired at earlier ages for the same model year. As noted above, the
NHTS data used in the NPRM suffered from the same problem. To mitigate
this issue, the agencies applied a second filter when sampling the data
set: The agencies only included readings where the reading date of the
second reading in the pair is January 2015 or later. This reduces the
potential bias from the joint probability of usage and survival to only
those vehicles scrapped between January 2015 and the first quarter of
2017. This balances losing information for older, less represented ages
by excluding too much data on these vehicles and severely biasing the
estimates of usage by age.
For estimates within the CAFE model the average usage is the
relevant measure. Table VI-177 shows the average usage rates for cars
by age as well as linear, quadratic, and cubic polynomial fits on these
points.\1789\ The average usage rates follow a relatively smooth
pattern, but appear to decline at an accelerating rate for the oldest
ages. The linear equation captures this trend for older vehicles, but
underestimates early ages. The quadratic fit shows a diminishing
decrease in the usage of older vehicles which may overestimate their
use. The cubic fit captures the early age usage trends and the
accelerating decrease in the usage of older ages. For this reason, the
agencies used the cubic curve as the basis for the new VMT schedules by
age.
---------------------------------------------------------------------------
\1789\ In general, the objective of a polynomial regression is
to capture the nonlinear relationship between two variables. While
the fit produces a nonlinear curve, it is linear in the
coefficients. Choosing the lowest degree of the polynomial function
that captures the inflection points in the data preserved degrees of
freedom and ensures that applying the polynomial function to
observations outside the range of data (as done here for ages beyond
30) is well behaved.
[GRAPHIC] [TIFF OMITTED] TR30AP20.358
[[Page 24681]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.359
Table VI-178 shows the observed and predicted average usage rates
by age for SUVs/vans. All the polynomial fits predict the observed
average usage rates reasonably well. However, the linear fit under
predicts the usage of the oldest vehicles, and the cubic fit predicts
higher usage rates for vehicle ages beyond age 30. The quadratic fit
predicts reasonable usage rates for all observed and out-of-sample ages
through age 40. For this reason, the quadratic fit was used as the
basis for the SUV mileage schedule.
[[Page 24682]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.360
[[Page 24683]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.361
Table VI-179 shows the observed and predicted average usage rates
for pickups by age. The observed rates initially decline at an
increasing rate, the decline diminishes and appears to accelerate again
for the oldest ages. The linear fit underestimates the usage rates for
the youngest and oldest ages and overestimates middle-aged vehicles.
The quadratic fit reasonably predicts the observed average usage rates
but predicts an increase in usage rates for the oldest ages out of the
observed sample. The cubic fit reasonably predicts the observed
averages and appears to capture the diminishing decline of usage for
the oldest ages observed in the in-sample averages. For this reason,
the agencies used the cubic fit as the basis for the pickup VMT
schedules.
[GRAPHIC] [TIFF OMITTED] TR30AP20.362
[[Page 24684]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.363
As in the NPRM, the current schedule differs by body-style to
represent different usage profiles that the agencies observed in the
data. While more stratification is possible, it is unlikely to provide
much additional value. Table VI-180 shows the annual miles driven at
each age for passenger cars, SUVs (and CUVs and minivans), and pickup
trucks at each age of their useful life, conditional upon surviving to
that age.
[[Page 24685]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.364
(b) Benchmarking Total VMT
In order to assess the fuel consumption and environmental impacts
of regulatory alternatives, it is desirable to have a representation of
aggregate travel and fuel consumption that is both reasonable and
internally consistent. Some commenters suggested that the aggregate
totals presented in the NPRM deviated from other published estimates,
and argued that the entire analysis was therefore an unreliable source
of information for decision-makers to consider. For example, EDF
stated, ``the NHTSA model `projects' aggregate, nationwide VMT levels
for 2016 and 2017 that are about 20 percent lower than formal
government estimates by EIA and FHWA.'' \1790\ EDF further stated,
``[b]etween 2017 and 2025, fleetwide VMT grows by 3.1% per year in the
Volpe Model, while it only grows 0.5% per year in the 2018 Annual
Energy Outlook.'' \1791\ EDF also suggested, ``[o]ne obvious way to
assess the accuracy of the schedules is to compare the projections of
the Volpe Model of total fleetwide fuel consumption in a recent
calendar year with actual gasoline sales.'' \1792\
---------------------------------------------------------------------------
\1790\ EDF, Appendix A, NHTSA-2018-0067-12108, at 59.
\1791\ EDF, Appendix B (Rykowski comments), NHTSA-2018-0067-
12108, at 44.
\1792\ Id. at 43.
---------------------------------------------------------------------------
The Federal Highway Administration (FHWA) publishes annual VMT
estimates for the light-duty vehicle fleet, the most recent of which is
calendar year 2017. The NPRM estimate of total light-duty VMT was 2.22
trillion miles in calendar year 2016. The FHWA estimate for light duty
VMT in 2016 was 2.85 trillion miles.\1793\ While the definitions of
light-duty are not identical in the two cases (where FHWA excludes
trucks with 10,000 lbs. GVW, the agencies' analysis excludes trucks
with GVW greater than 8,500 lbs. from its light duty definition), that
definitional discrepancy is not significant enough to account for the
difference in the total VMT. While some commenters suggested that the
agencies compare simulated fuel consumption to published estimates from
EIA to determine the validity of our VMT assumptions, such a comparison
requires accurate assumptions about the true on-road fuel efficiency of
registered vehicles over forty model years in addition to their annual
usage. Comparing simulated VMT directly to FHWA measurements requires
fewer assumptions and is a more meaningful comparison.
---------------------------------------------------------------------------
\1793\ See Highway Statistics 2017, Table VM-1, available at
https://www.fhwa.dot.gov/policyinformation/statistics/2017/vm1.cfm.
---------------------------------------------------------------------------
Substituting the updated mileage accumulation schedules for the
NPRM schedules, and using the calendar year 2016 fleet from the NPRM,
produces an estimate of total light duty VMT in 2016 that is about 2.85
trillion miles--nearly identical to the FHWA estimate for 2016, despite
the use of different estimation methods and data sources. FHWA's
estimate of total light-duty VMT in 2017 is 2.88 trillion miles,\1794\
while the estimate produced by the simple product of the mileage
accumulation schedule on the estimated on-road fleet is 2.94 trillion
miles, a difference of about two percent. While not as close as the
estimate for calendar year 2016, the discrepancy is still small
considering that the estimates are obtained through entirely different
methods. One important source of
[[Page 24686]]
discrepancy with FHWA's 2017 VMT estimate is the fact that the CAFE
model simulation assumes all of the vehicles produced in a given model
year are driven for the entire calendar year matching the
vintage.\1795\ This means, for calendar year 2017, the initial year of
the simulation used to support this rule, MY2017 vehicles are assumed
to have been both registered and driven for the entirety of CY2017. As
a result, it naturally overestimates the true VMT for calendar year
2017. The analysis accounts for this discrepancy by adjusting calendar
2017 total VMT downward by one percent, and the discrepancy in total
VMT caused by conflating model years and calendar years dissipates over
time.
---------------------------------------------------------------------------
\1794\ Id.
\1795\ The CAFE model uses an annual timestep, meaning that each
time period represents one year. Because calendar years are
(obviously) years, and all of the other inputs (discounting and
inflation, macroeconomic variables, fuel prices, VMT, etc.)
represent annual values, the timestep in the CAFE model is a
calendar year. However, model years start prior to the calendar year
for which they are named, and new model year sales continue (albeit
only slightly) after their calendar year ends. In order to account
for model year sales on their true timing relative to calendar
years, the model would need to be restructured to use a quarterly
timestep. While this would improve the fidelity between calendar
year and model year for sales, obtaining quarterly projections of
nearly every other variable in the analysis would be complicated (if
not impossible). For this reason, the model conflates ``model year''
and ``calendar year'' for the analysis, even though it is a
simplification.
---------------------------------------------------------------------------
While the agencies have established that the years for which they
have data are sufficiently similar to published VMT estimates, the
question of projection still remains. FHWA, in its forecasts of VMT
(Spring 2019),\1796\ forecasts a compound annual growth rate of 0.8
percent for light-duty vehicles between 2017 and 2047 in its baseline
economic outlook. However, that projection uses a different set of
macroeconomic conditions and fleet assumptions than this analysis. To
compare CAFE model simulations of total VMT to the FHWA projections,
the agencies ran the FHWA model with a comparable set of assumptions to
the greatest extent possible.\1797\ \1798\ Using similar economic
growth assumptions, our reference case total light-duty VMT grows at a
compound rate of 0.63 percent per year between 2017 and 2050. Using
comparable assumptions in the FHWA model produce an annual growth rate
of 0.66 percent. Again, these differences are remarkably low for models
created with different methods, and lead to trivial variances, for the
purposes of our analysis, in total VMT. The relevant annual projections
for the baseline scenario appear in Table VI-181.
BILLING CODE 4910-59-P
---------------------------------------------------------------------------
\1796\ See ``FHWA Forecasts of Vehicle Miles Traveled (VMT):
Spring 2019,'' Office of Highway Policy Information, available at
https://www.fhwa.dot.gov/policyinformation/tables/vmt/vmt_forecast_sum.pdf.
\1797\ See ``FHWA Travel Analysis Framework: Development of VMT
Forecasting Models for Use by the Federal Highway Administration,''
Volpe, available at https://www.fhwa.dot.gov/policyinformation/tables/vmt/vmt_model_dev.pdf.
\1798\ In particular, we ran the FHWA VMT forecasting model with
the same: Personal disposable income, population, fuel prices (all
of which come from AEO2019), and simulated on-road fleet fuel
economy in the baseline.
---------------------------------------------------------------------------
[[Page 24687]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.365
[[Page 24688]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.366
BILLING CODE 4910-59-C
(c) Preserving Total VMT Across Regulatory Alternatives
In the NPRM, the combined effect of the sales and scrappage
responses created small percentage differences in total VMT across the
range of regulatory alternatives.\1799\ However, as the Environmental
Group Coalition noted, even a 0.4 percent difference can result in
``692 billion additional VMT under the CAFE standards and 894 billion
under the CO2 program.'' \1800\ Since VMT is related to many
of the costs and benefits of the program, VMT of this magnitude can
have meaningful impacts on the incremental net benefit analysis. This
point was made by a number of commenters who were concerned about the
magnitude and direction of differences in VMT between regulatory
alternatives (IPI, EDF, CBD, CARB, EPA's Science Advisory Board).\1801\
---------------------------------------------------------------------------
\1799\ The agencies explained in the NPRM that some amount of
this difference was due to the rebound effect, and that ``non-
rebound'' VMT between alternatives differed by as much as 0.4
percent. See 83 FR at 43099 (Aug. 24, 2018).
\1800\ Environmental Group Coalition, Appendix A, NHTSA-2018-
0067-12000, at 180.
\1801\ See, e.g., id.; EDF, Appendix B (Rykowski comments),
NHTSA-2018-0067-12108, at 42-46; IPI, Appendix, NHTSA-2018-0067-
12213; at 79; CARB, Detailed Comments, NHTSA-2018-0067-11873, at
237-242.
---------------------------------------------------------------------------
More generally, commenters argued that non-rebound VMT should be
held constant across regulatory alternatives, regardless of the number
of new vehicles sold and registered vehicles scrapped. For example, CBD
commented that the ``total number of VMT should be determined based on
demand for travel, not arbitrarily driven by fleet size.'' CARB added
that fleet size can change across the alternatives ``as long as the VMT
schedules are adjusted to account for overall travel activity that is
distributed over a larger number of vehicles.'' \1802\ NCAT, Global,
Auto Alliance, EDF, IPI, and Honda made similar arguments.\1803\
---------------------------------------------------------------------------
\1802\ CARB, Detailed Comments, NHTSA-2018-0067-11873, at 238
(internal citation omitted).
\1803\ See, e.g., Global, Attachment A, NHTSA-2018-0067-12032,
at A-26-A-30; NCAT, Comments, NHTSA-2018-0067-11969, at 28-32; EDF,
Appendix A, NHTSA-2018-0067-12108, at 30; IPI, Appendix, NHTSA-2018-
0067-12213, at 80-85; Honda, NHTSA-2018-0067-12111.
---------------------------------------------------------------------------
While commenters generally provided few specific recommendations
about the level to which VMT should be constrained among alternatives,
several of them argued that VMT projections would benefit from
consideration of travel demand modeling. UCS, CBD, NCAT, and others
suggested that the overall level of light-duty VMT in a given year
should reflect the broader economic context in which travel
occurs.\1804\ For example, Honda stated, ``[i]ncreasing VMT is closely
associated with increased economic activity.'' \1805\
---------------------------------------------------------------------------
\1804\ See, e.g., NCAT, Comments, NHTSA-2018-0067-11969, at 31-
32; Environmental Group Coalition, Appendix A, NHTSA-2018-0067-
12000, at 175-76; and, UCS, Technical Appendix, NHTSA-2018-0067-
12039, at 60-61.
\1805\ Honda, Supplemental Analysis, NHTSA-2018-0067-1211, at 4.
---------------------------------------------------------------------------
The agencies agree that the total demand for VMT should not vary
excessively across alternatives and stated as much in the NPRM.\1806\
That said, it is reasonable to assume that fleets with differing age
distributions and inherent cost of operation will have slightly
different annual VMT, absent VMT associated with rebound miles;
however, the difference could conceivably be small. To address these
comments and to take an intentionally conservative approach, the
agencies decided to constrain ``non-rebound'' VMT (defined more
explicitly below) to be identical across regulatory alternatives in
this analysis using the FHWA VMT demand model to determine the
constraint; therefore, the only difference in total VMT between
regulatory alternatives is the rebound miles attributable to
differences in fuel economy resulting from the regulatory alternatives.
Nevertheless, as explained in the NPRM and revealed in the extensive
quantitative results published with the NPRM, setting aside the rebound
effect, aggregate VMT as estimated in the NPRM was roughly constant
across alternatives. Although differences may have appeared large in
absolute terms, especially when aggregated across many calendar years
and ignoring the underlying annual total quantities, the differences
were nevertheless very small in relative terms--small enough to be well
within the range of measurement or estimation error for virtually any
of the other inputs to, or outputs of, the agencies' analysis. It is
unclear whether a 0.4 percent change in highway travel can be measured
with any degree of confidence.
---------------------------------------------------------------------------
\1806\ See 83 FR at 43099 (Aug. 24, 2018).
---------------------------------------------------------------------------
To constrain non-rebound VMT, the agencies needed to create a
definition of non-rebound VMT and a method for calculating it. The
agencies used the FHWA VMT forecasting model to produce a forecast of
non-rebound VMT, to which total non-rebound VMT in every regulatory
alternative is constrained in each year, regardless of the fleet size
or distribution of ages in the fleet. In calendar years where total
non-rebound VMT determined by the size of the fleet and assumed usage
of each vehicle is lower than the constraint produced from the FHWA
model, VMT is added to that total and allocated across vehicles to
match the non-rebound forecast (preserving the constraint). These
additional miles are then carried throughout the analysis as vehicles
accrue costs and benefits. Because non-rebound VMT is being held
constant for the FRM analysis across the set of regulatory alternatives
in each calendar year, the only difference in VMT among the
[[Page 24689]]
alternatives in any calendar year results from differences in fuel
economy improvement relative to MY2016 that occur as a result of the
standards. Finally, in Section VII, the agencies calculate the changes
in total VMT attributable to fuel economy, otherwise known as the
rebound VMT.
(i) Defining Non-Rebound VMT
In order to constrain non-rebound VMT, it is first necessary to
define ``non-rebound VMT'' more precisely. The NPRM defined the rebound
effect as the overall elasticity of travel with respect to changes in
the cost per mile (CPM). CPM has two components. The first component of
CPM is fuel prices--the agencies expect vehicles to be driven less if
fuel prices go up, all else equal. The second component of CPM is fuel
economy. Therefore, the NPRM defined the percentage change in CPM, for
a given scenario, model year, and calendar year, as: \1807\
---------------------------------------------------------------------------
\1807\ See 83 FR at 43091 (Aug. 24, 2018).
Equation VI-7--Full change in cost per mile of travel
[GRAPHIC] [TIFF OMITTED] TR30AP20.367
Where FP is fuel price, FE is fuel economy, and REF refers to the
reference FE value of a given age (in particular, FE
2016-(CY-MY), which is the FE of the MY cohort that was
age CY-MY in CY 2016). In the equation above, FESN,MY,CY
refers to the observed fuel economy of the MY cohort (typically
applied at the vehicle level) for a given scenario (SN) in calendar
year CY.
The CAFE model uses one value, the value specified as the rebound
effect, to measure CPM elasticity. Naturally, the CAFE model produces
the same magnitude of change in travel for equivalent changes in fuel
prices and fuel economy. Constructing such a projection of future VMT
(from 2017 to 2050) that sets aside the rebound effect required
constructing inputs that were consistent with that perspective. In
particular, it was necessary to separate the price response associated
with the change in fuel prices relative to the year on which the
agencies based the mileage accumulation schedule (end of CY2016), and
the change in VMT associated with only the improvements in fuel
economy, relative to MY2016, that occur for future model years at the
forecasted fuel price.
As vehicles age, the agencies expect their VMT to decrease in the
presence of a non-zero rebound effect if rising fuel prices over time
increase the per-mile cost of travel, and the rebound effect represents
the degree to which their travel is reduced for a percentage change
increase in operating cost. It is intuitive that, as the cost of fuel
rises over time, a vehicle with a fixed fuel economy would be driven
less if gasoline costs $3.50/gallon than it would be if gasoline costs
$2.50/gallon. Such a response is also consistent with economic
principles (and literature),\1808\ and so it is included in the ``non-
rebound'' VMT that the agencies constrain across alternatives in each
calendar year.
---------------------------------------------------------------------------
\1808\ See, e.g., Goodwin, P., J. Dargay, and M. Hanly.
Elasticities of road traffic and fuel consumption with respect to
price and income: A review. Transport Reviews, 24:275-292, 2004.
---------------------------------------------------------------------------
Similarly, the annual mileage accumulation of cohorts in the
inherited fleet is clearly affected by fuel price, but also by
evolution. Setting aside any fuel economy improvements in vehicles sold
and entering the on-road fleet between 2017 and 2050, the average fuel
economy of each age cohort is going to improve over that period. The
travel behavior of the on-road fleet was last observed through calendar
year 2016 in the Polk data (discussed in (a)(ii)), when a 20-year-old
car was part of the model year 1997 cohort, and had an average fuel
economy of 23.4 MPG. However, the fleet continually turns over. In
2035, the 20-year-old car will be a member of the model year 2016
cohort, and have an average fuel economy of 29.2 MPG (assumed to be the
average fuel economy of MY2016 vehicles when they were new).\1809\ If
fuel prices persist at 2016 levels (in real dollars), then that 25
percent improvement in fuel economy would reduce the cost per mile of
travel for 20-year-old vehicles relative to the observed values in
calendar year 2016, and lead to an increase in travel demand for
vehicles of that age. Importantly, this transition to more efficient
age cohorts occurs in all of the regulatory alternatives. Considering
only the fuel economy levels of vehicles that exist prior to the first
year of simulation (2017), a secular improvement in the fuel economy of
the on-road fleet would occur with no further improvements in fuel
economy from new vehicles in model years 2017 to 2050. As the fleet
turns over, its fuel efficiency will gradually resemble that of the
model year 2016 cohort, up to the point at which each age cohort is as
efficient as the model year 2016 cohort.\1810\
---------------------------------------------------------------------------
\1809\ In practice, vehicles will scrap at different rates over
time, even within a body-style. Some nameplates and manufacturers
have reputations for longevity and individual vehicle models with
different fuel economies may seem like better candidates for repairs
under particular fuel price scenarios. In light of this, the fuel
economy for a given body-style will likely not continue to be the
sales-weighted average fuel economy when the cohort was new, even
without accounting for degradation and changes to the on-road gap
over time. The agencies make this assumption here out of necessity.
\1810\ Vehicles scrap at different rates over time, and there
are important differences by body style for both scrappage rates and
mileage accumulation. This discussion is intended to provide
intuition, without all of the computational nuance that exists in
the model's implementation.
---------------------------------------------------------------------------
The notion of ``non-rebound'' VMT is a construct necessary to
support this regulatory analysis by controlling for VMT attributable to
reasons other than rebound driving, but present only in theory. Using
our symmetrical definition of rebound to represent the expected
response to changes in CPM, regardless of whether those changes occur
as a result of changes in fuel price or fuel economy, it is well
established that demand for VMT responds to the cost of travel. To
isolate the change in VMT for which the regulatory alternatives are
responsible, the agencies have also included the VMT attributable to
secular fleet turnover (through MY2016) in the total ``non-rebound''
VMT projection. In particular, this means that the conventional rebound
definition used in previous analyses, is replaced in the ``non-
rebound'' VMT estimation with a more limited definition:
Equation VI-8--Fuel price and secular improvement component of
elasticity
[[Page 24690]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.368
Where FP is fuel price, FE is fuel economy, and REF refers to the
reference FE value of a given age (in particular, FEREF =
FE 2016-(CY-MY), which is the average FE of the MY cohort
that was age (CY-MY) in CY 2016). In Equation VI-8,
FEMIN(2016,MY)
refers to the observed fuel economy of the model year being evaluated
up to and including the 2016MY cohort. This construction explicitly
accounts for the improvement in fuel economy between MY2016 and all the
historical ages (through MY1977) with respect to the change in (real)
fuel price relative to calendar year 2016. Thus, the VMT associated
with the rebound effect in this analysis only accounts for changes to
CPM that result from the amount of fuel economy improvement that occurs
relative to MY2016. The full elasticity definition (in Equation VI-7)
differs from that in Equation VI-8 in only one way; the fuel economy in
the denominator of the first term is the fuel economy of the model year
being evaluated, rather than being the minimum of the actual model year
and model year 2016.
Combining this demand elasticity with the endogenously estimated
vehicle population and the mileage accumulation schedule provides an
initial estimate of non-rebound VMT, as in Equation VI-9.
Equation VI-9--Unadjusted total non-rebound VMT in a calendar year
[GRAPHIC] [TIFF OMITTED] TR30AP20.369
In Equation VI-9, VMT represents the non-rebound mileage
accumulation schedule (by age, A, and body style, S), Population is the
on-road vehicle population simulated by the CAFE Model (in calendar
year CY, for each age, A, and body style, S), [egr] is the elasticity
of demand for travel (the rebound effect, assumed to be -0.2 in this
analysis).
However, there are factors beyond the CPM that affect light-duty
demand for VMT. The FHWA VMT forecasting model includes additional
parameters that can mitigate or increase the magnitude of the effect of
fuel price changes on demand for VMT. In particular, the model accounts
for changes to per-capita personal disposable income (and U.S.
population) over time. This means that even if fuel prices are
increasing over the study period (as they are in the central case), and
fleetwide fuel economy improves only through fleet turnover (as it does
in the simulated ``non-rebound'' case), total demand for VMT can still
grow as a result of increases in these other relevant factors. Not only
does the forecast of non-rebound VMT continue to grow in the non-
rebound case, it does so at a faster rate than Equation VI-9 produces.
Thus, in order to preserve non-rebound VMT in a way that represents
expected VMT demand, the agencies must constrain non-rebound VMT in
each alternative to match the forecast produced by the FHWA model using
the fuel price series from the central analysis, AEO2019 Reference case
assumptions for per-capita personal disposable income, and fleetwide
fuel economy values produced by simulating the effect of fleet turnover
(only) in the CAFE model.\1811\
---------------------------------------------------------------------------
\1811\ Non_rebound_VMT_forecasting.xls in Docket No. NHTSA-2018-
0067.
---------------------------------------------------------------------------
Constraining Non-Rebound VMT
For this final rule, total `non-rebound' VMT is calculated for each
calendar year and reported in Section VI.D.1.b)(5)(d). In any future
calendar year, ``non-rebound'' VMT is calculated as a product of the
initial CY2017 total and a series of compound growth rates:
Equation VI-10--Total non-rebound VMT constraint in each calendar year
[GRAPHIC] [TIFF OMITTED] TR30AP20.370
Where CY is calendar year, r is the compound annual growth rate
(unique to each CY), and TotalVMT is the calendar year total light-
duty VMT estimated by the CAFE Model using the annual VMT for each
body style and age in the mileage accumulation schedule (defined in
Table VI-180), the population of each age/style cohort in CY2017,
and the initial difference between operating costs in 2016 and 2017.
The compound annual growth rates, rCY, in Equation VI-10
are derived from the inter-annual differences in the forecast of
total non-rebound VMT that the agencies created using the FHWA
model.
The agencies used the FHWA forecasting model to produce two
distinct VMT forecasts (both of which appear in Table VI-182). The
first of these is identical to the forecast of total VMT reported in
Table VI-181, and represents the AEO2019 Reference case assumptions
with the exception of average on-road fuel economy, which was simulated
using the CAFE model to simulate new vehicle fuel economy, new vehicle
sales, and vehicle retirement under the baseline standards. The
forecast in the second column of Table VI-182 is identical to the
first, except that the average on-road fuel economy accounts for only
the effect of fleet turnover on fuel economy
[[Page 24691]]
improvements (new vehicles are assumed to be only as fuel efficient as
the MY2016 cohort, discussed above).
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.371
[[Page 24692]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.372
BILLING CODE 4910-59-C
The third column is the non-rebound VMT constraint produced by the
CAFE model, to which non-rebound VMT is constrained to in every
regulatory alternative (under central analysis assumptions regarding
fuel prices and economic growth). The non-rebound VMT constraint is
produced endogenously by the model in each run based on the estimated
VMT for calendar year 2017 and a series of growth rates intended to
reproduce the general growth trend in light-duty VMT under the set of
``non-rebound'' assumptions in the FHWA model (Equation VI-10).\1812\
It differs from the ``non-rebound'' forecast produced by the FHWA model
by one to three percent in any year. This adjustment was both an
attempt to match the FHWA model's projection of total VMT (including
rebound) in the baseline, and an acknowledgment that differing levels
of modeling resolution and construction are likely to produce slightly
different projections. In general, the one to three percent difference
in non-rebound VMT is within the range of projections based on the
confidence intervals of the coefficients that define the FHWA
forecasting model.
---------------------------------------------------------------------------
\1812\ This ensures internal consistency with the set of
assumptions provided by the user, but can lead to differences
between the non-rebound VMT constraint in the central analysis and
one that is generated under a different set of assumptions (as in
the sensitivity analysis, for example).
---------------------------------------------------------------------------
[[Page 24693]]
The fourth column in Table VI-182 represents the unadjusted ``non-
rebound'' VMT produced by the CAFE Model using Equation VI-9. The
reader will observe that in every calendar year, this total is lower
than the non-rebound VMT constraint. This occurs because the projected
fuel prices in the central analysis increase much faster than the
fleetwide fuel economy (in the non-rebound case). This increases CPM
and, as a consequence, reduces demand for VMT based on the price
elasticity of demand for travel (rebound effect). However, the FHWA
model accounts for additional variables that recognize the economic
context in which this fuel price projection occurs. In particular, the
model accounts for changes in the U.S. (human) population and changes
to personal disposable income over the same period. These factors act
to attenuate the demand response to rising fuel prices, producing a
rising demand for VMT even as the CPM rises for several years.
In order to constrain non-rebound VMT to be identical in each year
across regulatory alternatives, it is necessary to add VMT to the
unadjusted total, endogenously calculated by the CAFE Model in each
calendar year. These additional miles, denoted [Delta]miles for this
discussion, represent the simple difference between the annual VMT
constraint (column 3 of Table VI-182) and the unadjusted VMT defined in
Equation VI-9 (above) in each calendar year.
[GRAPHIC] [TIFF OMITTED] TR30AP20.373
Because each regulatory scenario produces a unique on-road fleet
(in terms of the number of vehicles, the distribution of ages among
them, and the resulting distribution of fuel economies), the total
unadjusted VMT in each calendar year (given by Equation VI-9) will be
unique to each regulatory scenario. As a corollary,
[Delta]milescy will also be unique to each regulatory
scenario. By distributing [Delta]milescy across the vehicle
fleet in each calendar year, the CAFE Model scales up the unadjusted
non-rebound VMT to equal the non-rebound VMT constraint in each
calendar year, for each regulatory alternative. While there are a
number of ways to reallocate [Delta]milescy across the on-
road fleet in order to match the non-rebound VMT constraint, the fact
that unadjusted VMT is always lower suggests an obvious approach.
The primary goal of reallocation is to adjust total non-rebound VMT
so that it is identically equal to the VMT constraint in every calendar
year for each regulatory alternative, while conserving the general
trends of the mileage accumulation schedule--which represents a good
estimate of observed usage at the start of the simulation. In
particular, the reallocation approach should preserve the basic ideas
that annual mileage decreases with vehicle age because newer (and more
efficient) vehicles are more likely to be driven additional miles than
their older counterparts, and mileage accumulation varies by body
style. To accomplish the reallocation, the CAFE Model computes a ratio
that varies by body style, calendar year, and regulatory alternative.
The ratio captures the share of additional VMT that can be absorbed by
the registered vehicle population of each body style based on their
relative representation in the fleet, so that per-vehicle totals across
ages remain sensible (even if the distribution of body styles should
change over time as the new vehicle market evolves). Then this quantity
is further scaled by the total VMT for a given body style in the
calendar year for which [Delta]miles has been computed. The resulting
ratio is then used to scale the unadjusted miles from Equation VI-9, so
that the new sum of annual (non-rebound) VMT across all of the vehicles
in the on-road fleet equals the constraint. For a single calendar year,
CY, and a single body style, S, the scaling ratio, R, is computed as:
[GRAPHIC] [TIFF OMITTED] TR30AP20.374
In Equation VI-12, Population, refers to the on-road vehicle
population for a given age and body style (summed over the full range
of ages in the simulation, where vehicles are modeled to survive for,
at most, forty years). The fraction in the numerator calculates the
fleet composition by body type.\1813\ As long as the unadjusted non-
rebound VMT produced by the CAFE Model is smaller than the VMT
constraint for all years and regulatory alternatives (and it is), this
scaling ratio allows the CAFE Model to add miles to the annual total in
a way that preserves the basic ideas of the mileage accumulation
schedule and achieves equality with the constraint. In particular, the
total adjusted non-rebound VMT is then calculated as:
---------------------------------------------------------------------------
\1813\ We also considered basing this ratio on each body style's
share of total VMT in that calendar year. However, that approach has
the potential to result in allocations that add (or remove) too many
miles per vehicle, depending on the age distribution and size of
each body style cohort. While that approach better preserves the age
distribution of VMT within a style, capturing the differences in age
distribution of the population in each scenario is an objective of
the VMT accounting. In testing, the differences in approach were
small (about 0.1 percent difference).
---------------------------------------------------------------------------
[[Page 24694]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.375
To make each alternative match the VMT constraint, Equation VI-13
allocates miles (in this case, adds) to each vehicle in a calendar year
by multiplying the product of the mileage accumulation schedule (for
that style vehicle, at that age), the %[Delta]NrbdCPM (described in
Equation VI-8), and the elasticity (the rebound effect of -0.2) with
the appropriate scaling ratio (defined in Equation VI-12). The
``Allocated Miles'' in Table VI-176 are the result of this calculation
for a passenger car in CY2020.
Unlike some of the accounting, which focuses on the impacts to a
model year cohort of vehicles over the course of its useful life, the
rebound constraint and reallocation are calendar year concepts. The
constraint represents demand for VMT absent ``rebound miles'' (defined
more explicitly above) in a specific calendar year. Thus, this
reallocation occurs in every calendar year, and a vehicle of a model
year cohort will likely experience many of these reallocation events
during its simulated useful life. The resulting survival weighted
mileage accumulation is discussed in detail in the discussion of VMT
Resulting From Simulation found in Section (d), but an example of the
annual reallocation is provided here.
In the baseline alternative, the non-rebound VMT constraint in
CY2020 is about 3.068T miles, but the endogenously computed ``non-
rebound'' VMT is only 2.955T miles. This creates a difference,
[Delta]miles2020, of 112.6B miles that must be added to the
total unadjusted non-rebound VMT in calendar year 2020 and allocated
across the on-road fleet in that year to preserve total non-rebound
VMT. Over time, this discrepancy between the FHWA model's projection
and the unadjusted total non-rebound VMT grows to about 230 billion
miles. While the other classes operate identically, this example uses
the reallocation that occurs to passenger cars to illustrate the
mechanics of reallocation. Rising fuel prices depressing non-rebound
VMT (relative to the mileage schedule) over time is a general trend
that emerges for all body styles, as shown for passenger cars in Table
VI-183.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.376
[[Page 24695]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.377
BILLING CODE 4910-59-C
The number of miles added to each age vehicle is generally less
than the difference between the unadjusted non-rebound VMT (for a given
age) and the mileage schedule. Thus, adding the requisite miles to each
age does not distort either the shape of the schedule with age, nor
does it create annual usage estimates that are out of line with
observed usage. The example shown here uses the baseline alternative to
illustrate the reallocation of VMT in 2020, but this reallocation
differs by alternative. In less stringent regulatory alternatives, new
vehicles are less expensive; this increases new vehicle sales and
accelerates the retirement of older vehicles (relative to the
baseline). In those cases, the unadjusted non-rebound VMT is higher,
[Delta]miles smaller, and corresponding allocation of [Delta]miles
smaller--though still consistently positive.
Commenters encouraged us to use a demand model to avoid creating
unrealistic VMT projections that failed to account for factors that
exogenously influence total demand for VMT, which the agencies have
done here.\1814\ Had baseline case been used instead, regardless of
whether it happens to be the most or least stringent alternative, as
the non-rebound VMT constraint, both the non-rebound VMT and VMT with
rebound would have differed meaningfully from both other government
forecasts and from the projections produced by the demand models
underlying those forecasts. By producing and enforcing a non-rebound
constraint based on results from a travel demand model, the agencies
ensure realism in the projections of total VMT under each regulatory
alternative and ensure that the costs and benefits associated with
rebound VMT result only from fuel economy improvements in the
regulatory alternatives considered.
---------------------------------------------------------------------------
\1814\ See, e.g., NCAT, Comments, NHTSA-2018-0067-11969, at 31-
32; Environmental Group Coalition, Appendix A, NHTSA-2018-0067-
12000, at 175-76; UCS, Technical Appendix, NHTSA-2018-0067-12039, at
59; Honda, Supplemental Analysis, NHTSA-2018-0067-1211, at 4.
---------------------------------------------------------------------------
(d) VMT Resulting From Simulation
This section has already demonstrated that total VMT projections
from the simulation are consistent with FHWA projections of total light
duty VMT using the same set of economic assumptions. Lifetime mileage
accumulation is now a function of the sales model, scrappage model,
mileage accumulation schedules (described in Table VI-180), and the
redistribution of VMT across the age distribution of registered
vehicles in each calendar year to preserve the non-rebound VMT
constraint.
The definition of ``non-rebound'' VMT in this analysis determines
the
[[Page 24696]]
additional miles associated with secular fleet turnover and fuel price
changes. Conversely, rebound miles measure the VMT difference due to
fuel economy improvements relative to MY2016 (independent of changes in
fuel price, or secular fleetwide fuel economy improvement resulting
from the continued retirement of older vehicles and their replacement
with newer ones). In order to calculate total VMT with rebound, the
agencies apply the rebound elasticity to the full change in CPM and the
initial VMT schedule, but apply the rebound elasticity to the
incremental percentage change in CPM between the non-rebound and full
CPM calculations to the miles applied to each vehicle during the
reallocation step that ensured adjusted non-rebound VMT matched the
non-rebound VMT constraint.
[GRAPHIC] [TIFF OMITTED] TR30AP20.378
Where VMTA,S is the initial VMT schedule by age and body-style,
%[Delta]NonReboundCPM and %[Delta]CPM are defined in Equation VI-8
and Equation VI-7, respectively, and [Delta]MilesA,S,CY is the per-
vehicle miles added by the reallocation described in Equation VI-13.
The additional miles that are added to each vehicle in the
reallocation step ([Delta]MilesA,S,CY) are multiplied by the
difference between the percentage changes in CPM (full and non-
rebound, respectively) because the %[Delta]NonRbdCPM was used to
derive the allocated miles and using the full CPM change to scale
the allocated miles would count that change twice. Taking the
difference avoids overestimating the total mileage in the presence
of the rebound effect. The ``rebound miles'' will be the difference
between Equation VI-14 and Equation VI-10 for each alternative. To
the extent that regulatory scenarios produce comparable numbers of
rebound miles in early calendar years, the impacts associated with
those miles net out across the alternatives in the benefit cost
analysis.
BILLING CODE 4910-59-P
Table VI-184 displays the annual survival-weighted VMT at each age
of a MY2025 vehicle, by regulatory class including and reallocation
needed to preserve the VMT constraint and all rebound miles (using a 20
percent rebound effect).\1815\
---------------------------------------------------------------------------
\1815\ Annual survival-weighted VMT is calculated by dividing
the annual VMT of a MY cohort by the total population of the cohort
purchased. As such, Table VI-183 and Table VI-184 report different
types of values.
[GRAPHIC] [TIFF OMITTED] TR30AP20.379
[[Page 24697]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.380
BILLING CODE 4910-59-C
As earlier portions of this section have shown, the second decade
of useful life now shows significantly higher utilization than the NPRM
analysis for both passenger cars and light trucks. While the current
lifetime accumulation is similar to the values produced in the 2012
final rule, those values were simulated to occur under fuel prices that
were consistently 40 percent higher than the prices in this analysis
(when adjusted for inflation).\1816\ Under comparable prices, lifetime
mileage accumulation would have been considerably higher.
---------------------------------------------------------------------------
\1816\ The 2012 final rule also assumed a 10 percent rebound
effect, which would have further affected lifetime mileage
accumulation.
---------------------------------------------------------------------------
(e) Sales, Scrappage and VMT Integration
The VMT construct described above, while an improvement over the
version presented in the NPRM for the reasons explained, does not
represent the fully integrated model of ownership, usage, and
retirement decisions that some commenters argued would be preferred or
even required to assess properly the impacts of CAFE/CO2
standards. In particular, RFF commented that integrating sales,
scrappage and VMT would ``make the analysis internally consistent and
will account for the fact that households do not make scrappage and
vehicle use decisions in isolation.'' \1817\ IPI concurred and expanded
in their comment, stating `` `a
[[Page 24698]]
unified model of vehicle choice and usage' is necessary.'' \1818\
---------------------------------------------------------------------------
\1817\ RFF, Comments, NHTSA-2018-0067-11789 at 14.
\1818\ IPI, Appendix, NHTSA-2018-0067-12213, at 80 (internal
citation omitted).
---------------------------------------------------------------------------
The implication of such commenters is that the agencies have
ignored important benefits of more stringent standards by not
explicitly considering household decisions at the level of household
vehicle fleet management. However, the opposite may be true. A recent
National Bureau of Economic Research (``NBER'') paper finds that
households engage in attribute substitution while managing the set of
attributes in their vehicle portfolios.\1819\ In particular, the
authors argue that attribute substitution within a household's vehicle
portfolio may erode up to 60 percent of the intended fuel economy
benefits of the footprint-based CAFE/CO2 standards, as the
higher fuel economy of owned vehicles reduces demand for efficiency in
the next bought vehicle, all else equal. This suggests that examining
effects at the household level may not be as beneficial, or as
meaningful, as some commenters might hope.
---------------------------------------------------------------------------
\1819\ Archsmith, J., Gillingham, K., Knittel, C., Rapson, D.
(Sept. 2017), Attribute Substitution in Household Vehicle
Portfolios. NBER Working Paper No. NBER Working Paper No. 23856.
Available at https://www.nber.org/papers/w23856 (last accessed Feb.
4, 2020).
---------------------------------------------------------------------------
While commenters have suggested ambitious models of dynamic
relationships at the household level, moreover, it is not clear that
such a model is currently possible. Capturing the heterogeneous
preferences of households across purchase, usage, and retirement
decisions at the same level of detail required to produce meaningful
estimates of regulatory compliance costs is beyond the current scope of
this analysis. While the agencies agree that expected usage influences
the household decision of which vehicle to purchase, how long to hold
it, and how to manage the usage and retirement of other vehicles within
a household fleet, the agencies do not agree that such a detailed model
is a necessary prerequisite to assess the impacts of CAFE and tailpipe
CO2 emissions standards, nor that it is necessarily
appropriate to do so given that the agencies are examining aggregate
national fleetwide effects of such standards. Furthermore, in the most
recent peer review of the CAFE Model, one reviewer remarked that while
the sales and VMT would benefit from a household choice model, ``the
decision to scrap a vehicle (remove it from the national in-use fleet)
and the decision to purchase a new vehicle often are not made by the
same household. No U.S. national-level transportation demand models
(that this reviewer is aware of) tackle the issue with this level of
complexity.'' \1820\
---------------------------------------------------------------------------
\1820\ CAFE Model Peer Review, DOT HS 812 590, Revised (July
2019), pp. B19-B29, available at https://www.regulations.gov/contentStreamer?documentId=NHTSA-2018-0067-0055&attachmentNumber=2&contentType=pdf.
---------------------------------------------------------------------------
Each iteration of these regulatory analyses has endeavored to
improve the accuracy and breadth of modeling to capture better the
relevant dynamics of the markets affected by these policies. The
agencies intend to address current limitations in future rulemakings,
and meanwhile believe that the scope of the current analysis is
reasonable and appropriate for informing decision-makers as to the
effects of different levels of CAFE and tailpipe CO2
emissions stringency.
(6) What is the mobility benefit that accrues to vehicle owners?
(s) Mobility Benefits in the NPRM Analysis
As the proposal noted, the increase in travel associated with the
rebound effect provides benefits that reflect the value to drivers and
other vehicle occupants of the added--or more desirable--social and
economic opportunities that become accessible with additional travel.
The fact that drivers and their passengers elect to make more frequent
or longer trips to gain access to these opportunities when the cost of
driving declines demonstrates that the benefits they gain by doing so
exceed the costs they incur, including the economic value of their
travel time, fuel and other vehicle operating costs, and the economic
cost of safety risks drivers assume. The amount by which the benefits
of this additional travel exceeds its economic costs measures the net
benefits drivers and their passengers experience, usually referred to
as increased consumer surplus.
Under the proposal, the fuel cost of driving each mile would have
increased as a consequence of the lower fuel economy levels it
permitted, thus reducing the number of miles that buyers of new cars
and light trucks would drive as the well-documented fuel economy
rebound effect operates in reverse.\1821\ The agencies' analysis of the
proposed rule described the resulting loss in consumer surplus, and
calculated its annual value using the conventional approximation, which
is one half of the product of the increase in vehicle operating costs
per vehicle-mile and the resulting decrease in the annual number of
miles driven. Because the value of this loss depends on the extent of
the change in fuel economy, it varied by model year, and also differed
among the alternative standards that the NPRM considered.
---------------------------------------------------------------------------
\1821\ Normally, the fuel economy rebound effect refers to an
increase in vehicle use that results when increased fuel economy
reduces the fuel cost for driving each mile.
---------------------------------------------------------------------------
The agencies' analysis specifically recognized that the economic
value of any additional travel prompted by the fuel economy rebound
effect must exceed the additional fuel costs drivers incur, plus the
economic cost of safety risks they and their passengers assume.\1822\
Thus, when vehicle use was projected to decline in response to lower
fuel economy, the agencies noted that the resulting loss in benefits
must have more than offset both the savings in fuel costs and the value
of drivers' and passengers' reduced exposure to safety risks. In the
accounting of benefits and costs for the preferred alternative, the
loss of benefits associated with reduced mobility was recognized by
reporting losses in travel benefits that exactly offset the value of
reduced risks of being involved in both fatal and non-fatal crashes.
---------------------------------------------------------------------------
\1822\ Although it did not attempt to estimate operating costs
other than those for fuel or the value of drivers' and passengers'
travel time, the benefits from any additional travel that occurs
voluntarily must also at least compensate for these costs.
---------------------------------------------------------------------------
In addition, the accounting reported a loss in mobility benefits
from reduced use of new cars and light trucks, which included a
component that exactly offset the fuel savings from reduced driving,
together with the loss in consumer surplus that foregone travel would
otherwise have provided. Including this first component was necessary
to offset the fact that the savings in fuel costs had already been
recognized elsewhere in the accounting, by deducting those savings from
the increase in fuel costs resulting from lower fuel economy to arrive
at the reported net increase in fuel costs. Thus, the resulting value
of the net loss in travel benefits was exactly equal to the loss in
consumer surplus that any travel foregone in response to higher fuel
costs would otherwise have provided.
(b) Comments on the Agencies' Treatment of Mobility Benefits in the
NPRM
The agencies received only two comments referring to their
treatment of mobility benefits in the analysis supporting the proposed
CAFE and CO2 standards. The California Air Resources Board
(CARB) noted that the accounting of benefits and costs resulting from
the proposal included losses in mobility benefits that offset the
reduction in fatality costs related to the decline in
[[Page 24699]]
new vehicle use from the fuel economy rebound effect. While CARB did
not comment on the agencies' inclusion of losses in mobility benefits
in their accounting, it did object to the fact that the agencies also
reported the numerical change in fatalities that could be ascribed to
the rebound effect, and considered the improvement in safety it
reflected when selecting their proposed alternative.\1823\ Similarly,
the Institute for Policy Integrity (IPI) termed the agencies' reliance
on the estimated change in the number of fatalities as partial
justification for selecting their preferred alternative as arbitrary,
while at the same time arguing that the reduction in driving due to the
rebound effect had no net welfare impact.\1824\
---------------------------------------------------------------------------
\1823\ California Air Resources Board (CARB), NHTSA-2018-0067-
11873, at pp. 121.
\1824\ Institute for Policy Integrity (IPI), NHTSA-2018-0067-
12213, at pp. 11. In fact, the agencies did not treat the reduction
in driving as having no net impact on welfare, since as explained
immediately above, the loss in consumer surplus benefits on the
foregone driving was not accompanied by any offsetting cost savings.
Therefore, the decline in driving in response to the rebound effect
resulted in a net loss in welfare.
---------------------------------------------------------------------------
In response to these comments, the agencies observe that
considering changes in the actual number of fatalities as well as the
welfare effects of changes in drivers' and passengers' exposure and
valuation of the risks of being involved in fatal crashes represents a
sound approach to assessing the impacts of proposed CAFE and
CO2 standards. The safety implications of alternative future
standards are clearly a legitimate and highly visible consequence for
the agencies to consider when evaluating their relative merits, as are
the implications of changes in the safety risks for the economic
welfare of car and light truck users. Thus the agencies see no
inconsistency or duplication in separately considering both factors as
part of their assessment of alternative future standards.
(c) Mobility Benefits in the Final Rule
The analysis supporting this final rule continues to treat losses
in mobility benefits in the same manner the agencies previously did
when analyzing the alternatives considered for the proposed rule.
Because there are several subtleties in this treatment, Figure VI-75 is
included below to clarify its details. In the figure, the demand curve
shows the relationship of annual use of new cars (and light trucks),
which can be thought of as their total or average annual vehicle-miles
driven, to the cost per mile of driving.
[GRAPHIC] [TIFF OMITTED] TR30AP20.381
The initial cost per mile OC0 consists of the per mile
economic costs of the risks of being involved in fatal and non-fatal
crashes, shown by the heights of Og and gd on the vertical axis,
together with per-mile fuel costs at the baseline level of fuel
economy, the height of segment dC0.\1825\ Annual miles
driven at this initial per-mile cost are shown by the distance
OM0 on the horizontal axis in Figure VI-75. When fuel
economy declines from its baseline level under one of the regulatory
alternatives considered, fuel costs per mile increase from
dC0 to dC1, but the per-mile economic costs of
crash risks (both fatal and non-fatal) are unaffected, so total costs
per mile driven rise to OC1. In response to this increase in
the per-mile fuel and total cost of driving, annual use declines to
OM1.
---------------------------------------------------------------------------
\1825\ Per-mile fuel costs are equal to the dollar price of fuel
per gallon, divided by fuel economy in miles per gallon. For
simplicity, this figure omits non-fuel operating costs, vehicle
maintenance and depreciation, and the value of occupants' travel
time. Including them would not change the analysis.
---------------------------------------------------------------------------
The resulting loss in total benefits when vehicle use declines from
OM0 to OM1 is the trapezoidal area
M1acM0, but most of this loss is offset by cost
savings from reduced driving, so the net welfare loss is considerably
smaller. Specifically, the rectangle M1hiM0
represents a reduction in the total economic costs of the risk that
drivers and passengers will be involved in fatal crashes when the
decline in driving
[[Page 24700]]
reduces their exposure to that risk. The dollar value of this area thus
appears in the agencies' accounting of costs and benefits as both a
benefit from that reduction in risk and an exactly offsetting loss in
benefits from reduced mobility. The same is true of the rectangle hefi,
the dollar value of which corresponds to both the reduction in the
economic cost of non-fatal crash risks and an identical loss in
mobility benefits.
Total fuel costs for driving OM0 miles are initially the
rectangular area dC0cf, and the decline in driving to
OM1 that results as per-mile fuel and total driving costs
rise changes total fuel costs to the rectangle dC1ae.
Because these two areas share rectangle dC0be, the net
change in fuel costs reported in the agencies' accounting consists of
the dollar value of rectangle C0C1ab, minus that
of rectangle ebcf. The economic value of the loss in mobility benefits
the agencies report in their accounting is the trapezoid eacf, but part
of that area consists of rectangle ebcf, and is thus exactly equal to
the savings in fuel costs from reduced driving. Since this savings has
been already incorporated in the reported change in total fuel costs,
and it offsets part of the reported loss in mobility benefits, leaving
only the loss in consumer surplus that travelers would otherwise have
experienced on foregone reduced driving, the value of triangle bac, as
the net loss in mobility benefits.\1826\
---------------------------------------------------------------------------
\1826\ Thus the change in driving is not welfare-neutral, as IPI
asserted in the comment cited previously; instead, it results in a
net loss in welfare.
---------------------------------------------------------------------------
This discussion assumes that drivers correctly estimate and
consider--or ``internalize''--the risks of being involved in both fatal
and non-fatal crashes that are associated with their additional
driving. However, as is noted in the discussion of the potential
effects of the rule on the mass of vehicles and its resulting impact on
safety, consumers may value safety risks imperfectly. This possibility
is accounted for in the final rule analysis by assuming the portion of
the added safety risk that consumers internalize to be 90 percent. In
Figure VI-75 above, this would be reflected by including a total social
cost per mile that is higher than the C0 and C1
values for the baseline and reduced MPG cases shown in the graphic by
10 percent of the combined cost of fatal and non-fatal crash risks (the
distance Od on the figure's vertical axis), while reducing the costs of
safety risks that drivers do consider to 90 percent of the values
shown. The higher social costs would offset a portion of the consumer
surplus associated with additional mobility (in each case), and result
in a small ``deadweight loss'' over the region where the social cost of
driving exceeds the demand curve. These impacts are also fully
accounted for in the final rule analysis.
(7) What is the sales surplus that accrues to vehicle owners?
Buyers who would not have purchased new models with the baseline
standards in effect but decide to do so in response to the changes in
new vehicles' prices with less demanding standards in place will also
experience increased welfare. Collective benefits to these ``new''
buyers are measured by the consumer surplus they receive from their
increased purchases.
At the proposed rule stage, the agencies elected to exclude the
consumer surplus associated with new vehicle purchases because ``it is
not entirely certain that sales of new cars and light trucks [would]
increase in response to [the] proposed action.'' \1827\ Consumer
surplus is a fundamental economic concept and represents the net value
(or net benefit) a good or service provides to consumers. It is
measured as the difference between what a consumer is willing to pay
for a good or service and the market price. OMB circular A-4 explicitly
identifies consumer surplus as a benefit that should be accounted for
in cost-benefit analysis. For instance, OMB Circular A-4 states the
``net reduction in total surplus (consumer plus producer) is a real
cost to society,'' and elsewhere elaborates that consumer surplus
values be monetized ``when they are significant.'' \1828\
---------------------------------------------------------------------------
\1827\ See PRIA at 954.
\1828\ OMB Circular A-4, at 37-38.
---------------------------------------------------------------------------
The decision to exclude consumer surplus for new vehicles at the
proposed rule stage was an error and inconsistent with OMB's guidance
on regulatory analysis. The agencies are confident that lower vehicle
prices, holding all else equal, should stimulate new vehicle sales and
by extension produce additional consumer surplus. That preliminary
decision was also inconsistent with other parts of the agencies'
analysis. For instance, the agencies calculate the lost consumer
surplus associated with reductions in driving owing to the increase in
the cost per mile in less stringent regulatory cases, as discussed in
Section VI.D.3. The surpluses associated with sales and additional
mobility are inextricably linked as they capture the direct costs and
benefits accrued by purchasers of new vehicles. The sales surplus
captures the savings to consumers when they purchase cheaper vehicles
and the additional mobility measures the cost of higher operating
expenses. It would be inappropriate to include one without the other.
The shaded area in Figure VI-76 reflects the consumer surplus
calculated for new vehicle sales. Line C0 reflects the baseline vehicle
cost. The final rule is expected to reduce the cost of light duty
vehicles, as represented by dotted line C '. Consistent with other
sections of the analysis, the agencies assume that consumers value 30
months of fuel savings. Under the final rule, consumers are expected to
experience higher fuel costs than they would under the baseline
scenario, shifting costs from line C ' to line C1. The consumer surplus
is equal to the area under the curve between Q0 and Q1.\1829\
---------------------------------------------------------------------------
\1829\ The exact calculation is 0.5 * the increase in sales *
the reduction in the cost of light duty vehicles net of the
increased fuel cost.
---------------------------------------------------------------------------
[[Page 24701]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.382
(8) Implicit Opportunity Cost
The agencies' central analysis assumes the selling price for new
vehicles will be reduced to fully reflect manufacturers' savings in
technology costs for complying with less stringent CAFE and
CO2 emission standards. Specifically, new car and light
truck prices are assumed to decline by the average savings in
technology costs per vehicle that manufacturers would realize from
complying with the standards this rule establishes, instead of with the
more demanding baseline standards. The agencies' analysis assumes that
under these final standards, attributes of new cars and light trucks
other than fuel economy would remain identical to those under the
baseline standards, so that changes in sales prices and fuel economy
would be the only sources of benefits or costs to new car and light
truck buyers. Furthermore, the agencies recognize that buyers may have
time preferences that cause them to discount the future at higher rates
than the agencies are directed to consider in their regulatory
evaluations. In either case, the agencies' central analysis may
overstate both the net private and social benefits from adopting more
stringent fuel economy and CO2 emissions standards. For
instance, Table VII-93 (Combined LDV Societal Net Benefits for MYs
1975-2029, CAFE Program, 7 percent Discount Rate) shows that the CAFE
final rule would generate $16.1 billion in total social net benefits
using a 7 percent discount rate, but without the large net private loss
of $26.1 billion, the net social benefits would equal the external net
benefits, or $42.4 billion. Therefore, given that government action
cannot improve net social benefits absent a market failure, if no
market failure exists to motivate the $26.1 billion in private losses
to consumers, the net benefits of these final standards are $42.2
billion.
As indicated earlier, EPA's Science Advisory Board urged the
agencies to account for ``consumer preferences for performance and
other vehicle attributes'' in their analysis.\1830\ To explore further
the possibility that the central analysis is incomplete regarding the
consumer benefits of other vehicle attributes, the agencies conducted a
sensitivity analysis using a conservative estimate of this value. In
the proposal, the agencies considered the lost value of other vehicle
attributes in two sensitivity cases that reduced the total consumer
benefit.\1831\ The agencies received several comments suggesting that
the analysis of other vehicle attributes lost could be improved. For
example, CARB commented that the ``analyses do not adequately model how
vehicle values will change in response to improving fuel economy, or
the competing effects of other attributes.'' \1832\ In response to
commenters, the agencies have revised their sensitivity analyses to
model better the impact of the standards on other vehicle attributes.
---------------------------------------------------------------------------
\1830\ SAB at 10.
\1831\ See PRIA at 954. See also, PRIA at 1539.
\1832\ CARB, Detailed Comments, NHTSA-2018-0067-11873 at 189.
---------------------------------------------------------------------------
The agencies considered, such as they did in the proposal,
offsetting the net private costs associated with enabling more choices
in fuel-saving technologies in a manner similar to rebound driving.
However, the agencies believe that this approach is unnecessary, as
such an analysis would produce nearly identical net benefits to the
external net benefits--which the primary analysis already generates.
Furthermore, given that consumers are free to choose more fuel-
efficient vehicles absent more stringent regulations, consumers who
prefer certain vehicle attributes instead of fuel economy necessarily
value those attributes more than the fuel efficiency technologies they
voluntarily forgo. As such, a sensitivity analysis including a value
for other vehicle attributes should more than offset the net private
costs to consumers from the primary analysis.
For the final rule, instead of keeping the same approach as the
preliminary analysis, the agencies have elected to estimate consumer
benefits of other vehicle attributes in a sensitivity case using
similar logic to that used for the sales and scrappage models. In those
models, the agencies assume that consumers value thirty months of
undiscounted fuel savings. Given this assumption, it would be
reasonable for the agencies then to assume that the value of other
vehicle attributes must be greater than the fuel savings for the
remaining term of the useful life of the vehicle--as these are fuel
economy savings that consumers are clearly
[[Page 24702]]
willing to forgo. The agencies acknowledge that vehicles are typically
sold more than once, but evidence suggests that fuel savings are
capitalized into sales prices in the used car market.\1833\ If this is
the case, new car purchasers would internalize the additional value on
resale owing to fuel efficiency technologies, and the fuel savings over
the remaining useful life less thirty months would be an appropriate
value to use for the value of other vehicle attributes. Nevertheless,
the agencies have elected to be conservative and, instead, opted to use
the fuel savings over the first seventy-two months (less the first
thirty months), which approximates the amount of time the first owner
typically holds a new vehicle.\1834\ This value is referred to as the
``implicit opportunity cost'' of forgoing other vehicle attributes in
favor of increased fuel economy (or using their scarce financial
resources to invest in savings or the purchase of other goods that they
prefer more than fuel economy),\1835\ showing a cost savings for less
stringent alternatives.\1836\ Unlike the sales surplus, which measures
the consumer surplus of new vehicle buyers entering the market, the
implicit opportunity cost contained in this sensitivity case represents
the forgone benefits to consumers the model assumes would have
purchased a vehicle regardless of the standards (but would prefer to
take the upfront cost of fuel economy technologies and invest that
money elsewhere, whether it be on different vehicle attributes or
different goods altogether). These results are shown in Table VII-91
through Table VII-95 (Combined LDV Societal Net Benefits (Accounting
for Implicit Opportunity Cost) for MYs 1975-2029 CAFE Program, 3
percent Discount Rate and 7 percent Discount Rate, as well as the C02
Program, 3 percent Discount Rate and 7 percent Discount Rate).
---------------------------------------------------------------------------
\1833\ For further discussion of the evidence, see section
VI.D.2 of the preamble.
\1834\ There are several reasons why 72 months is an appropriate
approximation. According to a report from the Federal Reserve bank
of Chicago the average new vehicle is owned for over 77 months as of
2015. From the same report, the average new car financing term was
over 67 months in 2016. (https://www.chicagofed.org/publications/working-papers/2019/2019-04; accessed: December 23, 2019). Data from
R.L. Polk suggest that the average new car is held for 71.4 months
(as cited in https://www.autotrader.com/car-shopping/buying-car-how-long-can-you-expect-car-last-240725). State Comptrollers and
Treasurers referred to an IHS Markit report that the average length
of time a consumer keeps a new car is approximately 6.6 years (78
months). EPA-HQ-OAR-2018-0283-4153, at 2. CFA commented that new
vehicle leases are running, on average, 68 months and new vehicles
are being held, on average, longer than 60 months. Comments, NHTSA-
2018-0067-12005, at 76. The agencies selection of 72 months is
comfortably within the range of these estimates, but errs towards
the lower-end and therefore provides a conservative estimate.
\1835\ These vehicle attributes may include any that consumers
may value and are not explicitly modeled to be neutral across
regulatory alternatives. For instance, trim levels, entertainment
systems, crash avoidance technologies, etc. may be sacrificed to pay
for higher fuel economy technology levels.
\1836\ The implicit opportunity cost must be considered a value
that consumers place on other vehicle attributes that is net of the
cost of those attributes. This is the forgone consumer surplus of
other vehicle attributes. As such it is appropriately additive to
the technology cost/savings estimated in the primary analysis.
---------------------------------------------------------------------------
The agencies note that the central analysis of the final rule
features a conservative treatment of private benefits and costs that
may bias the results in the favor of more stringent regulatory
alternatives. This bias arises from the agencies' treatment of rebound
driving. The agencies assume that drivers make a rational decision when
electing to drive additional miles, which considers not only the risks
the additional driving poses to their own lives and property, but also
most of the risks their behavior poses to their passengers as well as
the person and property of other road users. In such a case, drivers
``internalize'' most of these risks, and it can be assumed that
benefits to drivers must be more valuable to them than the risks they
considered when deciding whether to undertake the additional driving.
Therefore, the agencies have appropriately offset the loss in safety
benefits, which are associated with the increased cost of driving in
the final rule, with commensurate lost benefits of additional driving.
In contrast, the agencies can be assured the private benefits and
costs of fuel saving technologies (aside from the external
environmental damages) are internalized--as there is no doubt that the
owners of the vehicles will accrue the fuel costs/savings. The agencies
believe it would be entirely contradictory to assert that consumers are
rational, informed, and considerate enough to internalize the risks of
additional driving to themselves, their passengers, as well as other
drivers and passengers; but are not similarly rational and informed
enough to consider the additional fuel costs of purchasing a vehicle
without a particular fuel-saving technology. After all, existing
regulations require that the estimated annual fuel costs of a vehicle
are disclosed on the new vehicle a consumer intends to purchase--and no
such disclosure exists for the risks associated with driving a rebound
mile. The agencies' decision to offset rebound miles, but not net
private costs stemming from enabling more choices in fuel-saving
technologies, significantly favors more stringent alternatives.
Another possibility, however, is that manufacturers could redirect
some or all of their savings in technology costs to instead improve
other attributes of cars and light trucks--passenger comfort, safety,
carrying and towing capacity, or performance--that potential buyers
value. For example, they could redeploy the energy efficiency
improvements from some technologies that would otherwise have been used
to increase fuel economy to instead improve vehicles' performance, or
redirect spending on fuel economy technology to improve safety or
interior comfort. Producers could also offer combinations of price
reductions and more limited improvements in these other attributes on
some of their models, while continuing to offer high levels of fuel
economy on other models, and channeling their entire cost savings into
price reductions on yet other vehicles. Individual manufacturers would
presumably select different combinations of these strategies, each in
an effort to realize maximum additional sales and profits.
The agencies' analysis does not quantify specific improvements in
other attributes manufacturers could make, or identify potential
combinations of lower prices and improvements in other attributes they
might offer when they face less demanding fuel economy and
CO2 standards. Nevertheless, there is ample empirical
evidence that tradeoffs among fuel economy and other attributes that
buyers value are important considerations in vehicle design and
marketing strategy, and that manufacturers commonly offer combinations
of both higher fuel economy and improvements in other attributes when
standards do not require them to focus exclusively on improving fuel
economy.
Table VI-185 summarizes empirical estimates of the tradeoffs among
fuel economy, horsepower (for cars) or torque (for light trucks), and
weight derived from different authors' econometric estimates of the
``curvature'' of technology frontiers for cars and light trucks. Such
frontiers describe the combinations of fuel economy and other
attributes that manufacturers can provide with different levels of
spending on vehicle design and technology, accounting for the gradual
improvements in technology and energy efficiency that occur over time.
The entries in the table show different authors' estimates of the
percent increases in horsepower, torque, and weight that car and light
truck manufacturers could instead achieve if
[[Page 24703]]
they reduced fuel economy by one percent. (Although increased weight is
not desirable in and of itself, it is associated with features such as
a vehicle's passenger- and cargo-carrying capacity, interior volume,
comfort, and safety, which potential buyers do value.). It is important
to note that these tradeoffs apply to the overall average values of
each attribute for cars and light trucks produced during recent model
years, rather than to the features of specific individual models.
[GRAPHIC] [TIFF OMITTED] TR30AP20.383
For example, Table VI-185 shows that Klier & Linn estimate reducing
the average fuel economy of cars by one percent would enable producers
to increase their average horsepower by 0.24 percent, and Knittel's
estimate of that tradeoff is very similar (0.26 percent). Similarly,
those two studies estimate that reducing the average fuel economy of
cars and light trucks by one percent would enable their weight to be
increased by 0.34-0.39 percent, which would in turn enable
manufacturers to make modest improvements in their passenger- and
cargo-carrying capacity, interior volume, comfort, or safety. (Note
that reducing average fuel economy by one percent would permit either
power or weight to increase as indicated in the table, but not both at
the same time.).
The tradeoffs summarized in Table VI-185 provide some indication of
changes in attributes other than fuel economy that manufacturers are
likely to offer under the less demanding CAFE and CO2
standards. For example, the agencies estimate that the baseline CAFE
standards would have required increases in fuel economy approximately 5
percent annually over model years 2020-26 for cars, while this rule
reduces the required rate of increase to 1.5 percent annually. This
less demanding standard would thus enable producers to accompany higher
fuel economy with significant improvements in other features that new
car buyers also value, as an alternative to simply reducing prices to
reflect their savings in technology costs. As noted previously, they
would do so only if they thought such a strategy would be more
attractive to buyers, so the agencies' estimates of benefits to new car
and light truck buyers represents the minimum improvement in utility
they would realize.
The historical evolution of car and light truck characteristics
under CAFE standards may also provide some indication about how
manufacturers are likely to respond to the less aggressive standards
this rule establishes. Figure VI-77 and Figure VI-78 show that during
the period when CAFE standards remained unchanged or increased slowly--
approximately 1985-2010--manufacturers gradually improved cars' and
light trucks' average fuel economy as well as their power (or torque)
and weight, while only modestly increasing the average interior volume
of cars.
BILLING CODE 4910-59-P
[[Page 24704]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.384
[[Page 24705]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.385
BILLING CODE 4910-59-C
Table VI-186 summarizes the rates of change in fuel economy and
other attributes of cars and light trucks over that period. As it
shows, most advances in cars' drive train technology were used to
increase power and fuel economy, while most of the improvement in light
trucks' energy efficiency was channeled into higher torque and weight,
with relatively little used to improve fuel economy.
[GRAPHIC] [TIFF OMITTED] TR30AP20.386
The last column of Table VI-186 combines the actual historical
rates of increase in attributes other than fuel economy with the
tradeoffs between fuel economy and other attributes shown previously in
Table VI-185 to estimate the annual rates of increase in fuel economy
that could have been achieved if all technological progress had been
channeled into improving fuel economy. As it indicates, manufacturers
could have increased the fuel economy of both cars and light trucks
over the period spanned by Table VI-186 at almost exactly the 1.5
percent annual rate this rule requires, if they had believed that
sacrificing other improvements in the interest of achieving higher fuel
economy was the most effective strategy to meet potential customers'
demands.
[[Page 24706]]
While this result should be regarded as illustrative, it appears to
show that meeting even these relaxed standards may require
manufacturers to focus on improving fuel economy instead of other
vehicle attributes. It also suggests that meeting the more demanding
baseline standards may have required manufacturers to make significant
sacrifices in other attributes, rather than simply holding those other
features at or near their current levels. Viewed from this perspective,
while this rule might not enable manufacturers to improve other
desirable features of cars and light trucks at the same time as they
provide the improvements in fuel economy it requires, it may
nevertheless prevent them from having to sacrifice other improvements
that buyers regard as valuable in order to focus solely on complying
with more demanding CAFE and CO2 standards.
(9) Additional Consumer Purchase Costs
Some costs of purchasing and operating new and used vehicles scale
with the value of the vehicle. When fuel economy standards increase the
price of new vehicles, both taxes and registration fees increase, too,
because they are calculated as a percentage of vehicle price.
Increasing the price of new vehicles also affects the average amount
paid on interest for financed vehicles and the insurance premiums for
similar reasons. The agencies compute these additional costs as scalar
multipliers on the MSRP of new vehicles. These costs are included in
the consumer per-vehicle cost-benefit analysis, but, for the reasons
described below, are not included in the societal cost-benefit
analysis.
It is worth noting that these costs are not included in the sales
and scrappage models, discussed above. The agencies do not expect that
the omission of these costs affects the sales and scrappage models
because of how these additional costs are calculated in the modeling.
These costs are assumed to be a fixed scalar on the average MSRP of new
vehicles, so that their inclusion would simply scale the coefficients
in the sales and scrappage models. While these costs have not stayed
constant over time (particularly not over the times series from 1970 to
today), the agencies do not have a time series dataset to accurately
estimate these costs.
The agencies hope to reconsider including sales taxes, registration
fees, additional interest payments and insurance costs in the sales and
scrappage models in future research.
(a) Sales Taxes and Registration Fees
In the analysis, sales taxes and registration fees are considered
transfer payments between consumers and the government and are
therefore not considered a cost from the societal perspective. However,
these costs do represent an additional cost to consumers and are
accounted for in the private consumer perspective. To estimate the
sales tax for the analysis, the agencies weighted the auto sales tax of
each state by its population--using Census population data--to
calculate a national weighted-average sales tax of 5.46%.\1837\
---------------------------------------------------------------------------
\1837\ See Car Tax by State, FactoryWarrantyList.com, http://www.factorywarrantylist.com/car-tax-by-state.html (last visited June
22, 2018). Note: County, city, and other municipality-specific taxes
were excluded from weighted averages, as the variation in locality
taxes within states, lack of accessible documentation of locality
rates, and lack of availability of weights to apply to locality
taxes complicate the ability to reliably analyze the subject at this
level of detail. Localities with relatively high automobile sales
taxes may have relatively fewer auto dealerships, as consumers would
endeavor to purchase vehicles in areas with lower locality taxes,
therefore reducing the effect of the exclusion of municipality-
specific taxes from this analysis.
---------------------------------------------------------------------------
The agencies recognize that weighting state sales tax by new
vehicle purchases within a state would likely produce a better estimate
since new vehicle purchasers represent a small subset of the population
and may differ between states. The agencies explored using Polk
registration data to approximate new vehicle sales by state by
examining the change in new vehicle registrations across several recent
years. The results derived from this examination resulted in a national
weighted-average sales tax rate slightly above 5.5%, which is almost
identical to the rate calculated using population instead. The agencies
opted to utilize the population estimate, rather than the registration-
based proxy of new vehicle sales, because the results were negligibly
different and the analytical approach involving new vehicle
registrations has not been as thoroughly reviewed.
(b) Financing Costs
Consumers who purchase new vehicles with financing options incur an
additional cost above the new vehicle price--interest. Based off an
Experian data, \1838\ the analysis assumes 85% of automobiles are
purchased through financing options. The analysis used data from Wards
Automotive and JD Power on the average transaction price of new vehicle
purchases, average principle of new auto loans, and the average OEM-
offered incentive as a percent of MSRP to compute the ratio of the
average financed new auto principal to the average new vehicle MSRP for
calendar years 2011-2016. Table VI-187 shows that the average financed
auto principal was between 82% and 84% of the average new vehicle MSRP.
Applying the assumption that 85% of new vehicle purchases involve some
financing, the average share of the MSRP financed for all vehicles
purchased, including non-financed transactions, was computed. Table-II-
34 shows that the average percentage of MSRP financed ranges between
70% and 72%. From this, the agencies chose to assume that 70% of the
value of all vehicles' MSRP is financed. It is likely that the share
financed is correlated with the MSRP of the new vehicle purchased, but
for simplification purposes, it is assumed that 70% of all vehicle
costs are financed, regardless of the MSRP of the vehicle. The agencies
note that this simplification does not impact the accuracy of the
calculation of the average cost to consumers, but concede that it
obfuscates which consumers bear the additional financing burden when
vehicle prices increase (selection of specific vehicles is likely not
independent of consumer characteristics). For sake of simplicity, the
model also assumes that increasing the cost of new vehicles will not
change the share of new vehicle MSRP that is financed; the relatively
constant share from 2011-2016 when the average MSRP of a vehicle
increased 10% supports this assumption. The agencies recognize that
this is not indicative of average individual consumer transactions but
provides a useful tool to analyze the aggregate marketplace.
---------------------------------------------------------------------------
\1838\ A report by Experian found that 85.2% of 2016 new
vehicles were financed, as were 85.9% of 2015 new vehicle purchases.
Zabritski, M. State of the Automotive Finance Market: A look at
loans and leases in Q4 2016, Experian, https://www.experian.com/assets/automotive/quarterly-webinars/2016-Q4-SAFM-revised.pdf (last
visited June 22, 2018).
---------------------------------------------------------------------------
[[Page 24707]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.387
From Wards Auto data, the average 48- and 60-month new auto
interest rates were 4.25% in 2016, and the average finance term length
for new autos was 68 months. The agencies recognize that longer
financing terms generally include higher interest rates. The share
financed, interest rate, and finance term length are added as inputs in
the parameters file so that they are easier to update in the future.
Using these inputs the model computes the stream of additional
costs associated with financing options paid for the average financed
purchases as follows: \1839\
---------------------------------------------------------------------------
\1839\ As alluded to above, the principle portion of repayments
do not represent an additional cost to consumers since it represents
the sales price.
[GRAPHIC] [TIFF OMITTED] TR30AP20.600
0Note: The above assumes the interest is distributed evenly over
the period, when in reality more of the interest is paid during the
beginning of the term. However, the incremental amount calculated as
attributable to the standard will represent the difference in the
annual payments at the time that they are paid, assuming that a
consumer does not repay early. This will represent the expected
---------------------------------------------------------------------------
change in the stream of financing payments at the time of financing.
The above stream does not equate to the average amount paid to
finance the purchase of a new vehicle. In order to compute this amount,
the share of financed transactions at each interest rate and term
combination would have to be known. Without having projections of the
full distribution of the auto finance market into the future, the above
methodology reasonably accounts for the increased amount of financing
costs due to the purchase of a more expensive vehicle, on an average
basis taking into account non-financed transactions. Financing payments
are also assumed to be an intertemporal transfer of wealth for a
consumer; for this reason, it is not included in the societal cost and
benefit analysis. However, because it is an additional cost paid by the
consumer, it is calculated as a part of the private consumer welfare
analysis.
It is recognized that increased financing terms, combined with
rising interest rates, lead to longer periods before a consumer will
have positive equity in the vehicle to trade in toward the purchase of
a newer vehicle. This has impacts in terms of consumers either trading
vehicles with negative equity (thereby increasing the amount financed
and potentially subjecting the consumer to higher interest rates and/or
rendering the consumer unable to obtaining financing) or delaying the
replacement of the vehicle until they achieve suitably positive equity
to allow for a trade.
(c) Insurance Costs
More expensive vehicles will require more expensive collision and
comprehensive (e.g., fire and theft) car insurance. Actuarially fair
insurance premiums for these components of value-based insurance will
be the amount an insurance company will pay out in the case of an
incident type weighted by the risk of that type of incident occurring.
For simplicity of this calculation, the agencies assume that the
vehicle has the same exposure to harm throughout its lifetime. However,
the value of vehicles will decline at some depreciation rate so that
the absolute amount paid in value-related insurance will decline as the
vehicle depreciates. This is represented in the model as the following
stream of expected collision and comprehensive insurance payments:
[GRAPHIC] [TIFF OMITTED] TR30AP20.388
[[Page 24708]]
To utilize the above framework, estimates of the share of MSRP paid
on collision and comprehensive insurance and of annual vehicle
depreciations are needed to implement the above equation. Wards has
data on the average annual amount paid by model year for new light
trucks and passenger cars on collision, comprehensive and damage and
liability insurance for model years 1992-2003; for model years 2004-
2016, they only offer the total amount paid for insurance premiums. The
share of total insurance premiums paid for collision and comprehensive
coverage was computed for 1979-2003. For cars the share ranges from 49
to 55%, with the share tending to be largest towards the end of the
series. For trucks the share ranges from 43 to 61%, again, with the
share increasing towards the end of the series. It is assumed that for
model years 2004-2016, 60% of insurance premiums for trucks, and 55%
for cars, is paid for collision and comprehensive. Using these shares
the absolute amount paid for collision and comprehensive coverage for
cars and trucks is computed. Then each regulatory class in the fleet is
weighted by share to estimate the overall average amount paid for
collision and comprehensive insurance by model year as shown in Table
VI-188. The average share of the initial MSRP paid in collision and
comprehensive insurance by model year is then computed. The average
share paid for model years 2010-2016 is 1.83% of the initial MSRP. This
is used as the share of the value of a new vehicle paid for collision
and comprehensive in the future.
[GRAPHIC] [TIFF OMITTED] TR30AP20.389
2017 data from Fitch Black Book was used as a source for vehicle
depreciation rates; two- to six-year-old vehicles in 2016 had an
average annual depreciation rate of 17.3%.\1840\ It is assumed that
future depreciation rates will be like recent depreciation, and the
analysis used the same assumed depreciation. Table VI-189 shows the
cumulative share of the initial MSRP of a vehicle assumed to be paid in
collision and comprehensive insurance in five-year age increments under
this depreciation assumption, conditional on a vehicle surviving to
that age--that is, the expected insurance payments at the time of
purchase will be weighted by the probability of surviving to that age.
If a vehicle lives to 10 years, 9.9% of the initial MSRP is expected to
be paid in collision and comprehensive payments; by 20 years 11.9% of
the initial MSRP; finally, if a vehicle lives to age 40, 12.4% of the
initial MSRP.
---------------------------------------------------------------------------
\1840\ Fitch Ratings Vehicle Depreciation Report February 2017,
Black Book, http://www.blackbook.com/wp-content/uploads/2017/02/Final-February-Fitch-Report.pdf (last visited June 22, 2018).
---------------------------------------------------------------------------
[[Page 24709]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.390
The increase in insurance premiums resulting from an increase in
the average value of a vehicle is a result of an increase in the
expected amount insurance companies will have to pay out in the case of
damage occurring to the driver's vehicle. In this way, it is a cost to
the private consumer, attributable to the CAFE standard that caused the
price increase.
(10) Measuring Fuel Consumption
The procedure the agencies use to estimate fuel consumption assumes
that all vehicle models of the same body type--cars, SUVs and vans, and
light trucks--and age are driven identical amounts each year. Under
this assumption, the agencies' estimates of fuel consumption from
increasing the fuel economy of each individual model depend only on how
much its fuel economy is increased, and do not reflect whether its
actual use differs from other models of the same body type. Neither do
the agencies' estimates of fuel consumption account for variation in
how much vehicles of the same body type and age are driven each year,
which appears to be significant.
This assumption may cause the agencies' estimates of fuel
consumption from imposing stricter CAFE and CO2 standards to
be too large. Because the distribution of annual driving is wide using
its mean value to estimate fuel savings for individual car or light
truck models may overstate the fuel consumption likely to result from
tighter standards, even when the fuel economy of different models are
correctly averaged.\1841\ This will be the case even when increases in
fuel economy can be estimated reliably for individual models, as the
agencies' analysis does, because the reduction in a specific model's
fuel consumption depends on how much it is actually driven as well as
on the increase that stricter standards require.
---------------------------------------------------------------------------
\1841\ The correct average fuel economy of vehicles whose
individual fuel economy differs is the harmonic average of their
individual values, weighted by their respective use; for two
vehicles with fuel economy levels MPG1 and
MPG2 that are assumed to be driven identical amounts (as
in the agencies' analysis), their harmonic average fuel economy is
equal to 2/(1/MPG1 + 1/MPG2).
---------------------------------------------------------------------------
To illustrate, the agencies estimate that new automobiles are
driven about 17,000 miles on average during their first year. If the
17,000 mile figure represents the average of two different models that
are driven 14,000 and 20,000 miles annually, and the two initially
achieve, respectively, 30 and 40 miles per gallon--thus averaging 35
miles per gallon--they will consume a total of 967 gallons
annually.\1842\ Improving the fuel economy of each model by 5 miles per
gallon will reduce their total fuel use to 844 gallons, thus saving 123
gallons annually.\1843\ In contrast, the agencies' would estimate total
fuel consumption for the two vehicles using the 17,000 mile average
figure for both, thus yielding estimated fuel savings of 128 gallons
per year, about 5% above the correct value.\1844\
---------------------------------------------------------------------------
\1842\ Calculated as 14,000 miles/30 miles per gallon + 20,000
miles/40 miles per gallon = 467 gallons + 500 gallons = 967 gallons
(all figures in this calculation are rounded to whole gallons).
\1843\ Calculated as 14,000 miles/35 miles per gallon + 20,000
miles/45 miles per gallon = 400 gallons + 444 gallons = 844 gallons
(again, all figures in this calculation are rounded to whole
gallons).
\1844\ The agencies estimate of their combined initial fuel
consumption would be 17,000 miles/30 miles per gallon + 17,000
miles/40 miles per gallon, or 567 gallons + 425 gallons = 992
gallons. After the 5 mile per gallon improvement in fuel economy for
each vehicle, the agencies' estimate would decline to 17,000 miles/
35 miles per gallon + 17,000 miles/45 miles per gallon = 486 + 378 =
863 gallons, yielding an estimated fuel savings of 992 gallons--863
gallons = 128 gallons (as previously, all figures in this
calculation are rounded to whole gallons).
---------------------------------------------------------------------------
The magnitude of this potential overestimation of fuel savings
increases with any association between annual driving and fuel economy,
which seems likely to be strong. Acting in their own economic interest,
car and light truck buyers who anticipate driving more should be more
likely choose models offering higher fuel economy, because the number
of miles driven directly affects their fuel costs and thus the savings
from driving a model that features higher fuel economy.\1845\
Conversely, buyers who anticipate driving less are likely to purchase
models with lower fuel economy. Such behavior--whereby buyers who
expect to drive more extensively are likely to select models offering
higher fuel economy--cannot be fully accounted for in today's analysis,
because that analysis is necessarily based on
[[Page 24710]]
empirical estimates of average vehicle use. To the extent it occurs,
the agencies are likely to consistently overstate actual fuel savings
from requiring higher fuel economy, as well as to overstate increases
in fuel consumption resulting from lower standards. Thus, the agencies'
central analysis is likely to overestimate the final rule's impact on
consumer benefits such as reduced fuel consumption and increased
refueling time, as well as on the resulting environmental impacts of
fuel production and use.
---------------------------------------------------------------------------
\1845\ For example, some businesses, rental car firms, taxi
operators, and ride sharing drivers are likely to anticipate using
their vehicles significantly more than the average new car or light
truck buyer. Furthermore, their choices among competing models are
likely to be more heavily influenced by economics than by the
preferences for other attributes that motivate many other buyers,
making them more likely to select vehicles with higher fuel economy
in order to improve their economic returns.
---------------------------------------------------------------------------
A similar phenomenon may cause the agencies to overstate the value
of fuel savings resulting from requiring higher fuel economy as well.
As with miles driven, the agencies' analysis assumes all vehicle owners
pay the national average fuel price at any time. However, fuel prices
vary substantially among different regions of the U.S., and one would
expect buyers in regions with consistently higher fuel prices to
purchase vehicles with higher fuel economy, on average. To the extent
they actually do so, evaluating the savings from requiring higher fuel
economy identically in all regions using nationwide average fuel prices
is likely to overstate their actual dollar value; similarly, assessing
the increased fuel costs likely to result from lower standards using
national average fuel prices is likely to overstate their true value
insofar as car and light truck buyers facing above-average fuel prices
choose higher-mpg models.
As an illustration, suppose gasoline averages $3.00 per gallon
nationwide, but a buyer who expects to drive a new car 17,000 miles
during its first year (the same value used in the example above) faces
a local price of $4.00 per gallon, and chooses a model that achieves 40
mpg. That driver's cost of fuel during the vehicle's first year will
total $1,700 (calculated at 17,000 miles/40 miles per gallon x $4.00
per gallon). A buyer who plans to drive the same number of miles but
faces a lower price of $2.00 per gallon and thus chooses a vehicle that
offers only 30 mpg will have first-year fuel costs of $1,133
(calculated as 17,000 miles/30 miles per gallon x $2.00 per gallon), so
total annual fuel costs for these two vehicles will be $1,700 + $1,133
= $2,633. If the fuel economy of both vehicles increases by 5 mpg,
their actual fuel savings will be $189 and $162, or a total savings of
$351. However, evaluating total fuel savings using the national average
price of $3.00 per gallon yields savings of $382, thus overstating
actual savings by about 10%. This same phenomenon would cause the
agencies to overestimate of costs of increased fuel use when standards
are relaxed, as with this rule.
(11) Refueling Benefit
Increasing CAFE/CO2 standards, all else being equal,
affect the amount of time drivers spend refueling their vehicles in
several ways. First, they increase the fuel economy of ICE vehicles
produced in the future and, consequentially, decrease the number of
refueling events for those vehicles. Second, given increased production
costs, they reduce sales of new vehicles and scrappage of existing
ones, causing more VMT to be driven by older and less efficient
vehicles which require more refueling events for the same amount of VMT
driven. Finally, they may change the number of electric vehicles that
are produced, and shift refueling to occur at a charging station,
rather than at the pump--changing per-vehicle lifetime expected
refueling costs. While there are multiple ways that fuel economy
standards alter refueling costs, the proposal accounted for only the
first. Before the inclusion of the sales and scrappage models, which
first appeared in the NPRM analysis for the first time a CAFE/
CO2 rulemaking, the agencies did not have the means to
capture the other two effects. While the agencies modeled sales and
scrappage effects, they did not extend the results to refueling time.
This oversight was noted by commenters, and the final rule model now
includes these additional factors. The basic calculation for all three
effects is the same: The agencies multiply the additional amount of
time spent refueling by the value of time of passengers, which is
assumed to be the same for all three effects.
(a) Value of Time
The calculation of the value of time remains relatively unchanged
from the proposal and follows the guidance from DOT's 2016 Value of
Travel Time Savings memorandum (``VTTS Memo'').\1846\ The economic
value of refueling time savings is calculated by applying valuations
for travel time savings from the VTTS Memo to estimates of how much
time is saved across alternatives.\1847\
---------------------------------------------------------------------------
\1846\ United States Department of Transportation, The Value of
Travel Time Savings: Departmental Guidance for Conducting Economic
Evaluations, (2016), available at https://www.transportation.gov/sites/dot.gov/files/docs/2016%20Revised%20V.
\1847\ VTTS Memo Tables 1, 3, and 4.
---------------------------------------------------------------------------
IPI commented that the agencies used old data to calculate the
refueling benefit in the proposal. Specifically, IPI pointed out that
the data used in the proposal seemed ``to come from the 2003 version of
[the VTTS Memo].'' \1848\ For the final rule, the analysis uses the
most recent VTTS memo along with updated wages. The value of travel
time depends on average hourly valuations of personal and business
time, which are functions of annual household income and total hourly
compensation costs to employers. As designated by the 2016 VTTS memo,
the nationwide median annual household income, $56,516 in 2015, is
divided by 2,080 hours to yield an income of $27.20 per hour. Total
hourly compensation cost to employers, inclusive of benefits, in 2015$
is $25.40.\1849\ Table VI-190 demonstrates the agency's approach to
estimating the value of travel time ($/hour) for both urban and rural
(intercity) driving. This approach relies on the use of DOT-recommended
weights that assign a lesser valuation to personal travel time than to
business travel time, as well as weights that adjust for the
distribution between personal and business travel.\1850\ In accordance
with DOT guidance, wage valuations are estimated with base year 2015
dollars and end results are adjusted to 2018 dollars.
---------------------------------------------------------------------------
\1848\ IPI, Appendix, NHTSA-2018-0067-12213, at 51.
\1849\ Ibid at11.
\1850\ Business travel is higher than personal travel because an
employer has additional expenses, e.g. taxes and benefits costs,
above and beyond an employee's hourly wage. In the proposal, the
agencies erroneously used the same value for personal and business
travel, which was inconsistent with the VTTS Memo.
---------------------------------------------------------------------------
[[Page 24711]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.391
Estimates of the hourly value of urban and rural travel time
($14.14 and $20.40, respectively) shown in Table VI-190, must be
adjusted to account for the nationwide ratio of urban to rural
driving.\1851\ This adjustment, which gives an overall estimate of the
hourly value of travel time--independent of urban or rural status--is
shown in Table VI-191.
---------------------------------------------------------------------------
\1851\ Estimate of Urban vs. Rural travel weights from FHWA
December 2018 Traffic Volume Trends, Monthly Report, Table 2--
Cumulative Monthly Vehicle-Miles of Travel in Billions. Available at
https://www.fhwa.dot.gov/policyinformation/travel_monitoring/18dectvt/page3.cfm.
[GRAPHIC] [TIFF OMITTED] TR30AP20.392
[[Page 24712]]
Note that the calculations above consider the value of travel time
for only one occupant. To estimate fully the average value of vehicle
travel time per vehicle, the agencies must account for the presence of
all additional passengers during refueling trips. The agencies
estimated average vehicle occupancy using survey data gathered as part
of our 2010-2011 National Automotive Sampling System's Tire Pressure
Monitoring System (TPMS) study.\1852\ The study was conducted at
fueling stations nationwide and researchers made observations regarding
a variety of characteristics of thousands of individual fueling station
visits from August, 2010 through April, 2011. Among these
characteristics of fueling station visits, the total number of
occupants per vehicle were observed. Average vehicle occupancy was
calculated and multiplied by the value of travel time per occupant. As
shown in Table VI-192, this adjustment is performed separately for
passenger cars and for light trucks, yielding occupancy-adjusted
valuations of vehicle travel time during refueling trips for each
fleet. Lastly, the occupancy-adjusted value of vehicle travel time is
converted to 2018 dollars using the GDP deflator as shown in Table VI-
193.\1853\
---------------------------------------------------------------------------
\1852\ Docket for Peer Review of NHTSA/NASS Tire Pressure
Monitoring System, available at https://www.regulations.gov/docket?D=NHTSA-2012-0001.
\1853\ Bureau of Economic Analysis, NIPA Table 1.1.9 Implicit
Price Deflators for Gross Domestic Product, available at https://apps.bea.gov/iTable/index_nipa.cfm.
[GRAPHIC] [TIFF OMITTED] TR30AP20.393
IPI commented that the exclusion of children from the NPRM's
refueling time analysis was inconsistent with DOT's 2016 Value of
Travel Time Savings memorandum (``VTTS Memo''). IPI claimed that the
VTTS Memo ``consider[ed] whether the value of travel time is different
for parents versus children, but ultimately conclude[d] that `it must
be assumed that all travelers' VTTS are independent and additive.' ''
IPI also quoted language from page 13 of the VTTS Memo that
``[a]lthough riders may be a family with a joint VTTS or passengers in
a car pool or transit vehicle with independent values, these
circumstances can seldom be distinguished [. . .] therefore, all
individuals are assumed to have independent values,'' and that it is
``inappropriate to use different income levels or sources for different
categories of traveler.'' \1854\
---------------------------------------------------------------------------
\1854\ See IPI, Appendix, NHTSA-2018-0067-12213, at 52-53
(citing United States Department of Transportation (``DOT''), The
Value of Travel Time Savings: Departmental Guidance for Conducting
Economic Evaluations, (2016), available at https://www.transportation.gov/sites/dot.gov/files/docs/2016%20Revised%20V).
---------------------------------------------------------------------------
IPI further asserted that excluding passengers under age 16 from
the calculation of travel time savings was inconsistent with the best
practices of benefit-cost analysis. IPI noted that Circular A-4 does
not distinguish between children and adults except when monetizing
health effects. IPI then cited Dale Whittington and Duncan MacRae as
stating ``there is a clear consensus that children should be counted in
cost-benefit analysis.'' Finally, IPI commented that Congress intended
that the agencies consider the economic impact to children when setting
standards.\1855\
---------------------------------------------------------------------------
\1855\ See IPI, Appendix, NHTSA-2018-0067-12213, at 53-54
(internal citations omitted).
---------------------------------------------------------------------------
The agencies point out that the first passage from the VTTS Memo
cited by IPI does not conclude, or even deliberate, that the VTTS of
children is the same as adults, but instead states that the VTTS of
children, parents and other passengers should be independent and
additive.\1856\ Assuming that the opportunity cost of children's time
is zero is compatible with this practice. Likewise, IPI concluded from
the text on page 12 that it was inappropriate to use different incomes
for children. However, IPI's analysis suffers from two errors.
---------------------------------------------------------------------------
\1856\ See VTTS Memo at 5.
---------------------------------------------------------------------------
First, the two quotes from page 12 reside in a section of the VTTS
Memo
[[Page 24713]]
entitled Special Issues, which provides guidance on three distinct
topics. The first quoted text comes from a paragraph advising how to
treat vehicles with multiple passengers, while the second is from an
ensuing topic about passenger incomes. It is baseless to assume that
the conclusion of the second topic holds true for the first.
Second, assuming IPI intended to comment that age is a ``category
of traveler'' for which ``it is inappropriate to use different income
levels,'' the agencies note that such an interpretation is tenuous. The
VTTS Memo clearly recognizes that some categories of travelers should
have different levels of income,\1857\ and provides two examples.\1858\
As children are not part of the workforce, they do not have wage
incomes. Therefore, it is not wild speculation that they do not bear a
financial opportunity cost associated with their time spent in vehicles
during refueling.\1859\ As such, excluding children from the
calculation of the refueling benefit is consistent with DOT's guidance.
---------------------------------------------------------------------------
\1857\ The full text quoted by IPI reads, ``[e]xcept for
specific distinctions, we consider it inappropriate to use different
income levels or sources for different categories of traveler.''
VTTS Memo at 12 (emphasis added). The VTTS Memo further contemplates
that it is appropriate to assign different incomes if ``estimates
[of income are] derived by reliable and focused research [. . .] in
specific cases.'' Id.
\1858\ The VTTS Memo provides specific guidance on how to
differentiate between personal and business travel, and air or high
speed rail from other modes of transportation. See VTTS Memo at 12.
\1859\ The TMPS study affords the agencies the opportunity to
distinguish between adults and passengers, a luxury not available in
every instance. Furthermore, there may be certain instances where it
is appropriate to value the VTTS of children the same as adults,
e.g., rules focusing primarily on the VTTS of children.
---------------------------------------------------------------------------
Turning to IPI's comments on best practices and Congress' intent,
the agencies agree that the benefit-cost analysis should include
children when appropriate. The majority of the components of the CAFE
model (e.g., safety analyses) include children. However, children are
excluded from the analysis when it is appropriate (e.g., employment).
For this specific valuation, it is reasonable to assume the value of a
child's time is not equivalent to an adult's. Nonetheless, the agencies
have examined the impact of valuing children's time as equal to adults'
by including them in the average vehicle occupancy rates applied in the
refueling analysis and using the full VTTS for personal travel. Results
indicate that the effect of this issue is minor and impacts total
benefits by about one-quarter percent. The agencies will continue to
consider this issue in future CAFE and CO2 rulemakings. IPI
also noted that the only portion of the TPMS publicly available was the
``User's Coding Manual.'' Specifically, IPI argued that ``the agencies'
failure to make available the full data and methodology used to
calculate these average occupancy figures frustrates any meaningful
public review.'' The agencies disagree. IPI was able to submit a
meaningful comment about the agencies' decision to exclude children
from the occupancy-adjusted value of vehicle travel time. Furthermore,
commenters knew that the agencies intended to use occupancy estimates
to calculate the refueling benefit; however, the agencies did not
receive any alternative estimates or methodologies from commenters.
Nonetheless, the agencies have provided reference to the docket folder
containing peer review documents, analysis documentation, and data for
the 2011 TPMS survey.
(b) Accounting for Improved Fuel Economy of ICE Vehicles
The methodology for calculating the refueling benefits associated
with improved fuel economy in new vehicles remains unchanged from the
proposal. The CAFE model calculates the number of refueling events for
each ICE vehicle in a calendar year. This is calculated as the number
of miles driven by each vehicle in that calendar year divided by the
product of that vehicle's on road fuel economy, tank size, and an
assumption about the average share of the tank refueled at each event,
as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.394
The model then computes the cost of refueling as the product of the
number of refueling events, total time of each event and value of the
time spent on each event (computed as average salary), as below:
The event time of a vehicle is calculated by summing a fixed and
variable component. The fixed component is the number of minutes it is
assumed each event takes, independent of any assumptions about tank
size or share refueled at each event (the time it takes to get to and
from the pump). The variable component is the ratio of the average
number of gallons refueled for each event (the product of the tank size
and share refueled) and the rate at which gallons flow from the pump.
This is shown below:
[GRAPHIC] [TIFF OMITTED] TR30AP20.600
1In order to calculate the refueling time cost, as described above,
the CAFE model takes the following inputs: The value of time, the fixed
time component of each refueling event, share of the tank refueled at
each event, rate of flow of fuel from the pump, and vehicle tank size.
The first of these is taken from DOT guidance on travel time savings.
The fixed time component, share refueled, and rate of flow are
calculated from survey data gathered as part of our 2010-2011 National
Automotive Sampling System's Tire Pressure Monitoring System (TPMS)
study.\1860\ Finally, the vehicle fuel tank sizes are taken from
manufacturer specs for the reference fleet and historical averages are
calculated from popular models for the existing vehicle fleet, as
described, below, in discussion of the legacy fleet.
---------------------------------------------------------------------------
\1860\ Docket for Peer Review of NHTSA/NASS Tire Pressure
Monitoring System, available at https://www.regulations.gov/docket?D=NHTSA-2012-0001.
---------------------------------------------------------------------------
The agencies estimated the amount of saved refueling time using
survey data gathered as part of the aforementioned TPMS study. In this
nationwide study, researchers gathered information on the total amount
of time spent pumping and paying for fuel. From a separate sample (also
part of the TPMS study),
[[Page 24714]]
researchers conducted interviews at the pump to gauge the distances
that drivers travel in transit to and from fueling stations, how long
that transit takes, and how many gallons of fuel are purchased.
The agencies focused on the interview-based responses in which
respondents indicated the primary reason for the refueling trip was due
to a low reading on the gas gauge. Such drivers experience a cost due
to added mileage driven to detour to a filling station, as well as
added time to refuel and complete the transaction at the filling
station. The agencies believe that drivers who refuel on a regular
schedule or incidental to stops they make primarily for other reasons
(e.g., using restrooms or buying snacks) do not experience the cost
associated with detouring in order to locate a station or paying for
the transaction, because the frequency of refueling for these reasons
is unlikely to be affected by fuel economy improvements. This
restriction was imposed to exclude distortionary effects of those who
refuel on a fixed (e.g., weekly) schedule and may be unlikely to alter
refueling patterns as a result of increased driving range. The relevant
TPMS survey data on average refueling trip characteristics are
presented below in Table VI-194.
[GRAPHIC] [TIFF OMITTED] TR30AP20.395
The agencies assume that all of the round-trip time necessary to
travel to and from the fueling station is a part of the fixed time
component of each refueling event. However, some portion of the time to
fill and pay is also a part of the fixed time component. Given the
information in Table VI-194, the agencies assume that each refueling
event has a fixed time component of 3.5 minutes. E.g., (for passenger
cars) the sum of 2.28 minutes round trip time to/from fueling station
and roughly 1.2 minutes to select and pay for fuel, remove/recap fuel
tank, remove/replace fuel nozzle, etc. The time to fill the fuel tank
is the variable time component; e.g., about 2.9 minutes for passenger
cars (2.28 + 1.2 + 2.9 = 6.38 total minutes). However, the CAFE model
uses a different methodology to determine the variable time component,
which is explained below.
Cars have average tank sizes of about 15 gallons, SUVs/vans of
about 18 gallons, and pickups of about 27 gallons (see Table VI-195
through Table VI-197 in discussion of the legacy fleet). It is a
reasonable assumption that the average passenger car has a tank of 15
gallons and the average light truck has a tank of 20 gallons (there are
more SUVs/vans than pickups in the light truck fleet). From these
assumptions, it is calculated that the average refueling event fills
approximately 65 percent of the fuel tank for both passenger cars and
light trucks. This value is used as an input in the CAFE model for all
three body styles (cars, SUVs/vans, and pickups).
Finally, the rate of the pump flow can be calculated either as the
total gallons pumped over the assumed variable time component
(approximately 3 minutes) or as the difference in the average number of
gallons filled between light trucks and passenger cars over the
difference in the time to fill and pay between the two classes. The
first methodology implies a rate between 3 and 4 gallons per minute.
Although the second methodology implies a rate of 15 gallons per
minute, there is a legal restriction on the flow of gasoline from pumps
of 10 gallons per minute.\1861\ Thus, the agencies assume the rate of
gasoline pumps range between 4 and 10 gallons per minute, and use 7.5
gallons per minute--a value slightly above the midpoint of that range--
as the average flow rate in the CAFE model.
---------------------------------------------------------------------------
\1861\ 40 CFR 80.22(j), Regulation of Fuels and Fuel Additives--
subpart B. Controls and Prohibitions, available at https://www.law.cornell.edu/cfr/text/40/80.22.
---------------------------------------------------------------------------
The calculations described above are repeated for each future
calendar year that light-duty vehicles of each model year affected by
the CAFE standards considered in this rule would remain in service for
each regulatory alternative. The resulting cumulative lifetime
valuations of time savings account for both the reduction over time in
the number of vehicles of a given model year that remain in service and
the reduction in the number of miles (VMT) driven by those that stay in
service. After calculating the absolute value for each regulatory
alternative using the methodology and inputs described above, the model
calculates the incremental value relative to the baseline as the
refueling cost or benefit for that regulatory alternative. More
efficient vehicles have to be refueled less often and refueling costs
per vehicle decline. In previous rules this was sufficient to account
for the majority of any changes in cost of refueling under different
CAFE standards as the modelling permitted, since the volumes of new
vehicles and existing vehicles on the road was assumed to be constant
under all possible standards. However, when sales and scrappage models
are included the distribution of new and vehicles varies and a
different number of miles will be driven by new and used vehicles in
each regulatory alternative.
IPI commented that it was inappropriate for the agencies to
[[Page 24715]]
exclude benefits from reducing the frequency of refueling events where
the primary reason for stopping at a fuel station was not to refuel a
vehicle. IPI argued that fuel efficiency impacts from relaxed standards
would affect all drivers regardless of their rationale for refueling,
by requiring either more frequent or marginally longer refueling
events.\1862\ The agencies note that the language in the NPRM suggested
that the agencies eliminated 40 percent of the potential benefit from
fewer refueling stops--where 40 percent represents the fraction of
refueling stops that were routinely scheduled or otherwise not made in
response to a low fuel reading--and this appears to have been the
origin of IPI's concern.\1863\ In fact, the agencies did not apply a 40
percent discount factor to the refueling benefits; instead, the total
number of additional refueling events that would result from
alternative CAFE levels was calculated, and these were valued based on
an assumption that their characteristics (e.g., vehicle occupancy)
would match those of drivers who refueled due to a low fuel reading.
---------------------------------------------------------------------------
\1862\ IPI, Appendix, NHTSA-2018-0067-12213, at 54-55.
\1863\ See 83 FR 43088 (Aug. 24, 2018).
---------------------------------------------------------------------------
To the extent that lower fuel economy affects those who refuel on a
routine schedule or incidental to stops made primarily for other
reasons, the per-event cost would actually be limited to the extra time
spent pumping a slightly larger volume of fuel. However, the agencies
note that by assuming that all extra fuel consumed under lower CAFE
standards results in added refueling trips, the agencies are adopting a
conservative assumption, in the sense that it maximizes the disbenefits
of alternatives to the current standards.
IPI also expressed concern that the agencies may have excluded the
fuel costs and added emissions from additional miles driven in the
course of the more frequent refueling events that would be required
with more lenient CAFE standards, and correspondingly lower on-road
fuel economy.\1864\ In the NPRM, the agencies asserted that these added
costs are reflected in their overall estimates of fuel cost savings,
while any increase in emissions is also reflected in the reported
changes in total emissions. However, IPI noted that the agencies did
not clearly explain how these cost savings and emissions reductions are
actually accounted for in their methodology.
---------------------------------------------------------------------------
\1864\ IPI, Appendix, NHTSA-2018-0067-12213, at 55.
---------------------------------------------------------------------------
The agencies' methodology fully accounts for both of these impacts
through its calculation of changes in the use of new cars and light
trucks due to the fuel economy rebound effect, which captures the
impact on their aggregate use (VMT) that results from changes in the
fuel cost of driving each mile. Studies that estimate the rebound
effect analyze the relationship between VMT per time period and fuel
economy or per-mile fuel costs, using data for individual vehicles,
fleet-wide average values, or aggregate estimates for an entire fleet.
Regardless of the level of aggregation they employ, their measures of
vehicle use invariably include travel for all purposes, including any
extra miles driven in the course of refueling.
Thus, the estimates of the rebound effect--the response of vehicle
use to changes in fuel economy or per-mile fuel costs--inevitably
capture any change in the number of miles driven for the purpose of
refueling that occurs in response to higher or lower fuel economy. This
change reflects the net effect of more or less frequent refueling trips
required by their baseline or ``pre-rebound'' level of use, and any
change in the number of refueling trips associated with increased or
reduced driving in response to the rebound effect.
As a consequence, the agencies' estimates of changes in aggregate
fuel consumption and fuel costs incorporate--that is, are net of--the
volume and cost of fuel consumed by changes in vehicle use that result
from the rebound effect, including any change in driving associated
with more or less frequent refueling. Similarly, the agencies'
estimates of changes in emissions resulting from vehicle storage and
use (referred to as ``tailpipe'' or ``downstream'' emissions) are
derived by applying per-mile emission factors to changes in aggregate
vehicle travel, so they necessarily incorporate changes in vehicle use
for all purposes, including more or less frequent refueling.
Furthermore, as the agencies demonstrated in the proposal with a
practical example, the benefit associated with fewer miles spent
refueling is less than 23[cent] per year for new vehicles. The
cumulative impact of this benefit amounts to less than one tenth of
percent of the costs of the rule.\1865\
---------------------------------------------------------------------------
\1865\ See 83 FR at 43088. Also, note that the 23 cents estimate
was derived for a less stringent alternative than today's standards
and included taxes which would have been removed had the agencies
calculated this number separately.
---------------------------------------------------------------------------
Because all of the alternative standards evaluated in this
rulemaking would permit lower fuel economy levels than under the
baseline standard, per-mile driving costs would be higher and total
vehicle use would decline in response. Although some (perhaps most) new
vehicles would require more frequent refueling, the agencies' estimates
of the change in aggregate use of new vehicles reflects (i.e., is net
of) any increase in driving associated with more frequent refueling
stops. As a result, the agencies' estimates of changes in total fuel
consumption, aggregate fuel costs, and emissions resulting from the
lower fuel economy levels that relaxing CAFE standards would permit
reflect the net reduction in use of new cars and light trucks due to
the fuel economy rebound effect, after considering any additional miles
that would be driven in the course of more frequent refueling stops.
(c) Including the Legacy Fleet
Under more stringent regulatory alternatives, more miles will be
driven by older and less efficient vehicles, and the effect is to
reduce or eliminate any refueling benefit from increasing the fuel
efficiency of new vehicles. Failing to include the existing fleet makes
the costs of refueling artificially lower under more stringent
standards because new vehicle sales are lower and not only because new
vehicles are more efficient. This update to the calculation of the
absolute refueling costs corrects this oversight present in the NPRM
cost-benefit analysis by calculating fleet-wide absolute refueling
costs before considering the incremental change relative to the
baseline.
For other portions of the CAFE model, the agencies track the legacy
vehicles by body style and vintage, using average measures for fuel
economy, horsepower and curb weight. To estimate refueling costs for
these vehicles, measures of average fuel tank sizes by body style and
vintage are needed. The agencies are unaware of any data that directly
estimates this value, but an estimate can be derived from publicly
available data on fuel tank sizes of 17 high-volume nameplates with
long histories. The tank sizes are averaged by body style, and these
historical values are used as estimates of the average by body style
and vintage. The vehicles included, their fuel tank sizes, and the
averages are reported in Table VI-195 through Table VI-197 for cars,
vans/SUVs, and pickups, respectively. The averages are used to
represent the fuel tank sizes by vintage and vehicle body style. The
agencies used the fuel tank sizes from Table VI-195 to Table VI-196 to
determine the number of refueling events and time spent refueling to
[[Page 24716]]
compute refueling costs using the methodology described above.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.396
[[Page 24717]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.397
[[Page 24718]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.398
[[Page 24719]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.399
[[Page 24720]]
BILLING CODE 4910-59-C
(d) Including Electric Vehicle Recharging
In addition to adding the refueling costs associated with the
``legacy fleet,'' this update adds the cost to recharge electric
vehicles to the total refueling costs. Excluding the time spent
recharging ignores a real cost borne by owners of electric vehicles,
one which was noted by multiple commenters. For example, Ariel Corp.
and VNG.co LLC commented that, ``EVs require significant changes in
consumer fueling behavior given the need to park at recharging points
for long periods of time.'' \1866\
---------------------------------------------------------------------------
\1866\ Ariel Corp. and VNG.co LLC, Comment, NHTSA-2018-0067-
7573, at 13.
---------------------------------------------------------------------------
In order to do so, it is important to first understand how many
electric vehicle charging events will require the driver to wait and
for how long. The answer to this question depends on the range of the
electric vehicle and the length of the trip.\1867\ For trips shorter
than the range, the driver can recharge the vehicle at times that will
not require them to be actively waiting and thus there is no recharging
cost. Only for trips where the vehicle is driven more miles than the
range will the driver have to stop at mid-trip, a time that is assumed
to be inconvenient, to recharge the vehicle at least enough to reach
the intended destination.
---------------------------------------------------------------------------
\1867\ While the range of EVs is dependent on a number of
factors, such as that grade, acceleration, and weather, the agencies
take a conservative approach and assume a best-case scenario.
---------------------------------------------------------------------------
The agencies use trip data from the National Household
Transportation Survey (NHTS) to estimate the frequency and expected
length of trips that exceed the range of the electric vehicle
technologies in the simulation (200 and 300 mile ranges).
The NHTS data is collected from a representative random sample of
U.S. households. The survey collects data on individual trips by mode
of transportation. A trip is defined by the starting and ending point
for any personal travel, so that vehicle trips will capture any time a
car is driven. The survey includes identification numbers for
households, individuals, and vehicles, and mode of transportation
(including the body style of the vehicle for vehicle trips), and the
date of the trip. Although some trips made in the same day may allow
for convenient charging in between trips, the agencies assume that
travel in the same day exceeding the range will involve the driver
waiting for the vehicle to charge. Thus, the total number of miles
driven by the same vehicle in a single day is summed, and it is assumed
that charging stations are not conveniently available to the driver in
between.
Some of the trips in the NHTS have missing information about the
duration or length of the trip; these trips are excluded from the
dataset. The agencies subset the dataset into three body styles--cars,
vans/SUVs, and pickups--consistent groupings with how the VMT schedules
and scrappage rates are estimated. The agencies exclude data on taxis
and rental cars as the body style of the vehicle for these trips is not
specified (they make up only 0.3 percent of the dataset, so their
exclusion is unlikely to alter the estimate). Table VI-198, below,
shows the resulting quantiles of the distribution of daily travel for
all vehicles considered in the final dataset. This will include
multiple days of travel for the same vehicle if more than one day of
trip data is recorded in the NHTS.
[GRAPHIC] [TIFF OMITTED] TR30AP20.400
[GRAPHIC] [TIFF OMITTED] TR30AP20.401
The data in Table VI-198 shows that excluding taxis and rentals may
be the best choice even if their body styles were known. For taxi
trips, only the number of trips an individual driver makes in a day is
known. The number of trips that the taxi cab itself makes in a day is
unknown. As can be seen, the distribution of ``daily'' travel is to the
left for taxis because not all trips for those vehicles are reported.
Thus, including these vehicles would incorrectly skew the daily travel
rates downwards.
The distribution of trip lengths for rental cars, on the other
hand, is generally to the right of trips taken privately-owned
vehicles. This is likely because individuals are travelling longer
distances when they are on vacation or otherwise out-of-town. It seems
likely that individuals renting cars for longer trips will not choose
electric vehicles for such temporary travel. Thus,
[[Page 24721]]
including these trips in the dataset would likely overestimate the
number of mid-trip charging events necessary for ordinary travel in a
way that will not match what actually occurs.
From the final body style datasets, the agencies are able to
calculate two measures that allow for the construction of the value of
recharging time. First, the expected distance between trips that exceed
the range of 200-mile and 300-mile BEVs (BEV200 and BEV300,
respectively) is calculated. This is calculated as the quotient of the
sum of total miles driven by each individual body style and the total
number of trips exceeding the range, as shown below:
[GRAPHIC] [TIFF OMITTED] TR30AP20.402
This calculates the expected frequency of enroute recharging
events, or the amount\1868\ of miles traveled per inconvenient
recharging event. This is used later used to calculate the total
expected time to recharge a vehicle.
---------------------------------------------------------------------------
\1868\ The denominator counts the number of incontinent
recharging events by body style. It is not a measurment of VMT.
---------------------------------------------------------------------------
The second measure needed to calculate the total expected
recharging time is the expected share of miles driven that will be
charged in the middle of a trip (causing the driver to wait and lose
the value of time). In order to calculate this measure the difference
of the trip length and range is summed, conditional on the trip length
exceeding the range for each body style. This figure is then divided by
the sum of the length of all trips for that body style. See the
equation below:
[GRAPHIC] [TIFF OMITTED] TR30AP20.403
The calculated frequency of inconvenient charging events and share
of miles driven that require the driver to wait for BEV's with 200 and
300-mile ranges are presented in Table VI-199, below. As the table
shows, cars are expected to require less frequent inconvenient charges
and a smaller share of miles driven will require the driver to charge
the vehicle in the middle of a trip. Pickups and vans/SUVs have fairly
similar measures, with vans and SUVs requiring slightly more
inconvenient charging than pickups.
[GRAPHIC] [TIFF OMITTED] TR30AP20.404
The measures presented in Table VI-199, above, can be used to
calculate the expected time drivers of electric vehicles of a given
body style and range will spend recharging at a time that will require
them to wait. First the agencies calculate the expected number of
refueling events for a vehicle of a given style and range in a given
calendar year. This is shown below as the expected miles driven by a
vehicle in a given calendar year divided by the charge frequency of a
vehicle of that style and range (from Table VI-199).
[[Page 24722]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.405
Next the agencies calculate the number of miles charged for a
vehicle of a given style\1869\ and range in a specific calendar year.
This is the product of the number of miles driven by the vehicle and
the share of miles driven that require an inconvenient charge for a
vehicle of that style and range (from Table VI-199), as presented
below:
---------------------------------------------------------------------------
\1869\ Note that [Sigma]Trip [epsi] Style Trip Length and Miles
CY,Veh are different values. MilesCY,Veh is the estimated amount of
VMT predicted by VMT while [Sigma]Trip [epsi] Style Trip Length is
the sum of trips observed by the NHTS study.
---------------------------------------------------------------------------
Then, the expected time that a driver of an electric vehicle of a
given style and range will spend waiting for the vehicle to charge is
calculated. This is the product of the fixed amount of time it takes to
get to the charging station and the number of recharging events plus
the quotient of the expected miles that will require inconvenient
charging over an input assumption of the rate of which a vehicle of
that style and range can be charged in a given calendar year (expressed
in units of miles charged per hour). The fixed amount of time it takes
to get to a charging station is set equal to the average time it takes
for an ICE vehicle to get to a gas station for a refueling event, as
discussed above.\1870\ This is shown below:
---------------------------------------------------------------------------
\1870\ The agencies note that this is a conservative estimate.
Gas stations vastly outnumber publicly available recharging stations
and are often in more convenient locations.
[GRAPHIC] [TIFF OMITTED] TR30AP20.406
The expected time that a driver will wait for their vehicle to
charge can then be multiplied by the value of time estimate, as is done
with gasoline, diesel, and E85 vehicles (see description above of the
current approach to accounting for refueling time costs).
It is worth a final note to talk about how plug-in hybrids are
treated in the modelling (which remains unchanged from the NPRM).
Presumably, plug-in hybrids that are taken on a trip that exceeds their
electric range will be driven on gasoline and the driver will recharge
the battery at a time that is convenient. For this reason, the electric
portion of travel should be excluded from the refueling time
calculation. The gasoline portion of travel is treated the same as
other gasoline vehicles so that when the tank reaches some threshold,
the vehicles is assumed to be refueled with the same fixed event time
and the same rate of refueling flow.
The NPRM calculation of refueling benefits did not account for the
impacts of fleet turnover--specifically the impact on ``legacy'' fleet
vehicles and new electric vehicles. However, when the quantities of
vehicles on the road varies between scenarios it becomes important to
calculate the refueling costs for all vehicles since fuel economy and
tank sizes (and therefore range before refueling) vary with vintage.
This updated analysis adds these elements to the calculation of the
refueling time and costs and is thus a more accurate estimation of the
refueling benefit.
(12) Energy Security
By amending existing standards, the final rule is expected to
increase domestic consumption of gasoline by a relatively minimal
amount relative to the baseline standards finalized in 2012, producing
a correspondingly small increase in the Nation's demand for crude
petroleum, a commodity that is traded actively in a worldwide market.
Specifically, the agencies project that this rule will increase
gasoline consumption by cars and light trucks produced during model
years 1978 through 2029 by 3.1 percent.\1871\ Although the U.S.
accounts for a sufficient (albeit diminishing) share of global oil
consumption that the resulting increase in global petroleum demand will
exert some upward pressure on worldwide prices, the rule is projected
to increase global petroleum demand by less than one half of one
percent from 2017 through 2050, so its effects on global prices is
likely to be minimal.
---------------------------------------------------------------------------
\1871\ This includes fuel consumed by cars and light trucks
produced during model years 1978-2017 that are on the road today
during their remaining lifetimes, as well as fuel consumed by cars
and light trucks projected to be manufactured during model years
2018-2029 over their entire lifetimes.
---------------------------------------------------------------------------
U.S. consumption and imports of petroleum products has three
potential effects on the domestic economy that are often referred to
collectively as ``energy security externalities,'' and increases in
their magnitude are sometimes cited as possible social costs of
increased U.S. demand for petroleum.m First, any increase in global
petroleum prices that results from higher U.S. gasoline demand will
cause a transfer of revenue to oil producers worldwide from consumers
of petroleum, because consumers throughout the world are ultimately
subject to the higher global price that results. Although this transfer
is simply a shift of resources that produces no change in global
economic welfare, the financial drain it produces on the U.S. economy
is sometimes cited as an external cost of increased U.S. petroleum
consumption, because consumers of petroleum products are unlikely to
consider it.
As the U.S. approaches self-sufficiency in petroleum production
(the nation is expected to become a net exporter of petroleum by 2020),
this transfer is increasingly from U.S. consumers of refined petroleum
products to U.S. petroleum producers, so it not only leaves welfare
unaffected, but even ceases to be a financial burden on the U.S.
economy.\1872\ In fact, as the U.S. becomes a net petroleum exporter,
any transfer from global consumers to petroleum producers would become
a financial benefit to the U.S. economy. Nevertheless, uncertainty in
the nation's long-term import-export balance makes it difficult to
project precisely how these effects might change in response to
increased consumption.
---------------------------------------------------------------------------
\1872\ The United States became a net exporter of oil on a
weekly basis several times in late 2019, and EIA's AEO 2019 projects
that will do so on a sustained, long-term basis by 2020; see EIA,
AEO 2019 Reference Case, Table 21 https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=wttntus2&f=4.
---------------------------------------------------------------------------
Higher U.S. petroleum consumption can also increase domestic
consumers' exposure to oil price shocks and thus
[[Page 24723]]
increase potential costs to all U.S. petroleum users (including those
outside the light duty vehicle sector, whose consumption would be
unaffected by today's final rule) from possible interruptions in the
global supply of petroleum or rapid increases in global oil prices.
Because users of petroleum products are unlikely to consider the effect
of their increased purchases on these risks, their economic value is
often cited as an external cost of increased U.S. consumption. Finally,
some analysts argue that domestic demand for imported petroleum may
also influence U.S. military spending; because the increased cost of
military activities would not be reflected in the price paid at the gas
pump, this is often alleged to represent a third category of external
costs form increased U.S. petroleum consumption.
Each of these three costs could rise incrementally--albeit by a
very limited magnitude--as a consequence of increases in U.S. petroleum
consumption--likely to result from the final rule. This section
describes the extent to which each cost is expected to increase as a
result of this action, whether it represents a significant economic
cost (or simply a transfer of resources), and how the agencies have
measured each cost and incorporated it into their analysis.
(a) U.S. Petroleum Demand and Its Effect on Global Prices
Figure VI-79 illustrates the effect of the increase in U.S. fuel
and petroleum demand anticipated to result from reducing CAFE and
CO2 standards on global demand for petroleum and its market
price. The marginal increase in domestic demand can be represented as
an outward shift in the U.S. demand curve for petroleum from its
position at DUS,0 with the baseline standards for future
model years in effect, to DUS,1 with the final rule
standards replacing them. Because global demand is simply the sum of
what each nation would purchase at different prices, the outward shift
in U.S. demand causes an identical shift in the global demand schedule,
as the figure shows.\1873\
---------------------------------------------------------------------------
\1873\ The figure exaggerates the U.S. share of total global
consumption, which currently stands at 20 percent, for purposes of
illustration.
[GRAPHIC] [TIFF OMITTED] TR30AP20.407
The global supply curve for petroleum slopes upward, reflecting the
fact that it is progressively costlier for oil-producing nations to
explore for, extract, and deliver additional supplies of oil to the
world market.\1874\ Thus the upward shifts in the U.S. and world demand
schedules cause an increase in the global price for oil, from
P0 to P1 in the figure. U.S. purchases of
petroleum increase from QUS,0 to QUS,1, but the
resulting increase in global consumption from QG,0 to
QG,1 will be slightly smaller than the increase in U.S.
demand and purchases, because the amount of petroleum other nations
purchase will decline slightly in response to its higher price.
Spending on petroleum by U.S. buyers who purchase the additional oil
will increase by the area QUS,0acQUS,1, the
product of its new, higher price P1 and the increase in U.S.
consumption, QUS,1-QUS,0, while spending by U.S.
consumers whose purchases remain unchanged will increase by the product
of their previous purchases QUS,0 and the price increase
P1-P0, or the area P1abP0.
---------------------------------------------------------------------------
\1874\ The figure depicts the relationship between the global
supply of petroleum and its worldwide price during a single time
period. The global supply curve for petroleum has been shifting
outward over time in response to increased investment in
exploration, the ability of refineries to utilize feedstocks other
than conventional petroleum, and technological innovations in
petroleum extraction. The combination of these developments may also
have reduced its upward slope, meaning that global supply now
increases by more in response to increases in the world price than
it once did.
---------------------------------------------------------------------------
CARB asserted in their comments, that the NPRM analysis was biased
[[Page 24724]]
against the baseline standards because the fuel prices in the NPRM were
based on a unique run of DOE's NEMS model that included the
baseline.\1875\ They argued that the proposal would have reduced fleet
average fuel economy, leading to increased demand and subsequently
higher fuel prices faced by consumers. As a result, the additional fuel
costs associated with the proposal (relative to the baseline) should
have been even higher than estimated because the fuel price faced by
drivers in that scenario would have been higher than in the baseline.
However, while the difference between the baseline and preferred
alternative could create differences in fleet fuel economy in a manner
that could influence prices at the pump, those differences are likely
to be small. In response to CARB's comments, the agencies conducted
additional runs with NEMS to compare the fuel price under the baseline
standards and the fuel price under the proposed standards. Through
2050, the fuel price difference between the alternatives was never
higher than two percent. The standards being finalized in this rule are
considerably closer to the baseline than were those in the proposal.
---------------------------------------------------------------------------
\1875\ NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
SAFE commented that the United States is a ``price-taker'' in the
global market and ``must accept the prevailing global oil price since
it lacks sufficient market power to influence decisively this price.''
\1876\ This comment, however, is directly at odds with both the
economics of the world oil market shown in Figure VI-79 above and other
comments asserting that the increase in U.S. gasoline demand resulting
from this rule will increase U.S. and global petroleum demand, thus
increasing world oil prices. In response to the comment from SAFE, the
agencies utilized a forecast of fuel prices in today's analysis that
considers the effect of the revised standards on global petroleum
demand and prices. This assumption slightly increases the cost of
forgone fuel savings in the preferred alternative, compared to their
value under the assumption that U.S. demand cannot change global prices
and the nation acts as a price-taker.
---------------------------------------------------------------------------
\1876\ NHTSA-2018-0067-11981.
---------------------------------------------------------------------------
In Figure VI-79, the increase in the price of oil from
P0 to P1 will mean that global consumers who
previously purchased the quantity of oil QG,0 at its lower
price will now pay more for that same amount. Specifically, previous
purchasers will pay the additional area P1deP0,
whose value is the increase in price P1-P0
multiplied by the volume they originally bought, QG,0. Of
this increase in revenue to oil producers, the rectangular area
P1abP0--which as indicated above is the product
of the increase in price P1-P0 and previous U.S.
purchases QUS,0, and thus measures the increase in spending
by previous U.S. consumers--is simply transferred from U.S. consumers
to global oil suppliers.\1877\ The remaining fraction of increased
payments to producers, the rectangular area adeb, whose value is the
product of the price increase P1-P0 and previous
purchases by other nations, which were QG,0-
QUS,0, is a transfer from consumers outside the U.S. to
global oil producers.
---------------------------------------------------------------------------
\1877\ Note that global oil suppliers include domestic as well
as US-owned foreign suppliers.
---------------------------------------------------------------------------
The total increase in global spending--including the additional
spending by U.S. consumers as well as by those in other nations--on the
amount of oil they previously purchased is simply a transfer of revenue
from consumers of petroleum products to oil producers. This transfer
can be described as a ``pecuniary'' externality, since it describes the
effect of the price increase on wealth allocation, but is considered
separately from any effects on quantity produced and consumed. Some of
the increase in payments by U.S. consumers for the petroleum products
they originally consumed may be made to foreign-owned oil producers,
and thus represents a financial drain on the U.S. economy, while the
remainder is received by domestic producers and thus remains within the
U.S. economy.\1878\
---------------------------------------------------------------------------
\1878\ Neither transfer, however, has an effect on domestic or
global economic welfare.
---------------------------------------------------------------------------
To an increasing extent, however, the additional payments by U.S.
consumers that result from upward pressure on the world oil price are a
transfer entirely within the Nation's economy, because a growing
fraction of domestic petroleum consumption is supplied by U.S.
producers. The U.S. is projected to become a net exporter of petroleum
in 2020--and in fact became a net exporter in September 2019--and as
the Nation moves toward that status, an increasing share of any higher
costs paid by U.S. consumers of petroleum products becomes a gain to
U.S. oil producers.\1879\ When the U.S. becomes self-sufficient in
petroleum supply--which is now anticipated to occur in the year this
final rule publishes--the entire value of increased payments by U.S.
petroleum users that results from relaxing CAFE and CO2
standards will have the same effect as if it were simply a transfer
within the U.S. economy. As a consequence, the financial burden that
transfers from U.S. consumers to foreign producers places on the U.S.
economy will disappear.
---------------------------------------------------------------------------
\1879\ The U.S. Energy Information Administration EIA estimates
that the United States exported more total crude oil and petroleum
products in September and October of 2019, and expects the United
States to continue to be a net exporter. See Short Term Energy
Outlook November 2019, available at https://www.eia.gov/outlooks/steo/archives/nov19.pdf.
---------------------------------------------------------------------------
Over almost the entire time period spanned by the analysis of this
final rule, any increase in domestic spending for petroleum caused by
the effect of higher U.S. fuel consumption and petroleum use on world
oil prices is expected on balance to be a transfer within the U.S.
economy and thus produce no drain on domestic economic resources. For
this reason--and because in any case such transfers do not create real
economic costs or benefits--increased U.S. spending on petroleum
products that results from increased U.S. fuel demand and any resulting
upward pressure on petroleum prices stemming from this action is not
included among the economic costs accounted for in this final rule.
(b) Macroeconomic Costs of U.S. Petroleum Consumption
In addition to influencing global demand and prices, U.S. petroleum
consumption imposes further costs that are unlikely to be reflected in
the market price for petroleum, or in the prices paid by consumers of
refined products such as gasoline.\1880\ Petroleum consumption imposes
external economic costs by exposing the U.S. economy to increased risks
of rapid increases in prices triggered by global events that may also
disrupt the supply of imported oil, and U.S. consumers of petroleum
products are unlikely to take such costs into account when making their
decisions about how much to consume.
---------------------------------------------------------------------------
\1880\ See, e.g., Bohi, D.R. & W. David Montgomery (1982), Oil
Prices, Energy Security, and Import Policy Washington, DC--Resources
for the Future, Johns Hopkins University Press; Bohi, D.R., & M.A.
Toman (1993), ``Energy and Security--Externalities and Policies,''
Energy Policy 21:1093-1109; and Toman, M.A. (1993). ``The Economics
of Energy Security--Theory, Evidence, Policy,'' in A. V. Kneese and
J.L. Sweeney, eds. (1993), Handbook of Natural Resource and Energy
Economics, Vol. III, Amsterdam--North-Holland, pp. 1167-1218.
---------------------------------------------------------------------------
Sudden interruptions in oil supply and rapid increases in its price
can impose significant economic costs, because they raise the costs of
producing all commodities whose manufacturing and distribution consumes
petroleum, thus temporarily reducing the level of output that the U.S.
economy can produce using its available supplies of labor and capital.
The magnitude of any reduction in
[[Page 24725]]
economic output depends on the extent and duration of the increases in
prices for petroleum products that result from a disruption in global
oil supplies, as well as on whether and how rapidly prices return to
their pre-disruption levels--which in turn depends largely on the rest
of the world's capability to respond to interruptions by increasing
production elsewhere. Even if prices for oil return completely to their
original levels, however, economic output will be at least temporarily
reduced from the level that would have been possible with uninterrupted
oil supplies and stable prices, so the U.S. economy will bear some
transient losses it cannot subsequently recover.
Supply disruptions and price increases caused by global political
events tend to occur suddenly and unexpectedly, so they can also force
businesses and households to adjust their use of petroleum products
more rapidly than if the same price increase occurred gradually. Rapid
substitutions between energy derived from oil and other forms of
energy, as well as between energy and other inputs, and other changes
such as adjusting production levels and downstream prices, can be
costly for businesses to make. As with businesses, sudden changes in
energy prices and use are also difficult for households to adapt to
quickly or smoothly, and doing so may impose at least temporary costs
or losses in utility for the various adjustments they make.
Interruptions in oil supplies and sudden increases in petroleum
prices are both uncertain prospects, and the costs of the disruptions
they can cause must be weighted or adjusted by the probability that
they will occur, as well as for their uncertain duration. The agencies
estimate this expected cost of such disruptions by combining the
probabilities that price increases of different magnitudes and
durations will occur during the future period spanned by their analysis
with the costs of reduced U.S. economic output and abrupt adjustments
to sharply higher petroleum prices. Any change in the probabilistic
``expected value'' of such costs that can be traced to higher U.S. fuel
consumption and petroleum demand stemming from this final rule to
establish less demanding fuel economy standards is considered to be an
external cost of the adopting it.
A variety of mechanisms exist to ``insure'' against higher
petroleum prices and reduce their costs for adjusting to sudden price
increases, including making purchases or sales in oil futures markets,
adopting energy conservation measures, diversifying the fuel economy
levels within the set of vehicles owned by the household, locating
where public transit provides a viable alternative to driving, and
installing technologies that permit rapid fuel switching. Growing
reliance on such measures, coupled with continued improvements in
energy efficiency throughout the economy, has certainly reduced the
vulnerability of the U.S. economy to the costs of oil shocks in recent
decades.
Thus, there is now considerable debate about the magnitude and
continued relevance of potential economic damages from sudden increases
in petroleum prices. The petroleum intensity of the U.S economy has
declined considerably and global oil prices are dramatically lower than
when analysts first identified and quantified the risks they create to
the U.S. economy. Further, not only has the Nation dramatically
increased its own petroleum supply, but other new global supplies have
emerged as well, both of which reduce the potential impact of
disruptions that occur in unstable or vulnerable regions where oil is
produced.
As a consequence, the potential macroeconomic costs of sudden
increases in oil prices are now likely to be considerably smaller than
when they were original identified and estimated. Research by the
National Research Council (2009) argued that non-environmental
externalities associated with dependence on foreign oil are small, and
perhaps trivial.\1881\ Research by Nordhaus and by Blanchard and Gali
have also questioned how harmful to the economy oil price shocks have
been, noting that the U.S. economy actually expanded immediately after
the most recent oil price shocks, and that there was little evidence of
higher energy prices being passed through to higher wages or
prices.\1882\
---------------------------------------------------------------------------
\1881\ National Research Council, Hidden Costs of Energy--
Unpriced Consequences of Energy Production and Use, National Academy
of Sciences, Washington, DC (2009).
\1882\ Nordhaus argues that one reason for limited vulnerability
to oil price shocks is that monetary policy has become more
accommodating to the price impacts, while another is that U.S.
consumers and businesses may determine that such movements are
temporary and abstain from passing them on as inflationary price
increases in other parts of the economy. He also notes that changes
in productivity in response to recent oil price increases are have
been extremely modest, observing that ``energy-price changes have no
effect on multifactor productivity and very little effect on labor
productivity.'' at p. 19. Blanchard and Gali (2010) contend that
improvements in monetary policy, more flexible labor markets, and
the declining energy intensity of the U.S. economy (combined with an
absence of concurrent shocks to the economy from other sources)
lessened the impact of oil price shocks after 1980. They find that
``the effects of oil price shocks have changed over time, with
steadily smaller effects on prices and wages, as well as on output
and employment . . . The message . . . is thus optimistic in that it
suggests a transformation in U.S. institutions has inoculated the
economy against the responses that we saw in the past.'' at p. 414;
See William Nordhaus, ``Who's Afraid of a Big Bad Oil Shock?''
Available at http://aida.econ.yale.edu/~nordhaus/homepage/
Big_Bad_Oil_Shock_Meeting.pdf; and Blanchard, Olivier and Jordi
Gali, J., ``The Macroeconomic Effects of Oil price Shocks--Why are
the 2000s so Different from the 1970s?,'' in Gali, Jordi and Mark
Gertler, M., eds., The International Dimensions of Monetary Policy,
University of Chicago Press, February (2010), pp. 373-421, available
at http://www.nber.org/chapters/c0517.pdf.
---------------------------------------------------------------------------
Since these studies were issued in 2009 and 2010, the petroleum
intensity of the U.S. economy has continued to decline while domestic
energy production has increased in ways and to an extent that experts
failed to predict, so that the U.S. became the world's largest producer
in 2018.\1883\ The U.S. shale oil revolution has both established the
potential for energy independence and placed downward pressure on
prices. Lower oil prices are also a result of sustained reductions in
U.S. consumption and global demand resulting from energy efficiency
measures, many undertaken in response to previously high oil prices.
---------------------------------------------------------------------------
\1883\ See U.S. Energy Information Administration EIA, Today in
Energy August 20, 2019, available at https://www.eia.gov/todayinenergy/detail.php?id=40973; Today in Energy September 12,
2018, available at https://www.eia.gov/todayinenergy/detail.php?id=37053.
---------------------------------------------------------------------------
Reduced petroleum intensity and higher U.S. production have
combined to produce a decline in U.S. petroleum imports--to
approximately 20 percent of domestic consumption in 2017--which permits
U.S. supply to act as a buffer against artificial or natural
restrictions on global petroleum supplies due to military conflicts or
natural disasters. In addition, the speed and relatively low
incremental cost with which U.S. oil production has increased suggests
that both the magnitude and (especially) the duration of future oil
price shocks may be limited, because U.S. production offers the
potential for a large and relatively swift supply response.
And while some risk of price shocks certainly still exists, even
the potential for a large and swift U.S. production response may be
playing a role in limiting the extent of price shocks attributable to
external events. The large-scale attack on Saudi Arabia's Abqaiq
processing facility--the world's largest crude oil processing and
stabilization plant--on September 14, 2019 caused ``the largest single-
day [crude oil] price increase in the past decade,'' of between $7 and
$8 per
[[Page 24726]]
barrel, according to EIA.\1884\ The Abqaiq facility has the capacity to
process 7 million barrels per day, or about 7 percent of global crude
oil production capacity. EIA declared, however, that by September 17,
only three days after the incident:
---------------------------------------------------------------------------
\1884\ https://www.eia.gov/todayinenergy/detail.php?id=41413.
Saudi Aramco reported that Abqaiq was producing 2 million
barrels per day, and they expected its entire output capacity to be
fully restored by the end of September. In addition, Saudi Aramco
stated that crude oil exports to customers will continue by drawing
on existing inventories and offering additional crude oil production
from other fields. Tanker loading estimates from third-party data
sources indicate that loadings at two Saudi Arabian export
facilities were restored to the pre-attack levels. Likely driven by
news of the expected return of the lost production capacity, both
Brent and WTI crude oil prices fell on Tuesday, September 17.\1885\
---------------------------------------------------------------------------
\1885\ Id.
Thus, the largest single-day oil price increase in the past decade
was largely resolved within a week, and assuming very roughly that
average crude oil prices were $70/barrel in September 2019 (slightly
higher than actual), an increase of $7/barrel would represent a 10
percent increase as a result of the Abqaiq attack. Contrast this with
the 1973 Arab oil embargo, which lasted for months and raised prices
350 percent.\1886\ Saudi Arabia could have experienced increased
revenue resulting from higher prices following the Abqaiq attack, but
instead moved rapidly to restore production and tap reserves to control
the risk of resulting price increases. In doing so, the Saudis likely
recognized that sustained, long-term price increases would reduce their
ability to control global supply (and thus prices and their own
revenues) by relying on their lower cost of production.\1887\
---------------------------------------------------------------------------
\1886\ See Jeanne Whalen, ``Saudi Arabia's oil troubles don't
rattle the U.S. as they used to,'' Washington Post, September 19,
2019, available at https://www.washingtonpost.com/business/2019/09/19/saudi-arabias-oil-troubles-dont-rattle-us-like-they-used/.
\1887\ See, e.g., ``Dynamic Delivery: America's Evolving Oil and
Natural Gas Transportation Infrastructure,'' National Petroleum
Council (2019) at 18, available at: https://dynamicdelivery.npc.org/downloads.php.
---------------------------------------------------------------------------
Some commenters asserted that U.S. shale oil resources cannot serve
as ``swing supply'' to provide stability in the face of a sudden,
significant global supply disruption (Jason Bordoff,
SAFE).1888 1889 Despite its greater responsiveness to price
changes, commenters argued that lead time to bring new shale resources
to market (6-12 months) is inferior to ``true spare capacity'' (like
Saudi Arabia's large oil fields) because it cannot be deployed quickly
enough to mitigate the economic consequences resulting from rapidly
rising oil prices. Bordoff, however, also notes that shale oil
projects' lead times are still shorter--and possibly much shorter--than
conventional oil resource development. So, while new U.S. oil resources
may take some time to respond to supply disruptions, they are
nevertheless likely to provide a stabilizing influence on supply.
---------------------------------------------------------------------------
\1888\ NHTSA-2018-0067-11981.
\1889\ NHTSA-2018-0067-10718.
---------------------------------------------------------------------------
This is especially true for price increases that occur more slowly.
When Beccue and Huntington updated their 2005 estimates of supply
disruption probabilities in 2016,\1890\ they found that the probability
distribution was generally flatter--suggesting that supply disruptions
of most potential magnitudes were less likely to occur under today's
market conditions than they had estimated previously in 2005. In
particular, Beccue and Huntington find that supply disruptions of
between two and four million barrels per day are significantly less
likely than their previous estimates suggested. Although their recent
study also estimated that larger supply disruptions (nine or more
million barrels per day) are now slightly more likely to occur than in
previous estimates, disruptions of that magnitude are extremely
unlikely under either set of estimates.
---------------------------------------------------------------------------
\1890\ Beccue, Phillip, Huntington, Hillard, G., 2016. An
Updated Assessment of Oil Market Disruption Risks: Final Report.
Energy Modeling Forum, Stanford University.
---------------------------------------------------------------------------
Based on this review of the literature, the agencies concede that
shale resources may not be able to stabilize oil markets fully to
prevent a price increase associated with a large supply disruption
elsewhere in the world. However, if supply disruptions are small
enough, or move slowly enough, U.S. resources may be an adequate
stabilizer.
The agencies reviewed further research that emphasizes the
continued threat to the U.S. economy posed by the potential for sudden
increases in global petroleum prices.\1891\ For example, Ramey and Vine
(2010) note ``remarkable stability in the response of aggregate real
variables to oil shocks once we account for the extra costs imposed on
the economy in the 1970s by price controls and a complex system of
entitlements that led to some rationing and shortages.'' \1892\ In
contrast, another recent study found that while the likely effects of
sudden oil price increases have become smaller over time, the declining
sensitivity of petroleum demand to prices means that any future
disruptions to oil supplies will have larger effects on petroleum
prices, so that on balance their economic impact is likely to remain
significant.\1893\
---------------------------------------------------------------------------
\1891\ Hamilton (2012) reviewed the empirical literature on oil
shocks and concluded that its findings are mixed, noting that some
recent research (e.g., Rasmussen and Roitman, 2011) finds either
less evidence for significant economic effects of oil price shocks
or declining effects (Blanchard and Gali 2010), while other research
finds evidence of their continuing economic importance. See
Hamilton, J. D., ``Oil Prices, Exhaustible Resources, and Economic
Growth,'' in Handbook of Energy and Climate Change available at
http://econweb.ucsd.edu/~jhamilto/handbook_climate.pdfhttp://
econweb.ucsd.edu/~jhamilto/handbook_climate.pdf.
\1892\ Ramey, V. A., & Vine, D. J. ``Oil, Automobiles, and the
U.S. Economy--How Much have Things Really Changed?'' National Bureau
of Economic Research Working Paper 16067 (June 2010). Available at
http://www.nber.org/papers/w16067.pdf.
\1893\ Baumeister, C. and G. Peersman (2012), ``The role of
time-varying price elasticities in accounting for volatility changes
in the crude oil market,'' Journal of Applied Econometrics 28 no. 7,
November/December 2013, pp.1087-1109.
---------------------------------------------------------------------------
Some commenters (SAFE, CARB, Fuel Freedom Foundation, IPI)
expressed skepticism that the United States could become a net
petroleum exporter in the future without the continuation of the
baseline standards. They cautioned that the global oil market is
inherently uncertain, and Bordoff cautioned that America's shale
resources may not last as long, or be as easy to develop, as they
currently appear.\1894\ If the U.S. does not become a net exporter of
petroleum as anticipated, any wealth effects from a high price of oil
would continue to accrue to foreign owners of oil reserves. In
addition, several of these commenters (CARB, SAFE, Bordoff, Zozana)
argued that, regardless of whether or not the U.S. becomes a net
petroleum exporter, its levels of petroleum consumption make it still
vulnerable to price shocks arising in the global oil market.
---------------------------------------------------------------------------
\1894\ NHTSA-2018-0067-10718.
---------------------------------------------------------------------------
The agencies believe that the United States lacks the power
(significantly) to control the global oil price and as a consequence
remains vulnerable to the effects of oil price spikes, regardless of
our own oil output. Geopolitical factors influence the global oil
price--unstable regimes are often unreliable suppliers, large suppliers
attempt strategically to manage supply to influence price or retain
market share, and international negotiations around politically
sensitive topics can influence the production behavior of firms in oil-
rich nations. All of these factors, as well as wars and natural
disasters, can influence the
[[Page 24727]]
global supply and the market price for oil.
In this analysis, any increase in the expected value of potential
costs from economy-wide disruptions caused by sudden price increases
that results from higher U.S. fuel and petroleum demand is accounted
for separately from the direct cost for increased purchases of
petroleum products. Consumers of petroleum products are unlikely to
consider their contributions to these costs when deciding how much
energy to consume, because those costs will be distributed widely
throughout the economy, falling largely on businesses and households
other than those whose decisions impose them. Thus, they represent an
external (or ``social'') cost that users of petroleum energy such as
transportation fuel are unlikely to internalize fully, and the agencies
analysis includes the estimated increase in these costs among of the
social costs stemming from the final rule. While increased U.S.
petroleum production may impose some limits on their potential
magnitude, their underlying source continues to be domestic petroleum
use rather than imports.
Although the vulnerability of the U.S. economy to oil price shocks
depends on aggregate petroleum consumption rather than on the level of
oil imports, variation in U.S. oil imports may itself have some effect
on the frequency, size, or duration of sudden oil price increases. The
expected value of the resulting economic costs would also depend partly
on the fraction of U.S. petroleum use that is supplied by imports.
While total U.S. petroleum consumption is the primary determinant of
potential economic costs to the Nation from rapid increases in oil
prices, the estimate of these costs that have been relied upon on in
past regulatory analyses--and in this analysis--is nevertheless
expressed per unit (barrel) of imported oil. When they are converted to
a per-gallon basis, they thus apply to fuel that is either imported in
refined form, or refined domestically from imported crude petroleum.
Table VI-200 reports the per-barrel estimates of external costs
from potential oil price shocks this analysis uses to estimate the
increase in their total value likely to result from this final rule.
These values differ from those used in previous analysis of CAFE and
CO2 standards. In their comments on the NPRM, SAFE pointed
out recent studies that have updated the estimates of the oil security
premium since the study--on which the agencies relied upon in the
NPRM--had been published. They depend in part on projected future oil
prices, the elasticities of consumption with respect to price, income,
and U.S. GDP. Since the NPRM values were last updated by the agencies,
all of these factors have evolved in directions that would reduce the
magnitude of the oil security premium, so continuing to use the NPRM
values would have overestimated the increase in expected costs to the
U.S. economy from potential oil price shocks calculated in this
analysis, perhaps significantly.\1895\
---------------------------------------------------------------------------
\1895\ The costs reported in Table VI-188 also depend on the
probabilities or expected frequencies of supply interruptions or
sudden price shocks of different sizes and durations. The most
recent reassessment of the probabilities on which these estimates
are based (which were originally developed in 2005) was conducted in
2016; see Beccue, Phillip C. and Hillard G. Huntington, An Updated
Assessment of Oil Market Disruption Risks--Final Report EMF SR 10,
Stanford University Energy Modeling Forum (February 5, 2016)
available at https://emf.stanford.edu/publications/emf-sr-10-updated-assessment-oil-market-disruption-risks.
---------------------------------------------------------------------------
Specifically, the global petroleum prices projected in EIA's Annual
Energy Outlook 2018 Reference Case range from 33-57 percent below those
used to develop the estimates used in the NPRM and reported in Table
VI-200. U.S. petroleum consumption and imports are now projected to be
3-8 percent and 20-27 percent lower than the forecast values used to
construct the NPRM estimates in the table. Finally, total petroleum
expenditures are now projected to average 1.5-2.4 percent of U.S. GDP,
in contrast to the 3.8-4.0 percent shares reflected in those values.
Each of these differences suggests that the values in the NPRM
overstated the current magnitude of potential costs to the U.S. economy
from the risk of petroleum price shocks, and together they suggest that
this overstatement may be significant. Indeed, the values used to
support this final rule analysis are sourced from a recent paper by
Brown.\1896\ Brown updates the underlying parameters used to estimate
the oil security premium and finds a range of $0.60-$3.45 per barrel of
imported oil, with a mean of $1.26 per barrel. The study, which was
cited by SAFE, determines that the U.S. is less much less sensitive to
oil price shocks than earlier estimates imply.\1897\ The values used in
today's rule reflect that conclusion.
---------------------------------------------------------------------------
\1896\ See Brown, Stephen P.A., New estimates of the security
costs of U.S. oil consumption, Energy Policy, Volume 13, 2018, Pages
171-192.
\1897\ Another report cited by SAFE, Krupnick, et. al, similarly
conclude that the macroeconomic cost of oil price shocks has
diminished and that the oil security premium is lower than the
majority of the existing literature would suggest. See Krupnick,
Alan, Morgenstern, Richard, Balke, Nathan, Brown, Stephen P.A.,
Herrera, Ana Maria, and Mohan, Shashank, ``Oil Supply Shocks, US
Gross Domestic Product, and the Oil Security Premium,'' Resources
for the Future, November 2017, available at: https://media.rff.org/documents/RFF-Rpt-OilSecurity.pdf (last accessed 01/2020).
\1898\ In order to convert per-barrel costs into per-gallon
costs, we make the common assumption (used throughout the analysis)
that each barrel of petroleum produces 42 gallons of motor gasoline.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24728]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.408
[[Page 24729]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.409
BILLING CODE 4910-59-C
Because they are expressed per barrel of petroleum that is imported
(either in already-refined form as gasoline, or as crude petroleum to
be refined domestically), applying these estimates requires the
agencies to project of any changes in U.S. petroleum imports that are
likely to result from the higher level of fuel consumption anticipated
to occur as a result of this final rule. As discussed in detail in
Section VI.D.3.c(b)(i) of this final rule, the agencies have elected to
retain their previous assumptions that 50 percent of any increase in
fuel consumption attributable to the rule will be accounted for through
imports in refined form, and that 90 percent of the remaining increase
would be refined domestically from imported petroleum. As a
consequence, the oil security premiums shown in Table VI-200 are
considered to be an external cost associated with 95 percent of the
increase in gasoline consumption projected to result from this final
rule.\1899\
---------------------------------------------------------------------------
\1899\ The 95 percent figure is calculates at 50 percent plus 90
percent of the remaining 50 percent, or 50 percent plus 45 percent.
---------------------------------------------------------------------------
(c) Potential Effects of Fuel Consumption and Petroleum Imports on U.S.
Military Spending
A third potential effect of increasing U.S. demand for petroleum is
an increase in U.S. military spending to secure the supply of oil
imports from potentially unstable regions of the world and protect
against their interruption. If an increase in fuel consumption that
results from reducing CAFE and CO2 standards lead to higher
military spending to protect oil supplies, this increase in outlays
would represent an additional external or social cost of the agencies'
action. Such costs could also include increased costs to maintain the
U.S. Strategic Petroleum Reserve (SPR), because it is intended to
cushion the U.S. economy against disruptions in the supply of imported
oil or sudden increases in the global price of oil.
While several commenters argued that current U.S. military
expenditures are uniquely attributable to securing U.S. supplies of
petroleum from unstable regions of the globe--the Middle East, in
particular--should be considered as a cost of this action (CARB, SAFE,
Zonana), they seemed to confuse those costs with the marginal impact of
increased oil consumption (relative to the baseline) on U.S. military
activity and its costs. However, the agencies disagree with commenters
that incremental changes to domestic consumption of oil for light-duty
transportation could meaningfully change the scope or scale of the U.S.
Department of Defense mission in the Persian Gulf region. Instead, they
side with the Fuel Freedom Foundation, which noted in its comment,
``[i]ncrementally decreasing petroleum consumption does not
significantly
[[Page 24730]]
decrease the military spending to protect and ensure its flow around
the world.'' \1900\
---------------------------------------------------------------------------
\1900\ NHTSA-2018-0067-12016.
---------------------------------------------------------------------------
SAFE estimated a per-gallon cost of military externalities
associated with U.S. dependence on petroleum products, and imported
petroleum specifically.\1901\ Their low estimate of $0.28/gallon
assumes $81 billion per year for protection of the global petroleum
supply and divides those costs by the number of gallons consumed by
U.S. drivers. In contrast, a similar analysis by Crane et al. stated,
``our analysis addresses the incremental cost to the defense budget of
defending the production and transit of oil. It does not argue that a
partial reduction of the U.S. dependence on imported oil would yield a
proportional reduction in U.S. spending that is focused on this
mission. The effect on military cost from such changes in petroleum use
would be minimal.'' \1902\ The agencies thus do not believe that any
incremental petroleum consumption that may result from this final rule
will influence any fraction of U.S. defense spending that can be
ascribed to protecting the global oil network.
---------------------------------------------------------------------------
\1901\ NHTSA-2018-0067-11981.
\1902\ Crane, K., A. Goldthau, M. Toman, T. Light, S.E. Johnson,
A. Nader, A. Rabasa, & H. Dogo, Imported Oil and U.S. National
Security, Santa Monica, CA, The RAND Corporation (2009) available at
https://www.rand.org/pubs/monographs/MG838.html.
---------------------------------------------------------------------------
Eliminating petroleum imports (to both the U.S. and its national
security allies) entirely might permit the Nation to scale back its
military presence in oil-supplying regions of the globe to the extent
that such interventions are driven by narrow concerns for oil
production rather than other geopolitical considerations, but there is
little evidence that U.S. military activity and spending in those
regions have varied over history in response to fluctuations in the
Nation's oil imports, or are likely to do so over the future period
spanned by this analysis. Figure VI-80 shows that military spending as
a share of total U.S. economic activity has gradually declined over the
past several decades, and that any temporary--although occasionally
major--reversals of this longer-term decline have been closely
associated with U.S. foreign policy initiatives or overseas wars.
[GRAPHIC] [TIFF OMITTED] TR30AP20.410
Figure VI-81 superimposes U.S. petroleum consumption and imports on
the history of military spending shown in the previous figure. Doing so
shows that variation in U.S military spending throughout this period
has had little association with the historical pattern of domestic
petroleum purchases, changes in which instead primarily reflected the
major increases in global petroleum prices that occurred in 1978-79,
2008, and 2012-13. More important, Figure VI-81 also shows that U.S.
military spending varied almost completely independently of the
nation's imports of petroleum over this period. This history suggests
that U.S. military activities--even in regions of the world that have
[[Page 24731]]
historically represented vital sources of oil imports--serve a far
broader range of security and foreign policy objectives than simply
protecting oil supplies. Thus, reducing the nation's consumption or
imports of petroleum is unlikely by itself to lead to reductions in
military spending.
SAFE further argued in its comments that the America's involvement
in wars in the Persian Gulf region, starting with the first Gulf War
and continuing through the Iraq War, has been a direct consequence of
our dependence upon oil. In particular, they state that ``[w]hile there
is debate over the precise role of oil in America's wars in the greater
Middle East, several retired military members of SAFE's ESLC and other
defense budget experts that were consulted for this report believe the
connection is clear.'' \1903\ However, neither today's action, nor the
baseline standards, has the ability to change the historical wealth
transfer that created powerful nations in the Middle East. Attributing
the cost of the Iraq War, for example, to oil dependence does not
directly support an assertion that a marginal reduction in oil
dependence could have reduced the cost of that conflict.
---------------------------------------------------------------------------
\1903\ NHTSA-2018-0067-11981.
[GRAPHIC] [TIFF OMITTED] TR30AP20.411
Further, the agencies were unable to find a record of the U.S.
government attempting to calibrate U.S. military expenditures, force
levels, or deployments to any measure of the Nation's petroleum use and
the fraction supplied by imports, or to an assessment of the potential
economic consequences of hostilities in oil-supplying regions of the
world that could disrupt the global market.\1904\ Instead, changes in
U.S. force levels, deployments, and spending in such regions appear to
have been governed by purposeful foreign policy initiatives,
[[Page 24732]]
unforeseen political events, and emerging security threats, rather than
by shifts in U.S. oil consumption or imports.\1905\
---------------------------------------------------------------------------
\1904\ Crane et al. (2009) analyzed reductions in U.S. forces
and associated cost savings that could be achieved if oil security
were no longer a consideration in military planning, and disagree
with this assessment. After reviewing recent allocations of budget
resources, they concluded that ``. . . the United States does
include the security of oil supplies and global transit of oil as a
prominent element in its force planning'' at p. 74 (emphasis added).
Nevertheless, their detailed analysis of individual budget
categories estimated that even eliminating the protection of foreign
oil supplies completely as a military mission would reduce the
current U.S. defense budget by approximately 12-15 percent. See
Crane, K., A. Goldthau, M. Toman, T. Light, S.E. Johnson, A. Nader,
A. Rabasa, & H. Dogo, Imported Oil and U.S. National Security.,
Santa Monica, CA, The RAND Corporation (2009) available at https://www.rand.org/pubs/monographs/MG838.html.
\1905\ Crane et al. (2009) also acknowledge the difficulty of
reliably allocating U.S. military spending by specific mission or
objective, such as protecting foreign oil supplies. Moore et al.
(1997) conclude that protecting oil supplies cannot be distinguished
reliably from other strategic objectives of U.S. military activity,
so that no clearly separable component of military spending to
protect oil flows can be identified, and its value is likely to be
near zero. Similarly, the U.S. Council on Foreign Relations (2015)
takes the view that significant foreign policy missions will remain
over the foreseeable future even without any imperative to secure
petroleum imports. A dissenting view is that of Stern (2010), who
argues that other policy concerns in the Persian Gulf derive from
U.S. interests in securing oil supplies, or from other nations'
reactions to U.S. policies that attempt to protect its oil supplies.
See Crane, K., A. Goldthau, M. Toman, T. Light, SE Johnson, A.
Nader, A. Rabasa, and H. Dogo, Imported Oil and U.S. National
Security., Santa Monica, CA, The RAND Corporation (2009) available
at https://www.rand.org/pubs/monographs/MG838.html; Moore, John L.,
E.J. Carl, C. Behrens, and John E. Blodgett, ``Oil Imports--An
Overview and Update of Economic and Security Effects,''
Congressional Research Service, Environment and Natural Resources
Policy Division, Report 98, No. 1 (1997), pp. 1-14; Council on
Foreign Relations, ``Automobile Fuel Economy Standards in a Lower-
Oil-Price World,'' November 2015; and Stern, Roger J. ``United
States cost of military force projection in the Persian Gulf, 1976-
2007,'' Energy Policy 38, no. 6 (June 2010), pp. 2816-25, https://www.sciencedirect.com/science/article/pii/S0301421510000194?via%3Dihub.
---------------------------------------------------------------------------
The agencies thus conclude that U.S. military activity and
expenditures are unlikely to be affected by even relatively large
changes in consumption of petroleum-derived fuels by light duty
vehicles. Certainly, the historical record offers no suggestion that
U.S. military spending is likely to adjust significantly in response to
the increase in domestic petroleum use that would result from reducing
CAFE and CO2 standards.
Nevertheless, it is possible that more detailed analysis of
military spending might identify some relationship to historical
variation in U.S. petroleum consumption or imports. A number of studies
have attempted to isolate the fraction of total U.S. military spending
that is attributable to protecting overseas oil supplies.\1906\ These
efforts have produced varying estimates of how much it might be reduced
if the U.S. no longer had any strategic interest in protecting global
oil supplies. However, none has identified an estimate of spending that
is likely to vary incrementally in response to changes in U.S.
petroleum consumption or imports.
---------------------------------------------------------------------------
\1906\ These include Copulos, M R. ``America's Achilles Heel--
The Hidden Costs of Imported Oil,'' Alexandria VA--The National
Defense Council Foundation, September 2003-1-153, available at
http://ndcf.dyndns.org/ndcf/energy/NDCFHiddenCostsofImported_Oil.pdf; Copulos, M R. ``The Hidden Cost
of Imported Oil--An Update.'' The National Defense Council
Foundation (2007) available at http://ndcf.dyndns.org/ndcf/energy/NDCF_Hidden_Cost_2006_summary_paper.pdf; Delucchi, Mark A. & James
J. Murphy. ``US military expenditures to protect the use of Persian
Gulf oil for motor vehicles,'' Energy Policy 36, no. 6 (June 2008),
pp. 2253-64; and National Research Council Committee on Transitions
to Alternative Vehicles and Fuels, Transitions to Alternative
Vehicles and Fuels (2013).
---------------------------------------------------------------------------
Nor have any of these studies tracked changes in spending that can
be attributed to protecting U.S. interests in foreign oil supplies over
a prolonged period, so they have been unable to examine whether their
estimates of such spending vary in response to fluctuations in domestic
petroleum consumption or imports. The agencies conclude from this
review of research that U.S. military commitments in the Persian Gulf
and other oil-producing regions of the world contribute to worldwide
economic and political stability, and insofar as the costs of these
commitments are attributable to petroleum use, they are attributable to
oil consumption throughout the world, rather than simply U.S. oil
consumption or imports.
It is thus unlikely that military spending would rise in response
to any increase in U.S. imports that did result from this final rule.
As a consequence, the analysis of alternative CAFE and CO2
emission standards for future model years applies no increase in
government spending to support U.S. military activities as a potential
cost of allowing new cars and light trucks to achieve lower fuel
economy and thus increasing domestic petroleum use.
Similarly, while the ideal size of the Strategic Petroleum Reserve
from the standpoint of its potential stabilizing influence on global
oil prices may be related to the level of U.S. petroleum consumption or
imports, its actual size has not appeared to vary in response to either
of those measures. The budgetary costs for maintaining the SPR are thus
similar to U.S. military spending in that, while they are not reflected
in the market price for oil (and thus do not enter consumers' decisions
about how much to use), they do not appear to have varied in response
to changes in domestic petroleum consumption or imports.
As a consequence, the analysis does not include any potential
increase in the cost to maintain a larger SPR among the external or
social costs of the increase in gasoline and petroleum consumption
likely to result from reducing future CAFE and CO2
standards. This view aligns with the conclusions of most recent studies
of military-related costs to protect U.S. oil imports, which generally
conclude that savings in military spending are unlikely to result from
incremental reductions in U.S. consumption of petroleum products on the
scale of those that would resulting from adopting higher CAFE or
CO2 standards.
(13) Social Cost of Carbon
In the proposal, the agencies projected costs resulting from fuel
consumption and emissions of CO2 using estimates of
anticipated climate-related economic damages within U.S. borders per
ton of CO2 emissions, which the agencies referred to as the
domestic social cost of carbon (domestic SC-CO2). The
domestic SC-CO2 estimates, which were originally developed
by EPA for an earlier regulatory analysis, represent the monetary value
of damages to the domestic economy likely to be caused by future
changes in the climate that result from incremental increases in
CO2 emissions during a given year.\1907\ The agencies did
not consider climate-related damage costs resulting from emissions of
other greenhouse gases (GHGs), such as methane or nitrous oxide, in
their analysis supporting the proposal.
---------------------------------------------------------------------------
\1907\ For a description of the procedures EPA used to develop
these values, see U.S. Environmental Protection Agency, Regulatory
Impact Analysis for the Proposed Emission Guidelines for Greenhouse
Gas Emissions from Existing Electric Utility Generating Units;
Revisions to Emission Guideline Implementing Regulations; Revisions
to New Source Review Program, EPA-452/R-18-006, August 2018 (https://www.epa.gov/sites/production/files/2018-08/documents/utilities_ria_proposed_ace_2018-08.pdf), Section 4.3, at 4-2 to 4-7.
The sources and potential magnitude of uncertainties surrounding the
SC-CO2 estimates are described in Chapter 7 of that same
document, at 7-1 to 7-10.
---------------------------------------------------------------------------
Climate-related damages caused by emissions of CO2 and
other GHGs include changes in agricultural productivity, adverse
effects on human health, property damage from increased flood risk, and
changes in costs for managing indoor environments in commercial and
residential buildings (such as costs for heating and air conditioning),
among other possible damages.
The agencies described the SC-CO2 estimates used in the
NPRM analysis as interim values developed under Executive Order 13783,
which are to be used in regulatory analyses until revised values that
incorporate recommendations from NAS can be developed.\1908\ E.O. 13783
directed
[[Page 24733]]
agencies to ensure that estimates of the social cost of greenhouse
gases used in regulatory analyses are consistent with the guidance
contained in OMB Circular A-4, ``including with respect to the
consideration of domestic versus international impacts and the
consideration of appropriate discount rates.'' \1909\
---------------------------------------------------------------------------
\1908\ The guidance followed by EPA in developing the SC-
CO2 values used in the NPRM analysis appears in President
of the United States, Executive Order 13783, ``Promoting Energy
Independence and Economic Growth,'' March 28, 2017, Federal
Register, Vol. 82, No. 61, Friday, March 31, 2017, 16093-97.
(https://www.govinfo.gov/content/pkg/FR-2017-03-31/pdf/2017-06576.pdf) The recommendations of the National Academies are
reported in National Academies of Sciences, Engineering, and
Medicine, Valuing Climate Damages: Updating Estimation of the Social
Cost of Carbon Dioxide, Washington, DC, January 2017. Revised values
incorporating this guidance have not yet been developed.
https://www.nap.edu/catalog/24651/valuing-climate-damages-updating-estimation-of-the-social-cost-of.
\1909\ E.O. 13783, at 16096.
---------------------------------------------------------------------------
Circular A-4 states that analysis of economically significant
regulations ``should focus on benefits and costs that accrue to
citizens and residents of the United States,'' and the agencies
followed this guidance by using estimates of the SC-CO2 that
included only domestic economic damages. In response to Circular A-4's
further guidance that regulatory analyses ``should provide estimates of
net benefits using [discount rates of] both 3 percent and 7 percent,''
the agencies presented estimates of the proposed rule's economic
impacts--including the costs of climate damages likely to result from
increased CO2 emissions--that incorporated both discount
rates. The PRIA included a detailed discussion of the analyses used to
construct estimates of the domestic SC-CO2 using these
discount rates.\1910\
---------------------------------------------------------------------------
\1910\ See NHTSA and EPA, PRIA, Chapter 8, Appendix A.
---------------------------------------------------------------------------
The estimates of the domestic SC-CO2 the agencies used
in their analysis supporting the proposal increased over future years,
partly because emissions during future years are anticipated to
contribute larger incremental costs. Future values of the SC-
CO2 also increase because U.S. GDP is growing over time, and
many categories of climate-related damage are estimates as proportions
of GDP. The agencies' estimates of the domestic SC-CO2 for
emissions occurring in the year 2020 were $1 and $8 (in 2016$) per
metric ton of CO2 emissions using 7 and 3 percent discount
rates, and these values were projected to increase to $2 and $10 (again
in 2016$) by the year 2050.
As the agencies indicated in the NPRM, the SC-CO2
estimates are subject to several sources of uncertainty. In accordance
with guidance provided by OMB Circular A-4 for treating uncertainty in
regulatory analysis, the PRIA included a detailed discussion of how the
analysis used to develop the interim SC-CO2 estimates
incorporated sources of uncertainty that could be quantified. It also
demonstrated how considering the uncertainty introduced by applying
discount rates over extended time horizons could affect the estimated
values.\1911\ To reflect this uncertainty, the analysis supporting the
proposed rule examined the sensitivity of its estimated costs and
benefits to using higher values for the SC-CO2 ($9-14 per
metric ton), which were derived using a lower ``intergenerational''
discount rate of 2.5 percent.\1912\
---------------------------------------------------------------------------
\1911\ See PRIA, Chapter 8, Appendix A.
\1912\ PRIA, Tables 13-8 and 13-9, at 1547-50.
---------------------------------------------------------------------------
(a) Comments on the NPRM Value for the SC-CO2
The agencies received extensive comments on the values of the SC-
CO2 used in the NPRM analysis. Broadly, these comments
stressed the following concerns:
Using a domestic value for SC-CO2 systemically
underestimates the benefits of adopting stricter standards.
The agencies' SC-CO2 omits potential costs due
to foreign social and political disruptions caused by climate change
that can affect the U.S.
The 7 percent discount rate used in the agencies' main or
central analysis is inappropriate because it represents an opportunity
cost of capital rather than a rate of time preference for current
versus future consumption opportunities, and climate change will affect
future consumption.
(b) Domestic vs. Global Value for SC-CO2
Many commenters asserted that it was inappropriate for the agencies
to use a domestic SC-CO2 value for analyzing benefits or
costs from changing required levels of fuel economy in the NPRM
analysis, primarily because doing so could lead regulatory agencies to
adopt measures that provide inadequate reductions in emissions and
protection from potential climate change.
As noted in the NPRM and above, the SC-CO2 estimates the
agencies used to estimate climate-related economic costs from adopting
less demanding fuel economy and CO2 emission were developed
in response to the issuance of E.O. 13783. The agencies remind
commenters that E.O. 13783 directed federal agencies to ensure that
estimates of the social cost of greenhouse gases used in their
regulatory analyses are consistent with the guidance contained in OMB
Circular A-4, ``including with respect to the consideration of domestic
versus international impacts and the consideration of appropriate
discount rates.'' \1913\ Circular A-4 states that analysis of
economically significant proposed and final regulations ``should focus
on benefits and costs that accrue to citizens and residents of the
United States.'' \1914\ The agencies adhered closely to this guidance
in evaluating the economic costs and benefits in the proposal and this
final rule by using the domestic value of the SC-CO2 in our
central analysis.
---------------------------------------------------------------------------
\1913\ Executive Order 13,783, at 16096.
\1914\ White House Office of Management and Budget, Circular A-
4: Regulatory Analysis, September 17, 2003, at 15. (https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/circulars/A4/a-4.pdf).
---------------------------------------------------------------------------
Commenters argued that Circular A-4 allows the agencies to use a
global SC-CO2 in their central analysis. For example, IPI et
al. commented that ``Circular A-4's reference to effects `beyond the
borders' confirms that it is appropriate for agencies to consider the
global effects of U.S. greenhouse gas emissions.'' \1915\ While the
agencies agree that Circular A-4 authorizes the agencies to consider
foreign impacts in certain circumstances, the agencies would also like
to note that Executive Order 13783 stipulates ``when monetizing the
value of changes in greenhouse gas emissions resulting from
regulations, including with respect to the consideration of domestic
versus international impact [. . .] agencies shall ensure [. . .] any
such estimates are consistent with the guidance contained in OMB
Circular A-4.'' \1916\ Using a global SC-CO2 in our central
analysis would be inconsistent with Circular A-4's directive that any
non-domestic effects calculated ``should be reported separately.''
\1917\ As such, if the agencies had used a global SC-CO2,
this rulemaking would be compelled by Circular A-4 to separate the SC-
CO2 into domestic and foreign components, and to include
only the former in our central analysis.
---------------------------------------------------------------------------
\1915\ IPI et al., DEIS Joint SCC Comments, NHTSA-2017-0069-
0559, at 20.
\1916\ Executive Order 13,783, at 16096.
\1917\ Specifically, OMB Circular A-4 directs federal agencies
as follows: ``Where you choose to evaluate a regulation that is
likely to have effects beyond the borders of the United States,
these effects should be reported separately.'' OMB Circular A-4, at
15.
---------------------------------------------------------------------------
Furthermore, today's analysis will likely have global impacts
beyond climate change. For example, freeing manufacturers who compete
in the U.S. domestic automobile market from burdensome fuel efficiency
standards may enable them to dedicate time and resources to becoming
more competitive in global markets, and is thus likely to affect
product innovation and performance throughout the global auto
[[Page 24734]]
market.\1918\ It would be inconsistent to report the global SC-
CO2 while ignoring other global costs and benefits. The
agencies do not have a method for analyzing the comprehensive impacts
of CAFE and CO2 standards--including their many likely
impacts beyond climate change--on a global scale, and did not receive
any suggestions about how to conduct such an analysis from commenters.
Because it would be inconsistent to quantify only climate change and
none of these other potential global-scale impacts, the agencies have
decided to focus their attention on domestic impacts, which are more
readily measurable.
---------------------------------------------------------------------------
\1918\ Some commenters assert that weakening U.S. fuel economy
standards could make domestic auto companies less competitive in
international markets, since several other nations have also adopted
similar standards. For reasons discussed Section VIII.B.6. of this
rule, however, the agencies find these comments unpersuasive.
---------------------------------------------------------------------------
Several commenters argued that the agencies are still obligated to
report the global impacts of carbon. For example, the North Carolina
Department of Environmental Quality commented that ``by omitting any
analysis of the global social cost of carbon, [the agencies] failed to
adhere to OMB's Circular A-4.'' \1919\ The agencies note Circular A-4
grants agencies discretion to choose which impacts to report. However,
to be fully informed of the gamut of potential effects of today's rule,
the agencies have included two sensitivity cases analyzing the impacts
of the standards using a global SC-CO2.
---------------------------------------------------------------------------
\1919\ North Carolina Department of Environmental Quality,
Comments, NHTSA-2018-0067-12025, at 39.
---------------------------------------------------------------------------
(c) Scope of Domestic Climate Damages
Some commenters asserted that even if the agencies are required to
use a domestic SC-CO2, the specific value employed by the
agencies underestimated the domestic impacts of climate change. They
argued the agencies failed to incorporate economic costs associated
with social or economic disruptions caused by climate change in regions
of the world that were more vulnerable to its effects, but that could
``spill over'' to impose damages to the U.S. via their effects on
migration patterns, international trade flows, or other mechanisms that
connect nations. Other commenters argued that E.O. 13783 does not
prohibit the agencies from using the estimates or practices developed
by the IWG to develop new estimates of the SC-CO2, and
asserted that the IWG's methods and resulting estimates continue to
represent the best available practices.
However, all of the IWG's estimates measure the global SC-
CO2, and as discussed previously, E.O. 13783, in conjunction
with Circular A-4, directs the agencies to use a domestic SC-
CO2 which precludes the use of the IWG estimates. To develop
interim estimates of the domestic SC-CO2 that were
consistent with the IWG's procedures, EPA used the same three climate
economic models the IWG employed previously to calculate the domestic
SC-CO2. Two of those three models directly estimate the U.S.
domestic SC-CO2, which represents the economic costs
resulting from climate change that are likely to be borne within U.S.
borders.\1920\ The third model the IWG used previously does not
estimate the domestic SC-CO2 directly, but EPA approximated
domestic U.S. costs from future climate change as 10 percent of its
estimate of their global value, based on results from a companion model
developed by the same author.\1921\ Thus the agencies believed that the
SC-CO2 values they used in the NPRM analysis represented the
most reliable estimates of domestic economic costs from future climate
change that were available for use in evaluating the proposal.
---------------------------------------------------------------------------
\1920\ The Policy Analysis of the Greenhouse Effect (PAGE) model
is described in Hope, C., ``The marginal impact of CO2
from PAGE2002: An integrated assessment model incorporating the
IPCC's five reasons for concern,'' The Integrated Assessment
Journal, Vol. 6 No. 1 (2006), at 19-56; and Hope, C., ``Optimal
carbon emissions and the social cost of carbon under uncertainty,''
The Integrated Assessment Journal Vol. 8, No. 1 (2008), at 107-22.
The Climate Framework for Uncertainty, Negotiation, and Distribution
(FUND) model is documented in Tol, Richard, ``Estimates of the
damage costs of climate change. Part I: benchmark estimates,'' and
``Estimates of the damage costs of climate change. Part II: Dynamic
estimates.'' Environmental and Resource Economics Vol 21 (2002), at
47-73 and 135-60.
\1921\ The third model is the Dynamic Integrated model of
Climate and the Economy (DICE), described in Nordhaus, William,
``Estimates of the Social Cost of Carbon: Concepts and Results from
the DICE-2013R Model and Alternative Approaches.'' Journal of the
Association of Environmental and Resource Economists, Vol. 1, No. 2
(2014), at 273-312 (https://www.jstor.org/stable/pdf/10.1086/676035.pdf). The 10 percent figure is based on the results from a
regional version of that model (RICE 2010), as described in
Nordhaus, William D. 2017, ``Revisiting the social cost of carbon,''
Proceedings of the National Academy of Sciences of the United
States, 114 (7), at 1518-23, Table 2. (https://pdfs.semanticscholar.org/f83b/3a7431e0ae2d4e8be3d0ee5f3787a802c34c.pdf?_ga=2.211824467.636056015.1572384992-158339427.1562696454).
---------------------------------------------------------------------------
The agencies were unable to develop an estimate of the domestic
value for SC-CO2 that incorporated any of these alleged
spillover effects, due both to their speculative nature and to the
absence of credible empirical estimates of their potential magnitude.
Nor did commenters provide credible explanations for how such
spillovers might arise, or reliable empirical estimates of their
potential magnitude.
(d) Discount Rate Used To Construct the SC-CO2 Value
Many commenters also objected to the agencies use of an SC-
CO2 value that incorporated a 7 percent discount rate in the
NPRM analysis. Some of these comments reflected a misperception that
the agencies used such a value in their main or central analysis, when
in fact it was only used in a sensitivity analysis case as described
below. Other comments appeared to object to the agencies' use of an SC-
CO2 value incorporating a 7 percent discount rate even as a
sensitivity case.
E.O. 13783 directed agencies to ensure that any estimates of the
social cost of CO2 and other greenhouse gases they used for
purposes of regulatory analyses are consistent with OMB Circular A-4's
guidance ``with respect to the consideration of. . .appropriate
discount rates.'' \1922\ In turn, Circular A-4 refers agencies to OMB's
earlier guidance on discounting contained in its Circular A-94, noting
that ``[a]s a default position, OMB Circular A-94 states that a real
discount rate of 7 percent should be used as a base-case for regulatory
analysis.'' \1923\ OMB continues to use the 7 percent rate to estimate
the average pre-tax rate of return to private capital investment
throughout the U.S. economy. Because it is intended to approximate the
opportunity cost of capital, it is the appropriate discount rate for
evaluating the economic consequences of regulations that affect
private-sector capital investments.
---------------------------------------------------------------------------
\1922\ E.O. 13,783, at 16096.
\1923\ OMB Circular A-4, at 33.
---------------------------------------------------------------------------
At the same time, however, OMB's guidance on discounting also
recognizes that some federal regulations are more likely to affect
private consumption decisions made by households and individuals, such
as when they affect prices or other attributes of consumer goods. In
these cases, Circular A-4 advises that a lower discount rate is likely
to be more appropriate, and that a reasonable choice for such a lower
rate is the real consumer (or social) rate of time preference. This is
the rate at which individual consumers discount future consumption to
determine its present value to them.
OMB estimated that the rate of consumer time preference has
averaged 3 percent in real or inflation-adjusted terms over an extended
period, and continues to use that value. In summary, Circular A-4
reiterates the guidance provided in OMB's earlier Circular A-
[[Page 24735]]
94 that ``[f]or regulatory analysis, you should provide estimates of
net benefits using both 3 percent and 7 percent.'' \1924\
---------------------------------------------------------------------------
\1924\ OMB Circular A-4, at 34.
---------------------------------------------------------------------------
Finally, OMB's guidance on discounting indicates that it may be
appropriate for government agencies to employ an even lower rate of
time preference when their regulatory actions entail tradeoffs between
improving the welfare of current and future generations. Recognizing
this situation, Circular A-4 advises if the ``rule will have important
intergenerational benefits or costs [an agency] might consider a
further sensitivity analysis using a lower but positive discount rate
in addition to calculating net benefits using discount rates of 3 and 7
percent.'' \1925\
---------------------------------------------------------------------------
\1925\ OMB Circular A-4, at 36.
---------------------------------------------------------------------------
The agencies adhered closely to each of these provisions of OMB's
guidance on discounting future climate-related economic costs in their
analysis supporting the NPRM. Specifically, their central analysis
relied exclusively on a SC-CO2 value that was constructed by
applying a 3 percent discount rate to future climate-related economic
damages. This value ranged from $6 per metric ton in 2015 to nearly $11
per metric ton (both figures in 2016$) by the end of the analysis
period, the year 2050.
Throughout the NPRM central analysis, costs resulting from
increased emissions of CO2 were also discounted from the
year when those increases in emissions occurred to the present using a
3 percent rate, even when all other future costs and benefits were
discounted at a 7 percent rate. Thus the agencies' central analysis for
the NPRM did not use SC-CO2 values for future years that
were constructed by applying a 7 percent rate to discount distant
future climate-related economic damages, and did not use a 7 percent
rate to discount costs of increased CO2 from the years when
they were projected to occur to 2018 (the base year used in the
analysis).
Notwithstanding concerns raised by commenters about including a
sensitivity analysis that used a higher discount rate, OMB's guidance
clearly directs the agencies to report estimates of the present value
of the economic costs resulting from increased CO2 emissions
that reflect discount rates of both 3 and 7 percent. Thus to supplement
their central analysis, which as indicated previously employed a 3
percent discount rate throughout, the agencies also reported an
estimate of the economic costs of increased CO2 emissions
based on a value for the SC-CO2 that was constructed using a
7 percent discount rate as a sensitivity case, which they termed the
``Low Social Cost of Carbon'' sensitivity analysis.\1926\ The values
for the SC-CO2 used in the Low Social Cost of Carbon
sensitivity analysis varied from $1 per metric ton in 2015 to $3 per
metric ton (both figures in 2016$) by the end of the analysis period.
Using these values reduced the loss in total economic benefits
resulting from the proposed alternative by 1.1 percent, thus increasing
its net benefits by slightly less than 2 percent.\1927\
---------------------------------------------------------------------------
\1926\ PRIA, Table 13-1, at 1531-34.
\1927\ PRIA, Tables 13-8 and 13-9, at 1547-50. Using a lower
value for the SC-CO2 had opposite effects on the
proposal's total and net economic benefits, because its net benefits
represented the difference between the loss in benefits and the
savings in costs that would result from adopting the proposed rule,
compared to the baseline of adopting the Augural standards.
---------------------------------------------------------------------------
For the proposal, the agencies also included a second sensitivity
analysis using a value for the SC-CO2 that reflected a lower
``intergenerational'' discount rate of 2.5 percent, which is within the
1 to 3 percent range for discount rates that have previously been
applied to economic costs and benefits that span multiple generations,
as reported in OMB guidance.\1928\ Because using a lower discount rate
results in a higher value for the SC-CO2, this analysis was
termed the ``High Social Cost of Carbon'' sensitivity case.\1929\ The
values for the SC-CO2 used in this additional sensitivity
analysis varied from $8 per metric ton in 2015 to $14 per metric ton
(both figures in 2016$) in 2050, the last year of the analysis. Using
these higher values increased the magnitude of the estimated loss in
economic benefits resulted from adopting the proposed rule (versus
retaining the Augural standards) by 0.5 percent from that estimated in
the central analysis, thus reducing its net benefits by 1.0
percent.\1930\ Thus it appeared that when used to construct alternative
estimates of the SC-CO2, the range of discount rates
specified in OMB Circular A-4 had little or no effect on the estimated
total benefits of the proposed rule, and the sensitivity analyses
conducted in support of this Final Rule confirm this result.\1931\
---------------------------------------------------------------------------
\1928\ OMB Circular A-4, at 36.
\1929\ PRIA, Table 13-1, at 1531-34.
\1930\ PRIA, Tables 13-8 and 13-9, at 1547-50. As in the Low
Social Cost of Carbon sensitivity case, using a higher value for the
SC-CO2 had opposite effects on the total and net economic
benefits, because its net benefits were the difference between the
sacrifice in benefits and the savings in costs from adopting the
proposed rule, where both were measured against the baseline of
adopting the Augural standards.
\1931\ See section VII.B. of this Final Rule for results of the
``High Social Cost of Carbon'' sensitivity case.
---------------------------------------------------------------------------
(e) SC-CO2 for the Final Rule
After carefully considering the concerns raised by commenters, the
agencies decided to leave the SC-CO2 values unchanged for
the final rule. This means the SC-CO2 estimate used in this
analysis is still a domestic value that was constructed using a 3
percent discount rate, and that costs from increased CO2
emissions are discounted from the year those emissions occur to the
present using a 3 percent rate. The agencies have again included ``High
Social Cost of Carbon'' and ``Low Social Cost of Carbon'' sensitivity
analyses, which continue to use domestic SC-CO2 values that
incorporate alternative discount rates of 2.5 percent and 7 percent.
The agencies have also added two sensitivity cases using global
values for the SC-CO2, which reflect discount rates of 3
percent and 7 percent. Finally, the agencies have also included an
additional sensitivity case that incorporates estimates of the domestic
climate damage costs caused by emissions of the GHGs methane
(CH4) and nitrous oxide (N2O). Like the SC-
CO2 values used in this analysis, the estimates of the
domestic values for SC-CH4 and SC-N2O are interim
estimates developed by EPA for use in regulatory analyses conducted
under the guidelines specified in E.O. 13783 and OMB Circular A-4, and
incorporate a 3 percent discount rate.
(14) External Costs of Congestion and Noise
(a) Values Used To Analyze the Proposal
As explained in the proposal, changes in vehicle use affect the
levels and economic costs of traffic congestion and highway noise
associated with motor vehicle use.\1932\ Congestion and noise costs are
``external'' to the vehicle owners whose decisions about how much,
where, and when to drive more--
[[Page 24736]]
or less--in response to changes in fuel economy result in these costs.
Therefore, unlike changes in the costs incurred by drivers for fuel
consumption or safety risks they willingly assume, changes in
congestion and noise costs are not offset by corresponding changes in
the travel benefits drivers experience.\1933\
---------------------------------------------------------------------------
\1932\ The proposal estimated changes in congestion and noise
costs associated with the overall change in vehicle use, which
included changes in the use of new cars and light trucks associated
with the fuel economy rebound effect as well as with changes in the
use of older vehicles resulting from the effect of CAFE and
CO2 standards on turnover in the car and light truck
fleets. As discussed in more detail elsewhere in this final rule,
the current analysis assumes that total vehicle use (VMT) differs
between the baseline and regulatory alternatives only because of
changes in the use of cars and light trucks produced during the
model years affected by this rule that occur in response to the fuel
economy rebound effect.
\1933\ The potential contribution of increased vehicle use to
the costs of injuries and property damage caused by motor vehicle
crashes may also be partly external to drivers who elect to travel
more in response to the fuel economy rebound effect. However, these
costs are dealt with directly and in more detail than the external
costs of congestion and noise, in section VI.C.2. below.
---------------------------------------------------------------------------
Congestion costs are limited to road users; however, since road
users include a significant fraction of the U.S. population, changes in
congestion costs are treated as part of the rule's economic impact on
the broader U.S. economy instead of as a cost or benefit to private
parties. Costs resulting from road and highway noise are even more
widely dispersed, because they are borne partly by surrounding
residents, pedestrians, and other non-road users, and for this reason
are also considered as a cost to the U.S. economy as a whole.
To estimate the economic costs associated with changes in
congestion and noise caused by differences in miles driven, the
analysis supporting the NPRM used estimates of per-mile congestion and
noise costs from increased automobile and light truck use that were
originally developed by FHWA as part of its 1997 Highway Cost
Allocation Study.\1934\ The agencies previously employed these same
cost estimates in the 2010, 2011, and 2012 final rules.
---------------------------------------------------------------------------
\1934\ Federal Highway Administration, 1997 Highway Cost
Allocation Study, Chapter V, Tables V-22 and V-23, available at
https://www.fhwa.dot.gov/policy/hcas/final/five.cfm.
---------------------------------------------------------------------------
The marginal congestion cost estimates reported in the 1997 FHWA
study were intended to measure the costs of increased congestion
resulting from incremental growth in travel by different types of
vehicles (including autos and light trucks), and the delays it causes
to drivers, passengers, and freight shipments. As explained in the 1997
FHWA study, the distinction between marginal and average costs is
extremely important in considering congestion costs on a per-vehicle-
mile basis. Average congestion costs on a section of highway are
calculated as the total congestion costs experienced by all vehicles,
divided by total vehicle miles. In contrast, marginal congestion costs
are calculated as the increase in congestion costs resulting from an
incremental increase in vehicle miles.
Marginal congestion costs are significantly higher than average
congestion costs because each additional vehicle that enters a crowded
roadway slows travel speeds only slightly, thus adding only modestly to
the average travel time of vehicles already on the road. During
congested conditions, however, this modest increase is experienced by a
very large number of vehicles, so the resulting increase in total delay
experienced by all travelers using the road can be extremely large. As
a consequence, the increases in total delay and congestion costs
associated with additional driving are more than proportional to
changes in VMT that cause them.\1935\
---------------------------------------------------------------------------
\1935\ Such ``non-linearity'' is a common feature of complex
systems, such as computing or juggling. Each additional element
added to a computation, or ball to a cascade, makes performing the
task more difficult than the last addition.
---------------------------------------------------------------------------
The FHWA study's estimates of marginal noise costs reflected the
variation in noise levels resulting from incremental changes in travel
by autos, light trucks, and other vehicles, and the annoyance and other
adverse impacts caused by noise. These included adverse impacts on
pedestrians and residents of the surrounding area, as well as on
vehicle occupants themselves.
To calculate the incremental costs of congestion and noise, the
agencies multiplied FHWA's ``middle'' estimates of marginal congestion
and noise costs per mile of auto and light truck travel in urban and
rural areas by the annual increases in driving attributable to the
standards to yield increases in total congestion and noise externality
costs. Because the proposal, and other alternatives that were
considered, reduced the stringency of CAFE and CO2 standards
for model years 2021-2026, resulting in lower fuel economy for new cars
and light trucks produced during those years, the fuel economy rebound
effect resulted in fewer miles driven relative to the baseline, thus
generating savings in congestion and noise costs relative to their
levels under the baseline. Similarly, each of those alternatives also
reduced the total amount of travel by the used vehicle fleet,
generating additional savings in these costs.
(b) Comments on the NPRM Values
The agencies received few comments on the estimates of congestion
and noise costs they used to analyze the economic impacts of the
proposal. Almost all of these comments focused on the appropriateness
of the estimated magnitude of the fuel economy rebound effect they used
to estimate the change in use of new cars and light trucks or the
plausibility of the reduction in driving by used vehicles, rather than
to the unit costs estimates themselves. These included comments from
ICCT and CARB.\1936\
---------------------------------------------------------------------------
\1936\ ICCT, Comment, NHTSA-2018-0067-11741 at 121; CARB,
Comment, NHTSA-2018-0067-11873 at 316.
---------------------------------------------------------------------------
One individual commenter did suggest that recent growth in traffic
levels, resulting in part from increased use of home delivery services
for online purchases, has increased congestion and resulting
delays.\1937\ Although this commenter is correct, traffic growth is not
strictly a recent phenomenon, and longer-term growth in vehicle use--
combined with comparatively modest increases in road and highway
capacity--has contributed to increasing congestion levels. Because
congestion increases more than proportionately to growing traffic
volumes, this suggests that FHWA's estimates of congestion costs--now
more than two decades old--are likely to understate the contribution of
continuing increases in vehicle use to congestion, resulting delays to
vehicle occupants and freight shipments, and their associated costs.
Because noise levels also increase non-linearly with the volume of
traffic using roads and highways, FHWA's 1997 estimates of marginal
noise costs may also understate current values.
---------------------------------------------------------------------------
\1937\ Richard Carriere, NHTSA-2018-0067-12216.
---------------------------------------------------------------------------
(c) Values Used To Analyze the Final Rule
The agencies are retaining the same methodology employed in the
NPRM to estimate congestion and noise costs for the final rule. Like
other nominal estimates used throughout the analysis, the agencies have
updated the FHWA estimates to account for current economic and highway
conditions. The major determinants of marginal congestion costs imposed
by additional travel include baseline traffic volumes, which determine
current travel speeds and how they would change in response to further
increases in travel, together with vehicle occupancy and the value of
occupants' travel time. These last two factors interact to determine
the average hourly value of delays to vehicles, which is by far the
largest component of the total cost of delays that occur under
congested travel conditions.\1938\ Because travel speeds measure the
duration of congestion-related delays, while the
[[Page 24737]]
value of vehicle occupants' time determines their hourly cost, the
effects of changes in these variables on overall congestion costs is
approximately additive, as long as changes in the two are relatively
modest.
---------------------------------------------------------------------------
\1938\ Fuel consumption and other operating costs can also
increase during travel in congested conditions, but their
relationships to the frequent changes in speed that typically occur
in congested travel is less well understood, and in any case, they
vary by far smaller amounts than the value of vehicle occupants'
travel time.
---------------------------------------------------------------------------
The agencies approximated the effect of growth in traffic volumes
on travel speeds and congestion-related delays by increasing congestion
costs in proportion to the increase in annual vehicle-miles of travel
per lane-mile on major U.S. highways that occurred between 1997 and
2017.\1939\ Next, they estimated the increase in the value of travel
time per vehicle-hour over that same period by combining growth in the
value of travel time per person-hour--estimated in accordance with DOT
guidance \1940\--with the increase in average vehicle occupancy by
persons 16 years of age and older (the same measure of occupancy used
to estimate the value of refueling time elsewhere in this
analysis).\1941\ The agencies applied the increases in congestion-
related delays and the hourly value of travel time to FHWA's 1997
estimates of marginal congestion costs to update those original values
to reflect current conditions. The updated values of external
congestion costs are $0.154 per vehicle-mile of increased travel by
cars and $0.138 per vehicle-mile for light trucks (expressed in
constant 2018 dollars), and these values are assumed to remain constant
throughout the analysis period.
---------------------------------------------------------------------------
\1939\ Traffic volumes, as measured by the annual number of
vehicle-miles traveled per lane-mile of roads and highways
nationwide, rose by 53 percent between 1997 and 2017. Calculated
from FHWA, Highway Statistics, 1998 and 2018, Tables VM-1 and HM-48,
available at https://www.fhwa.dot.gov/policyinformation/statistics.cfm.
\1940\ See U.S. Department of Transportation, ``Revised
Departmental Guidance for the Valuation of Travel Time in Economic
Analysis,'' 2016, at 5-6 and Table 1 at 13.
\1941\ The average hourly value of travel time increased by 82
percent between 1997 and 2017; see U.S. Department of
Transportation, ``Departmental Guidance for the Valuation of Travel
Time in Economic Analysis,'' April 9, 1997, Table 4, and U.S.
Department of Transportation, ``Benefit-Cost Analysis Guidance for
Discretionary Grant Programs,'' December 2018, Table A-3. From 1995
to 2017, the average number of light-duty vehicle occupants 16 years
of age and older increased by 18 percent; values were tabulated from
FHWA, Nationwide Personal Transportation Survey, 2005 and 2017,
using online table designer available at https://nhts.ornl.gov/ and
https://nhts.ornl.gov/index9.shtml.
---------------------------------------------------------------------------
Similarly, the agencies revised the FHWA estimate of marginal noise
costs by adjusting for inflation--since the 1994 base year used to
express values in the FHWA study. Because marginal noise costs are so
small--less than $0.001 per mile of travel for both cars and light
trucks--this change did not have a significant impact on the agencies'
estimates of benefits and costs from the final rule.
(15) Labor Utilization Assumptions
In previous joint CAFE/CO2 rulemakings, the agencies
considered employment impacts on the automobile manufacturing industry,
but many of the considerations were qualitative. In the NPRM, the
agencies presented and took comment on a methodology to quantify
roughly the direct labor utilization impacts. The agencies recognize
there is significant uncertainty in any forward-looking
characterization of labor utilization, including effects resulting from
CAFE/CO2 rulemakings. Changes to other policies such as
trade policies and tariff policies are likely substantially to alter
underlying assumptions presented in the analysis for the rulemaking,
and these changes could dwarf any differences between policy
alternatives presented. In this section the agencies discuss the
assumptions made in the NPRM analysis, summarize comments received on
that work, and respond to these comments.
(a) Labor Utilization Baseline (Including Multiplier Effect) and Data
Description
In prior CAFE/CO2 rulemakings, the agencies considered
an analysis of employment impacts in some form in setting both CAFE and
tailpipe CO2 emissions standards; NHTSA conducted an
employment analysis in part to determine whether the standards the
agency set were economically practicable, that is, whether the
standards were ``within the financial capability of the industry, but
not so stringent as to'' lead to ``adverse economic consequences, such
as a significant loss of jobs or unreasonable elimination of consumer
choice.'' \1942\ EPA similarly conducted an employment analysis under
the authority granted to the agency under the Clean Air Act.\1943\ Both
agencies recognized the uncertainties inherent in estimating employment
impacts; in fact, both agencies dedicated a substantial amount of
discussion to uncertainty in employment analyses in the 2012 final rule
for MYs 2017 and beyond.\1944\ Notwithstanding these uncertainties, by
imposing costs on new light duty vehicles, CAFE and CO2
standards can have an impact on the demand for labor. Providing the
best analysis practicable better informs stakeholders and the public
about the standards' impact than would omitting any estimates of
potential labor impacts.
---------------------------------------------------------------------------
\1942\ 67 FR 77015, 77021 (Dec. 16, 2002).
\1943\ See George E. Warren Corp. v. EPA, 159 F.3d 616, 623-24
(D.C. Cir. 1998) (ordinarily permissible for EPA to consider factors
not specifically enumerated in the Act).
\1944\ See 77 FR 62624, 62952, 63102 (Oct. 15, 2012).
---------------------------------------------------------------------------
The NPRM quantified many of the effects that were previously
qualitatively identified, but not considered. For instance, in the PRIA
for the 2017-2025 rule EPA identified ``demand effects,'' ``cost
effects,'' and ``factor shift effects'' as important considerations for
labor, but the analysis did not attempt to quantify each of these
effects.\1945\
---------------------------------------------------------------------------
\1945\ U.S. EPA, ``Regulatory Impact Analysis: Final Rulemaking
for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards
and Corporate Average Fuel Economy Standards,'' at 8-24 to 8-32
(Aug. 2012).
---------------------------------------------------------------------------
The NPRM analysis considered direct labor effects on the automotive
sector. The NPRM evaluated how labor utilization in different facets of
the automobile manufacturing industry may be affected by the rule,
including (1) dealership labor related to new light-duty vehicle unit
sales; (2) assembly labor for vehicles, for engines and for
transmissions related to new vehicle unit sales; and (3) labor related
to mandated additional fuel savings technologies, accounting for new
vehicle unit sales. Importantly, this analysis did not consider whether
price reductions and regulatory savings associated with different
standards would, because price reductions would allow consumers to save
or spend that money on other things of value, increase the consumption
of other vehicle technologies or, more generally, generate growth in
other sectors of the overall economy. This means that the analysis is
inherently and artificially narrow in its focus, and does not represent
an attempt to quantify the overall labor or economic effects of this
rulemaking. All labor effects were estimated and reported at a national
level, in person-years, assuming 2,000 hours of labor per person-
year.\1946\
---------------------------------------------------------------------------
\1946\ The agencies recognize a few local production facilities
may contribute meaningfully to local economies, but the analysis
reported only on national effects.
---------------------------------------------------------------------------
The NPRM analysis estimated labor effects from the forecasted CAFE
model technology costs and from review of automotive labor for the MY
2016 fleet. For each vehicle in the CAFE model analysis, the locations
for vehicle assembly, engine assembly, and transmission assembly and
estimated labor in MY 2016 were recorded. The percent of U.S. content
for each vehicle was also recorded.\1947\ The analysis also
[[Page 24738]]
took into account the portion of parts that are made in the U.S. by
holding constant the percent of U.S. content for each vehicle as
manufacturers add fuel-savings technologies. The analysis further
assumes that the U.S. labor added would be proportional to U.S.
content, which means that the analysis assumes that U.S. labor inputs
would remain constant over time, but this does not reflect a prediction
that U.S. labor inputs actually will remain constant.\1948\ From this
foundation, the analysis forecasted automotive labor effects as the
CAFE model added fuel economy technology and adjusted future sales for
each vehicle.
---------------------------------------------------------------------------
\1947\ NHTSA provides reports under 49 CFR part 583, ``American
Automobile Labeling Act Reports'' with information NHTSA received
from vehicle manufacturers about the U.S./Canadian content (by
percentage value) of the equipment (parts) used to assemble
passenger motor vehicles. See https://www.nhtsa.gov/part-583-american-automobile-labeling-act-reports.
\1948\ This is a key assumption that should be revisited as
trade deals and tax or tariff policies materially change.
---------------------------------------------------------------------------
The NPRM analysis also accounted for sales projections in response
to the different regulatory alternatives; the labor analysis considers
changes in new vehicle prices and new vehicle sales (for further
discussion of the sales model, see Section VI.D.1.b(2)). As vehicle
prices rise, the analysis expected consumers to purchase fewer vehicles
than they would have at lower prices.\1949\ As manufacturers sell fewer
vehicles, the manufacturers may need less labor to produce the vehicles
and dealers may need less labor to sell the vehicles. However, as
manufacturers add equipment to each new vehicle, the industry will
require labor resources to develop, sell, and produce additional fuel-
saving technologies. The analysis also accounted for the possibility
that new standards could shift the relative shares of passenger cars
and light trucks in the overall fleet (see Section VI.D.1.b(2));
insofar as different vehicles involved different amounts of labor, this
shifting impacts the quantity of estimated labor. The labor analysis
took into account the anticipated reduction in vehicle sales, shifts in
the mix of passenger cars and light trucks, and addition of fuel-
savings technologies that result from the regulation--and,
subsequently, the anticipated increase in sales and reduction of fuel-
savings technologies that are expected to result from a reduction in
stringency.
---------------------------------------------------------------------------
\1949\ Many commenters contend that higher prices for more
efficient goods will have no effect on unit sales and hence
necessary production resources and employment. The sales aspect of
labor utilization is addressed in the sales section. NHTSA-2018-
0067-12000-35, Center for Biological Diversity, et al.
---------------------------------------------------------------------------
For the NPRM analysis, the agencies assumed that some observations
about the production of MY 2016 vehicles would carry forward, unchanged
into the future. For instance, assembly plants would remain the same as
MY 2016 for all products now, and in the future. The analysis assumed
the percent of U.S. content would remain constant, even as
manufacturers updated vehicles and introduced new fuel-saving
technologies. The analysis further assumed that assembly labor hours
per unit would remain at estimated MY 2016 levels for vehicles,
engines, and transmissions, and the factor between direct assembly
labor and parts production labors would remain the same. When
considering shifts from one technology to another, the analysis assumed
revenue per employee at suppliers and original equipment manufacturers
would remain in line with MY 2016 levels, even as manufacturers added
fuel-saving technologies and realized cost reductions from learning.
The NPRM analysis focused on automotive labor because adjacent
employment factors and consumer spending factors for other goods and
services are uncertain and difficult to predict. The analysis did not
consider how direct labor changes may affect the macro economy and
possibly change employment in adjacent industries. For instance, the
analysis did not consider possible labor changes in vehicle maintenance
and repair, nor did it consider changes in labor at retail gas
stations. The analysis did not consider possible labor changes due to
raw material production, such as production of aluminum, steel, copper,
and lithium, nor did the agencies consider possible labor impacts due
to changes in production of oil and gas, ethanol, and electricity. The
analysis did not analyze potential labor effects arising from
consumption of other products that would not have occurred but for
improved fuel economy, nor did the analysis assess the effects arising
from reduced consumption of other products that results from more
expensive fuel savings technologies at the time of purchase. The
effects of increased usage of car-sharing, ride-sharing, and automated
vehicles were not analyzed. The analysis did not estimate how changes
in labor from any of these industries could affect gross domestic
product and possibly affect other industries as a result.
Many commenters voiced concerns that the NPRM analysis only
included automotive direct employment, and did not explicitly consider
other important factors, and that these factors would be better
addressed with a macroeconomic model. For instance, the International
Council on Clean Transportation contended that the dollars saved at the
pump as a result of fuel saving technologies would be spent elsewhere
in the economy, creating jobs.\1950\ The Association of Global
Automakers also referenced macroeconomic studies that project long-term
job gains due to savings at the pump, but also highlight short-term
setbacks for jobs as money spent to purchase additional fuel saving
technologies on new vehicles is not spent in other job creating sectors
of the U.S. economy, which were not considered in an analysis that only
addresses direct automotive employment.\1951\ The Union of Concerned
Scientists and Environmental Defense Fund argued that the modeling of
short-term job losses in the macroeconomic models is incorrect, and
that purchasing a new vehicle, especially if financed, should increase
disposable income, because monthly savings at the pump outpace the
monthly financed cost of the fuel saving equipment, but also that
consumers will not choose this equipment unless a stringent standard is
chosen.\1952\ The Institute for Policy Integrity commented that an
analysis looking only at direct employment is incomplete, and
encouraged the agencies to include long-term and economy-wide effects
in scope on employment discussions.\1953\
---------------------------------------------------------------------------
\1950\ NHTSA-2018-0067-11741-145, ICCT.
\1951\ NHTSA-2018-0067-12032-30, Association of Global
Automakers.
\1952\ NHTSA-2018-0067-12039-38, Union of Concerned Scientists;
NHTSA-2018-0067-12397-4, Environmental Defense Fund, et al.
\1953\ NHTSA-2018-0067-12213-66, Institute for Policy Integrity.
---------------------------------------------------------------------------
The agencies have not quantified employment effects outside of
automotive sector direct employment for this final rule. The agencies
agree with commenters that the reductions in production costs of new
vehicles will free up resources for other productive pursuits. Some
producers may shift resources away from the development and production
of fuel saving technologies and into the development and production of
other vehicle attributes. In this case, there would be a transfer of
labor resources within a firm. Other producers may instead pass along
the reduction in production costs to consumers in the form of price
reductions or avoided price increases, allowing those consumers to
allocate those new funds between expenditure in other consumption
categories or savings. The increased expenditure in other consumption
categories would more efficiently create new employment in sectors
expanding to cover new market-based (as opposed to regulatory-
[[Page 24739]]
based) demand. Increased savings also creates additional investment in
new productive capital, which will generate employment opportunities in
the future. However, the extent and nature of these effects are all
highly uncertain, and the agencies have therefore not quantified the
effect of the rule on economy-wide employment in the final rule
analysis.
Many commenters expressed concern that America would cede
leadership in development and production of fuel saving technologies,
and fuel-saving technology investment would be gutted if augural
standards were not kept in place. For instance, the Mayor of the City
of Chillicothe, and Mayors of other Ohio cities, pointed out that many
light duty vehicles are built in Ohio and neighboring geographies, and
that workers designing and producing fuel economy equipment make an
average annual salary of $61,500, expressing concern that if standards
are lowered, some of these jobs may no longer be necessary.\1954\ The
BlueGreen Alliance pointed out that over the last twenty years,
manufacturers have invested billions of dollars into fuel saving
technologies, and that multinational companies may shift jobs to other
countries if the standards do not require continued, strong, additional
investment in even more fuel saving technologies.\1955\
---------------------------------------------------------------------------
\1954\ NHTSA-2018-0067-12318-2, Mayors of the City of
Chillicothe and other Ohio cities.
\1955\ NHTSA-2018-0067-12009-6, BlueGreen Alliance.
---------------------------------------------------------------------------
The agencies recognize that development of fuel saving technologies
can be capital intensive. However, high fuel economy standards do not,
per se, guarantee multinational companies will invest in American
research and development or production. For example, the larger percent
U.S. content in the MY 2017 light truck vs. the MY 2017 passenger car
new vehicle fleet may be tied to the so-called ``Chicken Tax,'' a long-
established tariff on the import of light duty trucks.\1956\ On
average, a light truck in the MY 2017 fleet contained 47.8 percent U.S.
content, while a passenger car contained 36.0 percent U.S. content. To
the extent that other policies encourage multi-national corporations to
build and invest in U.S. production facilities, these organizations
will need access to capital to do so. Notably, as part of the sales
module, as fuel economy of the fleet improves, the agencies assume
customers increasingly choose light trucks, meaning that a shift
towards light-trucks is already considered in the CAFE model under the
augural standards.
---------------------------------------------------------------------------
\1956\ On average, a light truck in the MY 2017 fleet contained
47.8 percent U.S. content, while a passenger car contained 36.0
percent U.S. content.
[GRAPHIC] [TIFF OMITTED] TR30AP20.412
Finally, no assumptions were made about part-time-level of
employment in the broader economy and the availability of human
resources to fill positions. When the economy is at full employment, a
fuel economy regulation is unlikely to have much impact on net overall
U.S. employment; instead, labor would primarily be shifted from one
sector to another. These shifts in employment impose an opportunity
cost on society, as regulation diverts workers from other market-based
activities in the economy. In this situation, any effects on net
employment are likely to be transitory as workers change jobs (e.g.,
some workers may need to be retrained or require time to search for new
jobs, while short-term labor shortages in some sectors or regions could
result in firms bidding up wages to attract workers). On the other
hand, if a regulation comes into effect during a period of less-than-
full employment, a change in labor demand due to regulation would
affect net overall U.S. employment because the labor market is not in
equilibrium. Schmalensee and Stavins point out that net positive
employment effects are possible in the near term when the economy is at
less than full employment due to the potential hiring of idle labor
resources by the regulated sector to meet new requirements (e.g., to
install new equipment) and new economic activity
[[Page 24740]]
in sectors related to the regulated sector longer run.\1957\ However,
the net effect on employment in the long run is more difficult to
predict and will depend on the way in which the related industries
respond to regulatory requirements. For that reason, this analysis does
not include multiplier effects but instead focuses on labor impacts in
the most directly affected industries, which would face the most
concentrated labor impacts.
---------------------------------------------------------------------------
\1957\ Schmalensee, Richard, and Robert N. Stavins. ``A Guide to
Economic and Policy Analysis of EPA's Transport Rule.'' White paper
commissioned by Excelon Corporation, March 2011 (Docket EPA-HQ-OAR-
2010-0799-0676).
---------------------------------------------------------------------------
(b) Estimating Labor for Fuel Economy Technologies, Vehicle Components,
Final Assembly, and Retailers
The following sections discuss the approaches to estimating factors
related to dealership labor, final assembly labor and parts production,
and fuel economy technology labor.
(i) Dealership Labor
The NPRM analysis evaluated dealership labor related to new light-
duty vehicle sales, and estimated the labor hours per new vehicle sold
at dealerships, including labor from sales, finance, insurance, and
management. The effect of new car sales on the maintenance, repair, and
parts department labor is expected to be limited, as this need is based
on the vehicle miles traveled of the total fleet. To estimate the labor
hours at dealerships per new vehicle sold, the agencies referenced the
National Automobile Dealers Association 2016 Annual Report, which
provides franchise dealer employment by department and function.\1958\
The analysis estimated that slightly less than 20 percent of dealership
employees' work relates to new car sales (versus approximately 80
percent in service, parts, and used car sales), and that on average
dealership employees working on new vehicle sales labor for 27.8 hours
per new vehicle sold. The analysis presented today retains assumptions
about dealership labor hours per vehicle sold.
---------------------------------------------------------------------------
\1958\ NADA Data 2016: Annual Financial Profile of America's
Franchised New-Car Dealerships, National Automobile Dealers
Association, https://www.nada.org/2016NADAdata/ (last visited
December 20, 2019).
---------------------------------------------------------------------------
(ii) Final Assembly Labor and Parts Production
As new vehicle sales increase or decrease, the amount of labor
required to assemble parts and vehicles changes accordingly. The NPRM
evaluated how the quantity of assembly labor and parts production labor
for MY 2016 vehicles would increase or decrease in the future as new
vehicle unit sales increased or decreased. Specific assembly locations
for final vehicle assembly, engine assembly, and transmission assembly
for each MY 2016 vehicle were identified. In some cases, manufacturers
assembled products in more than one location, and the analysis
identified such products and considered parallel production in the
labor analysis.
The analysis estimated average direct assembly labor per vehicle
(30 hours), per engine (four hours), and per transmission (five hours)
based on a sample of U.S. assembly plant employment and production
statistics and other publicly available information. The analysis used
the assembly locations and averages for labor per unit to estimate U.S.
assembly labor hours for each vehicle. U.S. assembly labor hours per
vehicle ranged from as high as 39 hours if the manufacturer assembled
the vehicle, engine, and transmission at U.S. plants, to as low as zero
hours if the manufacturer imported the vehicle, engine, and
transmission.
The analysis also considered labor for parts production. The
agencies surveyed motor vehicle and equipment manufacturing labor
statistics from the U.S. Census Bureau, the Bureau of Labor Statistics,
and other publicly available sources. The agencies found that the
historical average ratio of vehicle assembly manufacturing employment
to employment for total motor vehicle and equipment manufacturing for
new vehicles was roughly constant over the period from 2001 through
2013, at a ratio of 5.26.\1959\ Observations from 2001-2013 included
many combinations of technologies and technology trends, and many
economic conditions, yet the ratio remained about the same over time.
Accordingly, the analysis scaled up estimated U.S. assembly labor hours
by a factor of 5.26 to consider U.S. parts production labor in addition
to assembly labor for each vehicle. The estimates for vehicle assembly
labor and parts production labor for each vehicle scaled up or down as
unit sales scaled up or down over time in the CAFE model.
---------------------------------------------------------------------------
\1959\ NAICS Code 3361, 3363.
---------------------------------------------------------------------------
The analysis presented today retains assumptions about coefficients
for final assembly labor and parts production, and updates production
and final assembly locations for the MY 2017 fleet. As discussed in
Section VI.D.1.b(2), today's analysis also applies updated methods for
estimating the extent to which changes in CAFE and CO2
standards might lead to changes in quantities of new vehicles sold each
year. These estimated changes in sales lead to changes in estimated
changes in domestic employment.
(iii) Fuel Economy Technology Labor
As manufacturers spend additional dollars on fuel-saving
technologies, parts suppliers and manufacturers require labor to bring
those technologies to market. Manufacturers may add, shift, or replace
employees in ways that are difficult for the agencies to predict;
however, it is expected that the revenue per labor hour at original
equipment manufacturers (OEMs) and suppliers will remain about the same
as in MY 2016 even as manufacturers include additional fuel-saving
technology. To estimate the average revenue per labor hour at OEMs and
suppliers, the analysis looked at financial reports from publicly
traded automotive businesses.\1960\ Based on recent figures, it was
estimated that OEMs would add one labor year per each $633,066
increment in revenue and that suppliers would add one labor year per
$247,648 in revenue.\1961\ These global estimates are applied to all
revenues, and U.S. content is applied as a later adjustment. In today's
analysis, the agencies assume these ratios would remain constant for
all technologies rather than that the increased labor costs would be
shifted toward foreign countries. There are some reasons to believe
that this may be a conservative assumption. For instance, domestic
manufacturers may react to increased labor costs by searching for
lower-cost labor in other countries.
---------------------------------------------------------------------------
\1960\ The analysis considered suppliers that won the Automotive
News ``PACE Award'' from 2013-2017, covering more than 40 suppliers,
more than 30 of which are publicly traded companies. Automotive News
gives ``PACE Awards'' to innovative manufacturers, with most recent
winners earning awards for new fuel-savings technologies.
\1961\ The analysis assumed incremental OEM revenue as the
retail price equivalent for technologies, adjusting for changes in
sales volume. The analysis assumed incremental supplier revenue as
the technology cost for technologies before retail price equivalent
mark-up, adjusting for changes in sales volume.
---------------------------------------------------------------------------
The analysis presented today retains assumptions about coefficients
for fuel economy technology labor, and updates the percent of U.S.
content for the MY 2017 fleet.
(iv) Labor Calculations
The agencies estimated the total labor effect as the sum of three
components: changes to dealership hours, final assembly and parts
production, and labor for fuel-economy technologies (at OEMs and
suppliers) that are due to the final rule. The CAFE model calculated
[[Page 24741]]
additional labor hours for each vehicle, based on current vehicle
manufacturing locations and simulation outputs for additional
technologies, and sales changes. The analysis applied some constants to
all vehicles.\1962\ Other constants were vehicle specific, for all
years considered in the analysis.\1963\ Still, other constants were
year-specific for a vehicle.\1964\ While a multiplier effect of all
U.S. automotive related labor on non-auto related U.S. jobs was not
considered for the final rule's analysis, the analysis did incorporate
a ``global multiplier'' that can be used to scale up or scale down the
total labor hours. This parameter exists in the parameters file, and
for the final rule's analysis the analysis set the value at 1.00. The
results of this analysis can be found in Table VI-201 below.
---------------------------------------------------------------------------
\1962\ The analysis applied the same assumptions to all
manufacturers for annual labor hours per employee, dealership hours
per unit sold, OEM revenue per employee, supplier revenue per
employee, and factor for the jobs multiplier.
\1963\ The analysis made vehicle-specific assumptions about
percent of U.S. content and U.S. assembly employment hours.
\1964\ The analysis estimated technology cost for each vehicle,
for each year based on the technology content applied in the CAFE
model, year-by-year.
[GRAPHIC] [TIFF OMITTED] TR30AP20.413
Results of this analysis can be found in Section VII. Considering
that, all else equal, increases in new vehicle sales lead to increases
in domestic employment while decreases in technology outlays lead to
decreases in domestic employment, the agencies estimate that less
stringent standards could slightly reduce domestic employment. It is
important to note, however, that the reduction in person-years
described in this table merely reflects the fact that, when compared to
the standards set in 2012, fewer jobs will be specifically created to
meet regulatory requirements that, for other reasons, are not
economically practicable. It is also important to note that avoided
outlays for technology can be invested by manufacturers into other
areas, or passed on to consumers. Moreover, consumers can either take
those cost savings in the form of a reduced vehicle price, or used
toward the purchase of specific automotive features that they desire
(potentially including a more-efficient vehicle), which would increase
employment among suppliers and manufacturers.
2. Simulating Safety Impacts of Regulatory Alternatives
The primary objectives of CAFE and CO2 standards are to
achieve maximum feasible fuel economy and reduce CO2
emissions, respectively, from the light-duty vehicle fleet. In setting
standards to achieve these intended effects, the potential of the
standards to affect vehicle safety is also considered. As a safety
agency, NHTSA has long considered the potential for adverse safety
consequences when establishing CAFE standards, and under the CAA, EPA
considers factors related to public health and human welfare, including
safety, in regulating emissions of air pollutants from mobile sources.
Safety trade-offs associated with increases in fuel economy
standards have occurred in the past--particularly before CAFE standards
became attribute-based--because manufacturers chose to comply with
stricter standards by building smaller and lighter vehicles. In cases
where fuel economy improvements were achieved through reductions in
vehicle size and mass, the smaller, lighter vehicles did not protect
their occupants as effectively in crashes as larger, heavier vehicles,
on average. Although the agencies now use attribute-based standards, in
part to reduce the incentive to downsize vehicles to comply with CAFE
and CO2 standards, the agencies must continue to be mindful
of the possibility of safety-related trade-offs.
Although prior analyses acknowledged that CAFE and CO2
standards could influence factors that affect safety other than vehicle
mass, those impacts were not estimated quantitatively.\1965\ Instead,
the agencies focused exclusively on the safety impacts of changes in
vehicle mass. In the proposal, the safety analysis was expanded to
include a broader and more comprehensive measure of safety impacts. The
final rule retains this comprehensive approach and analyzes the safety
impact of three factors:
---------------------------------------------------------------------------
\1965\ The agencies included a quantification of rebound-
associated safety impacts in its Draft TAR analysis, but because the
scrappage model is new for this rulemaking, did not include safety
impacts associated with the effect of standards on new vehicle
prices and thus on fleet turnover. The fact that the scrappage model
did not exist prior to this rulemaking does not mean that the
effects that it aims to show were not important considerations,
simply that the agencies were unable to account for them
quantitatively prior to the current rulemaking.
---------------------------------------------------------------------------
(1)Changes in Vehicle Mass. Similar to previous analyses, the
agencies calculate the safety impact of changes in vehicle mass made to
reduce fuel consumption and comply with the standards. The agencies'
statistical analysis of historical crash data indicates reducing mass
in heavier vehicles generally improves safety, while reducing mass in
lighter vehicles generally reduces safety. NHTSA's crash simulation
modeling of vehicle design
[[Page 24742]]
concepts for reducing mass revealed similar effects.
(2)Impacts of Vehicle Prices. Vehicles have become safer over time
through a combination of new safety regulations and voluntary safety
improvements. The agencies expect this trend to continue as emerging
technologies, such as advanced driver assistance systems, are
incorporated into new vehicles. Safety improvements will likely
continue regardless of changes to CAFE standards. However, the pace of
such improvements may be modified if manufacturers choose to delay or
forgo investments in safety technology because of the demands that
complying with stricter CAFE and CO2 standards impose on
scarce research, development, and manufacturing resources.
As discussed in Section VI.D.1.b), technologies added to comply
with fuel economy standards have an impact on vehicle prices, and, by
extension, on the affordability of newer, safer vehicles, and therefore
on the rates at which newer vehicles are acquired and older, less safe
vehicles are retired from use. The delays in fleet turnover caused by
the effect of new vehicle prices on sales and scrappage rates affect
safety, by slowing the penetration of new safety technologies into the
fleet.
The standards also influence the composition of the light-duty
fleet. As the safety provided by light trucks, SUVs and passenger cars
responds differently to technology that manufacturers employ to meet
the standards--particularly mass reduction--fleets with different
compositions of body styles will have varying numbers of fatalities, so
changing the share of each type of light-duty vehicle in the projected
future fleet impacts safety outcomes.
(3)Increased driving because of better fuel economy.The ``rebound
effect'' predicts consumers will drive more when the cost of driving
declines. More stringent standards reduce vehicle operating costs, and
in response, some consumers may choose to drive more. Additional
driving increases exposure to risks associated with motor vehicle
travel, and this added exposure translates into higher fatalities and
injuries.
We measure the impact of these factors as differences in fatalities
across the alternatives. Fatalities are calculated by deriving a fleet-
wide fatality rate (fatalities per vehicle mile of travel)
incorporating the different factors and multiplying it by the
alternative's expected VMT. Fatalities are converted into a societal
cost by multiplying fatalities with the DOT-recommended value of a
statistical life (VSL). As with the NPRM, traffic injuries and property
damage are not modeled directly; \1966\ rather, traffic injuries and
property damage continue to be estimated using adjustment factors that
reflect the observed relationship between societal costs of fatalities
and costs of injuries and property damage.
---------------------------------------------------------------------------
\1966\ The agencies noted in the NPRM that traffic injuries and
property damage are not directly modeled because of insufficient
data. See PRIA at 43108.
---------------------------------------------------------------------------
All three factors influence predicted fatalities, but only two of
them--changes in vehicle mass and in the composition of the light-duty
fleet in response to changes in vehicle prices--impose increased risks
on drivers and passengers that are not compensated for by accompanying
benefits. In contrast, increased driving associated with the rebound
effect is a consumer choice that reveals the benefit of additional
travel. Consumers who choose to drive more have apparently concluded
that the utility of additional driving exceeds the additional costs for
doing so--including the crash risk that they perceive additional
driving involves. As discussed in Section VI.D.2.c), the agencies
account for the benefits of rebound driving by offsetting a portion of
the added safety costs.
Some commenters argued that the agencies should be measuring the
change in the fatality rate rather than the change in the number of
fatalities. For example, EDF argued that changes in fatalities was a
measurement of VMT and number of passengers rather than safety, and
that ``NHTSA's job is to decrease the fatality rate per mile, not to
decrease the number of miles people drive.'' \1967\ EDF also commented
that the agencies were required to report the ``fatality rate data for
the overall safety impacts.'' The agencies disagree with EDF. The
agencies are responsible for measuring the impacts of fuel economy and
CO2 standards, including changes to VMT. While other NHTSA
safety rules have minimal impacts upon aggregate VMT, CAFE standards
have a large impact on VMT and VMT-related costs, including fatalities.
---------------------------------------------------------------------------
\1967\ EDF, Appendix A, NHTSA-2018-0067-12108, at 7-9.
---------------------------------------------------------------------------
Although NHTSA often uses changes in fatality rates as a metric to
evaluate the impact of regulations on safety, these rates are just a
tool utilized to derive the relevant safety impact--namely the
estimated change in fatalities. Furthermore, as part of the cost-
benefit analysis required by Executive Order 12866 and specified in OMB
Circular A-4, the agencies must quantify and value safety impacts to
compare them to the costs of the regulation. The fundamental metric for
valuing loss of life is the VSL. To apply this metric, the agencies
must first produce estimates of any change in the number of fatalities
that results from the regulatory action. Fatalities prevented, as well
as other safety impacts such as non-fatal injuries prevented and
property damage crashes avoided, are appropriate measures of rules that
affect motor vehicle safety.
(a) Impact of Weight Reduction on Safety
Vehicle mass reduction can be one of the more cost-effective means
of increasing fuel economy and reducing CO2 emissions to
meet standards--particularly for makes and models not already built
with much high strength steel or aluminum closures or low mass
components. Manufacturers have stated that they will continue to reduce
vehicle mass to meet more stringent standards, and therefore, this
expectation is incorporated into the modeling analysis supporting the
standards. Safety trade-offs associated with mass-reduction have
occurred in the past, particularly before CAFE standards were
attribute-based; past safety trade-offs may have occurred because
manufacturers chose at the time, in response to CAFE standards, to
build smaller and lighter vehicles. In cases where fuel economy
improvements were achieved through reductions in vehicle size and mass,
the smaller, lighter vehicles did not fare as well in crashes as
larger, heavier vehicles, on average. Although the agencies now use
attribute-based standards, in part to reduce or eliminate the incentive
to downsize vehicles to comply with CAFE and CO2
standards,\1968\ the agencies must be mindful of the possibility of
related safety trade-offs.
---------------------------------------------------------------------------
\1968\ CAFE and CO2 standards are ``footprint-
based,'' with footprint being defined as a measure of a vehicle's
size, roughly equal to the wheelbase times the average of the front
and rear track widths. Footprint-based standards create a
disincentive for manufacturers to produce smaller-footprint
vehicles. This is because, as footprint decreases, the corresponding
fuel economy/CO2 emission target becomes more stringent.
We also believe that the shape of the footprint curves themselves is
such that the curves should neither encourage manufacturers to
increase nor decrease the footprint of their fleets.
---------------------------------------------------------------------------
Historically, as shown in FARS data analyzed by the agencies, mass
reduction concentrated among the heaviest vehicles (chiefly, the
largest LTVs, CUVs and minivans) is estimated to reduce overall
fatalities, while mass reduction concentrated among the lightest
vehicles (chiefly, smaller passenger cars) is estimated to increase
[[Page 24743]]
overall fatalities. Mass reduction in heavier vehicles is more
beneficial to the occupants of lighter vehicles than it is harmful to
the occupants of the heavier vehicles. Mass reduction in lighter
vehicles is more harmful to the occupants of lighter vehicles than it
is beneficial to the occupants of the heavier vehicles. In response to
questions of whether designs and materials of more recent model year
vehicles may have weakened the historical statistical relationships
between mass, size, and safety, the agencies updated our public
database for statistical analysis consisting of crash data. The
analysis considered the full range of real-world crash types.
The methodology used for the statistical analysis of historical
crash data has evolved over many years. The methodology used for the
NPRM and unchanged for the final rule reflects learnings and
refinements from: NHTSA studies in 2003, 2010, 2011, 2012, and 2016;
independent peer review of 23 studies by the University of Michigan
Transportation Research Institute;\1969\ two public workshops hosted by
NHTSA;\1970\ interagency collaboration among NHTSA, DOE and EPA; and
comments to CAFE and CO2 rulemakings in 2010, 2012, the 2016
Draft TAR, and the 2018 NPRM. As explained in greater detail below, the
methodology used for the statistical analysis of historical crash data
for the NPRM and final rule is the best and most up to date available.
---------------------------------------------------------------------------
\1969\ Green, Paul E., Kostyniuk, Lidia P., Gordon, Timothy J.,
and Reed, Matthew P., Independent Review of Statistical Analyses of
Relationship between Vehicle Curb Weight, Track Width, Wheelbase and
Fatality Rates, UMTRI-2011-12, University of Michigan of
Transportation Research Institute (2011). Available at http://www.umtri.umich.edu/our-results/publications/independent-review-statistical-analyses-relationship-between-vehicle-curb.
\1970\ The workshops were held on February 25, 2011 and May 13-
14, 2013. Video, transcripts, and presentations are available on the
NHTSA website (recommended search terms include ``workshop,''
``mass,'' ``safety,'' and the dates of the workshops).
---------------------------------------------------------------------------
Additionally, to assess whether future vehicle designs may impact
the relationship of vehicle mass reduction on safety, NHTSA sponsored a
fleet crash simulation study using future mass reduction vehicle design
concepts (see Fleet Simulation Study below). The results of the
simulation research showed that future mass reduction techniques
continue to exhibit impacts on safety and were consistent with the
statistical analysis of FARS crash data. The agencies considered the
findings of the study and concluded it was reasonable and appropriate
to continue to consider the impact of mass reduction on safety for
future vehicles because the data indicate the relationship between mass
and safety will continue in the future.
For the rulemaking analysis, the CAFE model tracks the amount of
mass reduction applied to each vehicle model, and then applies
estimated changes in societal fatality risk per 100 pounds of mass
reduction determined through the statistical analysis of FARS crash
data. This process allows the CAFE model to tally changes in fatalities
attributed to mass reduction across all of the analyzed future model
years. In turn, the CAFE model is able to provide an overall impact of
the final standards and alternatives on fatalities attributed to mass
reduction.
A number of comments were received on technical aspects of the
mass-safety analysis in the NPRM. The agencies carefully considered all
comments. Where warranted, the agencies conducted additional analyses
to determine whether commenters' suggestions would improve the
analysis. The agencies found that the methodology employed by the
proposal, which was developed over many years, subject to extensive
review and feedback, remains the most rigorous methodology. The
agencies found the alternative approaches raised in comments would
provide less likely estimates, were statistically problematic, or, in
some cases, advocated discarding or ignoring the most likely estimates
altogether. The agencies' assessments of comments are discussed in
detail in the subsections below.
Overall, consistent with prior analyses, the data show that mass
reduction concentrated in heavier vehicles is generally beneficial to
overall safety, and mass reduction concentrated in lighter vehicles is
harmful.
(1) Crash Data
The agencies use real-world crash data as the basis for projecting
the future safety implications for regulatory changes. To support the
2012 rulemaking, NHTSA created a common, updated database for
statistical analysis consisting of crash data. The initial iteration
contained crash data for model years 2000-2007 vehicles in calendar
years 2002-2008. NHTSA made the preliminary version of the new
database, which was the basis for NHTSA's 2011 preliminary report
(hereinafter 2011 Kahane report),\1971\ available to the public in May
2011, and an updated version in April 2012 (used in NHTSA's 2012 final
report, hereinafter 2012 Kahane report), \1972\ enabling other
researchers to analyze the same data and, hopefully, minimize
discrepancies in results caused by reporting inconsistencies across
databases.\1973\ NHTSA updated the crash and exposure databases for the
2016 Draft TAR analysis.
---------------------------------------------------------------------------
\1971\ Kahane, C, J. Relationships Between Fatality Risk, Mass,
and Footprint in Model Year 2000-2007 Passenger Cars and LTVs--Final
Report, National Highway Traffic Safety Administration (Aug. 2012).
Available at https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/811665.
\1972\ Kahane, C, J. Relationships Between Fatality Risk, Mass,
and Footprint in Model Year 2000-2007 Passenger Cars and LTVs--
Preliminary Report. Docket No. NHTSA-2010-0152-0023. Washington, DC:
National Highway Traffic Safety Administration.
\1973\ See 75 FR 25324, 25395-96 (May 7, 2010).
---------------------------------------------------------------------------
For the proposed rule and unchanged for today's final rule, the
crash and exposure databases were updated again. The databases are the
most up-to-date possible (MY 2004-2011 vehicles in CY 2006-2012), given
the processing time for crash data and the need for enough crash cases
to permit statistically meaningful analyses. As in previous analyses,
NHTSA has made the new databases available to the public on its
website.\1974\
---------------------------------------------------------------------------
\1974\ ftp://ftp.nhtsa.dot.gov/CAFE/2018_mass_size_safety/.
---------------------------------------------------------------------------
(2) Methodology
The relationship between a vehicle's mass, size, and fatality risk
is complex, and it varies in different types of crashes. The agencies
have been examining this relationship for more than two decades. The
basic analytical method used to analyze the impacts of weight reduction
on safety for the proposal, and unchanged for this final rulemaking, is
the same as in 2016 Puckett and Kindelberger report.\1975\ NHTSA
released the 2016 Puckett and Kindelberger report as a preliminary
report on the relationship between fatality risk, mass, and footprint
in June 2016 in advance of the Draft TAR. The 2016 Puckett and
Kindelberger report covered the same scope as previous NHTSA
reports,\1976\ offering a detailed description of the crash and
exposure databases, modeling approach, and analytical results on
relationships among vehicle size, mass, and fatalities that informed
the Draft TAR. The
[[Page 24744]]
modeling approach described in the 2016 Puckett and Kindelberger report
was developed with the collaborative input of NHTSA, EPA and DOE, and
subject to extensive public review, scrutiny in two NHTSA-sponsored
workshops, and a thorough peer review that compared it with the
methodologies used in other studies.\1977\
---------------------------------------------------------------------------
\1975\ Puckett, S.M. and Kindelberger, J.C. (2016, June).
Relationships between Fatality Risk, Mass, and Footprint in Model
Year 2003-2010 Passenger Cars and LTVs--Preliminary Report. (Docket
No. NHTSA-2016-0068). Washington, DC: National Highway Traffic
Safety Administration, available at https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/2016-prelim-relationship-fatalityrisk-mass-footprint-2003-10.pdf.
\1976\ The 2016 Puckett and Kindelberger report is an extension
of 2011 Kahane report and 2012 Kahane report.
\1977\ Previous reports from which the 2016 Puckett and
Kindelberger report was derived from, were also subject to extensive
peer reviews. Farmer, Green, and Lie, who reviewed the 2010 Kahane
report, also peer-reviewed the 2011 Kahane report. In preparing his
2012 report (along with the 2016 Puckett and Kindelberger report and
Draft TAR), Kahane also took into account Wenzel's assessment of the
preliminary report and its peer reviews, DRI's analyses published
early in 2012, and public comments such as the International Council
on Clean Transportation's comments submitted on NHTSA and EPA's 2010
notice of joint rulemaking. These comments prompted supplementary
analyses, especially sensitivity tests, discussed at the end of this
section.
---------------------------------------------------------------------------
In computing the impact of changes in mass on safety, the agencies
are faced with competing challenges. Research has consistently shown
that mass reduction affects ``lighter'' and ``heavier'' vehicles
differently across crash types. The 2016 Puckett and Kindelberger
report found mass reduction concentrated amongst the heaviest vehicles
is likely to have a beneficial effect on overall societal fatalities,
while mass reduction concentrated among the lightest vehicles is likely
to have a detrimental effect on fatalities.\1978\ To accurately capture
the differing effect on lighter and heavier vehicles, the agencies must
split vehicles into lighter and heavier vehicle classifications in the
analysis.\1979\ However, this poses a challenge of creating
statistically-meaningful results. There is limited relevant crash data
to use for the analysis. Each partition of the data reduces the number
of observations per vehicle classification and crash type, and thus
reduces the statistical robustness of the results. The methodology
employed by the agencies was designed to balance these competing forces
as an optimal trade-off to accurately capture the impact of mass-
reduction across vehicle curb weights and crash types while preserving
the potential to identify robust estimates.
---------------------------------------------------------------------------
\1978\ The findings of the 2016 Puckett and Kindelberger report
are consistent with the results of the 2012 Kahane report and Draft
TAR.
\1979\ If lighter and heavier vehicles are left undistinguished,
the agencies analysis would be restricted to identifying a single
effect of mass reduction for passenger cars and a single effect of
mass reduction for truck-based LTVs. As discussed below, distinct
effects have been estimated historically for lighter versus heavier
vehicles for cars and LTVs, confirming the validity of
distinguishing by curb weight where feasible.
---------------------------------------------------------------------------
For the proposal and the final rule, the agencies employed the
modeling technique developed in the 2016 Puckett and Kindelberger
report to analyze the updated crash and exposure data by examining the
cross sections of the societal fatality rate per billion vehicle miles
of travel (VMT) by mass and footprint, while controlling for driver
age, gender, and other factors, in separate logistic regressions for
five vehicle groups and nine crash types. ``Societal'' fatality rates
include fatalities to occupants of all the vehicles involved in the
collisions, plus any pedestrians, cyclists, or occupants of other
conveyances (e.g., motorcyclists). The agencies utilize the
relationships between weight and safety from this analysis, expressed
as percentage increases in fatalities per 100-pound weight reduction,
to examine the weight impacts applied in this CAFE analysis. The
effects of mass reduction on safety were estimated relative to
(incremental to) the regulatory baseline (augural standards) in the
CAFE analysis, across all vehicles for MYs 2018 and beyond.
As in the 2012 Kahane report, 2016 Puckett and Kindelberger report,
and the Draft TAR, the vehicles are grouped into three classes:
Passenger cars (including both two-door and four-door cars); CUVs and
minivans; and truck-based LTVs. The curb weight of passenger cars is
formulated, as in the 2012 Kahane report, 2016 Puckett and Kindelberger
report, and Draft TAR, as a two-piece linear variable to estimate one
effect of mass reduction in the lighter cars and another effect in the
heavier cars. The boundary between ``lighter'' and ``heavier'' cars is
3,201 pounds (which is the median mass of MY 2004-2011 cars in fatal
crashes in CY 2006-2012, up from 3,106 pounds for MY 2000-2007 cars in
CY 2002-2008 in the 2012 NHTSA safety database, and up from 3,197
pounds for MY 2003-2010 cars in CY 2005-2011 in the 2016 NHTSA safety
database). Likewise, for truck-based LTVs, curb weight is a two-piece
linear variable with the boundary at 5,014 pounds (again, the MY 2004-
2011 median, higher than the median of 4,594 pounds for MY 2000-2007
LTVs in CY 2002-2008 and the median of 4,947 pounds for MY 2003-2010
LTVs in CY 2005-2011). CUVs and minivans are grouped together in a
single group covering all curb weights of those vehicles; as a result,
curb weight is formulated as a simple linear variable for CUVs and
minivans. Historically, CUVs and minivans have accounted for a
relatively small share of new-vehicle sales over the range of the data,
resulting in less crash data available than for cars or truck-based
LTVs. In sum, vehicles are distributed into five groups by class and
curb weights: Passenger cars < 3,201 pounds; passenger cars 3,201
pounds or greater; truck-based LTVs < 5,014 pounds; truck-based LTVs
5,014 pounds or greater; and all CUVs and minivans.
There are nine types of crashes specified in the analysis for each
vehicle group: three types of single-vehicle crashes, five types of
two-vehicle crashes; and one classification of all other crashes.
Single-vehicle crashes include first-event rollovers, collisions with
fixed objects, and collisions with pedestrians, bicycles and
motorcycles. Two-vehicle crashes include collisions with: heavy-duty
vehicles; cars, CUVs, or minivans < 3,187 pounds (the median curb
weight of other, non-case, cars, CUVs and minivans in fatal crashes in
the database); cars, CUVs, or minivans >= 3,187 pounds; truck-based
LTVs < 4,360 pounds (the median curb weight of other truck-based LTVs
in fatal crashes in the database); and truck-based LTVs >= 4,360
pounds. Grouping partner-vehicle CUVs and minivans with cars rather
than LTVs is more appropriate because their front-end profile and
rigidity more closely resemble a car than a typical truck-based LTV. An
additional crash type includes all other fatal crash types (e.g.,
collisions involving more than two vehicles, animals, or trains).
Splitting the vehicles from this crash type involved in crashes
involving two light-duty vehicles into a lighter and a heavier group
permits more accurate analyses of the mass effect in collisions of two
vehicles.
For a given vehicle class and weight range (if applicable),
regression coefficients for mass (while holding footprint constant) in
the nine types of crashes are averaged, weighted by the number of
baseline fatalities that would have occurred for the subgroup MY 2008-
2011 vehicles in CY 2008-2012 if these vehicles had all been equipped
with electronic stability control (ESC). The adjustment for ESC, a
feature of the analysis added in 2012, takes into account results will
be used to analyze effects of mass reduction in future vehicles, which
will all be ESC-equipped, as required by NHTSA's safety regulations.
The agencies received multiple comments on how they distribute
vehicles into classifications. IPI, quoting a study by Tom Wenzel,
commented that sorting vehicles into footprint deciles shows positive
impacts from mass reduction for the majority of the
[[Page 24745]]
footprint deciles.\1980\ CARB commented that the agencies should have
used the curb weight of all vehicles to calculate the thresholds for
``lighter'' and ``heavier'' vehicle types rather than just the curb
weights of vehicles involved in fatal crashes.\1981\ CARB also
commented that pickup trucks and SUVs that are not subject to CAFE
regulation (i.e., Class 2b and Class 3 vehicles, such as \3/4\-ton and
one-ton pick-up trucks, vans and related SUVs) should not be included
in the assessment of the impact of mass on safety and doing so raises
the median weight of trucks.\1982\ CARB also commented that the median
weights are static values representing the historical fleet, but the
median weights and proportions of crash types involving given vehicle
weight categories should change with median weight of the fleet modeled
by the CAFE model.\1983\ Commenters generally believed that the
agencies' approach ``results in inappropriate apportioning of cars and
trucks into the corresponding lighter or heavier bins,'' which in turn
causes the agencies to overestimate the fatalities associated with mass
reduction.\1984\
---------------------------------------------------------------------------
\1980\ IPI, Detailed Comments, Docket No. NHTSA-2018-0067-12213,
at 127 (quoting Tom Wenzel, Assessment of NHTSA's Report
``Relationships Between Fatality Risk, Mass, and Footprint in Model
Year 2004-2011 Passenger Cars and LTVs,'' (LBNL Phase 1, 2018).
Available at https://escholarship.org/uc/item/4726g6jq.
\1981\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 276.
\1982\ Tom Wenzel of Lawrence Berkeley National Laboratories,
Comment, EPA-HQ-OAR-2018-0283-4118, at 1; see also CARB, Detailed
Comments, Docket No. NHTSA-2018-0067-11873, at 259.
\1983\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 260.
\1984\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 276.
---------------------------------------------------------------------------
Dividing vehicles into footprint deciles and excluding Class 2b and
3 vehicles pose sample size and data coverage issues. If vehicles were
grouped into footprint deciles, the sample sizes in each decile would
be approximately one-fifth as large as the corresponding sample sizes
in each of the agencies' four passenger car and LTV vehicle classes
(and one-tenth as large as the sample size for CUVs and minivans).
Smaller parameter estimates require correspondingly smaller standard
errors (i.e., relatively precise estimates) to achieve statistical
significance, but splitting the limited data into deciles yields larger
standard errors, restricting the ability to identify statistically-
significant estimates. Likewise, by extending the footprint-curb
weight-fatality data to include Class 2b and 3 trucks that are
functionally and structurally similar to corresponding \1/2\-ton models
that are subject to CAFE regulation,\1985\ the sample size and ranges
of curb weights and footprint are improved. Sample size is a challenge
for estimating relationships between curb weight and fatality risk for
individual crash types in the main analysis; dividing the sample
further or removing observations makes it exceedingly difficulty to
identify meaningful estimates and the relationships that are present in
the data.
---------------------------------------------------------------------------
\1985\ Class 2b and 3 pickup trucks, vans and SUVs have physical
characteristics and usage profiles that are substantially similar to
their Class 2a counterparts. For example, the Class 2a version of
the Ford F-150 has similar physical characteristics to and has a
similar usage profile to the Class 2b Ford F150. Same for the Class
2a Ford F150 relative to the Class 2b and 3 Ford F250, and for the
GMC Yukon relative to the Yukon XL. The Class 2b and 3 pickup trucks
in the sample generally have gross vehicle weight ratings of 10,000
pounds or less, and thus are subject to the same Federal motor
vehicle safety standards as their light-duty counterparts. Likewise,
these vehicles generally have similar physical dimensions (e.g.,
ground clearance, width) as related light-duty vehicles. Key
differentiating factors among these vehicles are height, payload,
and towing capacity. There are likely to be unobserved differences
in how these vehicles are driven relative to light-duty
alternatives; however, the crash data include a census of fatal
crashes involving case vehicles and the Class 2b and 3 vehicles
included in the analysis, in turn representing the relative risk of
differences in curb weight in crashes involving Class 2b and 3
vehicles. Despite being regulated by different fuel economy and
emissions regulations as they become heavier (i.e., once a given
model crosses a mass threshold changes classes), the vehicles may
continue to be used in similar ways over time; in turn, the safety
implications of the presence of these vehicles may continue to be
similar. In contrast, other types of heavy-duty vehicles, such as
box trucks, buses, refuse trucks, fire trucks, and other heavy-duty
commercial vehicles are substantially different from light duty
vehicles in their physical characteristics and usage profiles, and
it would not be appropriate to include them in the statistical
analysis to determine the impact of mass on crash fatalities.
---------------------------------------------------------------------------
Compounding the issue is the fact the analysis focuses on societal
fatality risk (i.e., all fatalities, including crash partners and
people outside of vehicles, such as pedestrians, cyclists, and
motorcyclists) rather than merely in-vehicle fatality risk, which
yields estimates that are smaller in magnitude (and thus more difficult
to identify meaningful differences from zero) than estimates
representing changes in in-vehicle fatality risk. That is, compared to
an analysis of in-vehicle fatality risk (which would tend to yield
relatively large estimated effects of mass reduction), the focus on
societal fatalities tends to yield relatively small (net) effects of
mass reduction on fatality risk.
Including Class 2b and 3 vehicles in the analysis to determine the
relationship of vehicle mass on safety has the added benefit of
improving correlation constraints. Notably, curb weight increases
faster than footprint for large light trucks and Class 2b and 3 pickup
trucks and SUVs, in part because the widths of vehicles are constrained
more tightly (i.e., due to lane widths) than their curb weights.
Including data from Class 2b and 3 pick-up truck and SUV fatal crashes
provides data over a wider range of vehicle weights, which improves the
ability to estimate the mass-crash fatality relationship. The agencies
believe the decision of whether to include Class 2b and 3 vehicles in
the analysis should be made based on whether the additional data
improves the estimate of the safety impact of mass reduction in light
trucks, and that the fatality data should not be simplistically
excluded because the vehicles are not regulated under the CAFE and
CO2 emissions programs. Ultimately, the agencies find that:
(1) The fundamental objective is to capture the strongest, meaningful
signal regarding societal fatality risk as a function of the mass of
light trucks; (2) that incorporating information on fatal incidents
involving Class 2b and 3 trucks improves the quality of the signal the
agencies can capture, and (3) including the vehicles provides the best
estimate of the impacts of mass on societal fatalities.
In assessing whether to calculate the median curb weight threshold
from all vehicles involved in accidents or on the road, the agencies
weighed changing the process used to establish the thresholds and the
potential impact on the robustness of the statistical analysis. From a
statistical perspective, using thresholds that allocate a similar
number of fatal crash cases to both the lower vehicle weight group and
the higher vehicle weight group for a given vehicle type will minimize
the average standard errors of estimates for both groups, which
provides the best estimates for each group. Because reducing average
standard errors strengthens the statistical analysis, the agencies
conclude using only the curb weight of vehicles involved in fatal
crashes to calculate the median curb weight threshold produces the best
estimate. This conclusion is the same that was reached previously when
considering the same issue for the 2011 Kahane, 2012 Kahane, and 2016
Puckett and Kindelberger analyses.
On a related note, the regression models are estimated based on
with respect to the total number of fatalities associated within each
vehicle weight group classification (referred to as vehicle group
below, for brevity). Shifting the threshold would change the estimated
incremental impact of changes in curb weight in each vehicle
[[Page 24746]]
group, but the net effects would offset each other across vehicle
groups, resulting in the same overall estimated effect of changes in
vehicle mass on societal fatality risk. For example, if one restricted
the ``lightest'' group for a vehicle type to include only the bottom
ten percentiles of vehicle weight, one would expect to identify a very
strong detrimental effect (or weakest beneficial effect) of mass
reduction for that group. However, the estimated effect of mass
reduction in that group has minimal implications for the fleet (i.e.,
because there are fewer vehicles in the group), and the corresponding
estimated effect of mass reduction for other groups would also mute the
impact (i.e., because there are many vehicles in the group that vary in
mass to a much larger degree than in the ``lighter'' group).
Ultimately, the mean effect of mass reduction across the lighter and
heavier groups would be the same as when using the median as the
threshold (or at least, similar, subject to limitations in statistical
optimization), but with a different point of reference when comparing
the groups. Thus, the agencies believe the selection of curb weight
threshold has a minimal impact on the estimated effects of mass
reduction across all vehicle types.
Full consideration of CARB's comment on mass thresholds, and
whether they should change as the median weight of the fleet modeled by
the CAFE model changes, requires a deeper look at each of the crash
types considered in the analysis. That is, the point estimates
presented in Table VI-202 represent weighted averages across nine
separate, mutually-exclusive and exhaustive crash models (analyzed
separately for cars, LTVs, and CUVs and minivans). For example, an
individual model for first-event rollovers yields estimates of the
percentage change in societal fatality risk per 100-pound mass
reduction for lighter and heavier (or, in the case of CUVs and
minivans, all) vehicles in the target vehicle class. The final, overall
point estimate for a given vehicle type is found by: (1) Multiplying
the estimate associated with an individual crash type by the estimated
share of societal fatalities involving the vehicle class (adjusting for
two-vehicle collisions that span vehicle classes to avoid double-
counting); and (2) summing the values estimated in (1) across all crash
types. In its comments, CARB noted that if the distribution of vehicles
in terms of curb weight changes through lightweighting, the shares of
(fatal) two-vehicle crashes involving a given pair of vehicles as
defined by weight class (e.g., car below a given threshold colliding
with a LTV above a given threshold) would change. In turn, the
appropriate weighting across the crash types modeled in the analysis
would likewise be different (involving an increasing share of vehicles
below a given curb weight threshold). Due to these potential
limitations, CARB questioned the stability of the summary point
estimates relative to changes in the shares of fatalities within each
crash type in the analysis.\1986\
---------------------------------------------------------------------------
\1986\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 278-79.
---------------------------------------------------------------------------
To evaluate CARB's concerns regarding future crash mixes and
definitions of vehicle weight classes, the agencies performed an
exploratory analysis examining the scope and impacts of potential model
changes. In doing so, the agencies examined the degree of change in the
median vehicle fleet weight in the NPRM analysis relative to the fixed
mass threshold values, and also how sensitive the curb weight safety
point estimates are to assumptions about the distribution of curb
weights in future vehicle fleets. The agencies also considered the
feasibility of changing the shares of fatalities by crash type as a
function of forthcoming or developing vehicle safety technologies. This
information would help inform adjustments to fatality rate impacts for
each vehicle type, because the likelihood of observing individual fatal
crash types could change in different ways across vehicle types in the
analysis as the vehicle mix changes. However, the agencies identified
no studies on the effectiveness of forthcoming or developing vehicle
safety technologies that could inform projections of shares of
fatalities across crash types, nor did the commenters reference any
such studies. Likewise, commenters provided no data that would enable
projections of these factors. Thus, for a given vehicle mix, the
agencies have no information available to justify changing the shares
of fatalities across crash types over time. Therefore, the agencies
decided to keep the distribution of fatality shares constant for:
First-event rollovers; fixed-object collisions; collisions with
pedestrians, bicyclists, and motorcycles; collisions with heavy
vehicles; collisions with one other light-duty vehicle (i.e., a
constant share across the sum of these crashes, but not constant for
any given type of crash partner); and all other crashes.
The agencies had sufficient information to evaluate the effects of
changes in the fatal crash mix for cases involving two light-duty
vehicles. The agencies agreed that it was internally consistent to
adjust fatality shares by crash type proportionally to the distribution
of vehicle types and curb weight classes for a given focal MY. An
important technical question associated with this approach is the level
of disaggregation. The agencies considered an alternative in which the
agencies would estimate and apply unique curb weight point estimates
for each calendar year in the analysis for each regulatory alternative.
This alternative would account for changes in the distribution of crash
types associated with changes in both vehicle type shares (i.e., shifts
from passenger cars to CUVs and LTVs) and vehicle mass shares (i.e.,
shifts from vehicles above the curb weight thresholds to vehicles below
the thresholds). As in the status quo analysis of curb weight and
fatality risk, the resulting point estimates would be weighted averages
across the individual crash type models as presented in the NPRM, but
re-weighted to reflect projected changes to the fleet.
The agencies investigated this alternative and identified several
concerns. A key functional constraint is that the curb weight safety
point estimates are applied in the CAFE model as a lump-sum, lifetime
effect to a given vehicle. This characteristic of the model limits the
ability to apply calendar-year-specific effects of changes in curb
weight and vehicle type distributions when evaluating safety impacts of
changes in curb weights. The safety point estimates also represent net
effects of changes in curb weights over the lifetime of a given vehicle
in the CAFE model; any changes in the calculation of safety point
estimates would need to preserve this characteristic. More broadly, the
vehicle fleet is not static over a vehicle's lifetime (i.e., the
distributions of curb weight and vehicle type change each year), so the
effective probabilities of each crash type over a given vehicle's
lifetime are a function of many calendar-year-level curb weight and
vehicle type distributions. To capture any effects of changes in
vehicle mass distributions over time within the current CAFE model
structure, the agencies would need to enact a method that: (1)
Identifies defensible changes in fatality risk associated with vehicle
mass as the distribution of vehicle mass changes (e.g., accounting for
changes in the likelihood of observing particular fatal crash types
that reflect projected changes in the distribution of vehicle types and
curb weights across vehicles); and (2) allocates calendar-year-specific
impacts of curb weight on fatality risk to each vehicle in the fleet
across the
[[Page 24747]]
analysis horizon. Identifying how best to achieve this would be
complex, and would require the development of an alternative analytical
approach that would be outside the scope of this rulemaking.
With these concerns in mind, the agencies explored an alternative
approach to test the sensitivity of the safety point estimates to
distributions of vehicles by curb weight and vehicle type. The starting
point for the alternative approach is maintaining the understanding
that the nine crash type models that are present in the curb weight
safety analysis represent the best statistical alternatives for
evaluating the crash data in the database (i.e., optimal statistical
precision conditional on the coverage of the data). Furthermore, the
nine crash type models are defined in terms of physical relationships
(i.e., crashes involving vehicles of particular curb weight ranges and
vehicle types) that are invariant to changes in the distributions of
vehicles for those same characteristics. That is, the estimated changes
in societal fatality risk as curb weights change for a focal vehicle
(i.e., of a particular type and weight range) that is involved in a
particular type of crash apply equally to any scenario involving such
vehicle, regardless of changes in the probability of observing such a
scenario. For example, the agencies would expect the societal fatality
risk for a crash involving a passenger car lighter than 3,201 pounds
colliding with a LTV heavier than 4,360 pounds to be the same
regardless of how many such collisions take place. Thus, the agencies
would expect the net effect of a given change in curb weight for a
given vehicle type in a given crash type to scale proportionally with
the probability of such crashes occurring. Put simply, if there are
half as many potential crash partners of a given type in a future year
compared to a base year, the agencies would expect a given curb weight
reduction to have half as large of a net effect on fatalities in the
future year relative to the base year. In the extreme, curb weight
changes would have no net effect on fatalities at all for a given crash
type if such crashes had a zero percent probability of occurring (i.e.,
if there are no potential crash partner vehicles).
Based on this maintained hypothesis, the agencies examined test
curb weight safety point estimates under alternative scenarios, in
which fatality shares by crash type were proportional to the
distribution of vehicle types and curb weight classes across a range of
outcomes reflecting different model years and policy alternatives
represented in the NPRM. The sensitivities of the safety point
estimates to changes in the distributions of vehicle curb weights and
vehicle types were tested by adjusting fatality shares across the
relevant crash types in the analysis (i.e., involving two light-duty
vehicles) in a manner consistent with potential changes in the vehicle
fleet, while holding the outputs of the individual crash type models
the same as in the NPRM.
For example, compare the safety point estimate for LTVs lighter
than 5,014 pounds in the NPRM with an alternative point estimate for an
extreme hypothetical future year where 80 percent of the LTV fleet is
lighter than the median curb weight for crash partners (4,360 pounds):
[GRAPHIC] [TIFF OMITTED] TR30AP20.414
[[Page 24748]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.415
The estimated net societal effect of a 100-pound mass reduction is
equal to: (1) The sum of the estimated net effects across all crash
types, divided by (2) the baseline estimate of annual fatalities
involving the vehicle class (adjusted to avoid double-counting) for the
most recent four MYs in the database (MYs 2008-2011), or 1,782
fatalities per year. In the NPRM, the estimated net societal effect of
a 100-pound mass reduction for lighter LTVs was a 5.5 fatality
increase, or a 0.31 percent increase relative to a baseline of 1,782
fatalities. Changing the share of crash fatalities involving heavier
LTVs to be consistent with a fleet with only 20 percent of LTVs above
the curb weight threshold yields: (1) An increase in incremental
fatalities in crashes involving lighter LTVs (from 0.5 fatality to 0.7
fatality); and (2) a decrease in incremental fatalities in crashes
involving heavier LTVs (from 1.5 fatalities to 0.7 fatality); for a
total net increment of 4.9 fatalities compared to the NPRM's estimate
of 5.5 fatalities. Thus, the agencies estimate that, in a future year
where the fleet differs from the baseline by having an extreme case of
80 percent of LTVs below the crash-partner curb weight threshold, the
net societal effect of a 100-pound mass reduction in LTVs lighter than
5,014 pounds would be 4.9 divided by 1,782, or 0.28 percent, versus
0.31 percent in the baseline.
This simple example confirms that the estimates do indeed change as
the distribution of curb weights changes. In this case, the change is
intuitive: As the LTV fleet becomes lighter, mass reduction among LTVs
below 5,014 pounds becomes less detrimental to society. However, the
incremental effect is estimated to be quite small: Shifting from an
even mix of LTVs above and below the threshold to an extreme 20 percent
to 80 percent split only changes the estimated net societal effect by
0.03 percent in absolute terms. Thus, the model results for lighter
LTVs appear relatively insensitive to the LTV curb weight distribution.
Indeed, in the limit, where all LTVs are below the crash-partner curb
weight threshold (and thus there are no fatality impacts for crashes
involving heavier LTVs), the estimated net societal effect of a 100-
pound mass reduction for LTVs below 5,014 pounds (i.e., all LTVs in
this case) is 0.25 percent, a difference of 0.06 percent in absolute
terms compared to the baseline. This result is driven by the dominating
effects of crash types involving either: (1) No crash partner (e.g.,
first-event rollovers); (2) one crash partner from a group not
associated with a given change in a curb weight distribution (e.g.,
heavy vehicles, bicyclists, passenger cars); or (3) multiple crash
partners (an element of ``all other crashes''). That is, even extreme
changes in the distribution of curb weights for a given vehicle type
will not change the role that vehicle mass plays in crashes for a focal
vehicle when that vehicle does not collide with another vehicle from
the distribution in question. In the above example involving lighter
LTVs, 90 percent of fatalities involve incidents that do not include a
single LTV crash partner, and 66 percent of fatalities involve
incidents that do not include a single light-duty crash partner.
Continuing with this example scenario, the point estimate for LTVs
heavier than 5,014 pounds becomes larger in magnitude (i.e., more
societally beneficial mass reduction) to a similar degree as the
reduction in magnitude for lighter LTVs when moving to an extreme 20
percent to 80 percent split of crash partner LTVs above (versus below
in the case above) the curb weight threshold:
[[Page 24749]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.416
In the NPRM and this analysis, the estimated net societal effect of
a 100-pound mass reduction for lighter LTVs was a 20.0 fatality
decrease, or a 0.61 percent decrease relative to a baseline of 3,304
fatalities. Changing the share of crash fatalities involving heavier
LTVs to be consistent with a fleet with only 20 percent of LTVs above
the curb weight threshold yields: (1) A larger reduction in fatalities
in crashes involving lighter LTVs per 100-pound mass reduction (from
4.0 fatalities to 6.1 fatalities); and (2) a decrease in incremental
fatalities in crashes involving heavier LTVs (from 1.6 fatalities to
0.7 fatality); for a total net change of -22.9 fatalities compared to a
baseline of -20.0 fatalities. Thus, the agencies estimate that, in a
future year where the fleet differs from the baseline by having 80
percent of LTVs below the crash-partner curb weight threshold, the net
societal effect of a 100-pound mass reduction in LTVs 5,014 pounds or
heavier would be -22.9 divided by 3,304, or -0.69 percent, versus -0.61
percent in the baseline. Consistent with the test results for lighter
LTVs, the model results for heavier LTVs appear relatively insensitive
to the LTV curb weight distribution. In the limit, where all LTVs
(except for one remaining heavier LTV in consideration) are below the
crash-partner curb weight threshold (and thus there are no effective
fatality impacts for crashes involving heavier LTVs), the estimated net
societal effect of a 100-pound mass reduction for the remaining LTV
above 5,014 pounds is -0.76 percent, a difference of 0.15 percent in
absolute terms compared to the baseline.
Expanding the analysis to account for changes in the relative sales
shares of each vehicle type dampens the net effects further. As the
fleet share of passenger cars decreases, the net effects of mass
reduction among LTVs become less societally beneficial. That is, as
there are fewer relatively vulnerable passenger cars in the fleet,
there become fewer opportunities to reduce fatalities in collisions
between LTVs and passenger cars through mass reduction. In some
scenarios considered in the exploratory analysis, the effects of sales
shifts from passenger cars to LTVs at least fully offset the estimated
improvements in net fatalities associated with mass reduction
summarized above as the LTV fleet becomes lighter.
Ultimately, the exploratory analysis using extreme example cases
confirmed that the baseline safety point estimates are very reasonable
for the feasible ranges of mixes of vehicle types and curb weights
across the model years in the CAFE model analysis. The sensitivities of
the point estimates are relatively low across relative shares of
lighter versus heavier LTVs (especially relative to the uncertainty in
the baseline estimates), and similarly low and offsetting across
decreasing fleet shares for passenger cars. Because shifts in mass in
the rulemaking analysis would have insignificant impacts on the safety
estimated values and therefore rulemaking decision making, the agencies
conclude no changes are warranted for this final rule analysis.
Mass Safety Results
Table VI-204 presents the estimated percent increase in U.S.
societal fatality risk per 10 billion VMT for each 100-pound reduction
in vehicle mass, while holding footprint constant, for each of the five
vehicle classes:
[[Page 24750]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.417
Techniques developed in the 2011 (preliminary) and 2012 (final)
Kahane reports have been retained to test statistical significance and
to estimate 95-percent confidence bounds (sampling error) for mass
effects and to estimate the combined annual effect of removing 100
pounds of mass from every vehicle (or of removing different amounts of
mass from the various classes of vehicles), while holding footprint
constant.
None of the estimated effects have 95-percent confidence bounds
that exclude zero, and thus are not statistically significant at the
95-percent confidence level. The NPRM reported that two estimated
effects are statistically significant at the 85-percent level. Societal
fatality risk is estimated to: (1) Increase by 1.2 percent if mass is
reduced by 100 pounds in the lighter cars; and (2) decrease by 0.61
percent if mass is reduced by 100 pounds in the heavier truck-based
LTVs. The estimated increases in societal fatality risk for mass
reduction in the heavier cars and the lighter truck-based LTVs, and the
estimated decrease in societal fatality risk for mass reduction in CUVs
and minivans are not significant, even at the 85-percent confidence
level. Although 85-percent statistical significance is not a
traditional metric of meaningful differences to zero, this result
confirms that the estimated effects for vehicles with curb weights most
dissimilar to the median vehicle are the most likely to be
significantly different to zero.
The agencies judge the central value estimates are the best and
most up-to-date estimates available; the estimates offer a stronger
statistical representation of relationships among vehicle curb weight,
footprint and fatality risk than an assumption of no correlation
whatsoever. The agencies appropriately present the statistical
uncertainty. For example, the central values for the highest vehicle
weight group (LTVs 5,014 pounds or heavier) and the lowest vehicle
weight group (passenger cars lighter than 3,201 pounds) (which, based
on fundamental physics, are expected to have the greatest impact of
mass reduction on safety) are economically significant,\1987\ and are
in line with the prior analyses used in past NHTSA CAFE and EPA
CO2 rulemakings. As shown in Table VI-205, the estimated
coefficients have trended to lower numerical values in successive
studies, but remain positive for lighter cars and negative for heavier
LTVs. The 85-percent confidence level was reported only to show the
scope of uncertainty at the first rounded (to five percent) threshold
where the coefficient estimates were significantly different to zero
for the two vehicle groups at the extremes of the curb weight
distribution. No preference was suggested for an 85-percent confidence
bound. Rather, the agencies found value in reporting confidence
intervals for all five coefficients at the threshold where the
estimates for the two extremes of the curb weight distribution were
significantly different to zero. The agencies determined it was better
to include the estimates, despite the slightly lower confidence level,
than knowingly omitting economically significant results.
---------------------------------------------------------------------------
\1987\ The agencies use ``economically significant results'' to
mean values that have an important, practical implication, but may
be derived from estimates that do not meet traditional levels of
statistical significance. For example, if the projected economic
benefit of a project equaled $100 billion, the agencies would
consider the impact economically significant, even if the estimates
used to derive the impact were not statistically significant at the
95-percent confidence level. Conversely, if the projected economic
benefit of a project equaled $1, the agencies would not consider the
impact economically significant, even if the estimates used to
derive the impact were statistically significant at the 99.99-
percent confidence level. In the case above, we considered the
results associated with the lightest and heaviest vehicle types to
be economically significant because the associated safety costs were
large and the estimates had magnitudes meaningfully different from
zero and were statistical significant at the 85-percent confidence
level.
---------------------------------------------------------------------------
The regression results are constructed to project the effect of
changes in mass, independent of all other factors, including footprint.
With each additional change from the current environment (e.g., the
scale of mass change, presence and prevalence of safety features,
demographic characteristics), the results may become less
representative. That is, although safety features and demographic
factors are accounted for separately, the estimated effects of mass are
identified under the specific mix of vehicles and drivers in the data.
The agencies note that the analysis accounts for safety features that
are optional but available across all MYs in the sample (most notably
electronic stability control, which was not yet mandatory for all model
years in the sample), and calibrates historical safety data to account
for future fleets with full ESC penetration to reflect the mandate.
The agencies considered the near multicollinearity of mass and
footprint to be a major issue in the 2010 Kahane report and voiced
concern about inaccurately estimated regression coefficients. High
correlations between mass and footprint and variance inflation factors
(VIF) have persisted from MY 1991-1999 to MY 2004-2011; large footprint
vehicles continued to be, on the average, heavier than small footprint
vehicles to the same extent as in the previous decade.
Nevertheless, multicollinearity appears to have become less of a
problem in the 2012 Kahane, 2016 Puckett and Kindelberger/Draft TAR,
and current analyses. Ultimately, only three of the 27 core models of
fatality risk by vehicle type in the current analysis indicate the
potential presence of effects of multicollinearity, with estimated
effects of mass and footprint
[[Page 24751]]
reduction greater than two percent per 100-pound mass reduction and
one-square-foot footprint reduction, respectively; these three models
include passenger cars and CUVs in first-event rollovers, and CUVs in
collisions with LTVs greater than 4,360 pounds. This result is
consistent with the 2016 Puckett and Kindelberger report, which also
found only three cases out of 27 models with estimated effects of mass
and footprint reduction greater than two percent per 100-pound mass
reduction and one-square-foot footprint reduction.
For comparison, Table VI-205 shows the fatality coefficients from
the 2012 Kahane report (MY 2000-2007 vehicles in CY 2002-2008) and the
2016 Puckett and Kindelberger report and Draft TAR (MY 2003-2010
vehicles in CY 2005-2011).
[GRAPHIC] [TIFF OMITTED] TR30AP20.418
The new results are directionally the same as in 2012; in the 2016
analysis, the estimate for lighter LTVs was of opposite sign (but small
magnitude). Consistent with the 2012 Kahane and 2016 Puckett and
Kindelberger reports, mass reductions in lighter cars are estimated to
lead to increases in fatalities, and mass reductions in heavier LTVs
are estimated to lead to decreases in fatalities.
---------------------------------------------------------------------------
\1988\ Median curb weights in the 2012 Kahane report: 3,106
pounds for cars, 4,594 pounds for truck-based LTVs. Median curb
weights in the 2016 Puckett and Kindelberger report: 3,197 pounds
for cars, 4,947 pounds for truck-based LTVs.
---------------------------------------------------------------------------
The estimated mass effect for heavier truck-based LTVs is stronger
in this analysis and in the 2016 Puckett and Kindelberger report than
in the 2012 Kahane report; both estimates are statistically significant
at the 85-percent confidence level, unlike the corresponding estimate
in the 2012 Kahane report. The estimated mass effect for lighter truck-
based LTVs is insignificant and positive in this analysis and the 2012
Kahane report, while the corresponding estimate in the 2016 Puckett and
Kindelberger report was insignificant and negative.
Multiple commenters, including the South Coast Air Quality
Management District and States and Cities, challenged the practical
value of using estimates with statistical significance at the 85-
percent level, arguing that below 95 (or 90) percent are insufficiently
reliable.\1989\ For example, CARB stated, ``[d]ue to the lack of
statistical significance, NHTSA should not be attributing any increase
in fatalities due to mass reduction'' and argues that the ``effect of
mass reduction on fatality risk should be set to zero since the
estimates are not statistically different to zero.'' \1990\
---------------------------------------------------------------------------
\1989\ See South Coast Air Quality Management District, Detailed
Comments, Docket No. NHTSA-2018-0067-11813, at 6 (internal citation
omitted); States and Cities, Detailed Comments, Docket No. NHTSA-
2018-0067-11735, at 95.
\1990\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 269.
---------------------------------------------------------------------------
The agencies believe the updated analysis that was presented in the
NPRM represents the most up to date and best estimate of the impacts of
mass reduction on crash fatalities; and, that it is appropriate for the
analysis to use the best and most likely estimates for safety, even if
the estimates are not statistically significant at the 95-percent
confidence level. Significance at the 85-percent confidence level is
important evidence that the relevant point estimates are meaningfully
different to zero (e.g., approximately five to six times more likely to
be non-zero than zero). The agencies believe it would be misleading to
ignore these data or to use values of zero for the rulemaking analysis,
as doing so would not properly inform decision makers on the safety
impacts of the regulatory alternatives and final standards. Similar to
past analyses, the NPRM and this final rule analysis use the best
available estimates. The agencies feel it is inappropriate to ignore
likely impacts of the standards simply because the best available
estimates have confidence levels below 95 percent; uniform estimates of
zero are statistically weaker than the estimates identified in the
analysis, and thus are not the best available. Because the point
estimates are derived from the best-fitting estimates for each crash
type (all of which are non-zero), the confidence bounds around an
overall estimate of zero would necessarily be larger than the
corresponding confidence bounds around the point estimates presented
here.
The sensitivity analysis in Section VII.C Sensitivity Analysis
provides an evaluation of extreme cases in which all of the estimated
net fatality rate impacts of mass reduction are either at their fifth-
or 95th-percentile values. The range of net impacts in the sensitivity
analysis not only covers the relatively more likely case that
uncertain, yet
[[Page 24752]]
generally offsetting, effects are distinct from the central estimates
considered here (e.g., in a plausible case where mass reduction in the
heaviest LTVs is less beneficial than indicated by the central
estimates, it would also be relatively likely that mass reduction in
the lightest passenger cars would be less harmful, yielding a similar
net impact), but also covers the relatively unlikely case that all of
the estimates are uncertain in the same direction.
At a broader level, multiple commenters asserted that the role of
safety-related estimates should be restricted because of what they
claim is a weak historical relationship between fuel economy and
vehicle safety. For example, the Green Energy Institute at Lewis &
Clark Law School commented, ``[o]ver the past 40 years, per-capita
vehicle fatalities decreased by 50%, while average fuel economy
doubled.'' \1991\ However, this statistic is misleading because it does
not account for vehicle safety factors and changes in driving behavior
external to fuel economy (e.g., FMVSS and other safe design advances,
reductions in drunk driving, increases in seat belt use). That is,
fatality rates have decreased due to a range of factors that are
unrelated to fuel economy efforts. The methodology in the 2012 Kahane
report, the 2016 Puckett and Kindelberger, the Draft TAR, the 2018 NPRM
analysis and today's final rule analysis addresses these other changes
in order to isolate the impacts of mass reduction alone. The role of
the safety analysis outlined in this document is to isolate incremental
effects on safety outcomes that are related to changes in fuel economy.
---------------------------------------------------------------------------
\1991\ Green Energy Institute at Lewis & Clark Law School,
Docket No. NHTSA-2018-0067-12213, at 3.
---------------------------------------------------------------------------
Multiple commenters disagreed with the results in Table VI-204,
maintaining that mass reduction need not reduce societal safety. EDF
cited a Michigan Manufacturing Technology Center (MMTC) review as
supporting that widespread lightweighting would decrease crash severity
through reduced kinetic energy in multiple-vehicle crashes. Similarly,
the Aluminum Association commented, ``[v]ehicle size, not weight, has
been shown to be the leading safety determinant.'' \1992\ Other
commenters cited Anderson and Auffhammer (2014), which finds that the
safety effects of mass reduction in one vehicle are offset by the
safety effects in the crash partner vehicle.\1993\ The South Coast Air
Quality Management District asserted that NHTSA and EPA appear to argue
``that fuel-efficient vehicles are lighter than other vehicles, and
therefore, less safe.'' The North Carolina Department of Environmental
Quality asserted that a takeaway from the preferred alternative is that
larger vehicles are safer than smaller vehicles. The agencies'
conclusion is that, at the societal level, it is the distribution of
changes in vehicle mass that matter (i.e., mitigating mass reduction in
the lightest vehicles is societally beneficial, while mitigating mass
reduction in the heaviest vehicles is societally harmful).
---------------------------------------------------------------------------
\1992\ The Aluminum Association, Detailed Comments, Docket No.
NHTSA-2018-0067-12213, at 3.
\1993\ Anderson, M.L. and M. Auffhammer (2014). ``Pounds that
Kill,'' Review of Economic Studies, Vol. 81, No. 2, pp. 535-71.
---------------------------------------------------------------------------
The 2012 Kahane report, the 2016 Puckett and Kindelberger, the
Draft TAR, the 2018 NPRM analysis and today's final rule analysis all
have shown that both mass and vehicle size impact societal safety.
Across recent rulemakings, the analyses have confirmed a protective
effect of vehicle size (i.e., societal fatality risk decreases as
footprint increases). As mentioned previously, the agencies believe
vehicle footprint-based standards help to discourage vehicle
manufacturers from downsizing their vehicles, and therefore assume
changes in CAFE and CO2 standards will not impact vehicle
size and size-related safety impacts. On the other hand, mass reduction
is a cost-effective technology for increasing fuel economy and reducing
CO2 emissions. Therefore, the agencies do include the
assessment of safety impacts related to mass reduction. As discussed
throughout this mass-safety subsection, the agencies present
comprehensive consideration of the various studies and workshops on the
impact of vehicle mass on safety, and conclude there is in fact a
relationship. The fleet simulation study, discussed in the next
subsection, further supports the existence of this relationship and
that this relationship will continue to exist in future vehicle
designs.
The principal difference between heavier vehicles, especially
truck-based LTVs, and lighter vehicles, especially passenger cars, is
that mass reduction has a different effect in collisions with another
car, LTV, or other object such as a lamp post. When two vehicles of
unequal mass collide, the change in velocity (delta-V) is greater in
the lighter vehicle. Through conservation of momentum, the degree to
which the delta-V in the lighter vehicle is greater than in the heavier
vehicle is proportional to the ratio of mass in the heavier vehicle to
mass in the lighter vehicle:
[GRAPHIC] [TIFF OMITTED] TR30AP20.419
[GRAPHIC] [TIFF OMITTED] TR30AP20.420
Because fatality risk is a positive function of delta-V, the
fatality risk in the lighter vehicle in two-vehicle collisions is also
higher. Vehicle design can reduce the magnitude of delta-V to some
degree (e.g., changing the stiffness
[[Page 24753]]
of a vehicle's structure could dampen delta-V for both crash partners).
These considerations drive the overall result: mass reduction is
associated with an increase in fatality risk in lighter cars, a
decrease in fatality risk in heavier LTVs, CUVs, and minivans, and has
smaller effects in the intermediate groups. Mass reduction may also be
harmful in a crash with a movable object such as a small tree, which
may break if hit by a high mass vehicle resulting in a lower delta-V
than may occur if hit by a lower mass vehicle which does not break the
tree and therefore has a higher delta-V. However, in some types of
crashes not involving collisions between cars and LTVs, especially
first-event rollovers and impacts with fixed objects, mass reduction
may not be harmful and may even be beneficial.
Ultimately, delta-V is a direct function of relative vehicle mass
for given vehicle structures. Removing some mass from the heavier
vehicle involved in an accident with a lighter vehicle reduces the
delta-V in the lighter vehicle, where fatality risk is higher,
resulting in a large benefit to the passengers of the lighter vehicle.
This is partially offset by a small increase in the delta-V in the
heavy vehicle; however, the fatality risk is lower in the heavier
vehicle and remains relatively low despite the increase in delta-V. In
sum, the change in mass and delta-V from mass reduction in heavier
vehicles results in a net societal benefit.
Multiple commenters claimed that the agencies' analysis does not
allow for the likely outcome that mass reduction would be concentrated
among relatively heavy vehicles.\1994\ For example, Global Automakers
commented that the agencies should not include weight reduction in
their safety analysis because ``very few vehicles [have] implemented
lightweight material substitution strategies.'' \1995\
---------------------------------------------------------------------------
\1994\ See also, e.g., South Coast Air Quality Management
District, Detailed Comments, Docket No. NHTSA-2018-0067-11813, at 6.
\1995\ Association of Global Automakers, Attachment A, Docket
No. NHTSA-2018-0067-12032, at A-32.
---------------------------------------------------------------------------
Neither CAFE standards nor this analysis mandate mass reduction, or
mandate mass reduction occur in any specific manner. However, mass
reduction is a highly cost effective technology for improving fuel
economy and CO2 emissions. The steel, aluminum, plastics,
composite, and other material industries are developing new materials
and manufacturing equipment and facilities to produce those materials.
In addition, suppliers and manufacturers are optimizing designs to
maintain or improve functional performance with lower mass.
Manufacturers have stated that they will continue to reduce vehicle
mass to meet more stringent standards, and therefore, this expectation
is incorporated into the modeling analysis supporting the standards to:
(1) Determine capabilities of manufacturers; and (2) to predict costs
and fuel consumption effects of CAFE standards. The CAFE and
CO2 rulemakings in 2012, the Draft TAR and EPA Preliminary
Determination, imposed an artificial constraint on vehicle mass
reduction to achieve a desired safety-neutral outcome. For the current
rulemaking, this artificial constraint is eliminated so the analysis
reflects manufacturers applying the most cost effective technologies to
achieve compliance with the regulatory alternatives and the final
standards; this approach allows mass reduction to be applied across the
fleet. This is consistent with industry trends.\1996\ To the extent
that mass reduction is only cost-effective for the heaviest vehicles,
the CAFE model would create the outcome predicted by commenters. In
reality, however, mass reduction is a cost-effective means of improving
fuel economy and does take place across vehicles of all sizes and
weights. Accordingly, the model reflects that manufacturers may reduce
vehicle mass--regardless of vehicle class--when doing so is cost
effective.
---------------------------------------------------------------------------
\1996\ The baseline MY 2016 (for the NPRM) and MY 2017 (for this
final rule analysis) vehicle fleet data show manufacturers have in
fact implemented mass reduction technology across vehicle types and
sizes- including smaller and lighter vehicles.
---------------------------------------------------------------------------
The National Tribal Air Association claimed the 2015 NAS study
found ``evidence suggest[ing] that the [2012] standards will lead the
nation's light-duty vehicle fleet to become lighter but not less
safe.'' \1997\ The agencies note the NAS quote is one phrase from the
press release that accompanied the NHTSA sponsored 2015 NAS
study,\1998\ and the agencies do not believe the phrase in isolation
reflects the findings of the NAS Committee, which are discussed in over
3 pages of the report.\1999\ The 2015 NAS report supported the
analytical methodology used for the 2012 NHTSA CAFE and EPA
CO2 rulemaking and found it reasonable. As discussed in the
subsections further above, a nearly identical methodology was used for
the NPRM analysis and for this final rule.
---------------------------------------------------------------------------
\1997\ National Tribal Air Association, Detailed Comments,
Docket No. NHTSA-2018-0067-11948, at 2.
\1998\ NAS (2015). Press Release. ``Analysis Used by Federal
Agencies to Set Fuel Economy and Greenhouse Gas Standards for U.S.
Cars Was Generally of High Quality; Some Technologies and Issues
Should Be Re-examined.'' June 18, 2015. Available at http://www8.nationalacademies.org/onpinews/newsitem.aspx?RecordID=21744.
\1999\ Key excerpts from the report include: ``[o]ccupants of
smaller vehicles are at a greater risk of fatality in crashes,
particularly in a crash with a vehicle of greater mass;'' and
``[t]he 2012 studies (by NHTSA, Lawrence Berkeley National
Laboratories, and Dynamic Research, Inc.) indicate that mass
reduction while holding footprint constant is associated with a
small increase in risk for lighter-than-average cars only; the
estimated effect on other vehicle types is not statistically
significant.'' National Research Council (2015). Cost,
Effectiveness, and Deployment of Fuel Economy Technologies for
Light-Duty Vehicles, available at https://doi.org/10.17226/21744.
pp. 224-28.
---------------------------------------------------------------------------
The agencies received several comments about the relationship
between mass and crash avoidance. The NRDC commented that the analysis
should account for the expected result that mass reduction makes it
easier to avoid crashes.\2000\ Conversely, IPI quoted a finding by LNL
that ``found that mass reductions may increase the number of accidents
but that each crash results in fewer fatalities.'' \2001\
---------------------------------------------------------------------------
\2000\ NRDC, Detailed Comments, Docket No. NHTSA-2018-0067-
11973.
\2001\ IPI, Detailed Comments, Docket No. NHTSA-2018-0067-12213,
at 129.
---------------------------------------------------------------------------
The phenomenon touched upon by IPI and NRDC has been identified in
past rulemakings as well, and highlights that the relationship between
mass reduction and societal fatality risk include two partially-
offsetting components (i.e., increased exposure to crashes is offset
partially by decreased risk in some vehicles conditional on a crash
occurring). The agencies note that this relationship, while not
reported separately, is in fact embedded within the analysis detailed
in this document, as the extent to which some vehicles are more
maneuverable and faster-braking, the crash data reflect those
characteristics through lower observed fatality rates. However, when
considering the purposes of estimating effects of mass reduction on
fatalities, it is immaterial what share of the effect is comprised of
crash avoidance factors and crashworthiness factors, the ultimate
effect is present within the data evaluated in the analysis. The mass-
safety impacts estimated by the statistical analysis of crash data are
based on the safety technologies and mass levels present among the
vehicle fleets for the calendar and model years in the data. As
discussed below in this section, the analysis separately accounts for
the effects of future safety technologies.
(4) Sensitivity Analysis
Table VI-206 shows the principal findings and includes sampling-
error
[[Page 24754]]
confidence bounds for the five parameters used in the CAFE model. The
confidence bounds represent the statistical uncertainty that is a
consequence of having less than a census of data. NHTSA's 2011, 2012,
and 2016 reports acknowledged another source of uncertainty: The
central (baseline) statistical model can be varied by choosing
different control variables or redefining the vehicle classes or crash
types, which for example, could produce different point estimates.
Beginning with the 2012 Kahane report, NHTSA has provided results
of 11 plausible alternative models that serve as sensitivity tests of
the baseline model. Each alternative model was tested or proposed by:
Farmer (IIHS) or Green (UMTRI) in their peer reviews; Van Auken (DRI)
in his public comments; or Wenzel in his parallel research for DOE. The
2012 Kahane and 2016 Puckett and Kindelberger reports provide further
discussion of the models and the rationales behind them.
Alternative models use NHTSA's databases and regression-analysis
approach but differ from the central model in one or more explanatory
variables, assumptions, or data restrictions. The agencies applied the
11 techniques to the latest databases to generate alternative CAFE
model coefficients. The range of estimates produced by the sensitivity
tests offers insight to the uncertainty inherent in the formulation of
the models, subject to the caveat that these 11 tests are, of course,
not an exhaustive list of conceivable alternatives.
The central and alternative results follow, ordered from the lowest
to the highest estimated increase in societal risk per 100-pound
reduction for cars weighing less than 3,201 pounds:
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.421
[[Page 24755]]
BILLING CODE 4910-59-C
The sensitivity tests illustrate both the fragility and the
robustness of central estimates. On the one hand, the variation among
the coefficients is quite large relative to the central estimate: In
the preceding example of cars < 3,201 pounds, the estimated
coefficients range from almost zero to almost double the central
estimate. This result underscores the key relationship that the
societal effect of mass reduction is small. In other words, varying how
to model some of these other vehicle, driver, and crash factors, which
is exactly what sensitivity tests do, can appreciably change the
estimate of the societal effect of mass reduction.
On the other hand, variations are not particularly large in
absolute terms. The ranges of alternative estimates are generally in
line with the sampling-error confidence bounds for the central
estimates. Generally, in alternative models as in the central model,
mass reduction tends to be relatively more harmful in the lighter
vehicles and more beneficial in the heavier vehicles, just as they are
in the central analysis. In all models, the point estimate of the
coefficient is positive for the lightest vehicle class, cars < 3,201
pounds. In 10 out of 11 models, the point estimate is negative for CUVs
and minivans, and in nine out of 11 models the point estimate is
negative for LTVs >= 5,014 pounds. The agencies believe the central
case uses the most rigorous methodology, as discussed further above,
and provides the best estimates of the impacts of mass reduction on
safety.
Tom Wenzel commented confirming a preference for the alternative
model with footprint separated into track width and wheelbase, and with
the induced exposure data limited to stopped vehicle cases.\2002\
Wenzel asserts that splitting footprint into its components reduces
multicollinearity with curb weight, and that limiting induced exposure
cases to stopped vehicles mitigates bias against driver-vehicle pairs
that are less likely to be involved in crashes. Based on this feedback
and the intuitiveness of the approach, the agencies further considered
the alternative model with footprint split into track width and
wheelbase. Consistent with previous analyses and assessments, there are
problems with splitting footprint into its components within the mass-
size-safety models because of strong correlations among curb weight,
track width and wheelbase. For all vehicle classes in the analysis,
curb weight is correlated either nearly as high or higher with track
width as with footprint. Track width and wheelbase are also highly
correlated with one another (ranging from around 0.64 to 0.80, with the
exceptions of smaller correlations for large pickups and minivans).
Viewed from another angle, wheelbase is almost perfectly correlated
with footprint (with correlations ranging from around 0.95 to 0.97).
---------------------------------------------------------------------------
\2002\ Wenzel, T., Lawrence Berkeley National Laboratories,
Docket No. EPA-HQ-OAR-2018-0283-4118.
---------------------------------------------------------------------------
Considered in concert, the track width and wheelbase model not only
essentially incorporates the full correlation issues from the baseline
model (curb weight highly correlated with another independent
variable), but also adds a further correlation issue (the variable that
is highly correlated with curb weight is also highly correlated with a
separate independent variable). The agencies examined supplementary
means of confirming the relative methodological merit of the footprint-
based model and the track-width-wheelbase-based alternative. The
supplementary analysis centered on the condition index, which
quantifies the invertibility of the matrix of independent variables in
a given model through its measure, the condition number.\2003\ A model
with a low condition number has relatively low correlations among its
independent variables, and thus its invertibility and the corresponding
model outputs are robust to variations in model input values. A model
with a high condition number has relatively high correlations among its
independent variables, and thus its invertibility and model outputs are
not robust to variations in model input values. That is, a model with a
high condition number is likely to be subject to the problems
associated with multicollinearity. Although there is no strict
threshold condition number value to indicate multicollinearity, higher
values indicate greater likelihood that the independent variables are
correlated to a problematic degree.
---------------------------------------------------------------------------
\2003\ See Belsley, D.A., Kuh, E., and Welsch, R.E. (1980).
``The Condition Number.'' Regression Diagnostics: Identifying
Influential Data and Sources of Collinearity. New York: John Wiley &
Sons; Freund, R.J. and Littell, R.C. (2000). SAS System for
Regression, Third Edition. Cary, NC: SAS Institute, Inc.; and
Hallahan, C. (1995). ``Understanding the Multicollinearity
Diagnostics in SAS/Insight and Proc Reg.'' SAS Conference
Proceedings, Washington, DC, October 8-10, 1995.
---------------------------------------------------------------------------
The condition index offers an alternative means of capturing the
same forces as the variance inflation factor (VIF), which the agencies
have used historically (including in this rulemaking) as a diagnostic
of multicollinearity. However, the condition index offers some
advantages relative to the VIF. Notably, the condition index applies
regardless of the econometric form of the model (i.e., the
decomposition of the independent variables is the same regardless of
how the variables are applied in the model). This is distinct from the
VIF, which is limited to a linear diagnostic of the data that may not
map well to non-linear econometric models, including the logistic
regression models that form the core of the curb weight-fatality risk
analysis. The condition index estimates the incremental effects of
individual variables, which is helpful in an analysis of which
independent variables are the most problematic. Conversely, the
diagnostic values from the VIF are not necessarily sensitive to
incremental correlated variables, as the VIF value (1/(1-R\2\) does not
necessarily change much once correlations are relatively high (i.e.,
when R\2\ is already high, the inclusion of one or more highly
correlated variables may not change R\2\, and in turn, the VIF, by
much.
An incremental comparison of VIF estimates for the data confirmed
the potential weakness of the VIF in this case. For the CUV-minivan
model data, the VIF decreases from 9.4 to 6.7 when: (1) Substituting
either track width or footprint for footprint that has an identical
correlation with curb weight as footprint; and (2) adding the other
component of footprint. This result is counterintuitive (i.e., the
simpler model should necessarily have fewer issues of
multicollinearity), and may be an artifact of differences in model fit
(e.g., a higher R\2\ in the simpler model could indicate better model
fit rather than anything problematic in terms of correlation
structure). This result led the agencies to question how well the VIF
identifies relative impacts of multicollinearity across related models,
especially in non-linear applications.
The calculated condition numbers for the curb weight-footprint
models and their corresponding curb weight-wheelbase-track width
alternatives were consistent with expectations regarding
multicollinearity, however. The condition numbers for the curb weight-
wheelbase-track width models are approximately two to three times
higher than the condition numbers for the curb weight-footprint models.
This indicates that the level of imprecision in model estimates using
track width and wheelbase would be expected to be between approximately
two to three times higher than in the baseline models using footprint.
Unlike the VIF, the condition index supports a hypothesis that
multicollinearity would not be mitigated in an alternative with
disaggregated variables that are highly
[[Page 24756]]
correlated with both the variable of interest and the variable they are
replacing. Considering these results, the agencies that using footprint
to represent vehicle size in the safety models provides a more reliable
estimate of safety impacts than splitting footprint into track width
and wheelbase.
The agencies also considered the use of stopped-vehicle data as an
alternative. The primary problem with this approach is that the
agencies do not observe as large of a share of cases on roads with
higher travel speeds (e.g., interstate highways) when including only
stopped vehicles; this relationship influences the extent to which the
induced exposure data reflect the distributions of driver attributes
and contextual effects across national VMT. Based on this assessment,
the agencies believe the methodology used for the analysis in the
proposal provides a more reliable and representative estimate of safety
impacts, and thus is not changing the methodology for today's final
rule.
In a related comment, Wenzel proposes that future analyses should
directly account for differences in curb weight between vehicles in
two-vehicle crashes. The agencies believe that would require the
development of a model that directly accounts for the relative weights
of vehicles in two-vehicle crashes, and that such a model would require
peer review. Key alternatives to test would vary in terms of the
functional form of the mass disparity between two crash partners (e.g.,
a relative mass ratio consistent with the delta-V calculation presented
above, linear mass difference, non-linear mass difference). The
agencies will consider initiating work to explore such a model in the
future.
DRI requested the agencies clarify whether the analysis accounts
for all road users (i.e., including pedestrians, bicyclists,
motorcyclists, and other crash partners), while the Pennsylvania
Department of Environmental Protection commented, ``[i]t is inadequate
for the agencies' analysis for this Proposed Rule to only focus on
frontal crashes while omitting near-frontal collisions, side-impact
collisions, rear-end collisions, rollover accidents, impacts with
stationary objects and accidents involving pedestrians.'' \2004\ The
agencies confirm that the analysis presented in this section continues
to apply the methodology developed by Kahane, which incorporates all
road users, without double-counting, to identify societal fatality rate
impacts. Because every fatal crash (across crash types) is included in
the analysis, not just frontal crashes, the agencies find this comment
lacks a basis. The agencies believe the commenter's confusion may stem
from the use of front-to-back crashes to generate estimates of the
proportions of all driving for each vehicle model associated with
particular characteristics of drivers (e.g., age, gender) and crashes
(e.g., urban/rural, day/night). These crashes represent the best
available trade-off among sample size, representativeness of overall
vehicle and driver exposure, and mitigating bias in a sample that is
intended to be effectively random (i.e., the probability of being
struck from behind by an at-fault driver is assumed to be a function of
characteristics of other drivers and travel demand, but not of the
struck driver or the struck vehicle).
---------------------------------------------------------------------------
\2004\ Pennsylvania Department of Environmental Protection,
Detailed Comments, Docket No. NHTSA-2018-0067-11956, at 9.
---------------------------------------------------------------------------
(5) Fleet Simulation Study
Commenters to recent CAFE rulemakings, including some vehicle
manufacturers, have suggested designs and materials of more recent
model year vehicles may have weakened the historical statistical
relationships between mass, size, and safety. NHTSA and EPA agreed that
the statistical analysis would be improved by using an updated crash
and exposure database reflecting more recent safety technologies,
vehicle designs and materials, and reflecting changes in the vehicle
fleet. As mentioned above, a new crash and exposure database was
created with the intention of capturing modern vehicle engineering and
has been employed for assessing safety effects for CAFE rules since
2012.
The agencies have traditionally relied solely on real-world crash
data as the basis for projecting the future safety implications for
regulatory changes. The agencies are required to consider relevant data
in setting standards.\2005\ Every fleet regulated by the agencies'
standards differs from the fleet used to establish said standard, and
as such, the light-duty vehicle fleet in the MY 2021-2026 timeframe
will be different from the MY 2004-2011 fleet analyzed above. This is
not a new or unique phenomenon, but instead is an inherent challenge in
regulating an industry reliant on continual innovation. This is the
agencies' sixth evaluation of effects of mass reduction and/or
downsizing,\2006\ comprising databases ranging from MYs 1985 to 2011.
Despite continual claims that modern lightweight engineering will
render current data obsolete, results of the six studies, while not
identical, have been generally consistent in showing a small, negative
impact related to mass reduction. The agencies strongly believe that
real-world crash data remains the best, relevant data to measure the
effect of mass reduction on safety.
---------------------------------------------------------------------------
\2005\ See Center for Biological Diversity v. NHTSA, 538 F.3d
1172, 1203 (9th Cir. 2008).
\2006\ As outlined throughout this section, NHTSA's six related
studies include the new analysis supporting this rulemaking, and:
Kahane, C.J. Vehicle Weight, Fatality Risk and Crash Compatibility
of Model Year 1991-99 Passenger Cars and Light Trucks, National
Highway Traffic Safety Administration (Oct. 2003), available at
https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/809662;
Kahane, C.J. Relationships Between Fatality Risk, Mass, and
Footprint in Model Year 1991-1999 and Other Passenger Cars and LTVs
(Mar. 24, 2010), in Final Regulatory Impact Analysis: Corporate
Average Fuel Economy for MY 2012-MY 2016 Passenger Cars and Light
Trucks, National Highway Traffic Safety Administration (Mar. 2010)
at 464-542; Kahane, C.J. Relationships Between Fatality Risk, Mass,
and Footprint in Model Year 2000-2007 Passenger Cars and LTVs--
Preliminary Report, National Highway Traffic Safety Administration
(Nov. 2011), available at Docket ID NHTSA-2010-0152-0023); Kahane,
C.J. Relationships Between Fatality Risk, Mass, and Footprint in
Model Year 2000-2007 Passenger Cars and LTVs: Final Report, NHTSA
Technical Report. Washington, DC: NHTSA, Report No. DOT-HS-811-665;
and Puckett, S.M., & Kindelberger, J.C. Relationships between
Fatality Risk, Mass, and Footprint in Model Year 2003-2010 Passenger
Cars and LTVs--Preliminary Report, National Highway Traffic Safety
Administration (June 2016), available at https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/2016-prelim-relationship-fatalityrisk-mass-footprint-2003-10.pdf.
---------------------------------------------------------------------------
However, because lightweight vehicle designs introduce fundamental
changes to the structure of the vehicle, there remains a persistent
question of whether historical safety trends will apply. To address
this concern and to verify that real-world crash data remain an
appropriate source of data for projecting mass-safety relationships in
the future fleet, in 2014, NHTSA sponsored research to develop an
approach to utilize experimental lightweight vehicle designs to
evaluate safety in a broader range of real-world representative
crashes.\2007\ NHTSA contracted with George Washington University to
perform a fleet simulation model to study the impact and relationship
of light-weighted vehicle design with injuries and fatalities.\2008\
The study involved simulating crashes on eight test vehicles, five of
which were equipped with lightweight materials
[[Page 24757]]
and advanced designs not yet incorporated into the U.S. fleet. The
study assessed a range of frontal crashes, including crashes with fixed
objects and other vehicles, across wide range of vehicle speeds, and
with mid-size male and mid-size female dummies.\2009\ In all, more than
440 vehicle crashes with 1,520 dummy passengers were simulated for a
range of crash speeds and crash configurations. Results from the fleet
simulation study showed the trend of increased societal injury risk for
light-weighted vehicle designs occurs for both single vehicle and two-
vehicle crashes. Results are listed in Table VI-207.\2010\
---------------------------------------------------------------------------
\2007\ See also 83 FR at 43133 (Aug 24, 2018).
\2008\ Samaha, R.R., Prasad, P., Marzougui, D., Cui, C., Digges,
K., Summers, S., Patel S., Zhao, L., & Barsan-Anelli, A. (2014,
August). Methodology for evaluating fleet protection of new vehicle
designs--Application to lightweight vehicle designs. Report No. DOT
HS 812 051A, Washington, DC--National Highway Traffic Safety
Administration.
\2009\ Regulatory and consumer information crash safety tests
are performed at high speeds, and the dummy occupant is generally a
mid-size male. In the real world, crashes occur at various impact
velocities and configurations; with various impact partners (e.g.,
rigid obstacles, lighter or heavier vehicles); and involve occupants
of various sizes and ages.
\2010\ This fleet simulation study does not provide information
that can be used to modify coefficients derived for the NPRM
regression analysis because of the restricted types of crashes and
vehicle designs. Additionally, the fleet simulation study assumed
restraint equipment to be as in the baseline model, in which
restraints/airbags are not redesigned to be optimal with light-
weighting.
[GRAPHIC] [TIFF OMITTED] TR30AP20.422
[GRAPHIC] [TIFF OMITTED] TR30AP20.423
The change in the safety risk from the fleet simulation study was
directionally consistent with results for passenger cars from the 2012
Kahane report,\2011\ the 2016 Puckett and Kindelberger report, and the
analysis used for the proposal and today's final rule. As noted, fleet
simulations were performed in frontal crash mode and did not consider
other crash modes such as rollover crashes.\2012\ The fleet simulation
analysis confirmed that real-world crash data were still a reliable
source for analyzing mass safety impacts.
---------------------------------------------------------------------------
\2011\ The 2012 Kahane study considered only fatalities,
whereas, the fleet simulation study considered severe (AIS 3+)
injuries and fatalities (DOT HS 811 665).
\2012\ The risk assessment for CUV in the regression model
combined CUVs and minivans in all crash modes and included belted
and unbelted occupants.
---------------------------------------------------------------------------
Despite the results of the fleet simulation analysis, which was
republished in the proposal, the agencies received additional comments
questioning the assumption that relationships among vehicle mass, size,
and fatality risk will continue in the future. For example, the
Alliance for Vehicle Efficiency asserted that using lighter frame
materials has no impact on safety, noting that any mass reduction
strategies are applied to components that are unrelated to crash safety
and crash ratings have not declined for vehicles over the past five
years.\2013\ CARB commented that the agencies did not account for new
vehicle improvements and claimed the data used for the analysis was
``not a good indicator of the safety performance of future purpose-
designed lightweighted vehicles.'' \2014\ Consumers Union offered a
similar appraisal, indicating that the MYs in the sample are ``unlikely
to capture the current and future mass/fatality relationship of modern
vehicles.'' \2015\ While the Aluminum Association commented vehicle
size, not mass, is the only physical feature that impacts safety.\2016\
The American Chemistry Council, Hyundai, and Tesla commented that it is
feasible to utilize
[[Page 24758]]
design improvements and technologies to offset the incremental risk for
vehicle occupants associated with mass reduction.\2017\ EDF said the
mass-safety analysis did not agree with conclusions from a study by the
Michigan Manufacturing Technology Center.\2018\ Comments from States
and Cities, American Honda, ICCT, and NRDC shared these
sentiments.\2019\
---------------------------------------------------------------------------
\2013\ Alliance for Vehicle Efficiency, Detailed Comments,
Docket No. NHTSA-2018-0067-11696, at 11.
\2014\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 270.
\2015\ Consumers Union, Detailed Comments, Docket No. NHTSA-
2018-0067-12068, at 18.
\2016\ Aluminum Association, Detailed Comments, Docket No.
NHTSA-2018-0067-11952, at 3.
\2017\ American Chemistry Council, Detailed Comments, Docket No.
EPA-HQ-OAR-2018-0283-1415, at 2-8; Hyundai-Kia America Technical
Center, Detailed Comments, Docket No. EPA-HQ-OAR-2018-0283-4411, at
13; Tesla, Detailed Comments, Docket No. EPA-HQ-OAR-2018-0283-4186,
at 21-23.
\2018\ Michigan Manufacturing Technology Center study ``Vehicle
Lightweighting: A Review of the Safety of Reduced Weight Passenger
Cars and Light Duty Trucks,'' October 2018, available at https://advocacy.consumerreports.org/wp-content/uploads/2018/10/CU-MMTC-Safety-Study-10-24-2018.pdf.
\2019\ States and Cities, Detailed Comments, Docket No. NHTSA-
2018-0067-11735 at 81 and 95; American Honda, Detailed Comments,
Docket No. NHTSA-2018-0067-11818, at 15; ICCT, Detailed Comments,
Docket No. NHTSA-2018-0067-11741, at II-10-11. National Resources
Defense Council, Detailed Comments, Docket No. EPA-HQ-OAR-2018-0283-
4410, at 11-14.
---------------------------------------------------------------------------
These comments and the MMTC study ignored the results of the fleet
simulation study and seem premised on the notion that a vehicles'
performance on NHTSA FMVSS, NHTSA voluntary NCAP, and IIHS voluntary
safety tests is the only measure for assessing societal safety impacts
for mass reduction. The regulatory and consumer information tests are
representative of real-world, single-vehicle crash configurations.
However, the tests are performed at constant speeds, and the dummy
occupant is generally a mid-size male. In the real world, crashes occur
at various impact velocities and configurations; with various impact
partners (e.g., rigid obstacles, lighter or heavier vehicles); and
involve occupants of various sizes and ages. The fleet simulation
study, summarized above, assessed additional types of frontal crashes,
including crashes with fixed objects and other vehicles at a wide range
of vehicle speeds, and with mid-size male and mid-size female dummies.
The fleet simulation study was more comprehensive and focused on the
need to assess overall societal safety impacts. The fleet simulation
study found that vehicle mass does impact safety with future
lightweight vehicle designs that perform well on regulatory and
consumer information tests.
The agencies received one comment regarding the fleet simulation
analysis. CARB commented that the analysis tested too few vehicles and
crash types, should have optimized restraints in the lightweighted
models to simulate future safety improvements instead of using modern
restraints, and lacked credibility because the results of the fleet
simulation analysis did not reproduce the same results of other
studies.\2020\ CARB's comments demonstrate a general misunderstanding
of the fleet simulation analysis; the analysis was not intended to
serve as a prediction of how the future vehicle fleet will perform, but
rather was an exploration of whether expected lightweighting techniques
would alter the dynamic between mass reduction and safety. The analysis
was not an attempt to model every potential vehicle construction or
crash scenario. Attempting to simulate every future crash would be
impractical and ineffective. The combination of vehicles and crash
simulations were purposely selected to provide the strongest insight
into the effective of lightweighting techniques. For passenger cars and
light trucks, frontal crashes account for 58 percent of fatal crashes;
\2021\ it is appropriate to focus research on understanding the effects
of mass reduction where the largest issue exists. For the study, the
use of generic restraint systems as the foundations for the models was
intentional so that the models would be more representative of a
vehicle class rather than a specific vehicle. The models of the
restraint systems represented designs currently in production at time
of the study in terms of pretensioners, load limiters and air bag
inflators. It is worth noting that in general, driver air bags are
similar in most vehicles. And finally, the analysis was not an attempt
to reproduce the 2012 Kahane report or any other study. The fact that
the fleet simulation analysis showed mass-reduction to be detrimental
in more types of vehicles than in the FARS data only further highlights
the need to consider how today's standards may impact mass-safety.
While in the future there may be resources and opportunity to expand
the fleet simulation approach to other crash scenarios and, if they
become available, to include additional vehicle mass reduction
concepts, the lack of potential future data does not justify ignoring
the data that currently exist.
---------------------------------------------------------------------------
\2020\ CARB, Detailed Comments, Docket No. NHTSA-2018-0067-
11873, at 272-73.
\2021\ Samaha, R.R., Prasad, P., Marzougui, D., Cui, C., Digges,
K., Summers, S., Patel S., Zhao, L., & Barsan-Anelli, A. (2014,
August). Methodology for evaluating fleet protection of new vehicle
designs--Application to lightweight vehicle designs. Report No. DOT
HS 812 051A, Washington, DC--National Highway Traffic Safety
Administration.
---------------------------------------------------------------------------
From a higher perspective, the comments, and in particular CARB's
comment, identify the problem with abandoning real-world crash data:
There is no alternate methodology or data that can account for the full
diversity of crash scenarios that occur in the real world. Real-world
crash data is the only data type that can achieve that. Therefore, the
agencies have determined that, while simulations can prove helpful to
understanding potential effects of key crash scenarios and as a check
on the agencies' preferred analysis, real-world data still is still the
best, most relevant data available for assessing safety.
(6) Summary of Mass Safety Impacts
Table VI-208 through Table VI-213 show results of NHTSA's vehicle
mass-size-safety analysis over the cumulative lifetime of MY 1977-2029
vehicles, for both the CAFE and CO2 programs, based on the
MY 2017 baseline fleet, accounting for the projected safety baselines.
Results are driven extensively by the degree to which mass is reduced
in relatively light passenger cars and in relatively heavy vehicles
because their coefficients in the logistic regression analysis have the
most significant values. The agencies assume any impact on fatalities
will occur over the lifetime of the vehicle, and the chance of a
fatality occurring in any particular year is directly related to the
weighted vehicle miles traveled in that year.
BILLING CODE 4910-59-P
[[Page 24759]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.424
[[Page 24760]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.425
[[Page 24761]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.426
[[Page 24762]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.427
[[Page 24763]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.428
[[Page 24764]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.429
As shown in the tables above, all of the alternatives are estimated
to lead to a decrease in the number of mass-related fatalities over the
cumulative lifetime of MY 1977-2029 vehicles. The effects of mass
changes on fatalities
[[Page 24765]]
range from a combined decrease (relative to the augural standards, the
baseline) of 143 fatalities for Alternative #7 to a combined decrease
of 288 fatalities for Alternatives #1 and #2. The difference in results
by alternative depends upon how much weight reduction is used in that
alternative and the types and sizes of vehicles to which the weight
reduction applies. The decreases in fatalities are driven by impacts
within passenger cars (decreases of between 167 and 380 fatalities) and
are offset by impacts within light trucks (increases of between 9 and
92 fatalities).
Changes in vehicle mass are estimated to decrease social safety
costs over the lifetime of the nine model years by between $2.5 billion
(for Alternative #7) and $5.1 billion (for Alternatives #1 and #2)
relative to the augural standards at a three-percent discount rate and
by between $1.5 billion and $3.1 billion at a seven-percent discount
rate. The estimated decreases in social safety costs are driven by
estimated decreases in costs associated with passenger cars, ranging
from $3.0 billion (for Alternative #7) to $6.7 billion (for
Alternatives #1 and #2) relative to the augural standards at a three-
percent discount rate and by between $1.8 billion and $4.0 billion at a
seven-percent discount rate. The estimated decreases in costs
associated with passenger cars are offset partially by estimated
increases in costs associated with light trucks, ranging from $0.1
billion (for Alternative #5) to $1.6 billion (for Alternatives #1 and
#2) relative to the Augural standards at a three-percent discount rate
and by between $0.1 billion and $0.9 billion at a seven-percent
discount rate.
In this analysis, the profile of mass reduction across vehicle
models leads to a small, but beneficial effect on fatalities as fuel
economy standards are tightened. Table VI-212 through Table VI-219
present average annual estimated safety effects of vehicle mass
changes, for CYs 2036-2045. The CY-level values offer a complementary
view of the impacts of fuel economy standards on mass-related
fatalities relative to model-year-level results. Effects by CY over the
interval selected (2036-2045) enable a summary view of (a flow of)
annual fatality impacts during a period where vehicles subjected to the
standards have not only fully entered the fleet, but also interact with
both older and newer vehicles. Conversely, the MY-level values offer a
summary view of (a stock of) the impacts of fuel economy standards for
the lifetime of a given MY:
[[Page 24766]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.430
[[Page 24767]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.431
[[Page 24768]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.432
[[Page 24769]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.433
[[Page 24770]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.434
[[Page 24771]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.435
For all light-duty vehicles, mass changes are estimated to lead to
an average annual decrease in fatalities in all alternatives evaluated
for CYs 2035-2045. The effects of mass changes on fatalities range from
a combined
[[Page 24772]]
decrease (relative to the augural standards) of 20 fatality per year
for Alternative #7 to a combined decrease of 37 fatalities per year for
Alternative #4. The difference in the results by alternative depends
upon how much weight reduction is used in that alternative and the
types and sizes of vehicles to which the weight reduction applies. The
decreases in fatalities are generally driven by impacts within
passenger cars (decreases of between 22 and 50 fatalities per year
relative to the augural standards) and are offset by impacts within
light trucks (increases of between 2 and 12 fatalities per year).
Changes in vehicle mass are estimated to decrease average annual
social safety costs in CY 2035-2045 by between $0.3 billion (for
Alternative #7) and $0.6 billion (for Alternative #4) at a three-
percent discount rate relative to the augural standards (decrease of
between $0.1 and $0.2 billion at a seven-percent discount rate).
Average annual social safety costs associated with passenger cars in CY
2035-2045 are estimated to decrease by between $0.3 billion and $0.7
billion at a three-percent discount rate (decrease of between $0.1
billion and $0.3 billion at a seven-percent discount rate), but this
effect is partially offset by a corresponding increase in costs
associated with light trucks (increase of $0.2 billion or less across
alternatives at three-percent and seven-percent discount rates).
To help illuminate effects at the model year level, Table VI-220
presents the lifetime fatality impacts associated with vehicle mass
changes for passenger cars, light trucks, and all light-duty vehicles
by model year under the preferred alternative, relative to the augural
standards for the CAFE Program. Table VI-221 presents an analogous
table for the CO2 Program.
[[Page 24773]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.436
Under the preferred alternative, passenger car fatalities
associated with mass changes are estimated to decrease relative to the
augural standards steadily from MYs 2018-19 (decrease of 5 fatalities)
through MY 2028 (decrease of
[[Page 24774]]
53 fatalities). Conversely, light truck fatalities associated with mass
changes under the preferred alternative are estimated to increase
relative to the augural standards from MY 2019 (increase of 2
fatalities) through MY 2029 (increase of 9 fatalities).
Table VI-222 and Table VI-223 present estimates of monetized
lifetime social safety costs associated with mass changes by model year
at three-percent and seven-percent discount rates, respectively for the
CAFE Program. Table VI-224 and Table VI-225 show comparable tables from
the perspective of the CO2 Program.
[[Page 24775]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.437
[[Page 24776]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.438
Lifetime social safety costs associated with mass change in
passenger cars are estimated to decrease by between $0.1 billion (for
MYs 2020-22) and $0.3 billion (for MYs 2026-29) at a three-percent
discount rate. At a seven-
[[Page 24777]]
percent discount rate, lifetime social safety costs associated with
mass change in passenger cars are estimated to decrease by between $0.1
billion and $0.2 billion from MY 2021 through MY 2029. Lifetime social
safety costs associated with mass change in light trucks are estimated
to increase by $0.1 billion or less for all MYs at three-percent and
seven-percent discount rates.
BILLING CODE 4910-59-P
[[Page 24778]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.439
[[Page 24779]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.440
[[Page 24780]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.441
[[Page 24781]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.442
[[Page 24782]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.443
[[Page 24783]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.444
As shown in the tables above, all of the alternatives are estimated
to lead to a decrease in the number of mass-related fatalities over the
cumulative lifetime of MY 1977-2029 vehicles. The effects of mass
changes on fatalities
[[Page 24784]]
range from a combined decrease (relative to the augural standards, the
baseline) of 126 fatalities for Alternative #7 to a combined decrease
of 253 fatalities for Alternatives #1 and #2. The difference in results
by alternative depends upon how much weight reduction is used in that
alternative and the types and sizes of vehicles to which the weight
reduction applies. The decreases in fatalities are driven by impacts
within passenger cars (decreases of between 146 and 33 fatalities) and
are offset by impacts within light trucks (increases of between 8 and
81 fatalities).
Changes in vehicle mass are estimated to decrease social safety
costs over the lifetime of the nine model years by between $2.2 billion
(for Alternative #7) and $4.5 billion (for Alternatives #1 and #2)
relative to the augural standards at a three-percent discount rate and
by between $1.3 billion and $2.7 billion at a seven-percent discount
rate. The estimated decreases in social safety costs are driven by
estimated decreases in costs associated with passenger cars, ranging
from $2.6 billion (for Alternative #7) to $5.9 billion (for
Alternatives #1 and #2) relative to the Augural standards at a three-
percent discount rate and by between $1.6 billion and $3.5 billion at a
seven-percent discount rate. The estimated decreases in costs
associated with passenger cars are offset partially by estimated
increases in costs associated with light trucks, ranging from $0.1
billion (for Alternative #5) to $1.4 billion (for Alternatives #1 and
#2) relative to the Augural standards at a three-percent discount rate
and by between $0.1 billion and $0.8 billion at a seven-percent
discount rate.
In this analysis, the profile of mass reduction across vehicle
models leads to a small, but beneficial effect on fatalities as fuel
economy standards are tightened. Table VI-232 through Table VI-237
present average annual estimated safety effects of vehicle mass
changes, for CYs 2035-2045:
BILLING CODE 4910-59-P
[[Page 24785]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.445
[[Page 24786]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.446
[[Page 24787]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.447
[[Page 24788]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.448
[[Page 24789]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.449
[[Page 24790]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.450
BILLING CODE 4910-59-C
For all light-duty vehicles, mass changes are estimated to lead to
an average annual decrease in fatalities in all alternatives evaluated
for CYs 2035-
[[Page 24791]]
2045. The effects of mass changes on fatalities range from a combined
decrease (relative to the augural standards) of 17 fatality per year
for Alternative #7 to a combined decrease of 34 fatalities per year for
Alternative #4. The difference in the results by alternative depends
upon how much weight reduction is used in that alternative and the
types and sizes of vehicles to which the weight reduction applies. The
decreases in fatalities are generally driven by impacts within
passenger cars (decreases of between 19 and 44 fatalities per year
relative to the augural standards) and are offset by impacts within
light trucks (increases of between 2 and 11 fatalities per year).
Changes in vehicle mass are estimated to decrease average annual
social safety costs in CY 2035-2045 by between $0.2 billion (for
Alternative #7) and $0.5 billion (for Alternative #4) at a three-
percent discount rate relative to the augural standards (decrease of
between $0.1 and $0.2 billion at a seven-percent discount rate).
Average annual social safety costs associated with passenger cars in CY
2035-2045 are estimated to decrease by between $0.3 billion and $0.6
billion at a three-percent discount rate (decrease of between $0.1
billion and $0.3 billion at a seven-percent discount rate), but this
effect is partially offset by a corresponding increase in costs
associated with light trucks (increase of $0.1 billion or less across
alternatives at three-percent and seven-percent discount rates).
To help illuminate effects at the model year level, Table VI-238
presents the lifetime fatality impacts associated with vehicle mass
changes for passenger cars, light trucks, and all light-duty vehicles
by model year under the preferred alternative, relative to the Augural
standards for the CAFE Program. Table VI-239 presents an analogous
table for the CO2 Program.
[[Page 24792]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.451
Under the preferred alternative, passenger car fatalities
associated with mass changes are estimated to decrease relative to the
augural standards steadily from MYs 2018-19 (decrease of 4 fatalities)
through MYs 2028-29
[[Page 24793]]
(decrease of 46 fatalities). Conversely, light truck fatalities
associated with mass changes under the preferred alternative are
estimated to increase relative to the augural standards from MY 2019
(increase of 1 fatality) through MY 2029 (increase of 8 fatalities).
Table VI-240 and Table VI-241 present estimates of monetized
lifetime social safety costs associated with mass changes by model year
at three-percent and seven-percent discount rates, respectively for the
CAFE Program. Table VI-242 and Table VI-243 show comparable tables from
the perspective of the CO2 Program.
[[Page 24794]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.452
[[Page 24795]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.453
BILLING CODE 4910-59-C
Lifetime social safety costs associated with mass change in
passenger cars are estimated to decrease by between $0.1 billion (for
MYs 2020-23) and $0.3
[[Page 24796]]
billion (for MYs 2026-29) at a three-percent discount rate. At a seven-
percent discount rate, lifetime social safety costs associated with
mass change in passenger cars are estimated to decrease by between $0.1
billion and $0.2 billion from MY 2022 through MY 2029. Lifetime social
safety costs associated with mass change in light trucks are estimated
to increase by less than $0.1 billion for all MYs at three-percent and
seven-percent discount rates.
b) Impact of Vehicle Prices on Fatalities
The sales and scrappage responses discussed above have important
safety consequences and influence safety outcomes through the same
basic mechanism, fleet turnover. In the case of the scrappage response,
delaying fleet turnover keeps drivers in older vehicles which are less
safe than newer vehicles.\2022\ Similarly, the sales response slows the
rate at which newer vehicles, and their associated safety improvements,
enter the on-road population. The sales response also influences the
mix of vehicles on the road--with more stringent CAFE standards leading
to a higher share of light trucks sold in the new vehicle market,
assuming all else is equal. Light trucks have higher rates of fatal
crashes when interacting with passenger cars and, as earlier sections
discussed, different directional responses to mass reduction technology
based on the existing mass and body style of the vehicle.\2023\
---------------------------------------------------------------------------
\2022\ See Passenger Vehicle Occupant Injury Severity by Vehicle
Age and Model Year in Fatal Crashes, Traffic Safety Facts Research
Note, DOT-HS-812-528, National Highway Traffic Safety
Administration, April, 2018, and The Relationship Between Passenger
Vehicle Occupant Injury Outcomes and Vehicle Age or Model Year in
Police-Reported Crashes, Traffic Safety Facts Research Note, DOT-HS-
(812-937), National Highway Traffic Safety Administration, March,
2020.
\2023\ See Section 6. Analytical Approach as Applied to
Regulatory Alternatives] for a full explanation of the sales and
scrappage effects and how they are modeled.
---------------------------------------------------------------------------
With an integrated fleet model now part of the analytical framework
for CAFE analysis, any effects on fleet turnover (either from delayed
vehicle retirement or deferred sales of new vehicles) will affect the
distribution of both ages and model years present in the on-road fleet.
Because each of these vintages carries with it inherent rates of fatal
crashes, and newer vintages are generally safer than older ones,
changing that distribution will change the total number of on-road
fatalities under each regulatory alternative. Similarly, the dynamic
fleet share model captures the changes in the fleet's composition of
cars and trucks. As cars and trucks have different fatality rates,
differences in fleet composition across the alternatives will affect
fatalities.
At the highest level, the agencies calculate the impact of the
sales and scrappage effects by multiplying the VMT of a vehicle by the
fatality risk of that vehicle. For this analysis, calculating VMT is
rather simple: the agencies use the distribution of miles calculated in
Section VI.D.1.b)(5)(b). The trickier aspect of the analysis is
creating fatality rate coefficients. The fatality risk measures the
likelihood that a vehicle will be involved in fatal accident per mile
driven. As explained below, the agencies' methodology changed from the
proposal to this final rule in response to comments, but the basic
analytical framework remains the same. The agencies calculate the
fatality risk of a vehicle based on the vehicle's model year, age, and
style, while controlling for factors which are independent of the
intrinsic nature of the vehicle, such as behavioral characteristics.
(1) How the Agencies Modeled Impacts of Vehicle Scrappage and Sales on
Fatalities in the NPRM
In the proposal, the sales-scrappage safety model comprised two
components.\2024\ First, the agencies estimated an empirical
relationship among vehicle age, model year or vintage, and fatalities
using the FARS database of fatal crashes, vehicle registration data
from Polk to represent the on-road vehicle population, and the mileage
accumulation schedules discussed in Section VI.D.1.b)(5) Vehicles Miles
Traveled to estimate total vehicle use.\2025\ These data were used to
construct per-mile fatality rates that varied by vehicle vintage, and
also accounted for the influence of vehicle age. To accomplish this,
the agencies used FARS data at a lower level of resolution; rather than
looking at each crash and the specific factors that contributed to its
occurrence, the agencies looked at the total number of fatal crashes
involving light-duty vehicles over time with a focus on the influence
of vehicle age and vehicle vintage. The model used in the proposal
incorporated a weighted quartic polynomial regression (with each
observation weighted by the number of registered vehicles it
represented) on vehicle age, and included fixed effects for each model
year present in the dataset. The model reproduced the observed
fatalities of a given model year, at each age, reasonably well with
more recent model years estimated with smaller errors. These estimates
were used to account for the inherent safety risks of the legacy fleet
and the influence of age on a vehicle's fatality rate.
---------------------------------------------------------------------------
\2024\ The derivation of the NPRM analysis is discussed in
detail in Section 7 of the FRIA.
\2025\ The analysis supporting the CAFE rule for MYs 2017 and
beyond did not account for differences in exposure or inherent
safety risk as vehicles aged throughout their useful lives. However,
the relationship between vehicle age and fatality risk is an
important one. In a 2013 Research Note, NHTSA's National Center for
Statistics and Analysis (NCSA) concluded a driver of a vehicle that
is 4-7 years old is 10% more likely to be killed in a crash than the
driver of a vehicle 0-3 years old, accounting for the other factors
related to the crash. This trend continued for older vehicles more
generally, with a driver of a vehicle 18 years or older being 71%
more likely to be killed in a crash than a driver in a new vehicle.
``How Vehicle Age and Model Year Relate to Driver Injury Severity in
Fatal Crashes,'' DOT HS 811 825, NHTSA NCSA, August 2013. While
there are more registered vehicles that are 0-3 years old than there
are 20 years or older (nearly three times as many) because most of
the vehicles in earlier vintages are retired sooner, the average age
of vehicles in the United States is 11.6 years old and has risen
significantly in the past decade.
---------------------------------------------------------------------------
In the proposal, the agencies noted that factors other than the
advent of new safety technologies have affected the historical trend in
fatality and injury rates and are likely to continue to do so in the
future. These include changes in driver behavior, including seat belt
use, driving under the influence of alcohol or drugs, and driver
distraction, particularly from the use of hand-held electronic devices
such as smartphones, all of which affect either the frequency with
which drivers are involved in crashes or the severity of accidents.
They also include changes in the demographic composition of driving,
since drivers of different ages, gender, income levels, and educational
attainment have differing accident-involvement rates, as well as in the
geographic distribution of motor vehicle travel, since road and driving
conditions (visibility, etc.) tend to be poorer in rural areas than in
urban locations, thus leading to more frequent and more severe crashes.
Other factors affecting safety trends include infrastructure
investments and road maintenance practices that improve road design and
travel conditions, thus reducing the frequency and severity of crashes,
improvements in accident response and emergency medical care, and
cyclical variation in economic activity, which affects the demographic
composition of drivers on the road.
Seat belts have historically been the single most effective safety
technology, preventing roughly half of all fatalities in the event of a
potentially fatal crash, and accounting for over half the lives
cumulatively saved by all FMVSS-related safety technologies since
1960.\2026\ While belts have been in passenger vehicles since the
1960s, few
[[Page 24797]]
drivers or passengers initially used them. Over the past 3 decades,
seat belt usage rates have steadily climbed from under 60 percent in
the early 1990s to roughly 90 percent in 2018 and has been the single
most significant factor in reducing fatality rates over time.
Additional changes in seat belt use are possible but challenging to
achieve, since the last drivers to buckle up are typically the most
likely to be risk takers and are often the most resistant to changing
their habits. Moreover, with usage rates already at 90 percent, there
is less potential for continued improvement.
---------------------------------------------------------------------------
\2026\ Kahane, C.J., Lives Saved by Vehicle Safety Technologies
and Associated Federal Motor Vehicle Safety Standards, 1960 to
2012--Passenger Cars and LTVs, National Highway Traffic Safety
Administration, Paper Number 15-0291. https://www-esv.nhtsa.dot.gov/Proceedings/24/files/24ESV-000291.PDF.
---------------------------------------------------------------------------
Overall, the agencies believe improvement in seat belt use is
unlikely to have the impact going forward that it has in the past.
Technological fixes are possible for seat belt use and impaired
driving, but would likely require the promulgation of new regulation,
and therefore cannot be assumed. Similarly, individual States could
take steps to address impaired driving, speeding, driver distraction,
seat belt use and roadway infrastructure improvements, but the pace and
impact of such improvements is speculative. The agencies also note that
improvements in roadway infrastructure and human factors such as belt
and alcohol use potentially affect both old and new vehicles alike. If
improvements in these non-vehicle factors are equally spread across
vehicles of all MY age groups, the differences in their fatality rates
would not change. In other words, these types of improvements might
shift the entire MY fatality rate curve down rather than change its
slope.
Nonetheless, the agencies stated that it was reasonable to expect
some continuation in the generalized trend from non-vehicle technology
factors such as these. In the analysis supporting the NPRM, our
statistical model controlled for non-vehicle safety factors by
accounting for the well-documented fact that older vehicles tend to be
owned and driven by drivers whose demographic characteristics,
behavior, and geographic location tends are associated with more
frequent or severe crashes.
Second, the agencies created estimates of future fatality rates.
The agencies noted that predicting future safety trends has an inherent
degree of uncertainty, which was amplified due to the dearth of
academic and empirical research available at the time of the proposal.
Although the agencies expected further safety improvements because of
advanced driver assistance systems, such as automatic braking and
eventually fully automated vehicles, the pace of development and extent
of consumer acceptance of these improvements was uncertain. Thus,
instead of attempting to model the impact of future safety features
directly, the agencies relied on two different trend models to predict
future safety trends. The first model relied on the results from a
previous NCSA study that measured the effect of known safety
regulations on fatality rates by performing statistical evaluations of
the effectiveness of motor vehicle safety technologies based on real
world performance in the on-road vehicle fleet to determine the
effectiveness of each safety technology.\2027\ The agencies used this
information to forecast future fatality rates. The second model
employed was simpler. The agencies used actual, aggregate fatality
rates measured from 2000 through 2016 and modeled the fatality rate
trend based on these historical data.
---------------------------------------------------------------------------
\2027\ Blincoe, L. and Shankar, U., ``The Impact of Safety
Standards and Behavioral Trends on Motor Vehicle Fatality Rates,''
National Highway Traffic Safety Administration, DOT HS 810 777,
Washington, DC, January, 2007.
---------------------------------------------------------------------------
The agencies noted that both models had significant limitations and
predicted significantly different safety trends. The NCSA study focused
on projections to reflect known technology adaptation requirements, but
it was conducted prior to the 2008 recession, which disrupted the
economy and changed travel patterns throughout the country, and
predated the emergence of newer technologies in the 2010s. The NCSA
anticipated continued improvement well beyond 2020. By contrast, the
historical fatality rate model reflected shifts in safety not captured
by the NCSA model, but gave arguably implausible results after 2020
because of an observed upward shift in fatalities between 2014 and
2015. It essentially represented a scenario in which economic, market,
or behavioral factors minimize or offset much of the potential impact
of future safety technology. To reconcile the two projections of safety
improvements beyond 2015, the agencies averaged the NCSA and historical
fatality rate models, accepting each as an illustration of different
and conflicting possible future scenarios.
The agencies received a number of comments on the provisional model
used in the NPRM, which focused mainly on its omission of variables
that change over time and can affect the safety of all vehicles in use,
regardless of their original model year or current age. As indicated
previously, these include changes in seat belt use, driving under the
influence of alcohol or drugs, use of hand-held electronic devices,
driver demographics, the geographic distribution of vehicle use, road
design and maintenance, emergency response and medical care, and
overall economic activity.
For example, CARB asserted that the NPRM modeling overestimated
fatality rates for older vehicles because it did not ``control for
factors that can have a significant influence on fatality risk, such as
crash circumstances and driver characteristics.'' Elsewhere, CARB
highlighted the omission of calendar year effects from the NPRM
analysis, adding ``the agencies only model fatality rate as a function
of model year, but fatality rate should be a function of both model
year and calendar year [. . .] [which] would account for systematic
safety improvements to the entire on-road fleet.'' \2028\ CARB also
argued that analysis should account for safety differences between body
styles, noting that passenger cars and other LTVs ``have historically
had different safety regulations.'' \2029\ Passenger cars and LTVs are
not always regulated at exactly the same pace and in some
circumstances, LTV regulations have differed from passenger car
regulations. However, with a few exceptions, both types of passenger
vehicles are equipped with safety technologies that address the same
basic safety hazards. Historically, these involve regulations that
preserve passenger compartment integrity and protect passengers in the
event of a crash. These include technologies such as air bags, seat
belts, stronger roof structures, side door beams, and fuel tank
integrity. Further, going forward, the agencies expect that both
vehicle types will eventually all be equipped with the same advanced
crash avoidance safety technologies that are currently being developed.
Whatever differences there are have influenced the fatality rates and
since this rulemaking uses combined average fatality rates (for PCs and
LTVs) for the model, the results should closely mirror the results from
an analysis that calculates the two vehicle types separately and then
adds them together.
---------------------------------------------------------------------------
\2028\ CARB, Detailed Comments, NHTSA-2018-0067-11873 at 263.
\2029\ CARB, Auken Fatality Report, NHTSA-2018-0067-11881, at
25.
---------------------------------------------------------------------------
Similarly, States and Cities noted the potential importance of
factors that can affect trends in vehicle safety over time, pointing
out that ``increased seat belt use over time, improvements in roadway
design and life-saving emergency response and treatment, and crash
compatibility with other vehicles improve the overall safety of
vehicles currently on the road'' and therefore
[[Page 24798]]
concluded that ``the CAFE model's assumption that the fatality rate of
a 1985 model year vehicle is 23.8 per billion vehicle miles traveled
for any calendar year is incorrect. That error increases the risk of
fatalities determined by the NPRM for scrappage by around 25 percent.''
\2030\ Consumers Union echoed this argument and suggested driver
characteristics and behavior may ``more strongly influence fatality
risk than a vehicle's model year.'' \2031\
---------------------------------------------------------------------------
\2030\ States and Cities, Detailed Comments, NHTSA-2018-0067-
11735, at 101 (internal citation omitted).
\2031\ Consumers Union, et al., NHTSA-2018-0067-11731,
Attachment 11, at 14.
---------------------------------------------------------------------------
IPI speculated that omitting the effect of variables that change
over time in ways that could affect fleet-wide safety may have caused
the agencies' analysis to over-emphasize the role of safety
improvements to new vehicles. Specifically, IPI observed that ``the
agencies could not adequately control for driver behavior trends. And a
decrease in fatalities could look like it was caused by vehicle
improvements over time rather than societal changes.'' \2032\
---------------------------------------------------------------------------
\2032\ IPI, NHTSA-2018-0067-12213, at 71.
---------------------------------------------------------------------------
The agencies also received a few comments on their modeling
choices. For example, CARB commented that the agencies equation for the
legacy fleet was ``either incorrect or [had] limited domain-of-validity
because it can potentially predict negative fatality rates'' and
because it was missing an intercept term.\2033\ CARB suggested a
logarithmic function would fix the problem. The agencies note that the
polynomial specification of the safety model the agencies developed for
the legacy fleet was extremely unlikely to predict negative fatality
rates in light of the estimated values of its coefficients, and that
its fixed-effects specification in effect included separate intercept
terms for each model year, with that for the earliest model year
serving as the ``reference case'' and thus performing the normal role
of the constant term.
---------------------------------------------------------------------------
\2033\ CARB, Auken Fatality Report, NHTSA-2018-0067-11881, at
25.
---------------------------------------------------------------------------
In electing to offset rebound-related safety consequences for the
NPRM, the agencies distinguished the rebound effect from mass and fleet
turnover impacts by describing the former as a voluntary consumer
choice and the latter as imposed by the standards on consumers.\2034\
The agencies acknowledged in the NPRM that a reasonable argument might
be made that consumers' decisions to purchase newer and safer cars or
light trucks and to keep older models in service are also voluntary
consumer choices, in which case changes in their decisions in response
to newly-adopted CAFE and CO2 standards might be accompanied
by offsetting gains or losses in benefits. The agencies dismissed this
argument in the NPRM by noting that new vehicle prices act as a barrier
to entry for some consumers, hence--at least ``marginal'' shoppers--
purchasing a more expensive vehicle is not a choice; and, without the
ability to determine how many potential purchasers are `priced out' of
the new vehicle market, it would be inappropriate to offset sales and
scrappage safety impacts.\2035\ The agencies sought comment on this
assumption.
---------------------------------------------------------------------------
\2034\ See 83 FR at 43107.
\2035\ The agencies further augmented the discussion by
explaining that less stringent standards encouraged new vehicle
purchases through lower vehicle prices while simultaneously
discouraging additional driving due to higher operating costs. See
id.
---------------------------------------------------------------------------
The agencies did not receive any suggestions for distinguishing
between consumers who voluntarily delayed purchases and those who were
forced to delay a purchase due to high vehicle prices. Thus, the
problem of deciphering the motives behind delayed purchases still
lingers. However, the agencies did receive several comments advocating
that the agencies offset fatalities attributable to sales and scrappage
as they do for the rebound effect. For example, NCAT commented that
``consumer purchases are voluntary and this effect should not be
attributed to the standards.'' \2036\ The environmental group coalition
commented that miles driven in older vehicles are ``a consumer choice,
not something the standards compel.'' \2037\ In comparing the decision
to retain and drive older vehicles to the decision to drive new
vehicles more, i.e. the rebound effect, EDF concluded, ``to treat these
identical choices in 180 degree different manners is of course
manifestly arbitrary.'' \2038\
---------------------------------------------------------------------------
\2036\ NCAT, Comments, NHTSA-2018-0067-11969, at 32-33.
\2037\ Environmental Group Coalition, Appendix A, NHTSA-2018-
0067-12000, at 40-41.
\2038\ EDF, Appendix B, NHTSA-2018-0067-12108, at 58.
---------------------------------------------------------------------------
On a rudimentary level, the agencies agree with commenters that
purchasing decisions are a consumer choice. While reducing the
stringency of the standards should make new vehicles more affordable,
nothing in today's rule requires consumers to purchase a new vehicle;
likewise, the analysis does not assume every older vehicle will be
replaced immediately. There is no strict requirement that the agencies
must offset consumer choices. In fact, such a viewpoint would be
untenable. Nothing in today's rule compels private parties to do
anything. If the agencies assumed all freely chosen or voluntary
actions, such as driving or manufacturing automobiles, were not
attributable to the rule, then each regulatory scenario would have the
same net benefit--zero. As such, the agencies explanation in the
proposal of freely chosen and voluntary was likely imprecise and led
commenters to an overly broad conclusion. Deciding which behavioral
responses are unambiguously attributable to a regulation and should
thus be quantified, and distinguishing them from responses that would
be anticipated to occur in its absence is inherently part of the
rulemaking process, and inevitably requires agencies considering new
regulations to apply careful judgment in making those distinctions.
To that end, the agencies felt it was appropriate to offset
rebound-related safety costs because of the benefit rebound miles
confer to society. As described in more detail in Section 1.b)(6),
additional driving that occurs as a consequence of the fuel economy
rebound effect is undertaken voluntarily, and the agencies can infer
from the fact that it is freely chosen that the mobility benefits it
provides necessarily exceed the additional operating costs and
increased exposure to safety risks it entails. Since reducing the
standards has the ancillary effect of reducing rebound miles, the
agencies concluded that including safety costs associated with rebound
driving would cause the agencies to underestimate the lost value of
rebound driving; therefore, it was appropriate to offset rebound safety
costs to account for the lost benefits.\2039\ Thus, the significance of
the terms freely chosen and voluntary was to signal that consumers'
actions were motivated in part by benefits that may not have been not
explicitly identified or accounted for, rather than to act as a
prohibitive characteristic.
---------------------------------------------------------------------------
\2039\ Arguably rebound fatalities and non-fatal injuries should
be included in today's analysis as a cost without an offset. While a
perfectly rational driver would fully and accurately internalize the
costs associated with driving on a per-mile basis and would only
drive if the expected benefits at least offset the expected costs,
it is difficult to ascertain how much of the risk a real person
internalizes. If not for the reduced standards, fatalities would
increase due to rebound driving.
---------------------------------------------------------------------------
When considering commenters' suggestion to offset fleet turnover
fatalities (as well as injury and ancillary costs), the agencies
attempted to identify specific benefits whose loss would be logically
attributable to the changes in standards this rule adopts, and were not
accounted for elsewhere in
[[Page 24799]]
their analysis. The agencies considered whether accelerated turnover of
the car and light truck fleet could cause mobility losses analogous to
those resulting from the rebound effect, but determined that on
balance, increasing the pace at which new vehicles replace older models
that are retired from use provides additional mobility and other
benefits.\2040\ In addition, the agencies considered whether consumers
experience some previously unidentified loss in welfare when they
purchase new vehicles, particularly when they do so to replace an older
model. As explained in in Section 1.b)(6) and 1.b)(8), the agencies
instead concluded that purchasers instead experience gains in welfare
as a result, but that the resulting benefits are already accounted for
elsewhere in their analysis.
---------------------------------------------------------------------------
\2040\ This occurs because newer vehicles are not only more
fuel-efficient on average than the older models they replace, but
also provide more reliable, comfortable, and otherwise higher-
quality transportation service, so they tend to be driven more than
those they replace.
---------------------------------------------------------------------------
Finally, the agencies contemplated whether--as commenters
contended-- owners of older vehicles derive some heretofore
unaccounted-for benefit from continuing to use them, which might be
reduced when the rule encourages more rapid retirement of older models.
Applying the same logic used to explain additional driving in response
to the rebound effect, an older vehicle will continue to be maintained
in working condition and driven when the benefits provided to the owner
is sufficient to offset the costs of maintenance and operation,
including the economic costs associated with additional exposure to
safety risks. Therefore, there is a benefit to driving an older
vehicle. But the relevant question is not whether a benefit exists but
how this rule might affect those benefits. With the very limited
exception of classic cars, it is unlikely that the benefit of driving
an older vehicle confers a greater benefit than driving a newer
vehicle.\2041\ Normally, when a vehicle is scrapped, it is replaced
with a newer vehicle. Hence mobility is not lost, but rather
transferred between vehicles--and with it, the associated
benefits.\2042\ In the limited instances where a retired vehicle is not
replaced with a newer vehicle, that action is freely taken and the
agencies can infer from that decision that the benefit derived from
scrapping the vehicle outweighed any possible loss, including lost
mobility. Offsetting the reduction in scrappage safety costs--realized
because of the standards--without a complementary benefit would be
directionally inconsistent.\2043\
---------------------------------------------------------------------------
\2041\ If the benefit of driving an older vehicle was higher
than the benefit of driving a newer vehicle, we would anticipate
consumers to forgo replacing older vehicles with newer vehicles.
\2042\ Since driving newer vehicles, including newer used
vehicles, likely confers greater benefits than would-be scrapped
vehicle, the agencies are likely underestimating the value of
increased scrappage.
\2043\ A similar argument could be made that consumers
`internalize' additional fuel costs, and therefore pre-tax fuel
savings should also be offset. However, this would also ignore that
benefits are remaining constant while the costs to obtain those
benefits is increasing.
---------------------------------------------------------------------------
The agencies reaffirm that off-setting safety costs attributable to
the sales and scrappage effects is inappropriate. Commenters' arguments
relied exclusively on the premise that driving older vehicles is freely
chosen and thus must have associated benefits, without considering the
impact of accelerating their retirement on the rule's overall net
safety and mobility benefits. Furthermore, the agencies remain
concerned that potential buyers may be ``frozen out'' of the new
vehicle market by prohibitively high prices; in which case enabling
access to newer, safer vehicles provides measurable safety benefits
that should be considered by the analysis.
However, in an abundance of caution, the agencies performed a
sensitivity analysis that applies the same safety offset to sales/
scrappage safety impacts that was applied to the rebound effect safety
impacts. The results are provided in Table VI-244 below. As might be
expected, this adjustment reduces net benefits in all scenarios, but
does not substantially shift the relative scope among alternatives.
BILLING CODE 4910-59-P
[[Page 24800]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.454
BILLING CODE 4910-59-C
Again, the agencies feel that this offset is inappropriate. The
sensitivity case disregards many of the tangible gains in safety
expected from increased sales and
[[Page 24801]]
scrappage. Furthermore, the agencies note that--even if they replaced
the central analysis' assumptions with this sensitivity case--the
anticipated changes in net benefits would not be enough to change their
decision.
(2) Revised Sales-Scrappage Safety Model
In response to the comments, the agencies have taken several steps
to revise the sales-scrappage safety model. First, the agencies
developed a revised statistical model to explain historical
improvements in the lifetime safety performance of each successive new
vintage of cars and light trucks, and used the results of this improved
model to project the future trend in the overall fatality rates. While
the revised historical trend model itself is more complex than the one
utilized in the proposal, the overall procedure is simpler; the
agencies have collapsed the two piecemeal components discussed above
into one model and eliminated the need to `reconcile' differences
between competing future projections. Next, the agencies applied
detailed empirical estimates of the market uptake and improving
effectiveness of crash avoidance technologies to estimate their effect
on the fleet-wide fatality rate, including explicitly incorporating
both the direct effect of those technologies on the crash involvement
rates of new vehicles equipped with them, as well as the ``spillover''
effect of those technologies on improving the safety of occupants of
vehicles that are not equipped with these technologies.
(a) Crash Avoidance
In the NPRM, the agencies took a very generalized approach to
estimating the pace of future safety trends. For reasons discussed
above, the agencies noted that there was uncertainty regarding actual
trends in fatality rates. This issue was addressed by numerous
commenters who took opposing positions. Among them, IPI stated that
``[t]he agencies have not provided an adequate explanation for why past
safety trends are likely to continue until the mid-2020s.'' IPI further
noted that ``crash avoidance technology may not be adopted as easily or
readily as crash mitigation technologies have been.'' \2044\ In
response, the agencies note that the trend the agencies adopted for the
NPRM was not a direct continuation of past trends. Rather, it was a
simple average of several possible models the agencies had examined,
accepting each as an illustration of different and conflicting possible
future scenarios.
---------------------------------------------------------------------------
\2044\ IPI, Appendix, NHTSA-2018-0067-12213, at 98.
---------------------------------------------------------------------------
By contrast, States and Cities asserted that fatality rates may be
lower in the future than the agencies estimated, noting that the NPRM
analysis did not ``account for safety benefits that new safety
technologies in future vehicles will have on the agencies predicted
outcome.'' \2045\ While the agencies agree that the NPRM analysis did
not analyze individual safety benefits of new technologies, the trends
included in the NPRM were intended, in part, as a proxy estimate of the
impact of these technologies. As discussed in the NPRM, these
technologies were cited as a justification for assuming a continued
downward trend in the fatality rate through roughly 2035.
---------------------------------------------------------------------------
\2045\ States and Cities, Detailed Comments, NHTSA-2018-0067-
11735, at 80.
---------------------------------------------------------------------------
Nonetheless, the agencies believe that further analysis of these
potential trends can now be ascertained for several explicit
technologies. In response to comments suggesting that the agencies
account more directly for new safety technologies, the agencies
augmented the sales-scrappage safety analysis for the final rule with
recent research into the effectiveness of specific advanced crash
avoidance safety technologies (also known as ADAS or advanced driver
assistance systems) that are expected to drive future safety
improvement to estimate the impacts of crash avoidance technologies.
The analysis analyzes six crash avoidance technologies that are
currently being produced and commercially deployed in the new vehicle
fleet. These include Frontal Collision Warning (FCW), Automatic
Emergency Braking (AEB), Lane Departure Warning (LDW), Lane Keep Assist
(LKA), Blind Spot Detection (BSD), and Lane Change Alert (LCA).\2046\
These are the principal technologies that are being developed and
adopted in new vehicle fleets and will likely drive vehicle-based
safety improvements for the coming decade. These technologies are being
installed in more and more new vehicles; in fact, 12 manufacturers
recently reported that they voluntarily installed AEB systems in more
than 75 percent of their new vehicles sold in the year ending August
31, 2019.\2047\ The agencies note that the terminology and the detailed
characteristics of these systems may differ across manufacturers, but
the basic system functions are common across all.
---------------------------------------------------------------------------
\2046\ A full description of these technologies and several
other technologies referenced below may be found in the
corresponding FRIA safety impacts discussion.
\2047\ NHTSA Announces Update to Historic AEB Commitment by 20
Automakers, NHTSA press release December 17, 2019. https://www.nhtsa.gov/press-releases/nhtsa-announces-update-historic-aeb-commitment-20-automakers.
---------------------------------------------------------------------------
These six technologies address three basic crash scenarios through
warnings to the driver or alternately, through dynamic vehicle control:
1. Forward collisions, typically involving a crash into the rear of
a stopped vehicle;
2. Lane departure crashes, typically involving inadvertent drifting
across or into another traffic lane; and
3. Blind spot crashes, typically involving intentional lane changes
into unseen vehicles driving in or approaching the driver's blind spot.
Unlike traditional safety features where the bulk of the safety
improvements were attributable to improved protection when a crash
occurs (crash worthiness), the impact of advanced crash avoidance
technologies (ADAS or advanced driver assistance systems) will have on
fatality and injury rates is a direct function of their effectiveness
in preventing or reducing the severity of the crashes they are designed
to mitigate. This effectiveness is typically measured using real world
data comparing vehicles with these technologies to similar vehicles
without them. While these technologies are actively being deployed in
new vehicles, their penetration in the larger on-road vehicle fleet has
been at a low, but growing level. This limits the precision of
statistical regression analyses, at least until the technologies become
more common in the on-road fleet.
Our approach in the final rule is to derive effectiveness rates for
these advanced crash-avoidance technologies from safety technology
literature. The agencies then apply these effectiveness rates to
specific crash target populations for which the crash avoidance
technology is designed to mitigate and adjusted to reflect the current
pace of adoption of the technology, including the public commitment by
manufactures to install these technologies. The products of these
factors, combined across all 6 advanced technologies, produce a
fatality rate reduction percentage that is applied to the fatality rate
trend model discussed below, which projects both vehicle and non-
vehicle safety trends. The combined model produces a projection of
impacts of changes in vehicle safety technology as well as behavioral
and infrastructural trends.
[[Page 24802]]
(i) Technology Effectiveness Rates
(a) Forward Crash Collision Technologies
For forward collisions, manufacturers are currently equipping
vehicles with FCW, which warns drivers of impending collisions, as well
as AEB, which incorporates the sensor systems from FCW together with
dynamic brake support (DBS) and crash imminent braking (CIB) to help
avoid crashes or mitigate their severity. Manufacturers have committed
voluntarily to install some form of AEB on all light vehicles by the
2023 model year (September 2022).\2048\
---------------------------------------------------------------------------
\2048\ See https://www.nhtsa.gov/press-releases/nhtsa-iihs-announcement-aeb.
---------------------------------------------------------------------------
Table VI-245 summarizes studies which have measured effectiveness
for various forms of FCW and AEB over the past 13 years. Most studies
focused on crash reduction rather than injury reduction. This is a
function of limited injury data in the on-road fleet, especially during
the early years of deployment of these technologies. In addition, it
reflects engineering limitations in the technologies themselves.
Initial designs of AEB systems were basically incapable of detecting
stationary objects at speeds higher than 30 mph, making them
potentially ineffective in higher speed crashes that are more likely to
result in fatalities or serious injury. For example, Wiacek et al. (2-
15) conducted a review of rear-end crashes involving a fatal occupant
in the 2003-2012 NASS-CDS data-bases to determine the factors that
contribute to fatal rear-end crashes.\2049\ They found that the speed
of the striking vehicle was the primary factor in 71 percent of the
cases they examined. The average Delta-V of the striking vehicle in
these cases was 46 km/h (28.5 mph), implying pre-crash travel speeds in
excess of this speed. While Table VI-245 includes studies going back to
2005, the agencies focus our discussion on more recent studies
conducted after 2012 in order to reflect more current safety systems
and vehicle designs.2050 2051 2052 2053 2054 2055 2056 2057
---------------------------------------------------------------------------
\2049\ Wiacek, C., Bean, J., Sharma, D., Real World Analysis of
Fatal Rear-End Crashes, National Highway Traffic Safety
Administration, 24th Enhanced Safety of Vehicles Conference, 150270,
2015.
\2050\ Sugimoto, Y., and Sauer, C., (2005). Effectiveness
Estimation Method for Advanced Driver Assistance System and its
Application to Collision Mitigation Brake systems, paper number 05-
148, 19th International Technical Conference on the Enhanced safety
of Vehicles (ESV), Washington DC, June 6-9, 2005.
\2051\ Page, Y., Foret-Bruno, J., & Cuny, S. (2005). Are
expected and observed effectiveness of emergency brake assist in
preventing road injury accidents consistent?, 19th ESV Conference,
Washington DC.
\2052\ Najm, W.G., Stearns, M.D., Howarth, H., Koopman, J. &
Hitz, J., (2006). Evaluation of an Automotive Rear-End Collision
Avoidance System (technical report DOT HS 810 569), Cambridge, MA:
John A. Volpe National Transportation System Center, U.S. Department
of Transportation.
\2053\ Breuer, JJ., Faulhaber, A., Frank, P. and Gleissner, S.
(2007). Real world Safety Benefits of Brake Assistance Systems,
Proceedings of the 20th International Technical Conference of the
Enhanced Safety of Vehicles (ESV) in Lyon, France June 18-21, 2007.
\2054\ Keuhn, M., Hummel, T., and Bende J., Benefit estimation
of advanced driver assistance systems for cars derived from real-
world accidents, Paper No. 09-0317, 21st International Technical
Conference on the Enhanced Safety of Vehicles (ESV)--International
Congress Centre, Stuttgart, Germany, June 15-18, 2009.
\2055\ Grover, C., Knight, I., Okoro, F., Simmons I., Couper,
G., Massie, P., and Smith, B. (2008). Automated Emergency Brake
Systems: Technical requirements, Costs and Benefits, PPR227, TRL
Limited, DG Enterprise, European Commission, April 2008.
\2056\ Kusano, K.G., and Gabler, H.C. (2015). Comparison of
Expected Crash Injury and Injury Reduction from Production Forward
Collision and Lane Departure Warning Systems, Traffic Injury
Prevention 2015; Suppl. 2: S109-14.
\2057\ HLDI (2011). Volvo's City Safety prevents low-speed
crashes and cuts insurance costs, Status Report, Vol. 46, No. 6,
July 19,2011.
[GRAPHIC] [TIFF OMITTED] TR30AP20.455
[[Page 24803]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.456
Doecke et al. (2012) created
2058 2059 2060 2061 2062 2063 simulations of 103 real world
crashes and applied AEB system models with differing specifications to
determine the change in impact speed that various AEB interventions
might produce. Their modeling found significant rear-end crash speed
reductions with various AEB performance assumptions. In addition, they
estimated a 29 percent reduction in rear-end crashes and that 25
percent of crashes over 10 km/h were reduced to 10 km/h or less.
---------------------------------------------------------------------------
\2058\ Docke, S.D., Anderson, R.W.G., Mackenzie, J.R.R., Ponte,
G. (2012). The potential of autonomous emergency braking systems to
mitigate passenger vehicle crashes. Australian Road Safety Research
Policing and Education Conference, October 4-6, 2012, Wellington,
New Zealand.
\2059\ Chauvel, C., Page, Y., Files, B.N., and Lahausse, J.
(2013). Automatic emergency braking for pedestrians effective target
population and expected safety benefits, Paper No. 13-0008, 23rd
International Technical Conference on the Enhanced Safety of
Vehicles (ESV), Seoul, Republic of Korea, May 27-30, 2013.
\2060\ Fildes B., Keall M., Bos A., Lie A., Page, Y., Pastor,
C., Pennisi, L., Rizzi, M., Thomas, P., and Tingvall, C.
Effectiveness of Low Speed Autonomous Emergency Braking in Real-
World Rear-End Crashes. Accident Analysis and Prevention, AAP-D-14-
00692R2.
\2061\ Cicchino, J.B. (2017). Effectiveness of forward collision
warning and autonomous emergency braking systems in reducing front-
to-rear crash rates. Accident Analysis and Prevention, V. 99, Part
A, February 2017, Pages 142-52.
\2062\ Kusano, K.D., and Gabler H.C. (2012). Safety Benefits of
Forward Collision Warning, Brake Assist, and Autonomous Braking
Systems in Rear-End Collisions, Intelligent Transportation Systems,
IEEE Transactions, Volume 13 (4).
\2063\ Leslie, A, Kiefer, R., Meitzner, M, and Flannagan, C.
(2019). Analysis of the Field Effectiveness of General Motors
Production Active Safety and Advanced headlighting Systems.
University of Michigan Transportation Research Institute, UMTRI-
2019-6, September, 2019.
---------------------------------------------------------------------------
Cicchino (2016) analyzed the effectiveness of a variety of forward
collision mitigation systems including both FCW and AEB systems.
Cicchino used a Poisson regression to compare rates of police-reported
crashes per insured vehicle year between vehicles with these systems
and the same models that did not elect to install them. The analysis
was based on crashes occurring during 2010 to 2014 in 22 States and
controlled for other factors that affected crash risk. Cicchino found
that FCW reduced all rear-end striking crashes by 27 percent and rear-
end striking injury crashes by 20 percent, and that AEB functional at
high-speeds reduced these crashes by 50 and 56 percent, respectively.
She also found that low speed AEB without driver warning reduced all
crashes by 43 percent and injury crashes by 45 percent. She also found
that even low-speed AEB could impact crashes at higher speed limits.
Reductions were found of 53 percent, 59 percent, and 58 percent for all
rear-end striking crash rates, rear-end striking injury crash rates,
and rear-end third party injury crash rates, respectively, at speed
limits of 40-45 mph. For speed limits of 35 mph or less, reductions of
40 percent, 40 percent, and 43 percent were found. For speed limits of
50 mph or greater, reductions of 31 percent, 30 percent, and 28
percent, were found. Further, Cicchino (2016) found significant
reductions (30 percent) in rear-end injury crashes even in crashes on
roadways where speed limits exceeded 50 mph.
Kusano and Gabler (2012) examined the effectiveness of various
levels of forward collision technologies including FCW and AEB based on
simulations of 1,396 real world rear end crashes from 1993-2008 NASS
CDS data-bases. The authors developed a probability-based framework to
account for variable driver responses to the warning systems. Kusano
and Gabler found FCW systems could reduce rear-end crashes by 3.2
percent and driver injuries in rear-end crashes by 29 percent. They
also found that full AEB systems with FCW, pre-crash brake assist, and
autonomous pre-crash braking could reduce rear-end crashes by 7.7
percent and reduce moderate to fatal driver injuries in rear-end
crashes by 50 percent.
Fildes et al. (2015) performed meta-analyses to evaluate the
effectiveness of low-speed AEB technology in passenger vehicles based
on real-world crash experience across six different predominantly
European countries. Data from these countries was pooled into a
standard analysis format and induced exposure methods were used to
control for extraneous effects. The study found a 38 percent overall
reduction in rear-end crashes for vehicles with AEB compared to similar
vehicles without this technology. The study also found no statistical
evidence for any difference in effectiveness between urban roads with
speed limits less than or equal to 60 km/h, and rural roads with speed
limits greater than 60 km/h. Fildes et al. (2015) found no statistical
difference in the performance of AEBs on lower speed urban or higher
speed rural roadways.
Kusano and Gabler (2015) simulated rear-end crashes based on a
sample of 1,042 crashes in the 2012 NASS-CDS. Modelling was based on 54
model year 2010-2014 vehicles that were evaluated in NHTSA's New Car
Assessment Program (NCAP). Kusano and Gabler found FCW systems could
prevent 0-67 percent of rear-end crashes and 2-69 percent of serious to
fatal driver injuries.
Leslie et al. (2019) analyzed the relative crash performance of
123,377 General Motors (GM) MY 2013 to 2017 vehicles linked to State
police-reported crashes by Vehicle Identification numbers (VIN). GM
provided VIN-linked safety content information for these vehicles to
enable precise identification of safety technology content. The authors
analyzed the effectiveness of a variety of crash avoidance technologies
including both FCW and AEB separately. They estimated effectiveness
comparing system-relevant crashes to baseline
[[Page 24804]]
(control group) crashes using a quasi-induced exposure method in which
rear-end struck crashes are used as the control group. Leslie et al.
found that FCW reduced rear-end striking crashes of all severities by
21 percent, and that AEB (which includes FCW) reduced these crashes by
46 percent.\2064\
---------------------------------------------------------------------------
\2064\ The agencies note that UMTRI, the sponsoring organization
for the Leslie et al. study, published a previous version of this
same study utilizing the same methods in March of 2018 (Flannagan,
C. and Leslie, A, Crash Avoidance Technology Evaluation Using real-
World crashes, University of Michigan Transportation research
Institute, March 22, 2018). The agencies focused on the more recent
2019 study because its sample size is significantly larger and it
represents more recent model year vehicles. The revised (2019) study
uses the same basic techniques but incorporated a larger data-base
of system-relevant and control cases (123,377 cases in the 2019
study vs. 35,401 in the 2018 study). Relative to the Flannagan and
Leslie (2018) findings, the results of the 2019 study varied by
technology. The revised study found effectiveness rates of 21% for
FCW and 46% for AEB, compared to 16% and 45% in the 2018 study. The
revised study found effectiveness rates of 10% for LDW and 20% for
LKA, compared to 3% and 30% for these technologies in the 2018
study. The revised study found effectiveness rates of 3% for BSD and
26-37% for LCA systems, compared to 8% and 19-32% for these
technologies in the 2018 study. Thus, some system effectiveness
estimates increased while others decreased.
---------------------------------------------------------------------------
For this analysis, the agencies based their projections on Leslie
et al. because they are the most recent study, and thus reflect the
most current versions of these systems in the largest number of
vehicles, and also because they arguably have the most precise
identification of the presence of the specific technologies in the
vehicle fleet. Furthermore, Leslie et al. was the only study to report
estimates for each of the six crash avoidance technologies analyzed for
the final rule, hence providing a certain level of consistency amongst
estimates. The agencies recognize that there is uncertainty in
estimates of these technologies effectiveness, especially at this early
stage of deployment. For this reason, the agencies examine a range of
effectiveness rates to estimate boundary outcomes in a sensitivity
analysis.
Leslie et al. measured effectiveness against all categories of
crashes, but did not specify effectiveness against crashes that result
in fatalities or injuries. The agencies examined a range of
effectiveness rates against fatal crashes using a central case based on
boundary assumptions of no effectiveness and full effectiveness across
all crash types. Our central case is thus a simple average of these two
extremes. Sensitivity cases were based on the 95th percent confidence
intervals calculated from this central case. Leslie et al. found
effectiveness rates of 21 percent for FCW and 46 percent for AEB. Our
central fatality effectiveness estimates will thus be 10.5 percent for
FCW and 23 percent for AEB. The calculated 95th percentile confidence
limits range is 8.11 to 12.58 percent effective for FCW and 20.85 to
25.27 for AEB. The agencies note that our central estimate is
conservative compared to averages of those studies that did
specifically examine fatality impacts; that is, the analysis assumes
reduced future fatalities less than most of, or the average of, those
studies, and thus minimizes the estimate of lives saved under
alternatives to the augural standards. Furthermore, the agencies note
that the estimates against fatal crashes is higher in the recent
studies in Table VI-245, which reflects the agencies' understanding
that earlier iterations of AEB and FCW may have been less effective
against crashes that result in fatalties than newer and improved
versions.\2065\
---------------------------------------------------------------------------
\2065\ As an example of improvements, the agencies note that the
Mercedes system described in their 2015 owner's manual specified
that for stationary objects the system would only work in crashes
below 31 mph, but that in their manual for the 2019 model, the
systems are specified to work in these crashes up to 50 mph.
---------------------------------------------------------------------------
(b) Lane Departure Crash Technologies
For lane departure crashes, manufacturers are currently equipping
vehicles with lane departure warning (LDW), which monitors lane
markings on the road and alerts the driver when their vehicle is about
to drift beyond a delineated edge line of their current travel lane, as
well as lane keep assist (LKA), which provides gentle steering
adjustments to help drivers avoid unintentional lane crossing. Table
VI-246 summarizes studies which have measured effectiveness for LDW and
LKA.
[GRAPHIC] [TIFF OMITTED] TR30AP20.457
Cicchino (2018) examined crash involvement rates per insured
vehicle 2066 2067 2068 2069 2070 year for
[[Page 24805]]
vehicles that offered LDW as an option and compared crash rates for
those that had the option installed to those that did not. The study
focused on single-vehicle, sideswipe, and head-on crashes as the
relevant target population for LDW effectiveness rates. The study
examined 5,433 relevant crashes of all severities found in 2009-2015
police-reported data from 25 States. The study was limited to crashes
on roadways with 40 mph or greater speed limits not covered in ice or
snow since lower travel speeds would be more likely to fall outside of
the LDW systems' minimum operational threshold. Cicchino found an
overall reduction in relevant crashes of 11 percent for vehicles that
were equipped with LDW. She also found a 21 percent reduction in injury
crashes. The result for all crashes was statistically significant,
while that for injury crashes approached significance (p<0.07).
Cicchino did not separately analyze LKA systems.
---------------------------------------------------------------------------
\2066\ Cicchino, J.B. (2018). Effects of lane departure warning
on police-reported crash rates, Journal of Safety Research 66
(2018), pp.61-70. National Safety Council and Elsevier Ltd., May,
2018.
\2067\ Sternlund, S., Strandroth, J., Rizzi, M., Lie, A., and
Tingvall, C. (2017). ``The effectiveness of lane departure warning
systems--A reduction in real-world passenger car injury crashes,''
Traffic Injury Prevention V. 18 Issue 2 (Jan 2017).
\2068\ Leslie et al., supra note 2063.
\2069\ Kusano & Gable, supra note 2056.
\2070\ Kusano, K., Gorman, T.I., Sherony, R., and Gabler, H.C.
Potential occupant injury reduction in the U.S. vehicle fleet for
lane departure warning-equipped vehicles in single-vehicle crashes.
Traffic Injury Prevention 2014 Suppl 1:S157-64.
---------------------------------------------------------------------------
Sternlund et al. (2017) studied single vehicle and head-on injury
crash involvements relevant to LDW and LKA in Volvos on Swedish
roadways. They used rear-end crashes as a control and compared the
ratio of these two crash groups in vehicles that had elected to install
LDW or LCA to the ratio in vehicles that did not have this content.
Studied crashes were limited to roadways with speeds of 70-120 kph and
not covered with ice or snow. Sternlund et al. found that LDW/LKA
systems reduced single vehicle and head-on injury crashes in their
crash population by 53 percent, with a lower limit of 11 percent, which
they determined corresponded to a reduction of 30 percent (lower limit
of 6 percent) across all speed limits and road surface assumptions.
Leslie et al. (2019) analyzed the relative crash performance of
123,377 General Motors (GM) MY 2013 to 2017 vehicles linked to state
police-reported crashes by Vehicle Identification numbers (VIN). GM
provided VIN-linked safety content information for these vehicles to
enable precise identification of safety technology content. The authors
analyzed the effectiveness of a variety of crash avoidance technologies
including both LDW and LKA separately. They estimated effectiveness
comparing system-relevant crashes to baseline (control group) crashes
using a quasi-induced exposure method in which rear-end struck crashes
are used as the control group. Leslie et al. found that LDW reduced
lane departure crashes of all severities by 10 percent, and that LKA
(which includes LDW) reduced these crashes by 20 percent.
Kusano et al. (2014) developed a comprehensive crash and injury
simulation model to estimate the potential safety impacts of LDW. The
model simulated results from 481 single-vehicle collisions documented
in the NASS-CDS data-base for the year 2012. Each crash was simulated
as it actually occurred and again as it would occur had the vehicles
been equipped with LDW. Crashes were simulated multiple times to
account for variation in driver reaction, roadway, and vehicle
conditions. Kusano et al. found that LDW could reduce all roadway
departure crashes caused by the driver drifting from his or her lane by
28.9 percent, resulting in 24.3 percent fewer serious injuries.
Kusano and Gabler (2015), simulated single-vehicle roadway
departure crashes based on a sample of 478 crashes in the 2012 NASS-
CDS. Modelling was based on 54 model year 2010-2014 vehicles that were
evaluated in NHTSA's New Car Assessment Program (NCAP). Kusano and
Gabler found LDW systems could prevent 11-23 percent of drift-out-of-
lane crashes and 13-22 percent of serious to fatally injured drivers.
As noted previously for frontal crash technologies, the agencies
will base our projections on Leslie et al. because they are the most
recent study, thereby reflecting the most current versions of these
systems in the largest number of vehicles, and because they arguably
have the most precise identification of the presence of the specific
technologies in the vehicle fleet. However, unlike forward crash
technologies, lane change technologies are operational at travel speeds
where fatalities are likely to occur. Both LDW and LKA typically
operate at speeds above roughly 35 mph. For this reason, and because
the research noted in Table VI-246 indicates similar effectiveness
against fatalities, injuries, and crashes, the agencies believe it is
reasonable to assume the Leslie et al. crash reduction estimates are
generally applicable to all crash severities, including fatal crashes.
Our central effectiveness estimates are thus 10 percent for LDW and 20
percent for LKA. For sensitivity analysis, the agencies adopt the 95
percent confidence intervals from Flannagan & Leslie. For LKA this
range is 14.95-25.15 percent. For LDW, the upper range was 4.95-13.93
percent.
(c) Blind Spot Crash Technologies
To address blind spot crashes, manufacturers are currently
equipping vehicles with BSD, which detects vehicles in either of the
adjacent lanes that may not be apparent to the driver. The system warns
the driver of an approaching vehicle's presence to help facilitate safe
lane changes and avoid crashes. A more advanced version of this, LCA,
also detects vehicles that are rapidly approaching the driver's blind
spot. Table VI-247 summarizes studies which have measured effectiveness
for BSD and LCA.2071 2072 2073
---------------------------------------------------------------------------
\2071\ Cicchino, J.B. (2017b). Effects of blind spot monitoring
systems on police-reported lane-change crashes. Insurance Institute
for Highway Safety, August 2017.
\2072\ Leslie et al., supra note 2063.
\2073\ Isaksson-Hellman, I., Lindman, M., An evaluation of the
real-world safety effect of a lane change driver support system and
characteristics of lane change crashes based on insurance claims.
Traffic Injury Prevention, February 28, 2018: 19 (supp. 1).
---------------------------------------------------------------------------
[[Page 24806]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.458
[GRAPHIC] [TIFF OMITTED] TR30AP20.459
Cicchino (2017) used Poisson regression to compare crash
involvement rates per insured vehicle year in police-reported lane-
change crashes in 26[thinsp]U.S. States during 2009-2015 between
vehicles with blind spot monitoring and the same vehicle models without
the optional system, controlling for other factors that can affect
crash risk. Systems designs across the 10 different manufacturers
included in the study varied regarding the extent to which the size of
the adjacent lane zone that they covered exceeded the blind spot area,
speed differentials at which vehicles could be detected, and their
ability to detect rapidly approaching vehicles, but these different
systems were not examined separately. The study examined 4,620 lane
change crashes, including 568 injury crashes. Cicchino found an overall
reduction of 14 percent in blind spot related crashes of all
severities, with a non-significant 23 percent reduction in injury
crashes.
Leslie et al. (2019) analyzed the relative crash performance of
123,377 2013-2017 General Motors (GM) vehicles linked to State police-
reported crashes by Vehicle Identification numbers (VIN). GM provided
VIN-linked safety content information for these vehicles to enable
precise identification of safety technology content. The authors
analyzed the effectiveness of a variety of crash avoidance technologies
including both BSD and LCA separately. They estimated effectiveness
comparing system-relevant crashes to baseline (control group) crashes
using a quasi-induced exposure method in which rear-end struck crashes
are used as the control group. Flannagan and Leslie found that BSD
reduced lane departure crashes of all severities by 3 percent (non-
significant), and that LCA (which includes BSD) reduced these crashes
by 26 percent.
Isaksson-Hellman and Lindman (2018) evaluated the effect of the
Volvo Blind Spot Information System (BLIS) on lane change crashes.
Volvo's BLIS functions as an LCA, detecting vehicles approaching the
blind spot as well as those already in it. The authors analyzed crash
rate differences in lane change situations for cars with and without
the BLIS system based on a population of 380,000 insured vehicle years.
The authors found the BLIS system did not significantly reduce the
overall number of lane change crashes of all severities, but they did
find a significant 31 percent reduction in crashes with a repair cost
exceeding $1250, and a 30 percent lower claim cost across all lane
change crashes, indicating a reduced crash severity effect.
Like lane change technologies, blind spot technologies are
operational at travel speeds where fatalities are likely to occur. The
agencies therefore assume the Leslie et al. crash reduction estimates
are generally applicable to all crash severities, including fatal
crashes. Our central effectiveness estimates are thus 3 percent for BSD
and 26 percent for LCA. For sensitivity analysis, the agencies adopt
the 95 percent confidence intervals from Flannagan & Leslie. For LCA
this range is 16.59-33.74 percent. For BSD, the upper range was 14.72
percent, but the findings were not statistically significant. The
agencies therefore limit the range to 0-14.72 percent.
Table VI-248 summarizes the effectiveness rates calculated in
Leslie et al. and used in this analysis. Differences between the rates
listed as ``Used in CAFE Fatality Analysis'' and those computed from
Leslie et al. are explained in the above discussion.
[[Page 24807]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.460
(ii) Target Populations for Crash Avoidance Technologies
The impact on fatality rates that will occur due to these
technologies will be a function of both their effectiveness rate and
the portion of occupant fatalities that occur under circumstances that
are relevant to the technologies function. The agencies base our target
population estimates on a recent study that examined these portions
specifically for a variety of crash avoidance technologies including
those analyzed here. Wang (2019) documented target populations for five
groups of collision avoidance technologies in passenger vehicles
including forward collisions, lane keeping, blind zone detection,
forward pedestrian impact, and backing collision avoidance. The first
three of these affect the light occupant target population examined in
this analysis. Wang separately examined crash populations stratified by
severity including fatal injuries, non-fatal injuries, and property
damaged only (PDO) vehicles. She based her analysis on 2011-2015 data
from NHTSA's Fatality Analysis Reporting System (FARS), National
Automotive Sampling System (NASS), and General Estimates System (GES).
FARS data was the basis for fatal crashes while nonfatal injuries and
PDOs were derived from the NASS and GES.
Wang followed the pre-crash typology concept initially developed by
the Volpe National Transportation Systems Center (Volpe). Under this
concept, crashes are categorized into mutually exclusive and distinct
scenarios based on vehicle movements and critical events occurring just
prior to the crash. Table VI-249 summarizes the portion of total annual
crashes and injuries for each crash severity category that is relevant
to the three crash scenarios examined.
[GRAPHIC] [TIFF OMITTED] TR30AP20.461
The relevant proportions vary significantly depending on the
severity of the crash. The rear-end crashes that are addressed by FCW
and AEB technologies tend to be low-speed crashes and thus account for
a larger portion of non-fatal injury and PDO crashes than for
fatalities. Only 4 percent of fatal crashes occur in front-to-rear
crashes, but over 30 percent of nonfatal crashes are this type. By
contrast, fatal crashes are highly likely to involve inadvertent lane
departure, 44 percent of all light vehicle occupant fatalities occur in
crashes that involve lane departure, but only 17 percent of non-fatal
injuries and 12 percent of PDOs involve this crash scenario. Blind spot
crashes account for only about 2 percent of fatalities, 7 percent of
MAIS1-5 injuries, and 12 percent of PDOs.
The target population of this analysis is occupants of the light
vehicles subject to CAFE. The values in Table VI-249 are portions of
all crashes that occur annually. These include crashes of motor
vehicles not subject to the current CAFE rulemaking such as medium and
large trucks, buses, motorcycles, bicycles, etc. To adjust for this,
the values in Wang were normalized to represent their portion of all
light passenger vehicle (PV) crashes, rather than all crashes of any
type. Wang provides total PV fatalities consistent with her technology
numbers which are used as a baseline for this process. Based on 2011-
2015 FARS data, Wang
[[Page 24808]]
found an average of 29,170 PV occupant fatalities occurred annually.
A second adjustment to Wang's results was made to make them
compatible with the effectiveness estimates found in Leslie et al. In
her target population estimate for lane departure warning, Wang
included both head-on collisions and rollovers, but Leslie et al. did
not. The Leslie et al. effectiveness rate is thus applicable to a
smaller target population than that examined by Wang. To make these
numbers more compatible, counts for these crash types were removed from
Wang's lane departure totals.
Electronic Stability Control (ESC) has been standard equipment in
all light vehicles in the U.S. since the 2012 model year. ESC is highly
effective in reducing roadway departure and traction loss crashes, and
although it will be present in all future model year vehicles, it was
present in only about 30 percent of the 2011-2015 on-road fleet
examined by Wang. To reflect the impact of ESC on future on-road fleets
therefore, the agencies further adjusted Wang's numbers to reflect a
100 percent ESC presence in the on-road fleet. The agencies allocated
the reduced roadway departure fatalities to the LDW target population,
and the reduced traction loss fatalities to the AEB target population.
This has the effect of reducing the total fatalities in both groups as
well as in the total projected fatalities baseline.
Table VI-250 summarizes the revised incidence counts and re-
calculated proportions of total PV occupant crash/injury. Revised
totals are derived from original totals referenced in Table 1-3 in Wang
(2019).
[GRAPHIC] [TIFF OMITTED] TR30AP20.462
(iii) Fleet Penetration Schedules
The third element of the rule's safety projections is the fleet
technology penetration schedules. Advanced safety technologies (ADAS)
will only influence the safety of future MY fleets to the extent that
they are installed and used in those fleets. These technologies are
already being installed on some vehicles to varying degrees, but the
agencies expect that over time, they will become standard equipment due
to some combination of market pressure and/or safety regulation. The
agencies adopt this assumption based on the history of most previous
vehicle safety technologies, which are now standard equipment on all
new vehicles sold in the U.S.
The pace of technology adoption is estimated based on a variety of
factors, but the most fundamental is the current pace of adoption in
recent years. These published data were obtained from Ward's Automotive
Reports for each technology.\2074\ Since these technologies are
relatively recent, only a few years of data--typically 2 or 3 years--
were available from which to derive a trend. This makes these
projections uncertain, but under these circumstances, a continuation of
the known trend is the baseline assumption, which the agencies modify
only when there is a rationale to justify it.
---------------------------------------------------------------------------
\2074\ Derived from Ward's Automotive Yearbooks, 2014 through
2018, % Factory Installed Electronic ADAS Equipment tables,
weighting domestic and imported passenger cars and light trucks by
sales volume.
---------------------------------------------------------------------------
The technologies were examined in pairs reflecting their mutual
target populations. Both FCW and AEB affect the same target
population--frontal collisions. Both systems have been installed in
some current MY vehicles, but their relative paces are expected to
diverge significantly due to a formal agreement brokered by NHTSA and
IIHS involving nearly all auto manufacturers, to have AEB installed in
100 percent of their vehicles by September 2022 (MY 2023).\2075\ Wards
first published installation rates for FCW and AEB for the 2016 model
year and as of this analysis the 2017 MY is the latest data they have
published. The agencies thus have data indicating that FCW was
installed in 17.6 percent of MY 2016 vehicles and 30.5 percent of MY
2017 vehicles. AEB was installed in 12.0 percent of MY 2016 vehicles
and 27.0 percent of MY 2017 vehicles. AEB was installed in 12.0 percent
of MY 2016 vehicles and 27.0 percent of MY 2017 vehicles. More recent
reports submitted by manufacturers to the Federal Register indicate
that installation rates accelerated in MY 2018 and 2019
[[Page 24809]]
vehicles. Four manufacturers, Tesla, Volvo, Audi, and Mercedes, have
already met their voluntary commitment of 100 percent installation 3
years ahead of schedule. During the period September 1, 2018 through
August 31, 2019, 12 of the 20 manufacturers equipped more than 75
percent of their new passenger vehicles with AEB, and overall
manufacturers equipped more than 9.5 million new passenger vehicles
with AEB.\2076\
---------------------------------------------------------------------------
\2075\ See https://www.nhtsa.gov/press-releases/nhtsa-iihs-announcement-aeb.
\2076\ See NHTSA Announces Update to Historic AEB Commitment by
20 Automakers. December 17, 2019. https://www.nhtsa.gov/press-releases/nhtsa-announces-update-historic-aeb-commitment-20-automakers.
---------------------------------------------------------------------------
Because of the NHTSA/IIHS agreement, the agencies assume that AEB
will be in 100 percent of light vehicles by the 2023 MY. To derive
installation rates for MYs 2018 through 2022, the agencies interpolate
between the MY 2017 rate of 27 percent and the MY 2023 rate of 100
percent. To derive a MY 2015 estimate, the agencies modelled the
results for MYs 2016-2023 and calculated a value for year x=0,
essentially extending the model results back one year on the same
trendline.
For FCW, the agencies used the same interpolation/modeling method
as was used for AEB to derive an initial baseline trend. However, while
both systems are available on some portion of the current MY fleet, the
agencies anticipate that by MY 2023, all vehicles will have AEB systems
that essentially encompass both FCW and AEB functions. The agencies
therefore project a gradual increase in both systems until the sum of
both systems penetration rates exceeds 100 percent. At that point, the
agencies project a gradual decrease in FCW only installations until FCW
only systems are completely replaced by AEB systems in MY 2023.
For LDW, Wards penetration data were available as far back as MY
2013, giving a total of 5 data points through MY 2017. The projection
for LDW was derived by modelling these data points. The data indicate a
near linear trend and our initial projections of future years were
derived directly from this model. Wards did not report any of the more
advanced LKA systems until MY 2016, leaving only 2 data points. The
agencies modelled a simple trendline through these data points to
estimate the pace of future LKA installations. As with Frontal crashes,
the agencies assume a gradual phase-in of the most effective
technology, LKA, will eventually replace the lesser technology, LDW,
and the agencies allow gradual increases in both systems penetration
until their sum exceeds 100 percent, at which point LDW penetration
begins to decline to zero while LKA penetration climbs to 100 percent.
For blind spot crashes, Wards data was available for MYs 2013-2017
for BSD, but no data was available to distinguish LCA systems. LCA
systems were available as optional equipment on at least 10 MY 2016
vehicles.\2077\ In addition, Flannagan and Leslie found numerous cases
in State data-bases involving vehicles with LCA. Because LCA data is
not specifically identified, the agencies will estimate its frequency
based on the samples found in Flannagan & Leslie. In that study, 62
percent of vehicles with blind spot technologies has BSD alone, while
38 percent had LCA (which includes BSD). The agencies employ this ratio
to establish the relative frequency of these technologies in our
projections. As with frontal and lane change technologies, the agencies
assume a gradual phase-in of the most effective technology, LCA, will
eventually replace the lesser technology, BSD, and the agencies allow
gradual increases in both systems penetration until their sum exceeds
100 percent, at which point BSD penetration begins to decline to zero
while LCA penetration climbs to 100 percent.
---------------------------------------------------------------------------
\2077\ See, e.g. https://www.autobytel.com/car-buying-guides/features/10-cars-with-lane-change-assist-using-cameras-or-sensors-130847.
---------------------------------------------------------------------------
(iv) Impact Calculations
Table VI-251, Table VI-252, and Table VI-253 summarize the
resulting estimates of impacts on fatality rates for frontal crash
technologies, lane change technologies, and blind spot technologies
respectively for MYs 2016-2035. All previously discussed inputs are
shown in the tables. The effect of each technology is the product of
its effectiveness, it's percent installation in the MY fleet, and the
portion of the total light vehicle occupant target population that each
technology might address. Since installation rates for each technology
apply to different portions of the vehicle fleet (i.e., vehicles have
either the more basic or more advanced version of the technology), the
effect of the two technologies combined is a simple sum of the two
effects. Likewise, since each crash type addresses a unique target
population, there is no overlap among the three crash types and the sum
of the normalized crash impacts across all three crash types represents
the total impact on fatality rates from these 6 technologies for each
model year. These cumulative results are shown in the last column of
Table VI-253. As technologies phase in to newer MY fleets,\2078\ their
impact on the light vehicle occupant fatality rate increases
proportionally to roughly 8.5 percent before levelling off. That is,
eventually, by approximately MY 2026, these technologies are expected
to reduce fatalities and fatality rates for new vehicles by roughly 8.5
percent below their initial baseline levels.
---------------------------------------------------------------------------
\2078\ While it is technically possible to retrofit these
systems into the on-road fleet, such retrofits would be
significantly more expensive than OEM installations. The agencies
thus assume all on-road fleet penetration of these technologies will
come through new vehicle sales.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24810]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.463
[GRAPHIC] [TIFF OMITTED] TR30AP20.464
[[Page 24811]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.465
BILLING CODE 4910-59-C
[[Page 24812]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.466
(b) Fatality Trend Model
The revised fatality trend model differs from the model employed in
the NPRM in four main respects:
The fatality rates for individual model years and ages
were re-calculated to correct the counts of fatalities to occupants of
light-duty vehicles and to reflect the revised VMT estimates, the
latter of which incorporate revisions to both vehicle registration
counts and the estimated relationship between vehicle age and annual
use; \2079\
---------------------------------------------------------------------------
\2079\ These revised estimates of the number of miles traveled
by vehicles of each model year during past calendar years were
developed from the expanded sample of vehicles' odometer readings
obtained by NHTSA.
---------------------------------------------------------------------------
In response to comments on the version used in the NPRM, t
model adds controls for changes to factors (such as driver demographics
and behavior, and geographic patterns of travel) that can affect
fatality rates for vehicles of all model years and ages;
The revised analysis clusters past model years into
``safety cohorts,'' which are groups of successive model years that
exhibit similar fatality rates during their first years of use, in
order to represent the actual historical pattern of safety improvements
more realistically; and
The model employs a slightly less complex mathematical
relationship between a model year's age and its fatality rate
(fatalities per mile driven), which still describes the observed
relationship accurately.
Similar to the fatality trend model employed in the proposal, the
revised estimates of annual travel were combined with tabulations of
annual fatalities occurring among occupants of light-duty vehicles of
each model year during past calendar years, tabulated from NHTSA's FARS
data. Fatalities occurring in vehicles produced during each model year
making up a calendar year's light-duty vehicle fleet are divided by the
estimated number of miles they were driven during that calendar year to
calculate historical fatality rates by model year and calendar year,
measured as fatalities per billion miles traveled. These data represent
the dependent variable in the revised statistical model of fatality
rates.
Longitudinal or time-series analyses such as the model of
historical variation in fatality rates for individual model years need
to incorporate three separate effects to account for all potential
sources of variation. First, they need to employ model year in some
form as an explanatory variable, to account for improvements in the
safety of vehicles produced during successive model years that persist
throughout their lifetimes in the vehicle fleet. This is an example of
a ``cohort effect'' in the age-period-cohort framework that is widely
used to of analysis of population-wide behavior.\2080\ Second, such a
model must account for the effect of age on the safety of each
individual model year as it grows older, accumulates mileage, and in
most cases changes ownership one or more times during its expected
service lifetime (the ``aging effect'' in age-period-cohort analysis).
---------------------------------------------------------------------------
\2080\ For a detailed explanation of the rationale and methods
for age-period-cohort analysis, see for example Columbia University
Mailman School of Public Health, Population Health Methods: Age-
Period-Cohort Analysis, available at https://www.mailman.columbia.edu/research/population-health-methods/age-period-cohort-analysis (accessed February 12, 2020); and Kupper,
Lawrence L. et al., ``Statistical age-period-cohort analysis: A
review and critique,'' Journal of Chronic Diseases 38:10 (1985), at
811-830, available at https://www.sciencedirect.com/science/article/abs/pii/0021968185901055#! (accessed February 12, 2020).
---------------------------------------------------------------------------
Finally, most longitudinal analyses, including the historical
safety model developed here, need to account explicitly for factors
that vary over time--in this case, calendar years. By doing so, they
can affect the safety of vehicles of all model years and ages making up
the fleet during successive calendar years, or change the composition
of total travel by vehicles of different model years and ages. In
either case, such time-related factors--often referred to as ``period
effects''--can change the overall safety performance of the entire
fleet from one calendar year to the next, independently of and in
addition to the changes that would result from the combination of new
model years entering the fleet while older ones are retired from
service (the cohort effect), and the aging of all model years making up
the fleet. For example, an increase in seat belt use among all drivers
during a calendar year would be expected to reduce the fatality rates
of vehicles of all model years and ages in use during that year, while
an economic recession may change the composition of drivers and
vehicles on the road during a calendar year. In either case, one result
will be a change in the fleet-wide composite fatality rate for that
calendar year.
Figure VI-83 below illustrates the contributions of cohort, aging,
and time-period effects to changes over time in population-wide
behavior. As the figure indicates, these effects are conceptually
independent, but interact in ways that combine to produce the observed
historical evolution of the fleet-wide fatality rate for light-duty
vehicle occupants. Again, calendar year or time-period factors can
affect the safety performance of the entire fleet independently of the
effect that would result from the combination of changes in the
specific model years making up the fleet and the advancing ages of all
model years, and any ``period effect'' effect attributable to factors
that vary over time is in addition to cohort and aging effects.
[[Page 24813]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.467
To introduce such period effects into the fatality trend model,
which were absent from the NPRM analysis, the agencies obtained
historical data on factors that varied by calendar year, and were
expected to be responsible for such effects. As indicated previously,
these included the following:
Seat belt use, as measured by the fraction of drivers
observed to be wearing lap and shoulder belts, estimated by NHTSA's
National Occupant Protection Survey (NOPUS);
Driving under the influence of alcohol or drugs, measured
by the fraction of drivers reporting having recently done so in surveys
conducted by the U.S. Centers for Disease Control (CDC); \2081\
---------------------------------------------------------------------------
\2081\ The agencies also experimented with measures of drivers
appearing to be under the influence of alcohol or drugs included in
NHTSA's NOPUS, available at https://crashstats.nhtsa.dot.gov/#/PublicationList/18.
---------------------------------------------------------------------------
Use of hand-held electronic devices, measured by the
fraction of drivers visually observed to be doing so in NHTSA's NOPUS;
The fraction of licensed drivers who are male and under
the age of 25 (historically the riskiest cohort of drivers), as
reported by the FHWA's annual Highway Statistics publication; \2082\
---------------------------------------------------------------------------
\2082\ Federal Highway Administration, Highway Statistics,
various years, Table DL-20, available at https://www.fhwa.dot.gov/policyinformation/statistics.cfm.
---------------------------------------------------------------------------
The fraction of miles traveled in rural areas, also as
reported by FHWA; \2083\ and
---------------------------------------------------------------------------
\2083\ Federal Highway Administration, Highway Statistics,
various years, Table VM-1, available at https://www.fhwa.dot.gov/policyinformation/statistics.cfm.
---------------------------------------------------------------------------
The overall performance of the U.S. economy, as measured
by the annual rate of unemployment.\2084\
---------------------------------------------------------------------------
\2084\ See Bureau of Labor Statistics, historical data series
LNS14000000, available at https://data.bls.gov/cgi-bin/surveymost?ln.
---------------------------------------------------------------------------
The agencies were unable to obtain useful measures of roadway
design parameters or road conditions that would be expected to affect
safety. Although such measures exist, they tend to be reported for
individual road and highway segments or routes, and it is difficult to
combine these data into meaningful, aggregate measures that describe
overall driving conditions that are likely to vary by calendar year.
Nor could they identify satisfactory measures of incident response time
or the effectiveness of emergency medical treatment in reducing the
consequences of injuries occurring in motor vehicle crashes.
An important challenge to incorporating these time-period effects
into the fatality trend model arose from the fact that their patterns
of variation over the historical period the agencies analyzed (which
extended from calendar year 1995 to 2017) were extremely closely
correlated, making it virtually impossible to distinguish their
independent contributions to improvements in fleet-wide safety over
time. Table VI-254 below reports the pairwise correlation coefficients
among the potential measures of period effects listed above. As it
suggests, patterns of variation about their respective mean values over
the period analyzed were very similar (with the exception of the
unemployment rate), and the resulting high statistical correlations (or
``collinearity'') among them made it nearly impossible to identify
their independent effects on variation in safety over time, even when
controlling for the effects of model year and vehicle age.
[[Page 24814]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.468
To address this difficulty, the agencies substituted a time trend--
that is, a variable that takes the value of one in the first calendar
year and increases by one in each successive calendar year--in an
effort to capture the joint movements in the variables that were
intended to measure time-period effects on safety. The agencies
experimented with both linear and more complex time trends to capture
the apparently declining rate of improvement in fleet-wide safety over
time, but found that the linear trend captured the combined effects
most reliably. Because the model's dependent variable is the natural
logarithm of model year and age-specific fatality rates, using a linear
time trend corresponds to assuming a constant percentage decline in
fatality rates each year (rather than a constant absolute decline each
year), and this pattern appeared to provide the best fit to the
observed historical pattern of safety improvements. Finally, after
noting that the linear time trend did not fully capture the effects on
fleet-wide safety associated with the economic recessions in 2001 and
2007-11, the agencies supplemented the time trend with indicator (or
``dummy'') variables for these years, finding that only those for 2008,
2009, and 2010 improved its explanatory power significantly.
Another significant improvement to the NPRM analysis was to group
model years into ``safety cohorts'' on the basis of similarity in their
fatality rates when new (that is, during their first year in service),
rather than treating each model year as a separate cohort. Groupings
were created through a combination of identifying years when new safety
regulations initially took effect or were phased in, examining of
first-year fatality rates, and limited statistical experimentation.
Grouping successive model years reduces the number of cohorts
significantly, since similar fatality rates were typically observed for
at least five, and sometimes as many as ten, consecutive model years
over the historical period the agencies examined. Grouping model years
into a smaller number of cohorts rather than treating each model year
as a separate cohort offers the advantage of introducing some variation
in the ages of vehicles making up the same cohort during a calendar
year, which improves the statistical reliability with which the
independent effect of age itself can be estimated.
Figure VI-84 below shows historical variation in the fatality rates
of past model years when each one was newly-introduced (i.e., during
its first year in use).\2085\ It clearly displays the significant
improvement in the safety of new vehicles over time in response to
improvements in safety features, including those required by NHTSA's
safety regulations. The figure also clearly documents the natural
clustering of fatality rates for successive model years that was used
to identify and define the safety cohorts used in the revised model. In
the panel structure of the model, which combines time-series and cross-
section variation in fatality rates for individual model years as their
ages vary across calendar years, the clustering of first-year fatality
rates for successive model years is captured by using separate ``fixed
effects'' for each safety cohorts illustrated in the figure. Some
judgment is inevitably required to distinguish between successive
cohorts and identify when the fatality rate for new model years has
changed significantly; the agencies experimented with using from five
to eight cohorts, ultimately finding that the agencies could
distinguish most reliably among the fatality rates for five cohorts.
---------------------------------------------------------------------------
\2085\ For simplicity, the figure assumes that each model year's
first year of use was the calendar year identical to its designated
model year; for example, the first full year of use for model year
2000 was assumed to be calendar year 2000. In fact, new vehicles
frequently become available for purchase during the calendar year
preceding their designated model year and continue to be sold
through the calendar year following it, although most sales occur
during the calendar year matching their designated model year.
---------------------------------------------------------------------------
[[Page 24815]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.469
A final revision to the NPRM model was to employ a slightly less
complex mathematical relationship between a model year's age and its
fatality rate than had been used in the NPRM version. Specifically, the
revised model relates fatality rates to age itself as well as the
second and third powers of age (that is, age squared and age cubed),
but omits the fourth power of age, which was included in the model
developed for the NPRM. This slightly simpler relationship proved
adequate to capture fully the complex--but strongly recurring--pattern
of fatality rates for past model years as they aged. Specifically, as
Figure VI-85 below shows, fatality rates have tended to remain
approximately constant for the first few years of most recent model
years' lifetimes, before increasing steadily through age 15-20 and then
declining gradually over the remainder of their lifetimes.
As discussed previously, the increase in fatality rates through
approximately age 20 is generally thought to result primarily from the
fact that used vehicles are commonly purchased and driven by members of
households whose demographic characteristics, driving behavior, and
geographic locations are associated with more risky driving behavior
and thus more frequent or severe crashes. Of course, increased
frequency of mechanical failures as vehicles age and accumulate mileage
also seems likely to contribute to this pattern. In contrast, the
consistent tendency for fatality rates to decline after about age 20 is
less well understood, but may owe partly to the demographic
characteristics and driving behavior of owners of very old vehicles.
Whatever its source, the number of vehicles remaining in service past
age 20 is so small and their use typically so limited that their
contribution to the fleet-wide fatality rate is minimal.
[[Page 24816]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.470
Figure VI-85 documents the relationship between age and fatality
rate for selected past model years.\2086\ As it shows, fatality rates
for recent model years follow a complex but strikingly similar pattern
of increase and subsequent decline with increasing age, although the
figure also shows that the earliest model years included in the sample
(1975-1980) tended not to display increasing fatality rates in the
first half of their lifetimes. At the same time, the figure illustrates
the gradual downward shift in fatality rates at all ages for successive
past model years, although there is considerable variation in the
extent of this shift for individual model years, particularly when they
are examined at specific ages. That is, the downward shift in fatality
rates for successive model years is not necessarily ``monotonic,''
particularly when it is examined at specific individual ages.
---------------------------------------------------------------------------
\2086\ For a color version, see the corresponding safety
discussion in the accompanying FRIA.
---------------------------------------------------------------------------
The agencies believe that the increase in fatality rates for cars
and light trucks produced during recent model years through
approximately age 20 reflects the fact that as aging vehicles change
ownership via the used car market, they are often purchased and driven
by households whose demographic characteristics and locations are
associated with riskier driving behavior and conditions. The decline in
vehicles' fatality rates after this age is not well understood, but
seems likely to reflect the fact that the relatively small fraction of
those originally produced in a model year that survive beyond age 20-25
are owned and driven by households that maintain them carefully, are
likely to reside in areas where driving conditions are safest, and
whose members engage in less risky driving behavior.
After examining the information summarized in Figure VI-85, the
agencies conclude that the effect of increasing age on vehicle safety
appears to be largely independent of the improvement in new cars'
fatality rates over successive model years, and appears to operate
similarly for all except the earliest model years in our historical
sample (which includes model years 1975-2017).\2087\ As a formal
statistical test, the agencies experimented with allowing the aging
effect to change across model years when the agencies estimated the
revised model, anticipating that newer safety technologies and vehicle
designs might ``flatten'' the relationship between fatality rates and
age--that is, reduce the degree to which fatality rates increased over
the 5-20 year range of vehicle ages--for newer model years. However,
the agencies found no evidence that the effect of age on safety changed
significantly for more recent model years compared to older ones, so
the agencies retained the assumption of identical aging effects for all
model
[[Page 24817]]
years in the revised model.\2088\ Thus the revised model shows
progressively lower fatality rates for more recent model years when
they are new, but fatality rates for all model years increase with age
and subsequently decline according to the same non-linear pattern
displayed in Figure VI-85. On a related question, the agencies also
found that including the squared and cubed values of age in addition to
age itself as explanatory variables in the model, while excluding the
fourth power of age, which had been included in the NPRM model, proved
adequate to capture the pattern of variation in fatality rates with
increasing age that most past model years have exhibited. Table VI-255
below reports the estimated parameter values for alternative
specifications of the model, together with various goodness-of-fit and
other diagnostic measures. The analysis described in the following
section uses the estimated time trend from Model 2 in the table, which
implies annual reduction in fatality rates for all model years of 2.14
percent.
---------------------------------------------------------------------------
\2087\ Of course, the agencies cannot observe the safety
performance of all model years included in the agencies' data sample
over their entire lifetimes, because the data the agencies use to
estimate the model start in calendar year 1990, by which time all
model years before 1990 were no longer new--for example, MY1975 cars
are already 15 years old by then--while the newest model years in
the agencies' sample are still very ``young'' when the agencies'
data ends in calendar year 2017. Thus, the agencies have only
incomplete information about the relationship of fatality rates to
age over the entire lifetimes of these model years, so it is
possible that this relationship differs at particularly early or
advanced ages for the oldest and newest model years in the agencies'
sample.
\2088\ Specifically, the agencies tested for interactions
between the age and model year variables, which would reveal changes
in the relationship between fatality rates and age for more recent
model years, but found that such interaction effects were generally
not statistically significant. Allowing for interactions between age
and the indicator variables for safety cohorts (recall that these
represent groupings of successive model years) produced this same
result--few of the interaction effects were statistically
significant.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24818]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.471
BILLING CODE 4910-59-C
BILLING CODE 4910-59-P
[[Page 24819]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.472
[[Page 24820]]
BILLING CODE 4910-59-C
Using the Model and Technology Analysis to Forecast Fatality Rates
The newest safety cohort includes model years from 2009 to 2017, so
in effect the agencies estimate that all those model years have
essentially the same fatality rate in their first year of use. The
agencies apply the estimated effectiveness of crash avoidance
technologies in reducing fatal crashes to the observed fatality rate
for model years 2009 to 2017 vehicles during their first year in use to
estimate fatality rates for future model years during the first year
each one is introduced. Figure VI-86 below shows the result of this
process; as it indicates, fatality rates for new model years decline
gradually through 2035 and then stabilize, reflecting the fact that the
agencies are only able to project the effectiveness of emerging crash
avoidance technologies on the safety of new vehicles through that year.
[GRAPHIC] [TIFF OMITTED] TR30AP20.473
The next step in constructing the forecast of fleet-wide fatality
rates is to apply the age-related increases in the fatality rate for
each model year making up the previous calendar year's fleet. For
example, the agencies assume that the fatality rates for all model
years comprising the light-duty vehicle fleet in 2017 increase with age
according to the relationship captured by the estimated coefficients on
the age variables in the preferred model specification shown in Table
VI-255. The same assumption is applied to all new model years
introduced in subsequent years. Finally, the agencies also assume that
the historical decline in fatality rates observed over past calendar
years (the ``period effect'' captured by the time trend variable) will
continue into the future. This implies that fatality rates for all
model years and ages will decline by an additional 2.41 percent in each
successive future calendar year from the rates that would have resulted
from the combined effects of continuing improvements in the safety of
newly-introduced model years and the effect of increasing age.\2089\
---------------------------------------------------------------------------
\2089\ The agencies do not apply this trend reduction to the
fatality rates for the newest model year in each calendar year's
fleet, because it is assumed to be independent of both the decline
in new-car fatality rates and the aging effect.
---------------------------------------------------------------------------
This process produces an estimate of the fatality rate for each
model year making up the fleet during each future calendar year. That
estimate reflects the combination of (1) reductions in fatality rates
for new cars, reflecting the continued improvements in their safety due
to crash avoidance technologies (through MY2035); (2) increases in the
fatality rates for each model year in the fleet from the previous
calendar year, which represent the effect of age estimated by the
historical model; and (3) the continuing downward trend in fatality
rates for all vehicles except the newest model year in each calendar
year's fleet, which is derived from the historical model.
The agencies then weight the fatality rate for each model year
making up a future year's fleet by the fraction of total fleet-wide VMT
it accounts for, and sum
[[Page 24821]]
the results to produce an estimate of the fleet-wide fatality rate. The
CAFE model does not actually use this fleet-wide fatality rate, because
all of the fatality calculations are performed separately for each
individual model year making up the fleet, which are then aggregated;
nevertheless, the agencies provide the fleet-wide rate as a useful
check on the reasonableness of our fatality rate forecasts for
individual model years as they enter the fleet and age over their
respective lifetimes. Figure VI-87 displays the projected fleet-wide
fatality rates for future calendar years, as well as the trend in their
recent historical values.
[GRAPHIC] [TIFF OMITTED] TR30AP20.474
(d) Impact of Advanced Technologies on Older Vehicle Fatality Rates
In the NPRM, the agencies calculated the potential safety impacts
of delayed purchases of vehicles with new safety technology that might
result from higher vehicles prices associated with more stringent CAFE
standards. A number of commenters noted that since these improvements
will be driven by crash avoidance technologies, they will also benefit
older vehicles and reduce their fatality rates as well. For example,
CARB noted that ``safety improvements generally provide systematic
safety benefits to all vehicles in the on-road fleet, not only to new
vehicles. However, NHTSA's safety model assigns safety coefficients to
vehicles solely based on their model year and it fails to incorporate
the effect that new safety designs and technologies will have on
systematically improving fleet-wide on-road safety.'' IPI similarly
noted that should ``new safety technologies be adopted, the predicted
fatalities for all the older vehicle vintages will have to be lowered
as well because effective crash avoidance technologies will lower all
vehicles' fatality costs.''
The agencies agree that the users of older vehicles will also
benefit from crash avoidance technologies on newer vehicles. In
response, the agencies have modified our methodology to reflect lower
fatality rates on older vehicles resulting from the new crash avoidance
technologies. Crash avoidance technologies prevent crashes from
happening and thus benefit both the vehicle with the technology and any
other vehicles that it might have collided with. However, the scope of
these impacts on older vehicle's fatality rates are somewhat limited
due to several factors:
Single vehicle crashes, which make up about half of all fatal
crashes, will not be affected. Only multi-vehicle crashes involving a
newer vehicle with the advanced technology and an older vehicle will be
affected. Multi-vehicle crashes account for roughly half of all light
vehicle occupant fatalities.
For a new safety technology to benefit an older vehicle in
a multi-vehicle crash, the vehicle with the technology must have been
in a position to control, or prevent the crash. For example, in front-
to-rear crashes which can be addressed by FCW and AEB, the older
vehicle would only benefit if it was the vehicle struck from behind. If
the struck vehicle were the newer vehicle, its AEB technology would not
prevent the crash. Logically this would occur in roughly half of two-
vehicle crashes and a third of all three-vehicle crashes. Since most
multi-vehicle crashes involve only two vehicles, roughly half of all
multi-vehicle crashes might qualify.
The benefits experienced by older vehicles are
proportional to the probability that the vehicles they collide with are
newer vehicles with advanced crash avoidance technology. The
[[Page 24822]]
agencies estimate that the probability that this would occur is a
function of the relative exposure of vehicles by age, measured by the
portion of total VMT driven by vehicles of that age. Based on VMT
schedules (see CY 2016 example in Table VI-256), new (current MY)
vehicles account for about 9.6 percent of annual fleet VMT. The
relevant portion would increase over time as additional MY vehicles are
produced with advanced technologies. However, the portion of older
vehicle crashes that might be affected by newer technologies is
initially very small--only about 2 percent (.5*.5*.096) of older
vehicles involved in crashes might benefit from advanced crash
avoidance technologies in other vehicles in the first year.
[GRAPHIC] [TIFF OMITTED] TR30AP20.475
[[Page 24823]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.476
To reflect this safety benefit for older vehicles, the agencies
calculated a revised fatality rate for each older MY vehicle on the
road based on its interaction with each new MY starting with MY 2021
vehicles based on the following relationship:
Revised fatality rate = Fm-((x-y)mnp) + F(1-m)
Where: F = initial fatality rate for each MY
x = baseline MY fatality rate
y = current MY fatality rate
m = proportion of occupant fatalities that occur in multi-vehicle
crashes (52 percent)
n = probability that crash is with a new MY vehicle containing
advanced technologies
p = probability that new vehicle is ``striking'' vehicle
The initial fatality rate for each vehicle MY (F) was derived by
combining fatality counts from NHTSA's Fatality Analysis Reporting
System (FARS) with VMT data from IHS/Polk.
The baseline MY fatality rate (x) represents the baseline rate over
which the impact of new crash avoidance technologies should be measured
It establishes the baseline rate for each MY that will be compared to
the most current MY rate to determine the change in fatality rate (FR)
for each MY. The relative effectiveness of new crash-avoidance
technologies in modifying the fatality rate of older model vehicles is
measured differently depending on the age of the older vehicle. The
fatality rate is a historical measure that reflects safety differences
due to both crashworthiness technologies such as air bags and crash
avoidance technologies such as electronic stability control, but up
through MY 2017, crashworthiness standards are the predominant cause of
these differences.
The most recent significant crashworthiness safety standard, which
upgraded roof strength standards which was effective in all new
passenger vehicles in MY 2017. Crashworthiness standards would not have
secondary benefits for older MY vehicles. Post MY 2017, the agencies
believe crash avoidance technologies will drive safety improvements. To
isolate the added crash avoidance safety expected in newer vehicles,
the marginal impact of the difference between the MY 2017 fatality rate
and the most current MY fatality rate represents the added marginal
effectiveness of new crash-avoidance technologies of each subsequent MY
for MYs 2017 and earlier. Beginning with MY 2018, the difference
between the older MY fatality rate and most current MY rate determines
the potential safety benefit for the older vehicles.
[[Page 24824]]
The current MY fatality rate (y), represents the projected fatality
rate of future MY vehicles after adjustment for the impacts of the
advanced crash avoidance technologies and projected improvements in
non-technology factors examined in this analysis. This process was
discussed in detail in the previous section.
The proportion of passenger vehicle occupant fatalities that occur
in multi-vehicle crashes (m), was derived from an analysis of occupants
of fatal passenger vehicle crashes from 2002-2017 FARS. The analysis
indicated that 47.8 percent of fatal crash occupants were in single
vehicle crashes, 40.2 percent were in two vehicle crashes, and 12
percent were in crashes involving 3 or more vehicles. Overall, 52.2
percent were in multi-vehicle crashes.
The portion of older vehicle crashes involving newer vehicles
containing advanced crash avoidance technologies (n), is assumed to be
equal to the cumulative risk exposure of vehicles that have these
technologies. This exposure is measured by the product of annual VMT by
vehicle age and registrations of vehicles of that age. The CAFE model
calculates this dynamically, but as an example, based on 2016
registration data (see Table VI-256 above), the most current MY would
represent 9.6 percent of all VMT in a calendar year, implying a 9.6
percent probability that the vehicle encountered would be from the most
current MY. This percentage would increase for each CY as more MY
vehicles adopt advanced crashworthiness technologies. The agencies note
that other factors such as uneven concentrations of newer vs. older
vehicles or improved crash avoidance in the younger vehicles already on
the road that are the basis for the agencies' VMT proportion table
might disrupt this assumption, but it is likely that this would only
serve to slow the probability of these encounters, making this a
conservative assumption in that it maximizes the probability that older
vehicles might benefit from newer technologies.
The probability that the vehicle with advanced crash avoidance
technology is the controlling or striking vehicle (p), was calculated
using the relative frequency of fatal crash occupants in multi-vehicle
crashes. As noted previously, 40.2 percent were in two vehicle crashes,
and 12 percent were in crashes involving 3 or more vehicles. The
agencies assume a probability of 50 percent for two vehicle crashes and
33 percent for crashes with 3 or more vehicles. Weighted together the
agencies estimate a 46.1 percent probability that, given a multi-
vehicle crash involving a vehicle with advanced technologies and an
older vehicle without them, the newer vehicle will be the striking
vehicle or in a position where its crash avoidance technologies might
influence the outcome of the crash with the older vehicle.
This process is illustrated in Table VI-257 below for adjustments
due to improvements in MY 2021 vehicles back through MY 1995. In Table
VI-257, the actual model year fatality rate is shown in the second
column. As noted above, the base fatality rate, shown in column 3, is
the MY 2017 rate for all MYs prior to 2018, after which it becomes the
actual MY rate. Column 4 shows the difference between the fatality rate
for MY 2021 and the base rate for each MY. Column 5 shows the resulting
revised fatality rate that would be used for each older MY, and column
6 and 7 list the change in that rate. The various factors noted in the
above formula are applied in column 5. The results indicate a 0.006
decrease in pre-2018 MY vehicles fatality rates, with declining impacts
going forward to MY 2021. In subsequent years, this impact would grow
to reflect the both the increased probability that an older vehicle
would crash with vehicles containing advanced technology, as well as
the increased technology levels in progressively newer vehicles. This
table was created using NPRM inputs and is provided for explanatory
purposes only. The actual impacts are dynamically calculated within the
Volpe model and reflect revised fatality rate trends going forward and
cover even older model years.
[[Page 24825]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.477
(e) Dynamic Fleet Composition
As described in the sales discussion in Section Dynamic Fleet Share
(DFS), the standards may impact the distribution of cars and trucks
purchased. As light trucks, SUVs and passenger cars respond differently
to technology applied to meet the standards--namely mass reduction--
fleets with different compositions of body styles will have varying
amounts of fatalities. Since mass-safety fatalities are calculated by
multiplying mass point-estimates by VMT, which implicitly captures the
impact of the dynamic fleet share model, the estimates of mass-safety
fatalities in the previous section include the impact of vehicle prices
on fleet composition.
(c) Impact of Rebound Effect on Fatalities
The ``rebound effect'' is a measure of the additional driving that
occurs when the cost of driving declines. More stringent standards
reduce vehicle operating costs, and in response, some consumers may
choose to drive more. Driving more increases exposure to risks
associated with on-road transportation, and this added exposure
translates into higher fatalities. The agencies have calculated this
impact by estimating the change in VMT that results from alternative
standards.
As noted previously, rebound miles are not imposed on consumers by
regulation. They are a freely chosen activity resulting from reduced
vehicle operational costs. As such, the agencies believe a large
portion of the safety risks associated with additional driving are
[[Page 24826]]
offset by the benefits drivers gain from added driving. For the
proposal, the agencies assumed that, in deciding to drive more, drivers
internalize the full cost to themselves and others, including the cost
of accidents, associated with their additional driving.
In response to the NPRM, EDF noted that consumers may not fully
value the added safety risk, such as risk to other drivers.\2090\ In
making this point, EDF suggested a value of 50 percent would be
conservative, but did not provide supporting evidence for that value.
The agencies agree that the level of risk internalized by drivers is
uncertain, and for the final rule have revised the portion of the added
monetized safety risk that consumers internalize to 90 percent, which
mostly offsets the societal impact of any added fatalities from this
voluntary consumer choice.
---------------------------------------------------------------------------
\2090\ EDF, Appendix B, NHTSA-2018-0067-12108, at 101.
---------------------------------------------------------------------------
The actual portion of risk from crashes that drivers internalize is
unknown. The agencies suspect that drivers are more likely to
internalize serious crash consequences than minor ones, and some
drivers may not perfectly internalize injury consequences to other
individuals, especially occupants of other vehicles and pedestrians.
However, legal consequences from crash liability, both criminal and
civil, should also act as a caution for drivers considering added crash
risk exposure. The agencies considered several approaches to estimating
internalized crash risk. The first assumes that drivers value harm to
themselves as well as legal liability for causing harm to others. It
considers that all fatalities in single vehicle crashes are fully
valued, that there is roughly a 50 percent chance that each driver
would be the one killed in multi-vehicle crashes, and that there is
roughly a 50 percent chance that each driver would be at-fault in a
multi-vehicle crash that they survived. This produces an estimate of
roughly 87 percent. Another approach assumes that drivers fully value
all damage in single vehicle crashes, and only discount property damage
incidents in multi-vehicle crashes. Based on data in Blincoe, et al.
(2015),\2091\ multi-vehicle property-damage-only crashes account for
about 7 percent of all societal crash costs, leaving 93 percent
recognized under this approach. Yet another approach would assume
drivers value injury crashes, but discount non-injury related costs
such as property damage and traffic congestion. This approach results
in roughly an 88 percent estimate of costs internalized. Overall, while
the agencies recognize this proportion is uncertain, the agencies
believe it is reasonable to assume that drivers internalize 90 percent
of the crash risk that results from added driving.
---------------------------------------------------------------------------
\2091\ Blincoe, L., Miller, T.R., Zaloshnja, E., Lawrence, B.A.,
(May 2015, Revised) The Economic and Societal Impact of Motor
Vehicle Crashes, 2010, (DOT HS 812 012), National Highway Traffic
Safety Administration, Washington, DC.
---------------------------------------------------------------------------
IPI commented that additional mileage attributable to the scrappage
and dynamic fleet model is ``inexplicably and unjustifiably not offset
by countervailing mobility benefits in the benefit cost analysis--and
the agencies inappropriately claim that these traffic fatalities--which
comprise the other half of the 12,700 projection--also justify the roll
back.'' \2092\ In this comment, IPI has erroneously conflated the
rebound effect and the scrappage effect. The agencies have
appropriately accounted for the additional value consumers get out of
increases in fuel efficiency, which manifest in two ways: Reductions in
fuel costs, and the additional driving resulting from the reductions in
per-mile fuel costs. The agency cannot appropriately consider one
without the other, as the two effects trade off, one against the other,
according to consumer preferences between the two.
---------------------------------------------------------------------------
\2092\ IPI, Appendix, NHTSA-2018-0067-12213, at 12 (internal
citation omitted).
---------------------------------------------------------------------------
The scrappage effect represents the behavior of consumers when
their choices are restricted by more stringent fuel economy standards.
For instance, the consumer loses lower-price and less fuel-efficient
bundles of vehicle attributes that would be available in the absence of
more stringent alternatives. If anything, these consumers experience an
un-estimated cost regarding the lost utility from being priced out of
the new car market and being forced to drive an older, less safe--and
likely less fuel efficient--vehicle. That the agencies have assessed
the benefits of the rebound effect by assuming they are at least as
great as 90 percent of the additional safety costs of rebound driving,
does not mean that other channels of safety effects must be offset.
However, the agencies did evaluate whether the sales, scrappage, and
dynamic fleet share model could lead to changes in fuel economy in the
legacy fleet that may result in significant changes in VMT and/or fuel
economy. Upon further review, the agencies determined that such an
effect--if it were to exist--would be very small and would not impact
the analysis meaningfully, so the agencies declined to include this
effect in the final rule's analysis.
d) Fatalities by Source
For the NPRM, the agencies calculated rebound fatalities by running
the model with a 20 percent rebound assumption and again with a 0
percent rebound assumption. The following difference was assumed to
assign the change in fatalities of the rule due to rebound:
Rebound Fatalities = (FatalitiesAlt,20 -
FatalitiesAlt,0) -
(FatalitiesAug,20 -
FatalitiesAug,0)
Similarly, the agencies calculated mass reduction fatalities by
running the model using the central assumptions about coefficients on
delta curb weight and again setting these coefficients to 0, so that a
change in mass reduction would not affect the fatality rate of a
vehicle. The following difference assigned the change in fatalities of
the rule due to changes in mass reduction levels:
[Delta]CW Fatalities = (FatalitiesAlt,MR - FatalitiesAlt,NoMR) -
(FatalitiesAug,MR) - (FatalitiesAug,NoMR)
Where ``Alt'' represents the alternative being estimated, ``Aug'' is
the augural or baseline, ``MR'' stands for mass reduction, and
``NOMR'' means no mass reduction or mass reduction equaling zero.
The NPRM modeling then assumed that the remaining incremental
fatalities were due to changes in sales, scrappage, and the dynamic
fleet share. This can be represented by the following:
Sales/Scrap Fatalities = (FatalitiesAlt - FatalitiesAug) - Rebound
Fatalities - [Delta]CW Fatalities
The changes to the VMT model (mainly the constraint that fixes
total non-rebound VMT to be constant across alternatives) necessitated
revising how fatalities are partitioned by source. The number of
vehicles of each regulatory class and age changes in each regulatory
alternative. Because of this, taking the increment of the rebound
fatalities solved in each scenario as described above would capture
changes both to the usage per vehicle from rebound, but also
differences in the number of vehicles. This would wrongly attribute
some of the sales and scrappage fatalities to rebound. Similarly,
taking the increment of the mass reduction fatalities solved in each
scenario as described above would capture the changes both to the
fatality rate for vehicles (from mass reduction) and the difference in
the number of vehicles across alternatives. This would likewise have
the potential of wrongly attributing the source of sales and scrappage
fatalities to mass reduction.
[[Page 24827]]
Instead of computing the fatalities due to rebound in each scenario
and then taking the incremental values across alternatives, the
agencies compute rebound fatalities by taking the difference in per
vehicle rebound miles in the regulatory alternative and the augural
case multiplied by the augural fatality rate per mile and augural
vehicle count. Holding the number of vehicles constant addresses the
concern about the NPRM fatality allocation method wrongly attributing
rebound fatalities to the sales and scrappage models. Fatalities due to
rebound are computed as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.478
Where ``RVMT'' is VMT including rebound miles, ``NRVMT'' is VMT
excluding rebound miles, ``Veh'' is the quantity of vehicles, and
``Alt'' and ``Aug'' have the same meaning described above. The
rebound fatalities will show as zero for the augural scenario, and
all alternatives will show fatalities due to rebound miles using the
augural vehicle counts.
The fatalities due to mass reduction will use the augural vehicle
counts, augural per vehicle VMT including rebound--this simplifies to
total VMT including rebound, as shown below. Using a constant vehicle
count addresses the concern of the NPRM method wrongly assigning some
mass reduction fatalities to the sales and scrappage models. As with
the fatalities attributable to rebound, the fatalities attributable to
changes in mass reduction are calculated inherently as incremental
values, relative to the augural standards (the values will appear as
zero for augural standards in the outputs). The equation used to
calculate the fatalities due to curb weight changes is as follows:
[Delta]CW FatalitiesAlt = (Fatality RateAlt - Fatality RateAug) * R
VMTAug
The agencies then computed the sales/scrappage fatalities as the
remainder, as was done in the NPRM.
Sales/Scrap Fatalities = (FatalitiesAlt-FatalitiesAug)-Rebound
Fatalities-[Delta]CW Fatalities
(e) Adjustment for Non-Fatal Crashes
Fatalities are valued as a societal cost within the CAFE models'
cost and benefit accounting. Their value is based on the comprehensive
value of a fatality, which includes lost quality of life and is
quantified in the value of a statistical life (VSL) as well as economic
consequences such as medical and emergency care, insurance
administrative costs, legal costs, and other economic impacts not
captured in the VSL alone. These values were derived from data in
Blincoe et al. (2015), adjusted to 2018 economics, and updated to
reflect the official DOT guidance on the value of a statistical life.
This gives a societal value of $10.4 million for each fatality, which
is an update to the value used in the NPRM.\2093\ The CAFE safety model
estimates traffic fatalities but does not directly estimate the
corresponding non-fatal injuries and property damage that would result
from the same factors that influence fatalities. To address this, the
agencies developed an adjustment factor applied to fatality costs that
accounts for these crashes and related costs. The agencies' approach to
estimating non-fatal costs remains relatively unchanged from the
proposal, however the agencies have made one minor adjustment to
account for advance crash technologies as advocated by commenters.
---------------------------------------------------------------------------
\2093\ The NPRM used a societal value of $9,900,000 in 2016
dollars.
---------------------------------------------------------------------------
In the proposal, development of this factor was premised on the
assumption that non-fatal crashes would be affected by the standards in
proportion to their current nationwide rate of incidence and severity.
The agencies assumed the injury profile--the relative number of crashes
of each injury severity level that occur nationwide--would increase or
decrease congruent with changes in fatalities, meaning that the ratio
between fatal and non-fatal costs remained constant across
alternatives. The agencies recognized that this may not be the case,
but did not have data to support individual injury estimates across
injury severities. The agencies provided several explanations as to why
a proportionality assumption may be an oversimplification.\2094\ For
example, the agencies reviewed NHTSA's separate analysis of traffic
crash data showing that older model year vehicles are generally less
safe than newer vehicles, meaning fatalities would comprise a larger
portion of the total injury picture for older vehicles. This would
imply lower ratios across the non-fatal injury and property damage only
(PDO) crash profiles and would imply the adjustment overstates total
societal impacts.
---------------------------------------------------------------------------
\2094\ See 83 FR 43146 (Aug. 24, 2018).
---------------------------------------------------------------------------
As noted previously, in response to requests by commenters, the
agencies have added the estimated impact of six advanced crash
avoidance technologies that are currently being deployed commercially
to their analysis of future fatality rates. The same data and methods
described previously in this section to compute the impact of advanced
crash avoidance technologies on fatalities can also be used to examine
the effectiveness of these technologies against non-fatal and PDO
crashes. The inputs and results are summarized for nonfatal injuries in
Table VI-258 through Table VI-260, and for PDOs in Table VI-261 through
Table VI-263.\2095\
---------------------------------------------------------------------------
\2095\ See previous discussion in this section for the studies
and methodology used to create these estimates.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24828]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.479
[GRAPHIC] [TIFF OMITTED] TR30AP20.480
[[Page 24829]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.481
[GRAPHIC] [TIFF OMITTED] TR30AP20.482
[[Page 24830]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.483
[GRAPHIC] [TIFF OMITTED] TR30AP20.484
[[Page 24831]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.485
[GRAPHIC] [TIFF OMITTED] TR30AP20.486
[[Page 24832]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.487
BILLING CODE 4910-59-C
Based on a comparison of the combined average effectiveness impacts
for the three crash severity groups (fatalities, non-fatal injuries,
and property damage), it is apparent that these advanced crash
avoidance technologies would reduce non-fatal injuries and property
damage crashes by even more than they would fatalities.\2096\ To
explore the scope of this impact, the agencies developed an adjustment
factor that reflects the ratio of the decline in the rate of non-fatal
crashes to that of fatal crashes. This factor would hypothetically
affect the portion of safety improvement that is attributable to safety
technologies. The adjustments were based on the cumulative fatality
rates (for all three technology groups) by model year, noted in Table
VI-251 (Phased Impact of Crashworthiness Technologies on Fatality
Rates, Forward Collision Crashes) for fatalities, Table VI-260 for non-
fatal injuries, and Table VI-263 for PDOs, which are listed by MY in
the last column of Table VI-260 and Table VI-263. These factors would
modify the original non-fatal impacts--which were derived using an
assumption that they were proportional to fatal impacts--to reflect the
higher effectiveness of these technologies against non-fatal crashes.
---------------------------------------------------------------------------
\2096\ For example, for MY 2035, the combined effectiveness for
PDO crashes is .224784, as shown in the second to last column of
Table VI-6, which is 2.613 times the .0860 combined effectiveness
for fatalities, as seen in the final table from the Crash Avoidance
discussion above, which shows the disproportional impact of crash
avoidance technologies on non-fatal accidents.
---------------------------------------------------------------------------
The agencies considered including this additional adjustment factor
to account for the additional cost savings attributable to advance
crash avoidance technologies. The impact of such a factor would
decrease the incidence and severity, and thus the costs of nonfatal
crashes in regulatory alternatives where new vehicle sales increase,
including the preferred alternative. The agencies ultimately erred on
the side of caution for this rulemaking and have excluded this factor.
Therefore, today's analysis assumes that advance crash avoidance
technologies impact non-fatal and PDO crashes to the same extent as
fatal crashes. The agencies will consider including an adjustment for
non-fatal and PDO crashes in future rulemakings.
[[Page 24833]]
The original proportionality-based adjustment factor, which is
described in detail in the following paragraphs, was derived from
Tables 1-8 and I-3 in Blincoe et al. (2015). Incidence in Table I-3 in
Blincoe et al. reflects the Abbreviated Injury Scale (AIS), which ranks
nonfatal injury severity based on an ascending 5 level scale with the
most severe injuries ranked as level 5.\2097\
---------------------------------------------------------------------------
\2097\ More information on the basis for these classifications
is available from the Association for the Advancement of Automotive
Medicine at https://www.aaam.org/abbreviated-injury-scale-ais/.
---------------------------------------------------------------------------
Table 1-3 in Blincoe et al. lists injured persons with their
highest (maximum) injury determining the AIS level. This scale is
represented in terms of maximum abbreviated injury scale (MAIS) level.
MAIS0 refers to uninjured occupants in injury vehicles, MAIS1 injuries
are generally considered minor (e.g., a superficial laceration) with no
probability of death, MAIS2 injuries are generally considered moderate
(e.g., a fractured sternum) with a 1-2 percent probability of death,
MAIS3 injuries are serious (e.g., open fracture of the humerus) with an
8-10 percent probability of death, MAIS4 injuries are severe (e.g.,
perforated trachea) with a 5-50 percent probability of death, and MAIS5
injuries are critical (e.g., rupture liver with tissue loss) with a 5-
50 percent probability of death. Counts for PDO's refer to vehicles in
which no one was injured. From Table VI-264, ratios of injury
incidence/fatality are derived for each injury severity level as
follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.488
[GRAPHIC] [TIFF OMITTED] TR30AP20.489
For each fatality that occurs nationwide in traffic crashes, there
are 561 vehicles involved in PDOs, 139 uninjured occupants in crashes
which resulted in at least one injury,\2098\ 105 minor injuries, 10
moderate injuries, 3 serious injuries, and fractional numbers of the
most serious categories which include severe and critical nonfatal
injuries. For each fatality ascribed to the standards, it is assumed
there will be non-fatal crashes in these same ratios.
---------------------------------------------------------------------------
\2098\ Uninjured passengers incur a cost despite being
uninjured. For example, they are often transported to emergency care
even tough uninjured resulting in lost time and productivity;
furthermore, their vehicle might be damaged even though they are
uninjured.
---------------------------------------------------------------------------
Property damage costs associated with delayed fleet turnover must
be treated differently than rebound- and mass-related costs because
crashes that involve vehicles that are retained longer due to the
standards involve damage to older, used vehicles instead of newer
vehicles.\2099\ Used vehicles are worth less and will cost less to
repair, if they are repaired at all. The consumer's property damage
loss is thus reduced by longer retention of these vehicles. To estimate
this loss, average new and used vehicle prices were compared. New
vehicle transaction prices were estimated from a study published by
Kelley Blue Book.\2100\ Based on this data, the average new vehicle
transaction price in January 2017 was $34,968. Used vehicle transaction
prices were obtained from Edmonds Used Vehicle Market Report published
in February of 2017.\2101\ Edmonds data indicate the average used
vehicle transaction price was $19,189 in 2016. There is a minor timing
discrepancy in these data because the new vehicle data represent
January 2017, and the used vehicle price is for the average over 2016.
The agencies were unable to locate exact matching data, but believe the
difference is minor and negligible.
---------------------------------------------------------------------------
\2099\ The agencies note that property damage costs are the
costs realized given an accident has occurred. The disparity of
incidence rates between new and older vehicles is accounted for
above in the fatality calculations.
\2100\ Press Release, ``New-Car Transaction Prices Remain High,
Up More Than 3 Percent Year-Over-Year in January 2017, According to
Kelley Blue Book,'' February 1, 2017, available at https://mediaroom.kbb.com/2017-02-01-New-Car-Transaction-Prices-Remain-High-Up-More-Than-3-Percent-Year-Over-Year-In-January-2017-According-To-Kelley-Blue-Book.
\2101\ Edmonds Used Vehicle Market Report, February 2017.
Available at https://dealers.edmunds.com/static/assets/articles/2017_Feb_Used_Market_Report.pdf.
---------------------------------------------------------------------------
Based on these data, new vehicles are on average worth 82 percent
more than used vehicles. To estimate the effect of higher property
damage costs for newer vehicles in crashes, the per unit property
damage costs from Table I-9 in Blincoe et al. (2015) were multiplied by
this factor.\2102\ Results are illustrated in Table VI-265.
---------------------------------------------------------------------------
\2102\ The original unit costs were derived from vehicles
involved in crashes, which are predominately used vehicles. While
not precise, we assume this average cost is a reasonable proxy for
the property damage to a used vehicle.
---------------------------------------------------------------------------
[[Page 24834]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.490
The total property damage cost reduction was then calculated as a
function of the number of increased fatalities due to stricter CAFE and
CO2 standards as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.491
Where:
S = total property damage reductions from retaining used
vehicles longer
F = increase in fatalities estimated due to used vehicles
being retained longer because of stricter standards
r = ratio of non-fatal injuries or PDO vehicles to
fatalities
p = value of property damage prevented by retaining older
vehicle
n = the 8 injury severity categories
The number of fatalities ascribed to the standards because of
slower fleet turnover was multiplied by the unit cost per fatality from
Table I-9 in Blincoe et al. (2015) to determine the societal impact of
fatalities.\2103\ After subtracting the total reductions in property
damage from this value, the agencies divided the fatality cost by it to
estimate that overall, fatalities account for 39 percent of the total
costs that would result from older vehicle retention.
---------------------------------------------------------------------------
\2103\ Note--These calculations used the original values in the
Blincoe et al. (2015) tables without adjusting for economics. These
calculations produce ratios and are thus not sensitive to
adjustments for inflation.
---------------------------------------------------------------------------
These calculations are summarized as follows:
SV = Fv/x-s
Where:
SV = Value of societal impacts of all crashes resulting
from changes to fleet turnover
F = Increase in fatalities estimated due to retaining used
vehicles longer because of stricter standards
v = Comprehensive societal value of preventing 1 fatality
x = Percent of total societal loss from crashes
attributable to fatalities
S = total property damage reductions from retaining used
vehicles longer
For the fatalities that occur because of mass effects or to the
rebound effect, the calculation was more direct, a simple application
of the ratio of the portion of costs produced by fatalities to the
change in fatalities; there is no need to adjust for property damage
because all impacts were derived from the mix of vehicles in the on-
road fleet. Again, from Table I-8 in Blincoe et al. (2015), the
agencies derived this ratio based on all cost factors including
property damage to be 36 percent.
For purposes of application in the CAFE model, these two factors
(the factor for sales/scrappage, and the factor for mass and rebound)
were combined based on the relative contribution to total fatalities of
different factors. As noted previously, although a safety impact from
the rebound effect is calculated, these impacts are considered to be
freely chosen rather than imposed by the standards and imply personal
benefits at least equal to the sum of their added operational costs and
the portion of safety consequences internalized. However, the agencies
still calculate and report the impacts of the rebound effect to provide
a comprehensive view of the impacts of the standards. There are two
different factors depending on which metric is considered (total
impacts or CAFE imposed impacts). The agencies created these two
adjustment factors by weighting components by the relative contribution
to changes in fatalities associated with each component. This process
and results are shown in Table VI-266. Note that due to programming
constraints, the agencies applied the average weighted factor to all
fatalities. This will tend to overstate costs slightly because of sales
and scrappage and to understate costs associated with mass and rebound.
[[Page 24835]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.492
f) Summary of Safety Impacts
Table VI-267 through Table VI-270 summarize the safety effects of
CAFE standards across the various alternatives under the 3 percent and
7 percent discount rates.
Table VI-271 through Table VI-274 summarize these impacts for
CO2 standards. As noted in Section VI.D.2.e), societal
impacts are valued using a $10.4 million value per statistical life
(VSL). Note that fatalities in these tables are undiscounted--only the
monetized societal impact is discounted.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.493
[GRAPHIC] [TIFF OMITTED] TR30AP20.494
[[Page 24836]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.495
[GRAPHIC] [TIFF OMITTED] TR30AP20.496
[[Page 24837]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.497
[[Page 24838]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.498
[[Page 24839]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.499
[[Page 24840]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.500
[[Page 24841]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.501
[[Page 24842]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.502
[[Page 24843]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.503
BILLING CODE 4910-59-C
These tables present aggregations or averages of results for
calendar years through 2050. Underlying model output files provide
results for each model year in each calendar year.\2104\ These results
can be used for more detailed review and analysis of estimated trends.
For example, for each calendar year through 2050, the following two
tables--one for CAFE standards and one for CO2 standards--
show (a) the number of light-duty vehicles in service, (b) the travel
accumulated by those vehicles,
[[Page 24844]]
and (c) the total number fatalities among the types included in today's
analysis.
---------------------------------------------------------------------------
\2104\ FOOTNOTE 2104???
---------------------------------------------------------------------------
The analysis shows the annual number of fatalities for the final
standards growing more slowly than under the baseline standards,
reflecting the combined effects of fleet turnover, mass reduction, and
shifts between passenger cars and light trucks in the new vehicle
fleet.
Table VI-274 summarizes the non-fatal safety impacts under
alternative CAFE and CO2 standards:
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.504
[[Page 24845]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.505
BILLING CODE 4910-59-C
The Pennsylvania Department of Environmental Protection commented
that the agencies did not fully account for safety improvements
associated with the augural standards.\2105\ The agencies note that the
analysis accounts for the safety impacts of mass reduction, sales and
scrappage, rebound, vehicle model year and vehicle age for each of the
alternatives relative to the augural baseline. The commenter did not
provide any specific items that were omitted from the analysis. The
agencies believe the analysis thoroughly assesses the safety effects of
all the alternatives.
---------------------------------------------------------------------------
\2105\ NOT ON MANUSCRIPT.
---------------------------------------------------------------------------
Simulating Environmental Impacts of Regulatory Alternatives
This final rulemaking predominantly addresses fuel economy of the
light-duty vehicle fleet in the United States through different
technologies to improve efficiency. Inherently, these technologies will
reduce the fuel consumed and therefore impact CO2 and other
greenhouse gases foremost. Certain technologies will also impact air
quality through changes to criteria pollutants and air toxics emitted
at the tailpipe as well as upstream of the fuel source. Upstream
emissions for conventional fuels occur during crude oil extraction,
transportation, refining, and the transportation, storage, and
distribution of the finished fuel. For electricity, upstream emissions
are dependent on the mix of feedstocks such as coal, natural gas,
nuclear, and renewable sources for power generation. Similarly,
specific hydrogen production pathways such as natural gas reforming or
electrolysis of water molecules will determine the upstream emissions
of hydrogen fuel. Emission impacts are described in greater detail in
the following sections.\2106\
---------------------------------------------------------------------------
\2106\ NHTSA also uses the results of the CAFE model to analyze
the potential environmental impacts of the regulatory alternatives
in its Environmental Impact Statement (EIS). That EIS informs the
agency's decision-making process.
---------------------------------------------------------------------------
The impacts of both greenhouse gases (GHGs) and criteria pollutant
emissions that result from changes in vehicle usage and fuel
consumption were estimated and considered as part of this analysis.
GHGs are gaseous constituents in the atmosphere, both natural and
anthropogenic, and absorb infrared radiation. Primary GHGs in the
atmosphere are water vapor, CO2, nitrous oxide
(N2O), methane (CH4), and ozone. Criteria air
pollutants include carbon monoxide (CO), nitrogen dioxide
(NO2) (one of several oxides of nitrogen), ozone, sulfur
dioxides (SO2), particulate matter (including fine
particulate matter, or PM2.5), and lead. Vehicles do not
directly emit ozone, but ozone impacts are evaluated based on emissions
of the ozone precursor pollutants nitrogen oxides (NOX) and
volatile organic compounds (usually referred to as VOC). These
pollutants are emitted during vehicle storage and use, as well as
throughout the fuel production and distribution system. While increases
in domestic fuel refining, storage, and distribution that result from
higher fuel consumption will increase emissions of these pollutants,
reduced vehicle use associated with the fuel economy rebound effect
will decrease their emissions. The net effect of CAFE and
CO2 standards on total emissions of each criteria pollutant
depends on the relative magnitudes of increases in its emissions during
fuel refining and distribution, and decreases in its emissions
resulting from vehicle use. Because the relationship between emissions
in fuel refining and vehicle use is different for each criteria
pollutant, the net effect of fuel consumption on total emissions of
each pollutant differs between regulatory alternatives.
Climate Change and CO2 Emissions Considered in This Rule
The NPRM described how both agencies consider climate change and
GHG emissions under their respective programs for fuel economy and
CO2. As noted in the NPRM, ``In 1988, NHTSA included climate
change concepts in its CAFE notices and prepared its first
environmental assessment addressing that subject.'' \2107\
Additionally, NHTSA ``cited concerns about climate change as one of its
reasons for limiting the extent of its reduction of the CAFE standard
for MY 1989 passenger cars.'' \2108\ As stated in the NPRM, ``Since
then, NHTSA has considered the effects of reducing tailpipe emissions
of CO2 in its fuel economy rulemakings pursuant to the need
of the United States to conserve energy by reducing petroleum
consumption.\2109\
---------------------------------------------------------------------------
\2107\ 83 FR 43211 (citing 53 FR 33080, 33096 (Aug. 29, 1988)).
\2108\ Id. (citing 53 FR 39275, 39302 (Oct. 6, 1988)).
\2109\ 83 FR 43211.
---------------------------------------------------------------------------
Similarly, in the NPRM, EPA described that ``the primary purpose of
Title II of the Clean Air Act is the protection of public health and
welfare. EPA's light-duty vehicle GHG standards serve this purpose, as
the GHG emissions from light-duty vehicles have been found by EPA to
endanger public health and welfare (see EPA's 2009 Endangerment Finding
for on-highway motor vehicles), and the goal of these standards is to
reduce these emissions that contribute to climate change.'' \2110\ In
the NPRM, EPA summarized its purpose for establishing CO2
standards as follows:
---------------------------------------------------------------------------
\2110\ 83 FR 4228 (citing 74 FR 66496 (Dec. 15, 2009)).
Section 202(a)(1) of the Clean Air Act (CAA) states that ``the
Administrator shall by regulation prescribe (and from time to time
revise) . . . standards applicable to the emission of any air
pollutant from any class or classes of new motor vehicles . . . ,
which in his judgment cause, or contribute to, air
[[Page 24846]]
pollution which may reasonably be anticipated to endanger public
health or welfare.'' If EPA makes the appropriate endangerment and
cause or contribute findings, then section 202(a) authorizes EPA to
issue standards applicable to emissions of those pollutants. Indeed,
EPA's obligation to do so is mandatory: Coalition for Responsible
Regulation, 684 F.3d at 114; Massachusetts v. EPA, 549 U.S. at
533.\2111\
---------------------------------------------------------------------------
\2111\ 83 FR 43228.
The agencies modeled the estimated physical changes in quantity of
CO2, CH4, and NO2 emissions in the
NPRM analysis, and conducted additional modeling of climate-related
impacts, including sea-level rise, global temperate increases, and
ocean pH changes in the Draft EIS accompanying the NPRM. The Draft EIS
also included a comprehensive discussion of climate change impacts,
drawing from various Intergovernmental Panel on Climate Change (IPCC)
reports, the U.S. Global Change Research Program (USGCRP) National
Climate Assessment (NCA) reports, and other peer-reviewed reports and
assessment reports. The agencies also considered the increase in
climate damages from an increase in CO2 emissions,\2112\
also known as the social cost of carbon and discussed previously in
Section VI.D.1, above.
---------------------------------------------------------------------------
\2112\ 83 FR 43106.
---------------------------------------------------------------------------
Many commenters expressed a desire for more information on the
rule's potential climate impacts, so the discussion has been expanded
here and in the Final EIS. Specifically, commenters stated that the
agencies failed to address climate change in the proposal, and that the
proposal ignored ``scores of studies and reports'' on climate change
published since EPA's 2009 Endangerment Finding and promulgation of the
existing CO2 and CAFE standards.\2113\ Several commenters
presented summaries of climate impacts, citing IPCC, USGCRP, and other
reports explicitly relied on in the DEIS, on temperature increases,
increases in extreme weather events, ocean warming, acidification, and
sea level rise, impacts on the United States' water supply, human
health impacts, impacts to crop productivity and global food security,
potential increases in the spread of infectious disease, national
security impacts, and impacts to animal and plant species, including
Federally protected species, among other impacts.\2114\
---------------------------------------------------------------------------
\2113\ NHTSA-2018-0067-12088.
\2114\ NHTSA-2018-0067-11735; NHTSA-2018-0067-11926; NHTSA-2018-
0067-11972; NHTSA-2018-0067-12088; NHTSA-2018-0067-12127; NHTSA-
2018-0067-12303; NHTSA-2018-0067-12378; NHTSA-2018-0067-12436.
---------------------------------------------------------------------------
In addition to comments stating the agencies had presented too
little information on climate change in the NPRM, some commenters
disagreed with how the agencies framed the impact of the rule on
climate change. Many commenters cited IPCC and USGCRP to reinforce
their understanding that human activities are the dominant cause of
global warming since the mid-20th century. NHTSA considered both the
IPCC and USGCRP reports in the DEIS accompanying the NPRM and in this
final rule, and did not dispute those findings. Commenters also cited
IPCC and the National Climate Assessments, among other reports, as
support to their understanding that regardless of the perceived
magnitude of the rule on total CO2 emissions, any additional
actions taken now to reduce CO2 emissions would affect the
degree of climate impacts in the future. Further discussion of these
comments occurs in Section VIII.
Just as NHTSA does with both the draft and final EIS, and as EPA
did for its Endangerment and Cause or Contribute Findings for
Greenhouse Gases under the Clean Air Act, for this rule, both agencies
relied on existing studies and reports to summarize the current state
of climate science and provide a framework for the analysis of impacts.
The agencies drew primarily on panel-reviewed synthesis and assessment
reports from the Intergovernmental Panel on Climate Change (IPCC) and
the U.S. Global Change Research Program (GCRP), supplemented with past
reports from the U.S. Climate Change Science Program (CCSP), the
National Research Council, and the Arctic Council and EPA's Technical
Support Document for Endangerment and Cause or Contribute Findings for
Greenhouse Gases under the Clean Air Act,\2115\ which, as stated above,
relied on past major international or national scientific assessment
reports.
---------------------------------------------------------------------------
\2115\ EPA Technical Support Document for Endangerment and Cause
or Contribute Findings for Greenhouse Gases under Section 202(a) of
the Clean Air Act. December 7, 2009. U.S. Environmental Protection
Agency, Office of Atmospheric Programs, Climate Change Division:
Washington, DC. Available at: https://www.epa.gov/sites/production/files/2016-08/documents/endangerment_tsd.pdf.
---------------------------------------------------------------------------
Assessment reports assess numerous individual studies to draw
general conclusions about the potential impacts of climate change. Even
where assessment reports include consensus conclusions of expert
authors, uncertainty still exists, as with all assessments of
environmental impacts. Given the global nature of climate change and
the need to communicate uncertainty to a variety of decision-makers,
IPCC has focused considerable attention on developing a systematic
approach to characterize and communicate this information. The IPCC is
a United Nations panel, founded in 1988, which evaluates climate
science by assessing research on climate change and synthesizing
relevant research into major assessment reports. The IPCC provides
regular assessments on climate impacts and future risks, and options
for adaptation and risk mitigation. The agencies used the system
developed by IPCC to describe uncertainty associated with various
climate change impacts.
The IPCC reports communicate uncertainty and confidence bounds
using commonly understood but carefully defined words in italics to
represent likelihood of occurrence. The referenced IPCC documents
provide a full understanding of the meaning of those uncertainty terms
in the context of the IPCC findings. The IPCC notes that there are two
primary uncertainties with climate modeling: Model uncertainties and
scenario uncertainties: \2116\
---------------------------------------------------------------------------
\2116\ IPCC. Climate Change 2013: The Physical Science Basis.
Contribution of Working Group I to the Fifth Assessment Report of
the Intergovernmental Panel on Climate Change. Stocker, T.F., D.
Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels,
Y. Xia, V. Bex and P.M. Midgley (Eds.). Cambridge University Press:
Cambridge, United Kingdom and New York, NY, USA. pp. 1535. Available
at: http://www.ipcc.ch/report/ar5/wg1/. [hereinafter IPCC 2013].
---------------------------------------------------------------------------
Model uncertainties. These uncertainties occur when a
climate model might not accurately represent complex phenomena in the
climate system. For some processes, the scientific understanding could
be limited regarding how to use a climate model to ``simulate''
processes in the climate system.
Scenario uncertainties. These uncertainties arise because
of uncertainty in projecting future GHG emissions, concentrations, and
forcings (e.g., from solar activity).
According to IPCC, these types of uncertainties are described by
using two metrics for communicating the degree of certainty: Confidence
in the validity of findings, expressed qualitatively, and quantified
measures of uncertainties, expressed probabilistically.\2117\ The
confidence levels synthesize the judgments about the validity of the
findings, determined through evaluation of the evidence and the degree
of scientific agreement. The qualitative expression of confidence
ranges are described, in italics, from very low to very high, with
higher confidence levels assigned to findings that are supported by
high scientific agreement. The quantitative expression of confidence
ranges from exceptionally unlikely to
[[Page 24847]]
virtually certain, with higher confidence representing findings
supported by robust evidence. Table VI-276 shows that the degree of
confidence increases as evidence becomes more robust and agreement is
greater.
---------------------------------------------------------------------------
\2117\ IPCC 2013.
[GRAPHIC] [TIFF OMITTED] TR30AP20.506
As described in more detail in the Final EIS, the process known as
the greenhouse effect is responsible for trapping a portion of a
planet's heat in the planet's atmosphere, rather than allowing all of
that heat to be radiated into space. GHGs trap heat in the lower
atmosphere (the atmosphere extending from Earth's surface to
approximately 4 to 12 miles above the surface), absorb heat energy
emitted by Earth's surface and lower atmosphere, and reradiate much of
it back to Earth's surface, thereby causing warming. Human activities,
particularly fossil-fuel combustion, lead to the presence of increased
concentrations of GHGs in the atmosphere; this buildup of GHGs is
changing the Earth's energy balance. IPCC states the warming
experienced over the past century is due to the combination of natural
climatic forcers (e.g., natural GHGs, solar activity) and human-made
climate forcers.\2118\ IPCC concluded, ``[h]uman influence has been
detected in warming of the atmosphere and the ocean, in changes in the
global water cycle, in reductions in snow and ice, in global mean sea-
level rise, and in changes in some climate extremes. . . . This
evidence for human influence has grown since [the IPCC Working Group 1
(WG1) Fourth Assessment Report (AR4)]. IPCC reports that it is
extremely likely that human influence has been the dominant cause of
the observed warming since the mid-20th century.'' \2119\
---------------------------------------------------------------------------
\2118\ IPCC 2013.
\2119\ IPCC 2013.
---------------------------------------------------------------------------
Although the climate system is complex, IPCC has identified the
following drivers of climate change:
GHGs. Primary GHGs in the atmosphere are water vapor,
atmospheric CO2, N2O (nitrous oxide),
CH4 (methane), and ozone.\2120\
---------------------------------------------------------------------------
\2120\ IPCC 2013.
---------------------------------------------------------------------------
Aerosols. Aerosols are natural (e.g., from volcanoes) and
human-made particles in the atmosphere that scatter incoming sunlight
back to space, causing cooling. Some aerosols are hygroscopic (i.e.,
attract water) and can affect the formation and lifetime of clouds.
Large aerosols (more than 2.5 micrometers in size) modify the amount of
outgoing long-wave radiation.\2121\ Other particles, such as black
carbon, can absorb outgoing terrestrial radiation, causing warming.
Natural aerosols have had a negligible cumulative impact on climate
change since the start of the industrial era.\2122\ Further discussion
of black carbon and other aerosols is located in Chapter 4 of the FEIS.
---------------------------------------------------------------------------
\2121\ IPCC 2013.
\2122\ IPCC 2013.
---------------------------------------------------------------------------
Clouds. Depending on cloud height, cloud interactions with
terrestrial and solar radiation can vary. Small changes in the
properties of clouds can have important implications for both the
transfer of radiative energy and weather.\2123\
---------------------------------------------------------------------------
\2123\ IPCC 2013.
---------------------------------------------------------------------------
Ozone. Ozone is created through photochemical reactions
from natural
[[Page 24848]]
and human-made gases. In the troposphere, ozone absorbs and reemits
long-wave radiation. In the stratosphere, the ozone layer absorbs
incoming short-wave radiation.\2124\
---------------------------------------------------------------------------
\2124\ IPCC 2013.
---------------------------------------------------------------------------
Solar radiation. Solar radiation, the amount of solar
energy that reaches the top of Earth's atmosphere, varies over time.
Solar radiation has had a negligible impact on climate change since the
start of the industrial era compared to other main drivers.\2125\
---------------------------------------------------------------------------
\2125\ IPCC 2013.
---------------------------------------------------------------------------
Surface changes. Changes in vegetation or land surface
properties, ice or snow cover, and ocean color can affect surface
albedo.\2126\ The changes are driven by natural seasonal and diurnal
changes (e.g., snow cover) as well as human influences (e.g., changes
in vegetation type).\2127\
---------------------------------------------------------------------------
\2126\ Surfaces on Earth (including land, oceans, and clouds)
reflect solar radiation back to space. This reflective
characteristic, known as albedo, indicates the proportion of
incoming solar radiation the surface reflects. High albedo has a
cooling effect because the surface reflects rather than absorbs most
solar radiation.
\2127\ IPCC 2013.
---------------------------------------------------------------------------
Effects of emissions and the corresponding processes that affect
climate are highly complex and variable, which complicates the
measurement and detection of change. However, IPCC indicates that an
increasing number of studies conclude that anthropogenic GHG emissions
are affecting climate in detectable and quantifiable
ways.2128 2129 GHGs occur naturally and because of human
activity. Other GHGs, such as the fluorinated gases,\2130\ are
primarily anthropogenic in origin and are used in commercial
applications such as refrigeration and air conditioning and industrial
processes such as aluminum production.
---------------------------------------------------------------------------
\2128\ IPCC. Summary for Policymakers. In: Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth
Assessment Report of the Intergovernmental Panel on Climate Change.
Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J.
Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (Eds.).
Cambridge University Press: Cambridge, United Kingdom and New York,
NY, USA. 1535 pp. Available at: http://www.ipcc.ch/pdf/assessment-report/ar5/wg1/WG1AR5_SPM_FINAL.pdf.
\2129\ GCRP. 2017. Climate Science Special Report: Fourth
National Climate Assessment. U.S. Global Change Research Program.
[Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C.
Stewart, and T.K. Maycock (Eds.)]. U.S. Government Printing Office:
Washington, DC 477 pp. doi:10.7930/J0J964J6. Available at: https://science2017.globalchange.gov/downloads/CSSR2017_FullReport.pdf.
[hereinafter GCRP 2017].
\2130\ Fluorinated GHGs or gases include PFCs, HFCs,
SF6, and NF3.
---------------------------------------------------------------------------
In its most recent assessment of climate change (IPCC WG1 AR5),
IPCC states that, ``Warming of the climate system is unequivocal, and
since the 1950s, many of the observed changes are unprecedented over
decades to millennia. The atmosphere and ocean have warmed, the amounts
of snow and ice have diminished, sea level has risen, and the
concentrations of greenhouse gases have increased.'' \2131\ IPCC
concludes that, at continental and global scales, numerous long-term
changes in climate have been observed. To be more specific, IPCC and
the GCRP include the following trends observed over the 20th century as
further supporting the evidence of climate-induced changes:
---------------------------------------------------------------------------
\2131\ IPCC 2013.
---------------------------------------------------------------------------
Most land areas have very likely experienced warmer and/or
fewer cold days and nights along with warmer and/or more frequent hot
days and nights.2132 2133 From 1880 to 2016, the global mean
surface temperature rose by about 0.9 [deg]C (1.6 [deg]F).\2134\ Air
temperatures are warming more rapidly over land than over
oceans.2135 2136 Similar to the global trend, the U.S.
average temperature is about 1.8 [deg]F warmer than it was in 1895, and
this rate of warming is increasing--most of the warming has occurred
since 1970.\2137\ IPCC projects a continuing increase in surface
temperature between 2081 and 2100, with a likely range between 0.3
[deg]C (0.5 [deg]F) and 4.8 [deg]C (8.6 [deg]F), compared with 1986
through 2005, where the lower value corresponds to substantial future
mitigation of carbon emissions.\2138\
---------------------------------------------------------------------------
\2132\ IPCC Climate Change 2014: Impacts, Adaptation, and
Vulnerability. Part A: Global and Sectoral Aspects. Contribution of
Working Group II to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. Field, C.B., V.R. Barros,
D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee,
K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N.
Levy, S. MacCracken, P.R. Mastrandrea, and L.L. White (Eds.).
Cambridge University Press: Cambridge, United Kingdom and New York,
NY, USA, 1132 pp. Available at: http://ipcc-wg2.gov/AR5/report/.
[hereinafter IPCC 2014].
\2133\ GCRP 2017.
\2134\ GCRP 2017.
\2135\ IPCC 2013.
\2136\ GCRP 2017.
\2137\ GCRP 2017.
\2138\ IPCC 2013.
---------------------------------------------------------------------------
Cold-dependent habitats are shifting to higher altitudes
and latitudes, and growing seasons are becoming
longer.2139 2140 According to the IPCC, ``it is virtually
certain that there will be more frequent hot and fewer cold temperature
extremes over most land areas on daily and seasonal timescales'' and it
is very likely that heat wave frequency and duration will also
increase.\2141\
---------------------------------------------------------------------------
\2139\ IPCC 2014.
\2140\ GCRP 2017.
\2141\ IPCC 2014.
---------------------------------------------------------------------------
Sea level is rising, caused by thermal expansion of the
ocean and melting of snowcaps and ice sheets.2142 2143
Between 1971 and 2010, global ocean temperature warmed by approximately
0.25 [deg]C (0.45 [deg]F) in the top 200 meters (approximately 660
feet).\2144\ IPCC concludes that mountain glaciers, ice caps, and snow
cover have declined on average, further contributing to sea-level rise.
Losses from the Greenland and Antarctic ice sheets very likely
contributed to sea-level rise from 1993 to 2010, and satellite
observations confirm that they have contributed to sea-level rise in
subsequent years.\2145\ IPCC projects that the global temperature
increase will continue to affect sea level, causing a likely rise of
0.26 meter (0.85 foot) to 0.82 meter (2.7 feet) in the next
century.\2146\
---------------------------------------------------------------------------
\2142\ IPCC 2013.
\2143\ GCRP 2017.
\2144\ IPCC 2013.
\2145\ IPCC 2013.
\2146\ IPCC 2013.
---------------------------------------------------------------------------
More frequent weather extremes such as droughts, floods,
severe storms, and heat waves have been observed.2147 2148
Average atmospheric water vapor content has increased since at least
the 1970s over land and the oceans, and in the upper troposphere,
largely consistent with air temperature increases.\2149\ Because of
changes in climate, including increased moisture content in the
atmosphere, heavy precipitation events have increased in frequency over
most land areas.2150 2151 Observations of increased dryness
since the 1950s suggest that some regions of the world have experienced
longer, more intense droughts caused by higher temperatures and
decreased precipitation, particularly in the tropics and
subtropics.\2152\ Heavy precipitation events have increased globally
since 1951, with some regional and subregional variability.\2153\ A
warmer atmosphere holds more moisture and increases the energy
available for convection, causing stronger storms and heavier
precipitation.2154 2155
---------------------------------------------------------------------------
\2147\ IPCC 2013.
\2148\ GCRP 2017.
\2149\ IPCC 2013.
\2150\ IPCC 2013.
\2151\ Min, S.-K., Zhang, X., Zwiers, F.W., & Hegerl, G.C. 2011.
Human contribution to more-intense precipitation extremes. Nature,
470(7334), pp. 378-81. Available at: https://doi.org/10.1038/nature09763.
\2152\ IPCC 2013.
\2153\ IPCC 2013.
\2154\ GCRP 2017.
\2155\ Gertlet, C., O'Gorman, P. 2019. Changing available energy
for extratropical cyclones and associated convection in the Northern
Hemisphere summer, PNAS 116(10):4105-4110.
---------------------------------------------------------------------------
[[Page 24849]]
Many commenters urged the agencies to consider more stringent
standards to address GHG emissions. The Northeast States for
Coordinated Air Use Management (NESCAUM) stated that ``effectively
combatting climate change requires GHG reductions on a national and
international scale. Maintaining an aggressive downward trend in
transportation sector GHG emissions will not occur in the absence of
strong national GHG emission reductions.'' \2168\ Similarly, the Center
for Biological Diversity et al. stated ``the scientific record is now
overwhelming that climate change poses grave harm to public health and
welfare; that its hazards have become even more severe and urgent than
previously understood; and that avoiding devastating harm requires
substantial reductions in greenhouse gas emissions, including from the
critically important transport sector, within the next decade.'' \2169\
Minnesota Pollution Control Agency (MPCA), the Minnesota Department of
Transportation (MnDOT), and the Minnesota Department of Health (MDH)
stated ``Tackling climate change will require aggressive and immediate
action on reducing emissions from the transportation sector. The
existing GHG and CAFE standards are a critical piece to the
multifaceted and global effort to reduce GHG emissions.'' \2170\
---------------------------------------------------------------------------
\2168\ NESCAUM, NHTSA-2018-0067-11691.
\2169\ Center for Biological Diversity et al., NHTSA-2018-0067-
12000.
\2170\ MPCA, MnDOT, and MDH, NHTSA-2018-0067-11706.
---------------------------------------------------------------------------
Commenters also expressed concerns that the agencies did not
accurately consider the effects of climate change resulting from the
rulemaking. Pennsylvania Department of Environmental Protection (PA
DEP) stated ``the Proposed Rule does not fully consider the potential
effects of global climate change resulting from these forgone
reductions or the interests of states in preventing or mitigating the
impacts of climate change on their citizens and environment.'' \2171\
The Center for Biological Diversity et al. stated ``the agencies
callously disregard the demonstrated need to reduce emissions sharply
over the next decade if severe impacts of a destabilized climate are to
be avoided.'' \2172\ Similarly, the Joint Submission from the States of
California et al. and the Cities of Oakland et al. stated ``discussion
of the effect of the Proposed Rollback on GHG emissions significantly
understates the outcome,'' and ``the overwhelming scientific consensus
is that immediate and continual progress toward a near-zero GHG-
emission economy by mid-century is necessary to avoid truly
catastrophic climate change impacts.'' \2173\
---------------------------------------------------------------------------
\2171\ PA DEP, NHTSA-2018-0067-11956.
\2172\ Center for Biological Diversity et al., NHTSA-2018-0067-
12000.
\2173\ Joint Submission from the States of California et al. and
the Cities of Oakland et al., NHTSA-2018-0067-11735.
---------------------------------------------------------------------------
The agencies have carefully considered these comments in the
context of the information on climate change summarized in the NPRM and
DEIS, and have updated information for this final rule. The agencies
drew upon updates to climate science and impacts for the analysis from
reports and studies that were updated or released since the NPRM,
including IPCC's Global Warming of 1.5 degrees C report, Volume 2 of
the 4th National Climate Assessment, and IPCC's Special Report on
Climate Change and Land, and the IPCC's Special Report on the Ocean and
Cryosphere in a Changing Climate.
The following sections also provide additional context about
climate impacts from this final rule; the results of the agencies'
quantitative analysis presented in Section VII shows estimated
CO2, CH4, and N2O emissions resulting
from the rule, and the discussion of how each agency balanced climate
change as a factor considered in decision-making is presented in
Section VIII. The Final EIS accompanying today's rule also includes a
comprehensive discussion of climate impacts, and additional climate
modeling that estimates climate-related effects. As discussed in more
detail in the FEIS and following sections, but relevant for placing the
following discussion in context, climate modeling performed for this
final rule shows the following impacts as a result of the final
standards selected: CO2 Concentrations of 789.80 ppm in
2100, compared with 789.11 ppm under the augural standards; global mean
surface temperature increases of 3.487 [deg]C in 2100, compared with
3.484 [deg]C under the augural standards; sea-level rise increases of
76.34 cm in 2100, compared with 76.28 cm under the augural standards;
and ocean pH of 8.2172 in 2100, compared with 8.2176 under the augural
standards. These equal differences of 0.69 ppm, 0.003 [deg]C, 0.06 cm,
and -0.0004, respectively. Additionally, the agencies valued
anticipated climate-related economic effects in accordance with E.O.
13783, as discussed in Section VI.D.1.
(1) Global Greenhouse Gas Emissions
According to NOAA and IPCC, Global atmospheric CO2
concentrations have increased 46.4 percent, from approximately 278
parts per million (ppm) in 1750 \2174\ to approximately 407 ppm in
2018.\2175\ According to IPCC and WRI, in 2014, CO2
emissions \2176\ accounted for 76 percent of global GHG emissions on a
global warming potential (GWP)-weighted basis,\2177\ followed by
CH4 (16 percent), N2O (6 percent), and
fluorinated gases (2 percent).2178 2179 IPCC notes that
atmospheric concentrations of CH4 and N2O
increased approximately 150 and 20 percent, respectively, over roughly
the same period.\2180\
---------------------------------------------------------------------------
\2174\ IPCC 2013.
\2175\ NOAA. Globally Averaged Marine Surface Annual Mean
CO2 Data. Available at: ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_annmean_gl.txt.
\2176\ These global GHG estimates do not include contributions
from land-use change and forestry or international bunker fuels.
\2177\ Each GHG has a different radiative efficiency (the
ability to absorb infrared radiation) and atmospheric lifetime. To
compare their relative contributions, GHG emission quantities are
converted to carbon dioxide equivalent (CO2e) using the
100-year time horizon global warming potential (GWP) as reported in
IPCC's Second Assessment Report (AR2): The Science of Climate Change
in Sections B.7 Summary of Radiative Forcing and B.8 Global Warming
Potential.
\2178\ IPCC. 1996. Second Assessment: Climate Change 1995.
Inventories. Available at: https://www.ipcc.ch/site/assets/uploads/2018/06/2nd-assessment-en.pdf.
\2179\ WRI (World Resources Institute). 2018. Climate Analysis
Indicators Tool (CAIT) 2.0: WRI's Climate Data Explorer. Available
at: http://cait.wri.org/. [hereinafter WRI 2018].
\2180\ IPCC 2013.
---------------------------------------------------------------------------
According to WRI, developed countries, including the United States,
have been responsible for the majority of historical GHG emissions
since the mid-1800s and still have some of the highest GHG emissions
per capita.\2181\ While annual emissions from developed countries have
been relatively flat over the last few decades, world population
growth, industrialization, and increases in living standards in
developing countries are expected to cause global fossil-fuel use and
resulting GHG emissions to grow substantially. According to IPCC,
global GHG emissions since 2000 have been increasing nearly three times
faster than in the 1990s.\2182\ This is further illustrated in Figure
VI-88 showing carbon dioxide emissions since 1990 by world region:
\2183\
---------------------------------------------------------------------------
\2181\ WRI 2018.
\2182\ IPCC 2013.
\2183\ EPA's Climate Change Indicators in the United States,
2016: www.epa.gov/climate-indicators. Data source: WRI, 2015.
---------------------------------------------------------------------------
[[Page 24850]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.507
GHGs are emitted from a wide variety of sectors, including energy,
industrial processes, waste, agriculture, and forestry. According to
WRI, the energy sector is the largest contributor of global GHG
emissions, accounting for 72 percent of global emissions in 2014; other
major contributors of GHG emissions are agriculture (10 percent) and
industrial processes (6 percent).\2184\ Transportation CO2
emissions--from the combustion of petroleum-based fuels--account for
roughly 15 percent of total global GHG emissions, and have increased by
64 percent from 1990 to 2014.2185 2186
---------------------------------------------------------------------------
\2184\ WRI 2018.
\2185\ The energy sector is largely composed of emissions from
fuels consumed in the electric power, transportation, industrial,
commercial, and residential sectors. The 15 percent value for
transportation is therefore included in the 72 percent value for
energy.
\2186\ WRI 2018.
---------------------------------------------------------------------------
In general, global GHG emissions continue to increase, although
annual increases vary according to factors such as weather, energy
prices, and economics. Comparing observed carbon emissions to projected
emissions, the current global trajectory is similar to the most fossil
fuel-intensive emissions scenario (A1Fi) in the IPCC Special Report on
Emissions Scenarios (2000) and the highest emissions scenario (RCP8.5)
represented by the more recent Representative Concentration Pathways
(RCP).2187 2188
---------------------------------------------------------------------------
\2187\ The Representative Concentration Pathways (RCPs) were
developed for the IPCC AR5 report. They define specific pathways to
emission concentrations and radiative forcing in 2100. The RCPs
established four potential emission concentration futures, a
business-as-usual pathway (RCP8.5), two stabilization pathways
(RCP6.0, 4.5), and an aggressive reduction pathway (RCP2.6).
\2188\ IPCC 2013.
---------------------------------------------------------------------------
(2) U.S. Greenhouse Gas Emissions and the Transportation Sector
Most GHG emissions in the United States are from the energy sector,
with the majority of those being CO2 emissions coming from
the combustion of fossil fuels. Fossil fuel combustion CO2
emissions alone account for 76 percent of total U.S.GWP-weighted
emissions, with the remaining 24 percent contributed by other sources
such as industrial processes and product use, agriculture and forestry,
and waste.\2189\ CO2 emissions due to combustion of fossil
fuels are from fuels consumed in the transportation (37 percent of
fossil fuel combustion CO2 emissions), electric power (35
percent), industrial (16 percent), residential (6 percent), and
commercial (5 percent) sectors.\2190\ In 2017, U.S. GHG emissions were
estimated to be 6,456.7 MMTCO2e,\2191\ or approximately 14
percent of global GHG emissions.2192 2193
---------------------------------------------------------------------------
\2189\ EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks:
1990-2017. EPA 430-R-19-001. U.S. Environmental Protection Agency.
Washington DC Available at: https://www.epa.gov/sites/production/files/2019-04/documents/us-ghg-inventory-2019-main-text.pdf.
[hereinafter EPA 2019].
\2190\ EPA 2019.
\2191\ Most recent year for which an official EPA estimate is
available. EPA 2019.
\2192\ Based on global and U.S. estimates for 2014, the most
recent year for which a global estimate is available. Excluding
emissions and sinks from land-use change and forestry and
international bunker fuels.
\2193\ WRI 2018.
---------------------------------------------------------------------------
Similar to the global trend, CO2 is by far the primary
GHG emitted in the U.S.,
[[Page 24851]]
representing 82 percent of U.S. GHG emissions in 2017 (on a GWP-
weighted basis),\2194\ and accounting for 15 percent of total global
CO2 emissions.2195 2196 Although CO2
is the GHG with the largest contribution to warming, methane accounts
for 10.2 percent of U.S. GHGs on a GWP-weighted basis, followed by
N2O (5.6 percent) and the fluorinated gases (2.6
percent).\2197\
---------------------------------------------------------------------------
\2194\ EPA 2019.
\2195\ The estimate for global emissions from the World
Resources Institute is for 2014, the most recent year with available
data for all GHGs. It excludes emissions and sinks from land use
change and forestry.
\2196\ WRI 2018.
\2197\ EPA 2019.
---------------------------------------------------------------------------
When U.S. CO2 emissions are apportioned by end use,
transportation is the single leading source of U.S. emissions from
fossil fuels, causing over one-third of total CO2 emissions
from fossil fuels.\2198\ Passenger cars and light trucks account for 59
percent of total U.S. CO2 emissions from transportation, an
increase of 14 percent since 1990.\2199\ This increase in emissions is
attributed to about 50 percent increase in vehicle miles traveled (VMT)
because of population growth and expansion, economic growth, and low
fuel prices. Additionally, the rising popularity of sport utility
vehicles and other light trucks with lower fuel economy than passenger
cars has contributed to higher emissions.2200 2201 Although
emissions typically increased over this period, emissions declined from
2008 to 2009 because of decreased economic activity associated with the
most recent recession.\2202\
---------------------------------------------------------------------------
\2198\ Apportioning by end use allocates emissions associated
with electricity generation to the sectors (residential, commercial,
industrial, and transportation) where it is used. EPA 2019.
\2199\ EPA 2019.
\2200\ EPA 2019.
\2201\ DOT. 2016. Table 4-23: Average Fuel Efficiency of U.S.
Light Duty Vehicles. U.S. Department of Transportation, Bureau of
Transportation Statistics. Available at: https://www.rita.dot.gov/bts/sites/rita.dot.gov.bts/files/publications/national_transportation_statistics/html/table_04_23.html.
\2202\ EPA 2019.
---------------------------------------------------------------------------
Today's rule addresses light-duty vehicle fuel economy and
CO2 emissions from new-model passenger cars and light
trucks. Several commenters observed that the transportation sector
accounted for a large, if not the largest, portion of the United States
greenhouse gas emissions, and that light-duty vehicle emissions
contributed to a large fraction of that portion.\2203\ Many commenters
referenced the IPCC Report from 2018 on Global Warming of 1.5 Degrees
Celsius, which considered transportation sector greenhouse gas
emissions in describing pathways to limit climate impacts.
---------------------------------------------------------------------------
\2203\ NHTSA-2018-0067-11284; NHTSA-2018-0067-10966; NHTSA-2018-
0067-11691; NHTSA-2018-0067-11735; NHTSA-2018-0067-11765; NHTSA-
2018-0067-11921; NHTSA-2018-0067-12000; NHTSA-2018-0067-12021;
NHTSA-2018-0067-12022; NHTSA-2018-0067-12088; NHTSA-2018-0067-12303;
NHTSA-2018-0067-4159.
---------------------------------------------------------------------------
Graphically, historical trends in U.S. GHG emissions reported by
EPA appear as follows.\2204\
---------------------------------------------------------------------------
\2204\ Historical data from https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-emissions-and-sinks. The asterisk
indicates that the chart does not include reported emissions changes
attributable to land use, land use change, and forestry (LULUCF).
---------------------------------------------------------------------------
[[Page 24852]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.508
Notably, light-duty vehicle CO2 emissions outweigh other
GHG emissions from light-duty vehicles, and light-duty vehicle
CO2 emissions have been relatively stable over a nearly 30-
year period during which highway vehicles miles traveled has increased
by about 50 percent.\2205\ Without fuel economy increases that have
accumulated since EPCA's passage in 1975, recent light-duty vehicle
CO2 emissions would have been 50 percent greater than shown
above.\2206\
---------------------------------------------------------------------------
\2205\ https://www.fhwa.dot.gov/policyinformation/travel_monitoring/historicvmt.pdf.
\2206\ DOT reports fuel economy levels of the historical on-road
fleet at https://www.bts.gov/content/average-fuel-efficiency-us-light-duty-vehicles.
---------------------------------------------------------------------------
For fuel combustion, EIA's National Energy Modeling System (NEMS),
which EIA uses to produce its Annual Energy Outlook (AEO) forecasts of
U.S. energy consumption and supply, provides corresponding estimates of
CO2 emissions. For the final rule, modeling conducted by the
agencies using the AEO2019 version of NEMS shows the following levels
of future CO2 emissions from sectors other than light-duty
vehicles (which this rule impacts directly) and refineries (which this
rule is estimated to impact through changes in fuel consumption):
[[Page 24853]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.509
As this chart indicates, EIA's representation of laws and
regulations current as of AEO2019 shows aggregate emissions from these
sectors remaining remarkably stable through 2050, despite projected
growth in the U.S. population and economy.
The agencies agree with commenters that the transportation sector,
and specifically light-duty vehicle emissions, contribute to the
largest portion of the United States' greenhouse gas emissions.\2207\
However, the fuel economy and CO2 of vehicles, regulated in
this rulemaking, is not the only determining factor for whether the
light-duty transportation sector would see a rise or decline in
CO2 emissions. As discussed elsewhere in this rule, the
standards from the final rule affect only new vehicles, which are
responsible for approximately 3.5 percent of on-road VMT in any year.
The agencies recognize that the revised standards result in additional
CO2 emissions, and these emissions are accounted for in the
analysis. It is worthwhile to note that the difference between the
augural standard and the new standard is a small change to a small
fraction of total VMT, and it is important to consider in context the
different mechanisms that contribute to transportation sector
greenhouse gas emissions. These mechanisms are considered in the 2018
IPCC special report cited by commenters as well; in addition to vehicle
fuel efficiency, IPCC considers preventing (or reducing) the need for
transport,\2208\ as ``increasingly efficient fleets of vehicles over
time . . . does not necessarily limit the driven distance.'' (internal
citations omitted).\2209\
---------------------------------------------------------------------------
\2207\ See U.S. Energy Information Administration available at
https://www.eia.gov/todayinenergy/detail.php?id=29612 and EPA,
Sources of Greenhouse Gas Emissions available at https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions.
\2208\ IPCC 2018 at 349 (citing Gota et al., 2018).
\2209\ IPCC 2018 at 377 (citing Ajanovic and Haas, 2017; Sen et
al., 2017).
---------------------------------------------------------------------------
b) Air Quality
This section discusses the health and environmental effects
associated with exposure to some of the criteria and air toxic
pollutants impacted by the proposed vehicle standards. The agencies
note that these impacts are, compared to the impacts on vehicular fuel
consumption and CO2 emissions, small and mixed. CAFE and
CO2 standards directly impact vehicular fuel consumption and
CO2 emissions. Notwithstanding modest indirect impacts, such
as impacts on vehicle
[[Page 24854]]
sales, retention, and mileage accumulation, one can ``draw a direct
line'' between CAFE/CO2 standards and resultant changes in
overall fuel consumption and CO2 emissions, and these follow
the expected trends.
Changes in emissions of criteria pollutants due to these rules will
impact air quality. The Clean Air Act (CAA) is the primary federal
statute that addresses air quality. Pursuant to its CAA authority, the
EPA has established National Ambient Air Quality Standards (NAAQS) for
six criteria pollutants: CO, NO2, ozone, SO2,
particulate matter (PM), and lead. Vehicles do not directly emit ozone,
but ozone impacts are evaluated based on emissions of the ozone
precursor pollutants nitrogen oxides (NOX) and volatile
organic compounds (VOC). When the measured concentrations of a criteria
pollutant in a geographic region are less than those permitted by
NAAQS, EPA designates the region as an attainment area for that
pollutant; regions where concentrations of criteria pollutants exceed
Federal standards are called nonattainment areas. Former nonattainment
areas that are now in compliance with NAAQS are designated as
attainment areas and are commonly referred to as maintenance areas.
Each state with a nonattainment area is required to develop and
implement a State Implementation Plan (SIP) documenting how the region
will reach attainment levels within periods specified in the CAA. For
maintenance areas, the SIP must document how the State intends to
maintain compliance with NAAQS. When EPA changes a NAAQS, each State
must revise its SIP to address how it plans to attain the new standard.
In addition to analyzing criteria pollutants, the agencies considered
hazardous air pollutants emitted from vehicles that are known or
suspected to cause cancer or other serious health and environmental
impacts and are referred to as mobile source air toxics, as further
discussed in this section. Table VI-277 below provides an overview of
criteria pollutants and mobile source air toxics with a high level
overview of health effects. See further within this section for details
on the pollutants and toxics.
BILLING CODE 4910-59-P
[[Page 24855]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.510
[[Page 24856]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.511
BILLING CODE 4910-59-C
The CAA requires the EPA to review periodically the NAAQS and the
supporting science, and to revise the standards as appropriate.\2210\
Schedules for recently completed and ongoing reviews are summarized
here. In February 2019, the EPA issued a decision to retain the
existing primary NAAQS for SO2.\2211\ For the ongoing
reviews of the NAAQS for PM and ozone, the EPA intends to issue
proposed decisions in early 2020 and final decisions in late 2020.
---------------------------------------------------------------------------
\2210\ https://www.epa.gov/criteria-air-pollutants/naaqs-table.
\2211\ 84 FR 9866 (March 18, 2019).
---------------------------------------------------------------------------
Nationally, levels of PM2.5, ozone, NO2,
SO2, CO and air toxics have declined significantly in the
last 30 years. However, as of January 31, 2020, more than 130 million
people lived in counties designated nonattainment for one or more of
the NAAQS, and this figure does not include the people living in areas
with a risk of exceeding a NAAQS in the future. Many Americans continue
to be exposed to ambient concentrations of air toxics at levels which
have the potential to cause adverse health effects. In addition,
populations who live, work, or attend school near major roads
experience elevated exposure concentrations to a wide range of air
pollutants. As discussed in the FEIS, concentrations of many air
pollutants are elevated near high-traffic roadways. If minority
populations and low-income populations disproportionately live near
such roads, then an issue of environmental justice (EJ) may be present.
Comments were received from multiple entities expressing concern about
emissions and EJ communities. The agencies considered EJ when
considering the effects of this rule; EJ considerations and EJ-related
comments received on the NPRM and DEIS are discussed in Section X and
the FEIS.
Total emissions from on-road mobile sources (highway vehicles) have
declined dramatically since 1970 because of pollution controls on
vehicles and regulation of the chemical content of fuels, despite
continuing increases in vehicle miles traveled (VMT). From 1970 to
2016, emissions from on-road mobile sources declined 89 percent for CO,
71 percent for NOX, 59 percent for PM2.5, 40
percent for PM10, 93 percent for SO2, and 90
percent for VOCs.\2212\ The figure below further shows the highway
vehicle emissions trends that indicate reduced pollutants regulated
under NAAQS.
---------------------------------------------------------------------------
\2212\ See https://www.epa.gov/transportation-air-pollution-and-climate-change/accomplishments-and-success-air-pollution-transportation https://gispub.epa.gov/air/trendsreport/2019/#home.
---------------------------------------------------------------------------
[[Page 24857]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.410
Many commenters expressed concerns about the increase of emissions
leading to regions in nonattainment for ozone and particulate matter
and concerns regarding the inability to meet the NAAQS. The Center for
Biological Diversity et al., and a number of State and local
governments and government agencies asserted that State and local
jurisdictions would be at jeopardy of becoming nonattainment areas
under the proposed rule.\2213\ CARB and the joint submission from the
States of California and Cities of Oakland stated that the proposed
rule would result in ``increases in emissions [which] will undermine
state implementation plans'' and the proposed rule ``would create an
additional 1.24 tons per day of NOX emissions in the South
Coast basin.'' \2214\ The South Coast Air Quality Management District
(SCAQMD) stated ``[a]s a regional air quality district, we have limited
authority to control emissions from mobile sources, and rely on the
Federal government to take action,'' and they expressed concern about
meeting the NAAQS under the proposed rule because, to meet that
standard, the Basin would have to ``reduce NOX emissions by
45% beyond existing requirements.'' \2215\
---------------------------------------------------------------------------
\2213\ Center for Biological Diversity, et al., NHTSA-2018-0067-
12123.
\2214\ CARB, NHTSA-2018-0067-11873, Joint Submission from States
of California and Cities of Oakland, NHTSA-2018-0067-11735.
\2215\ SCAQMD, NHTSA-2018-0067-11813.
---------------------------------------------------------------------------
In particular, commenters including PA DEP, the Regional Air
Pollution Control Agency (RAPCA), and CARB, expressed the importance of
existing CAFE standards in meeting the NAAQS.\2216\ The Northeast
States for Coordinated Air Use Management (NESCAUM) also asserted that
regulation and reduction of GHG was necessary to meet the NAAQS, and
``[o]ur states recognize the urgent need to reduce GHG emissions across
all sectors of our economy.'' \2217\ Similarly, the agencies from
Minnesota stated that ``[t]he existing standards are critical for
states to attain and maintain the NAAQS because vehicles account for
about 24% of Minnesota's overall air pollution emissions.'' \2218\ The
Pima County Department of Environmental quality stated that
``[f]reezing emission reductions for six years could put this region in
jeopardy of being designated as non-attainment of the ozone standard
and impact the health of many of our most vulnerable residents.''
\2219\ The Washington State Department of Ecology stated that increases
in NOX and VOC would increase ozone levels in two areas at
rise of ozone nonattainment in the Puget Sound and the Tri-Cities.''
\2220\ The Pennsylvania Department of Environmental Protection stated
``[r]emoving currently realized emissions reductions and forgoing
future achievable emissions reductions may make it more difficult for
areas to attain and maintain the NAAQS. PADEP relies on emission
reductions from mobile sources as part of its SIP planning to attain
and maintain the
[[Page 24858]]
NAAQS.'' \2221\ The North Carolina Department of Environmental Quality
asserted that based on modeling analysis conducted by NCDEQ, ``we
believe that the fleet changes predicted by the CAFE modeling would
lead to emissions increases that would interfere with the ability of
some ozone maintenance areas to meet transportation conformity budgets
and maintain compliance with the NAAQS.'' \2222\
---------------------------------------------------------------------------
\2216\ PA DEP, NHTSA-2018-0067-11956, RAPCA NHTSA-2018-0067-
11620, and CARB NHTSA-2018-0067-11873.
\2217\ NESAUM, NHTSA-2018-0067-11691.
\2218\ Minnesota Pollution Control Agency(MPCA), the Minnesota
Department of Transportation (MnDOT), and the Minnesota Department
of Health(MDH), NHTSA-2018-0067-11706.
\2219\ Pima County Department of Environmental Quality, NHTSA-
2018-0067-11876.
\2220\ Washington State Department of Ecology, NHTSA-2018-0067-
11926.
\2221\ PA DEP, NHTSA-2018-0067-11956.
\2222\ North Carolina Department of Environmental Quality,
NHTSA-2018-0067-12025.
---------------------------------------------------------------------------
Many State commenters also expressed concern about their ability to
conform with their State Implementation Plan (SIP) after this rule, as
the Federal vehicle emissions standards previously set were
incorporated into the SIPs and a rollback could result in further
increased emissions.\2223\ CARB stated that its ``2016 SIP calls for
reducing NOX emissions by approximately 6 tons per day,''
and according to CARB, the proposed rule would not allow California to
achieve its South Coast SIP commitments without dramatic
countermeasures to reduce emissions elsewhere.\2224\ Similarly, other
agencies expressed concern about SIP requirements, such as PA DEP, who
stated that ``[b]y flatlining emissions standards at the MY 2020 level,
the agencies' Proposed Rule increases vehicle emissions. The Proposed
Rule would interfere with Pennsylvania's SIP planning requirements.''
\2225\
---------------------------------------------------------------------------
\2223\ CARB NHTSA-2018-0067-11873, SCAQMD NHTSA-2018-0067-11813,
NESCAUM NHTSA-2018-0067-11691, Joint Submission from Colorado local
governments NHTSA-2018-0067-11929, PA DEP NHTSA-2018-0067-11956, and
Joint Submission from the States of California et al. and the Cities
of Oakland et al. NHTSA-2018-0067-11735.
\2224\ CARB NHTSA-2018-0067-11873.
\2225\ PA DEP NHTSA-2018-0067-11956.
---------------------------------------------------------------------------
The commenters expressed concerns that this final rule will present
challenges in fulfilling existing SIP requirements and in attaining or
maintaining the NAAQS, resulting in the need for emission reductions to
offset increases due to this rule. This final rulemaking predominantly
addresses fuel economy and CO2 emissions of the light-duty
vehicle fleet. It does not affect EPA's Tier 3 vehicle and gasoline
(Tier 3) standards or California's low emission vehicle III (LEV III)
emission standards. Tier 3 and LEV III regulations are predominantly
responsible for regulating criteria pollutant emissions (e.g.
NOX, VOCs, and carbon monoxide) from light-duty vehicles.
While this final rulemaking will result in increases in the amount of
gasoline produced, the number of vehicle re-fueling events and
emissions of certain criteria pollutants and precursors the emissions
impact will vary from area to area depending on factors such as the
composition of the local vehicle fleet and the amount of gasoline
produced in the area. The agencies expect that states will evaluate any
adverse emissions or air quality impacts that result from the
finalization of this rule in the context of state implementation plan
development for relevant NAAQS, such as the relevant ozone and
PM2.5 NAAQS.
CARB, the joint submission from the States of California and Cities
of Oakland, and other commenters also stated that the rulemaking
``fails to meet the general conformity requirements under the Clean Air
Act.'' \2226\ Similarly, the Center for Biological Diversity, et al.,
stated ``it is highly unlikely that the Proposal would not violate
general conformity.'' \2227\ The states and cities expressed that the
General Conformity rule applies to this action because ``[f]irst, an
increase in criteria pollutants is reasonably foreseeable as the
agencies quantified those emissions as part of this rulemaking. Second,
the agencies can practically control those emissions as they possess
ultimate regulatory authority over standards that govern vehicle
operation.'' \2228\ CARB stated ``NHTSA's determination regarding its
own conformity obligations . . . does not address conformity-related
obligations EPA may have that flow from the joint rulemaking.'' \2229\
SCAQMD similarly stated that ``EPA counts as a federal agency that must
comply with general conformity requirements. The proposal leaves
unclear whether EPA also determined its actions comply with the general
conformity requirements under 40 CFR 93.150 and general conformity SIP
revisions allowed under 40 CFR 51.851.'' \2230\ SCAQMD concluded that
EPA must make its own conformity determination, ``and it is not clear
that EPA can rely on NHTSA's analysis given its dissimilar position in
having continuing program responsibility over mobile source
emissions.'' \2231\
---------------------------------------------------------------------------
\2226\ CARB, NHTSA-2018-0067-11873, Joint Submission from States
of California and Cities of Oakland, NHTSA-2018-0067-11735.
\2227\ Center for Biological Diversity, et al., NHTSA-2018-0067-
12123.
\2228\ Joint Submission from States of California and Cities of
Oakland, NHTSA-2018-0067-11735.
\2229\ CARB, NHTSA-2018-0067-11873.
\2230\ SCAQMD, NHTSA_2018-0067-11813.
\2231\ SCAQMD, NHTSA_2018-0067-11813.
---------------------------------------------------------------------------
EPA and NHTSA disagree with the commenters that this rule is
subject to the CAA section 176(c) conformity requirement and the
General Conformity regulations. A General Conformity evaluation is
required for a general Federal action proposed to occur within specific
nonattainment or maintenance areas. For a General Conformity evaluation
to be necessary, the action must cause emissions of the criteria and
precursor pollutants for which the areas are nonattainment or
maintenance, and the emissions must originate within those areas.
Further, the evaluation would require a demonstration that the action
conforms to a specific State Implementation Plan's strategy for air
pollution prevention and control applicable to the nonattainment and
maintenance areas. In addition, any mitigation or offsets required to
demonstrate conformity may require written commitments that must be
fulfilled, and offsets must occur during the same calendar year as the
emission increases from the action.
While the EPA established the framework of methods and procedures
that Federal agencies must follow when General Conformity applies to
their actions, it is the responsibility of each Federal agency to
prepare its own General Conformity evaluation for actions the agency
supports, funds, permits or approves. When the EPA functions as a lead
agency for actions that are subject to General Conformity, such as
water projects, and the agency may issue permits or approve actions
that require a General Conformity evaluation, EPA is responsible for
and sometimes is required to prepare its own General Conformity
evaluation. For the reasons specified here and in Section X.E.2, a
General Conformity evaluation is not necessary for either agency.
As stated in section 4.1.1.4 of the DEIS and in section 4.1.1.4 of
the FEIS, the agencies do not believe the proposed rule would result in
either direct or indirect emissions as defined for General Conformity
at 40 CFR 93.152 or as required for applicability of the rule under
section 93.153(b). Furthermore, as described in the proposal, emissions
from operation of vehicles produced during the model years covered by
this rule, while reasonably foreseeable, cannot be quantified with any
certainty in any particular nonattainment or maintenance area. In
addition, while the emissions rates from MY 2021-2026 vehicles are
projected for future years in this rule, neither NHTSA nor EPA has
control over where, when or how many of the vehicles will operate
during a given future year or within a certain geographical area.
Therefore, the emissions are not quantifiable. Furthermore, the General
Conformity
[[Page 24859]]
applicability analysis requires an analytical comparison of the
emissions from MY 2021-2026 vehicles in some specific nonattainment or
maintenance area in a specific future year, to the emissions projected
from the operation of vehicles produced in other model years that would
otherwise operate in that same area in the same future year. Without
the identity of the future year vehicle fleet by type/make/model (which
depends on a specific nonattainment or maintenance location and year),
the net emissions, or total of direct and indirect emissions, cannot be
quantified. Thus, this rule, in and of itself, is not subject to a
General Conformity evaluation.
CARB stated that this rulemaking would, if finalized, invalidate
the model underlying California's SIPs (the EMFAC 2014 model), which
would result in the SIPs being disapproved by EPA.\2232\ CARB expressed
further concern that as a result of the Clean Air Act's conformity
requirements, this disapproval would put significant limits on new
RTPs, TIPS, or regionally significant transportation projects being
adopted or approved in California.\2233\
---------------------------------------------------------------------------
\2232\ CARB, NHTSA-2018-0067-11873.
\2233\ CARB, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
The commenter expressed the opinion that if this rule is finalized,
EPA would disapprove its SIPs because its on-road emission factor model
(EMFAC) would be invalidated. The commenter also opined that such
disapprovals would limit the ability of metropolitan planning
organizations in California to make transportation conformity
determinations for metropolitan transportation plans, transportation
improvement programs and certain transportation projects. It is
premature to assume that EPA will disapprove SIPs because they are
based on EMFAC2014 or EMFAC2017. EPA will evaluate and address, as
appropriate, the impact of the SAFE action on future SIP approval
actions EMFAC2014 and EMFAC2017 remain approved emission factor models
for SIPs and transportation conformity analyses in California. EPA is
aware that California released adjustment factors to be applied to
EMFAC2014 and EMFAC2017 model results to account for impacts of the
SAFE Part 1 rule for on-road criteria pollutant emissions from light-
duty vehicles. EPA will work with CARB and DOT on the appropriate
implementation of federal requirements based on current and available
information.
Because passenger cars and light trucks are subject to gram-per-
mile emissions standards for criteria pollutants, more fuel-efficient
(and, correspondingly, less CO2-intensive) vehicles are not,
from the standpoint of air quality, ``cleaner'' vehicles. Therefore, to
the extent that CAFE/CO2 standards lead to changes in
overall quantities of vehicular emissions that impact air quality,
these are dominated by induced changes in highway travel. Changes in
overall fuel consumption do lead to changes in emissions from
``upstream'' processes involved in supplying fuel to vehicles.
Depending on how total vehicular emissions and total upstream emissions
change in response to less stringent standards, overall emissions could
increase or decrease. While small in magnitude, net impacts could also
vary considerably among different geographic areas. In other words,
CAFE and CO2 standards impact fuel consumption and
CO2 emissions in ways that are direct and unambiguous, and
impact air quality in ways that are indirect and ambiguous.
The following sections, included in prior rules setting fuel
economy and CO2 standards and updated based on EPA's latest
scientific assessments, describe the criteria and air toxics considered
in this rule, and their health and environmental effects. Additionally,
the section that follows describes how the estimated effects of each
pollutant were modeled in this rulemaking. Section VII discusses the
interactions between upstream, tailpipe, and highway travel that result
in the net emissions of criteria and air toxic pollutants estimated as
a result of this rule.
(1) Particulate Matter
(a) Background
Particulate matter (PM) is a complex mixture of solid particles and
liquid droplets distributed among numerous atmospheric gases which
interact with solid and liquid phases. Particles range in size from
those smaller than 1 nanometer (10-\9\ meter) to over 100
micrometers ([micro]m, or 10-\6\ meter) in diameter (for
reference, a typical strand of human hair is 50-70 [micro]m in diameter
and a grain of fine beach sand is about typically 90 [micro]m in
diameter). Atmospheric particles can be grouped into several classes
according to their aerodynamic and physical sizes. Generally, the three
broad classes of particles include ultrafine particles (UFPs, generally
considered as particulates with a diameter less than or equal to 0.1
[micro]m [typically based on physical size, thermal diffusivity or
electrical mobility]), ``fine'' particles (PM2.5; particles
with a nominal mean aerodynamic diameter less than or equal to 2.5
[micro]m), and ``thoracic'' particles (PM10; particles with
a nominal mean aerodynamic diameter less than or equal to 10 [micro]m).
Particles that fall within the size range between PM2.5 and
PM10 are referred to as ``thoracic coarse particles''
(PM10-2.5 particles with a nominal mean aerodynamic diameter
greater than 2.5 [micro]m and less than or equal to 10 [micro]m). EPA
currently has standards that regulate PM2.5 and
PM10.\2234\
---------------------------------------------------------------------------
\2234\ Regulatory definitions of PM size fractions and
information on reference and equivalent methods for measuring PM in
ambient air are provided in 40 CFR parts 50, 53, and 58. With regard
to national ambient air quality standards (NAAQS) which provide
protection against health and welfare effects, the 24-hour
PM10 standard provides protection against effects
associated with short-term exposure to thoracic coarse particles
(i.e. PM10--2.5).
---------------------------------------------------------------------------
Most particles are found in the lower troposphere, where they can
have residence times ranging from a few hours to weeks. Particles are
removed from the atmosphere by wet deposition, such as when they are
carried by rain or snow, or by dry deposition, when particles settle
out of suspension due to gravity. Atmospheric lifetimes are generally
longest for PM2.5, which often remains in the atmosphere for
days to weeks before being removed by wet or dry deposition. \2235\In
contrast, atmospheric lifetimes for UFP and PM10-2.5 are
shorter. Within hours, UFP can undergo coagulation and condensation
that lead to formation of larger particles in the accumulation mode, or
can be removed from the atmosphere by evaporation, deposition, or
reactions with other atmospheric components. PM10-2.5 are
also generally removed from the atmosphere within hours, through wet or
dry deposition.\2236\
---------------------------------------------------------------------------
\2235\ U.S. EPA. Integrated Science Assessment (ISA) for
Particulate Matter (Final Report. 2019), U.S. Environmental
Protection Agency, Washington DC, EPA/600/R-19/188, 2019. Table 2-1.
\2236\ U.S. EPA. Integrated Science Assessment (ISA) for
Particulate Matter (Final Report, 2019). U.S Environmental
Protection Agency, Washington, DC, EPA/600/R-19/188, 2019. Table 2-
1.
---------------------------------------------------------------------------
Particulate matter consists of both primary and secondary
particles. Primary particles are emitted directly from sources, such as
combustion-related activities (e.g., industrial activities, motor
vehicles, biomass burning), while secondary particles are formed
through atmospheric chemical reactions of gaseous precursors (e.g.,
sulfur oxides (SOx), nitrogen oxides (NOx) and volatile organic
compounds (VOCs) and ammonia). From 2000 to 2017, national annual
average PM2.5 concentrations have declined by over
40%,\2237\ largely reflecting reductions in emissions of precursor
gases.
---------------------------------------------------------------------------
\2237\ See https://www.epa.gov/air-trends/particulate-matter-pm25-trends and https://www.epa.gov/air-trends/particulate-matter-pm25-trends#pmnat for more information.
---------------------------------------------------------------------------
[[Page 24860]]
(b) Health Effects of PM
Scientific evidence spanning animal toxicological, controlled human
exposure, and epidemiologic studies shows that exposure to ambient PM
is associated with a broad range of health effects. The Integrated
Science Assessment for Particulate Matter (PM ISA) (U.S. EPA 2009)
synthesizes the toxicological, clinical and epidemiological evidence to
determine whether each pollutant is causally related to an array of
adverse human health outcomes associated with either acute (i.e., hours
or days-long) or chronic (i.e. years-long) exposure; for each outcome,
the ISA reports this relationship to be causal, likely to be causal,
suggestive of a causal relationship, inadequate to infer a causal
relationship or not likely to be a causal relationship.
In brief, the ISA for PM2.5 found acute exposure to
PM2.5 to be causally related to cardiovascular effects and
mortality (i.e., premature death), and respiratory effects as likely-
to-be-causally related. The ISA identified cardiovascular effects and
total mortality as being causally related to long-term exposure to
PM2.5 and respiratory effects as likely-to-be-causal; and
the evidence was suggestive of a causal relationship for reproductive
and developmental effects as well as cancer, mutagenicity and
genotoxicity. The ISA for ozone found acute exposure to ozone to be
causally related to respiratory effects, a likely-to-be-causal
relationship with cardiovascular effects and total mortality and a
suggestive relationship for central nervous system effects. Among
chronic effects, the ISA reported a likely-to-be-causal relationship
for respiratory outcomes and respiratory mortality, and suggestive
relationship for cardiovascular effects, reproductive and developmental
effects, central nervous system effects, and total mortality. DOT
follows EPA's approach of estimating the incidence of air pollution
effects for those health effects above where the ISA classified as
either causal or likely-to-be-causal.
EPA's more recent Integrated Science Assessment for Particulate
Matter (PM ISA), which was finalized in December 2019,\2238\ summarizes
the most recent health effects evidence for short- and long-term
exposures to PM2.5, PM10-2.5, and ultrafine
particles, characterizing the strength of the evidence and whether the
relationship is likely to be causal nature in nature. The 2019 P.M. ISA
reinforces the findings of the 2009 ISA, and supports the decision to
continue monetizing the respiratory and cardiovascular health endpoints
monetized in the current analysis. EPA is currently in the process of
considering how the 2019 ISA and eventual decision by the Administrator
regarding the National Ambient Air Quality Standards for particulate
matter will be used to update forthcoming regulatory impact analysis.
---------------------------------------------------------------------------
\2238\ U.S. EPA. Integrated Science Assessment (ISA) for
Particulate Matter (Final Report, 2019). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-19/188, 2019.
---------------------------------------------------------------------------
(c) Current Concentrations
There are two primary NAAQS for PM2.5: an annual
standard (12.0 micrograms per cubic meter ([mu]g/m\3\)) set in 2012 and
a 24-hour standard (35 [mu]g/m\3\) set in 2006, and two secondary NAAQS
for PM2.5: an annual standard (15.0 [mu]g/m\3\) set in 1997
and a 24-hour standard (35 [mu]g/m\3\) set in 2006.\2239\
---------------------------------------------------------------------------
\2239\ The EPA is currently reviewing the PM NAAQS and
anticipates completing this review in late 2020 Available at https://www.epa.gov/naaqs/particulate-matter-pm-air-quality-standards).
---------------------------------------------------------------------------
There are many areas of the country that are currently in
nonattainment for the annual and 24-hour primary PM2.5
NAAQS. As of January 31, 2020, more than 19 million people lived in the
4 areas that are designated as nonattainment for the 1997 annual
PM2.5 NAAQS. These PM2.5 nonattainment areas are
comprised of 14 full or partial counties. As of January 31, 2020, 6
areas are designated as nonattainment for the 2012 annual
PM2.5 NAAQS; these areas are composed of 16 full or partial
counties with a population of more than 20 million. As of January 31,
2020, 14 areas are designated as nonattainment for the 2006 24-hour
PM2.5 NAAQS; these areas are composed of 41 full or partial
counties with a population of more than 31 million. In total, there are
currently 17 PM2.5 nonattainment areas with a population of
more than 32 million people.
The EPA has already adopted many mobile source emission control
programs that are expected to reduce ambient PM concentrations. As a
result of these and other federal, state and local programs, the number
of areas that fail to meet the PM2.5 NAAQS in the future is
expected to decrease. However, even with the implementation of all
current state and federal regulations, there are projected to be
counties violating the PM2.5 NAAQS well into the future.
(2) Ozone
(a) Background
Ground-level ozone pollution is typically formed through reactions
involving VOC and NOX in the lower atmosphere in the
presence of sunlight. These pollutants, often referred to as ozone
precursors, are emitted by many types of sources, such as highway and
nonroad motor vehicles and engines, power plants, chemical plants,
refineries, makers of consumer and commercial products, industrial
facilities, and smaller area sources.
The science of ozone formation, transport, and accumulation is
complex. Ground-level ozone is produced and destroyed in a cyclical set
of chemical reactions, many of which are sensitive to temperature and
sunlight. When ambient temperatures and sunlight levels remain high for
several days and the air is relatively stagnant, ozone and its
precursors can build up and result in more ozone than typically occurs
on a single high-temperature day. Ozone and its precursors can be
transported hundreds of miles downwind from precursor emissions,
resulting in elevated ozone levels even in areas with low local VOC or
NOX emissions.
(b) Health Effects of Ozone
This section provides a summary of the health effects associated
with exposure to ambient concentrations of ozone.\2240\ The information
in this section is based on the information and conclusions in the
February 2013 Integrated Science Assessment for Ozone (Ozone ISA),
which formed the basis for EPA's revision to the primary and secondary
standards in 2015.\2241\ The Ozone ISA concludes that human exposures
to ambient concentrations of ozone are associated with a number of
adverse health effects and characterizes the weight of evidence for
these health effects.\2242\ The discussion below
[[Page 24861]]
highlights the Ozone ISA's conclusions pertaining to health effects
associated with both short-term and long-term periods of exposure to
ozone.
---------------------------------------------------------------------------
\2240\ Human exposure to ozone varies over time due to changes
in ambient ozone concentration and because people move between
locations which have notable different ozone concentrations. Also,
the amount of ozone delivered to the lung is not only influenced by
the ambient concentrations but also by the individuals breathing
route and rate.
\2241\ U.S. EPA. Integrated Science Assessment of Ozone and
Related Photochemical Oxidants (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013. The ISA
is available at http://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=247492#Download.
\2242\ The ISA evaluates evidence and draws conclusions on the
causal nature of relationship between relevant pollutant exposures
and health effects, assigning one of five ``weight of evidence''
determinations: causal relationship, likely to be a causal
relationship, suggestive of, but not sufficient to infer, a causal
relationship, inadequate to infer a causal relationship, and not
likely to be a causal relationship. For more information on these
levels of evidence, please refer to Table II in the Preamble of the
ISA.
---------------------------------------------------------------------------
For short-term exposure to ozone, the Ozone ISA concludes that
respiratory effects, including lung function decrements, pulmonary
inflammation, exacerbation of asthma, respiratory-related hospital
admissions, and mortality, are causally associated with ozone exposure.
It also concludes that cardiovascular effects, including decreased
cardiac function and increased vascular disease, and total mortality
are likely to be causally associated with short-term exposure to ozone
and that evidence is suggestive of a causal relationship between
central nervous system effects and short-term exposure to ozone.
For long-term exposure to ozone, the Ozone ISA concludes that
respiratory effects, including new onset asthma, pulmonary inflammation
and injury, are likely to be causally related with ozone exposure. The
Ozone ISA characterizes the evidence as suggestive of a causal
relationship for associations between long-term ozone exposure and
cardiovascular effects, reproductive and developmental effects, central
nervous system effects and total mortality. The evidence is inadequate
to infer a causal relationship between chronic ozone exposure and
increased risk of lung cancer.
Finally, inter-individual variation in human responses to ozone
exposure can result in some groups being at increased risk for
detrimental effects in response to exposure. In addition, some groups
are at increased risk of exposure due to their activities, such as
outdoor workers or children. The Ozone ISA identified several groups
that are at increased risk for ozone-related health effects. These
groups are people with asthma, children and older adults, individuals
with reduced intake of certain nutrients (i.e., Vitamins C and E),
outdoor workers, and individuals having certain genetic variants
related to oxidative metabolism or inflammation. Ozone exposure during
childhood can have lasting effects through adulthood. Such effects
include altered function of the respiratory and immune systems.
Children absorb higher doses (normalized to lung surface area) of
ambient ozone, compared to adults, due to their increased time spent
outdoors, higher ventilation rates relative to body size, and a
tendency to breathe a greater fraction of air through the mouth.
Children also have a higher asthma prevalence compared to adults.
(c) Current Concentrations
The primary and secondary NAAQS for ozone are 8-hour standards with
a level of 0.07 ppm. The most recent revision to the ozone standards
was in 2015; the previous 8-hour ozone primary standard, set in 2008,
had a level of 0.075 ppm.\2243\ As of January 31, 2020, there were 36
ozone nonattainment areas for the 2008 ozone NAAQS, composed of 153
full or partial counties, with a population of more than 99 million. As
of January 31, 2020, there were 51 ozone nonattainment areas for the
2015 ozone NAAQS, composed of 206 full or partial countries, with a
population of more than 122 million. In total, there are currently 59
ozone nonattainment areas with a population of more than 127 million
people.
---------------------------------------------------------------------------
\2243\ The EPA is currently reviewing the PM NAAQS and
anticipates completing this review in late 2020 Available at
(https://www.epa.gov/naaqs/ozone-o3-air-quality-standards).
---------------------------------------------------------------------------
States with ozone nonattainment areas are required to take action
to bring those areas into attainment. The attainment date assigned to
an ozone nonattainment area is based on the area's classification. The
attainment dates for areas designated nonattainment for the 2008 8-hour
ozone NAAQS are in the 2015 to 2032 timeframe, depending on the
severity of the problem in each area. Nonattainment area attainment
dates associated with areas designated for the 2015 NAAQS will be in
the 2021-2038 timeframe, depending on the severity of the problem in
each area.
EPA has already adopted many emission control programs that are
expected to reduce ambient ozone levels. As a result of these and other
federal, state and local programs, 8-hour ozone levels are expected to
improve in the future. However, even with the implementation of all
current state and federal regulations, there are projected to be
counties violating the ozone NAAQS well into the future.
(3) Nitrogen Oxides
(a) Background
Oxides of nitrogen (NOX) refers to nitric oxide and
nitrogen dioxide (NO2). For the NOX NAAQS,
NO2 is the indicator. Most NO2 is formed in the
air through the oxidation of nitric oxide (NO) emitted when fuel is
burned at a high temperature. NOX is also a major
contributor to secondary PM2.5 formation. NOX and
VOC are the two major precursors of ozone.
(b) Health Effects of Nitrogen Oxides
The most recent review of the health effects of oxides of nitrogen
completed by EPA can be found in the 2016 Integrated Science Assessment
for Oxides of Nitrogen--Health Criteria (Oxides of Nitrogen ISA).\2244\
The primary source of NO2 is motor vehicle emissions, and
ambient NO2 concentrations tend to be highly correlated with
other traffic-related pollutants. Thus, a key issue in characterizing
the causality of NO2-health effect relationships was
evaluating the extent to which studies supported an effect of
NO2 that is independent of other traffic-related pollutants.
EPA concluded that the findings for asthma exacerbation integrated from
epidemiologic and controlled human exposure studies provided evidence
that is sufficient to infer a causal relationship between respiratory
effects and short-term NO2 exposure. The strongest evidence
supporting an independent effect of NO2 exposure comes from
controlled human exposure studies demonstrating increased airway
responsiveness in individuals with asthma following ambient-relevant
NO2 exposures. The coherence of this evidence with
epidemiologic findings for asthma hospital admissions and ED visits as
well as lung function decrements and increased pulmonary inflammation
in children with asthma describe a plausible pathway by which
NO2 exposure can cause an asthma exacerbation. The 2016 ISA
for Oxides of Nitrogen also concluded that there is likely to be a
causal relationship between long-term NO2 exposure and
respiratory effects. This conclusion is based on new epidemiologic
evidence for associations of NO2 with asthma development in
children combined with biological plausibility from experimental
studies.
---------------------------------------------------------------------------
\2244\ U.S. EPA. Integrated Science Assessment for Oxides of
Nitrogen--Health Criteria (2016 Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-15/068, 2016.
---------------------------------------------------------------------------
In evaluating a broader range of health effects, the 2016 ISA for
Oxides of Nitrogen concluded evidence is ``suggestive of, but not
sufficient to infer, a causal relationship'' between short-term
NO2 exposure and cardiovascular effects and mortality and
between long-term NO2 exposure and cardiovascular effects
and diabetes, birth outcomes, and cancer. In addition, the scientific
evidence is inadequate (insufficient consistency of epidemiologic and
toxicological evidence) to infer a causal relationship for long-term
NO2 exposure with
[[Page 24862]]
fertility, reproduction, and pregnancy, as well as with postnatal
development. A key uncertainty in understanding the relationship
between these non-respiratory health effects and short- or long-term
exposure to NO2 is copollutant confounding, particularly by
other roadway pollutants. The available evidence for non-respiratory
health effects does not adequately address whether NO2 has
an independent effect or whether it primarily represents effects
related to other or a mixture of traffic-related pollutants.
The 2016 ISA for Oxides of Nitrogen concluded that people with
asthma, children, and older adults are at increased risk for
NO2-related health effects. In these groups and life stages,
NO2 is consistently related to larger effects on outcomes
related to asthma exacerbation, for which there is confidence in the
relationship with NO2 exposure.
(c) Current Concentrations
On April 6, 2018, based on a review of the full body of scientific
evidence, EPA issued a decision to retain the current primary NAAQS for
NO2. The EPA has concluded that the current NAAQS are
requisite to protect the public health, including the at-risk
populations of older adults, children and people with asthma, with an
adequate margin of safety. The primary NAAQS for NO2 are a
one-hour standard with a level of 100 ppb, based on the three-year
average of 98th percentile of the annual distribution of daily maximum
one-hour concentrations, and an annual standard at a level of 53 ppb.
(4) Sulfur Oxides
(a) Background
Sulfur dioxide (SO2), a member of the sulfur oxide
(SOX) family of gases, is formed from burning fuels
containing sulfur (e.g., coal or oil derived), extracting gasoline from
oil, or extracting metals from ore. SO2 and its gas phase
oxidation products can dissolve in water droplets and further oxidize
to form sulfuric acid which reacts with ammonia to form sulfates, which
are important components of ambient PM.
(b) Health Effects of SO2
This section provides an overview of the health effects associated
with SO2. Additional information on the health effects of
SO2 can be found in the 2017 Integrated Science Assessment
for Sulfur Oxides--Health Criteria (SOX ISA).\2245\
Following an extensive evaluation of health evidence from animal
toxicological, controlled human exposure, and epidemiologic studies,
the EPA has concluded that there is a causal relationship between
respiratory health effects and short -term exposure to SO2.
The immediate effect or SO2 on the respiratory system in
humans is bronchoconstriction. People with asthma are more sensitive to
the effects of SO2, likely resulting from preexisting
inflammation associated with this disease. In addition to those with
asthma (both children and adults), there is suggestive evidence that
all children and older adults may be at increased risk of
SO2-related health effects. In free-breathing laboratory
studies involving controlled human exposures to SO2,
respiratory effects have consistently been observed following 5-10 min
exposures at SO2 concentrations >= 400 ppb in people with
asthma engaged in moderate to heavy levels of exercise, with
respiratory effects occurring at concentrations as low as
200 ppb in some individuals with asthma. A clear
concentration-response relationship has been demonstrated in these
studies following exposures to SO2 at concentrations between
200 and 1000 ppb, both in terms of increasing severity of
respiratory symptoms and decrements in lung function, as well as the
percentage of individuals with asthma adversely affected. Epidemiologic
studies have reported positive associations between short-term ambient
SO2 concentrations and hospital admissions and emergency
department visits for asthma and for all respiratory causes,
particularly among children and older adults (>=65 years). The studies
provide supportive evidence for the causal relationship.
---------------------------------------------------------------------------
\2245\ U.S. EPA (2017). Integrated Science Assessment (ISA) for
Sulfur Oxides. Health Criteria (Final Report). EPA 600/R-17/451.
Washington, DC, U.S. EPA.
---------------------------------------------------------------------------
For long-term SO2 exposure and respiratory effects, the
EPA has concluded that the evidence is suggestive or a causal
relationship. This conclusion is based on new epidemiologic evidence
for positive associations between long-term SO2 exposure and
increases in asthma incidence among children, together with animal
toxicological evidence that provides a pathophysiologic basis for the
development of asthma. However, uncertainty remains regarding the
influence of other pollutants on the observed associations with
SO2 because these epidemiologic studies have not examined
the potential for copollutant confounding.
Consistent associations between short-term exposure to
SO2 and mortality have been observed in epidemiologic
studies, with larger effect estimates reported for respiratory
mortality than for cardiovascular mortality. While this finding is
consistent with the demonstrated effects of SO2 on
respiratory morbidity, uncertainty remains with respect to the
interpretation of these observed mortality associations due to
potential confounding by various copollutants. Therefore, the EPA has
concluded that the overall evidence is suggestive of a causal
relationship between short-term exposure to SO2 and
mortality.
(c) Current Concentrations
On February 25, 2019, the EPA announced its decision to retain,
without revision, the existing NAAQS for SOX of 75 ppb, as
the annual 99th percentile of daily maximum SO2
concentrations, averaged over three years (84 FR 9866, March 18, 2019).
The existing primary (health-based) standard provides health protection
for the at-risk group (people with asthma) against respiratory effects
following short-term (e.g., 5-minute) exposures to SO2 in
ambient air. The EPA has been finalizing the initial area designations
for the 2010 SO2 NAAQS in phases and completed designations
for most of the country in December 2017. The EPA is under a court
order to finalize initial designations by December 31, 2020, for a
remaining set of about 50 areas where states have deployed new
SO2 monitoring networks. As of January 31, 2020 there are 34
nonattainment areas for the 2010 SO2 NAAQS. As of January
31, 2020 there also remain eight nonattainment areas for the primary
annual SO2 NAAQS set in 1971.
(5) Carbon Monoxide
(a) Background
Carbon monoxide is a colorless, odorless gas emitted from
combustion processes. Nationally, particularly in urban areas, the
majority of CO emissions to ambient air come from mobile sources.\2246\
---------------------------------------------------------------------------
\2246\ U.S. EPA (2010). Integrated Science Assessment for Carbon
Monoxide (Final Report). U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-09/019F, 2010. Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=218686. See Section
2.1.
---------------------------------------------------------------------------
(b) Health Effects of Carbon Monoxide
Information on the health effects of CO can be found in the January
2010 Integrated Science Assessment for Carbon Monoxide (CO ISA)
associated with the 2010 evaluation of the
[[Page 24863]]
NAAQS.\2247\ The CO ISA presents conclusions regarding the presence of
causal relationships between CO exposure and categories of adverse
health effects. This section provides a summary of the health effects
associated with exposure to ambient concentrations of CO, along with
the ISA conclusions.\2248\
---------------------------------------------------------------------------
\2247\ U.S. EPA (2010). Integrated Science Assessment for Carbon
Monoxide (Final Report). U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-09/019F, 2010. Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=218686.
\2248\ Personal exposure includes contributions from many
sources, and in many different environments. Total personal exposure
to CO includes both ambient and nonambient components; and both
components may contribute to adverse health effects.
---------------------------------------------------------------------------
Controlled human exposure studies of subjects with coronary artery
disease show a decrease in the time to onset of exercise-induced angina
(chest pain) and electrocardiogram changes following CO exposure. In
addition, epidemiologic studies observed associations between short-
term CO exposure and cardiovascular morbidity, particularly increased
emergency room visits and hospital admissions for coronary heart
disease (including ischemic heart disease, myocardial infarction, and
angina). Some epidemiologic evidence is also available for increased
hospital admissions and emergency room visits for congestive heart
failure and cardiovascular disease as a whole. The CO ISA concludes
that a causal relationship is likely to exist between short-term
exposures to CO and cardiovascular morbidity. It also concludes that
available data are inadequate to conclude that a causal relationship
exists between long-term exposures to CO and cardiovascular morbidity.
Animal studies show various neurological effects with in-utero CO
exposure. Controlled human exposure studies report central nervous
system and behavioral effects following low-level CO exposures,
although the findings have not been consistent across all studies. The
CO ISA concludes the evidence is suggestive of a causal relationship
with both short-and long-term exposure to CO and central nervous system
effects.
A number of studies cited in the CO ISA have evaluated the role of
CO exposure in birth outcomes such as preterm birth or cardiac birth
defects. There is limited epidemiologic evidence of a CO-induced effect
on preterm births and birth defects, with weak evidence for a decrease
in birth weight. Animal toxicological studies have found perinatal CO
exposure to affect birth weight, as well as other developmental
outcomes. The CO ISA concludes the evidence is suggestive of a causal
relationship between long-term exposures to CO and developmental
effects and birth outcomes.
Epidemiologic studies provide evidence of associations between
short-term CO concentrations and respiratory morbidity such as changes
in pulmonary function, respiratory symptoms, and hospital admissions. A
limited number of epidemiologic studies considered copollutants such as
ozone, SO2, and PM in two-pollutant models and found that CO
risk estimates were generally robust, although this limited evidence
makes it difficult to disentangle effects attributed to CO itself from
those of the larger complex air pollution mixture. Controlled human
exposure studies have not extensively evaluated the effect of CO on
respiratory morbidity. Animal studies at levels of 50-100 ppm CO show
preliminary evidence of altered pulmonary vascular remodeling and
oxidative injury. The CO ISA concludes that the evidence is suggestive
of a causal relationship between short-term CO exposure and respiratory
morbidity, and inadequate to conclude that a causal relationship exists
between long-term exposure and respiratory morbidity.
Finally, the CO ISA concludes that the epidemiologic evidence is
suggestive of a causal relationship between short-term concentrations
of CO and mortality. Epidemiologic evidence suggests an association
exists between short-term exposure to CO and mortality, but limited
evidence is available to evaluate cause-specific mortality outcomes
associated with CO exposure. In addition, the attenuation of CO risk
estimates which was often observed in copollutant models contributes to
the uncertainty as to whether CO is acting alone or as an indicator for
other combustion-related pollutants. The CO ISA also concludes that
there is not likely to be a causal relationship between relevant long-
term exposures to CO and mortality.
(c) Current Concentrations
There are two primary NAAQS for CO: an 8-hour standard (9 ppm) and
a 1-hour standard (35 ppm). The primary NAAQS for CO were retained in
August 2011. There are currently no CO nonattainment areas; as of
September 27, 2010, all CO nonattainment areas have been predesignated
to attainment.
The past designations were based on the existing community-wide
monitoring network. EPA made an addition to the ambient air monitoring
requirements for CO during the 2011 NAAQS review. Those new
requirements called for CO monitors to be operated near roads in Core
Based Statistical Areas (CBSAs) of 1 million or more persons (76 FR
54294, August 31, 2011).
(6) Diesel Exhaust
(a) Background
Diesel exhaust consists of a complex mixture composed of
particulate matter, carbon dioxide, oxygen, nitrogen, water vapor,
carbon monoxide, nitrogen compounds, sulfur compounds, and numerous
low-molecular-weight hydrocarbons. A number of these gaseous
hydrocarbon components are individually known to be toxic, including
aldehydes, benzene and 1,3-butadiene. The diesel particulate matter
present in diesel exhaust consists mostly of fine particles (< 2.5
[micro]m), of which a significant fraction is ultrafine particles (<
0.1 [micro]m). These particles have a large surface area which makes
them an excellent medium for adsorbing organics, and their small size
makes them highly respirable. Many of the organic compounds present in
the gases and on the particles, such as polycyclic organic matter, are
individually known to have mutagenic and carcinogenic properties.
Diesel exhaust varies significantly in chemical composition and
particle sizes between different engine types (heavy-duty, light-duty),
engine operating conditions (idle, acceleration, deceleration), and
fuel formulations (high/low sulfur fuel). Also, there are emissions
differences between on-road and nonroad engines because the nonroad
engines are generally of older technology. After being emitted in the
engine exhaust, diesel exhaust undergoes dilution as well as chemical
and physical changes in the atmosphere. The lifetime for some of the
compounds present in diesel exhaust ranges from hours to days.
(b) Health Effects of Diesel Exhaust
In EPA's 2002 Diesel Health Assessment Document (Diesel HAD),
exposure to diesel exhaust was classified as likely to be carcinogenic
to humans by inhalation from environmental exposures, in accordance
with the revised draft 1996/1999 EPA cancer
guidelines.2249 2250 A number of
[[Page 24864]]
other agencies (National Institute for Occupational Safety and Health,
the International Agency for Research on Cancer, the World Health
Organization, California EPA, and the U.S. Department of Health and
Human Services) had made similar hazard classifications prior to 2002.
EPA also concluded in the 2002 Diesel HAD that it was not possible to
calculate a cancer unit risk for diesel exhaust due to limitations in
the exposure data for the occupational groups or the absence of a dose-
response relationship.
---------------------------------------------------------------------------
\2249\ U.S. EPA. (1999). Guidelines for Carcinogen Risk
Assessment. Review Draft. NCEA-F-0644, July. Washington, DC: U.S.
EPA. Retrieved on March 19, 2009 from http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=54932.
\2250\ U.S. EPA (2002). Health Assessment Document for Diesel
Engine Exhaust. EPA/600/8-90/057F Office of Research and
Development, Washington DC. Retrieved on March 17, 2009 from http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060. pp. 1-1 & 1-2.
---------------------------------------------------------------------------
In the absence of a cancer unit risk, the Diesel HAD sought to
provide additional insight into the significance of the diesel exhaust
cancer hazard by estimating possible ranges of risk that might be
present in the population. An exploratory analysis was used to
characterize a range of possible lung cancer risk. The outcome was that
environmental risks of cancer from long-term diesel exhaust exposures
could plausibly range from as low as 10-5 to as high as
10-3. Because of uncertainties, the analysis acknowledged
that the risks could be lower than 10-5, and a zero risk
from diesel exhaust exposure could not be ruled out.
Non-cancer health effects of acute and chronic exposure to diesel
exhaust emissions are also of concern to EPA. EPA derived a diesel
exhaust reference concentration (RfC) from consideration of four well-
conducted chronic rat inhalation studies showing adverse pulmonary
effects. The RfC is 5 [micro]g/m\3\ for diesel exhaust measured as
diesel particulate matter. This RfC does not consider allergenic
effects such as those associated with asthma or immunologic or the
potential for cardiac effects. There was emerging evidence in 2002,
discussed in the Diesel HAD, that exposure to diesel exhaust can
exacerbate these effects, but the exposure-response data were lacking
at that time to derive an RfC based on these then-emerging
considerations. The EPA Diesel HAD stated, ``With [diesel particulate
matter] being a ubiquitous component of ambient PM, there is an
uncertainty about the adequacy of the existing [diesel exhaust]
noncancer database to identify all of the pertinent [diesel exhaust]-
caused noncancer health hazards.'' The Diesel HAD also noted ``that
acute exposure to [diesel exhaust] has been associated with irritation
of the eye, nose, and throat, respiratory symptoms (cough and phlegm),
and neurophysiological symptoms such as headache, lightheadedness,
nausea, vomiting, and numbness or tingling of the extremities.'' The
Diesel HAD noted that the cancer and noncancer hazard conclusions
applied to the general use of diesel engines then on the market and as
cleaner engines replace a substantial number of existing ones, the
applicability of the conclusions would need to be reevaluated.
It is important to note that the Diesel HAD also briefly summarized
health effects associated with ambient PM and discusses EPA's then-
annual PM2.5 NAAQS of 15 [micro]g/m\3\. In 2012, EPA revised
the annual PM2.5 NAAQS to 12 [micro]g/m\3\. There is a large
and extensive body of human data showing a wide spectrum of adverse
health effects associated with exposure to ambient PM, of which diesel
exhaust is an important component. The PM2.5 NAAQS is
designed to provide protection from the noncancer health effects and
premature mortality attributed to exposure to PM2.5. The
contribution of diesel PM to total ambient PM varies in different
regions of the country and also, within a region, from one area to
another. The contribution can be high in near-roadway environments, for
example, or in other locations where diesel engine use is concentrated.
Since 2002, several new studies have been published which continue
to report increased lung cancer risk with occupational exposure to
diesel exhaust from older engines. Of particular note since 2011 are
three new epidemiology studies which have examined lung cancer in
occupational populations, for example, truck drivers, underground
nonmetal miners and other diesel motor-related occupations. These
studies reported increased risk of lung cancer with exposure to diesel
exhaust with evidence of positive exposure-response relationships to
varying degrees.2251 2252 2253 These newer studies (along
with others that have appeared in the scientific literature) add to the
evidence EPA evaluated in the 2002 Diesel HAD and further reinforces
the concern that diesel exhaust exposure likely poses a lung cancer
hazard. The findings from these newer studies do not necessarily apply
to newer technology diesel engines because the newer engines have large
reductions in the emission constituents compared to older technology
diesel engines.
---------------------------------------------------------------------------
\2251\ Garshick, Eric, Francine Laden, Jaime E. Hart, Mary E.
Davis, Ellen A. Eisen, and Thomas J. Smith. 2012. Lung cancer and
elemental carbon exposure in trucking industry workers.
Environmental Health Perspectives 120(9), 1301-06.
\2252\ Silverman, D.T., Samanic, C.M., Lubin, J.H., Blair, A.E.,
Stewart, P.A., Vermeulen, R., & Attfield, M.D. (2012). The diesel
exhaust in miners study: a nested case-control study of lung cancer
and diesel exhaust. Journal of the National Cancer Institute.
\2253\ Olsson, Ann C., et al. ``Exposure to diesel motor exhaust
and lung cancer risk in a pooled analysis from case-control studies
in Europe and Canada.'' American journal of respiratory and critical
care medicine 183.7 (2011): 941-48.
---------------------------------------------------------------------------
In light of the growing body of scientific literature evaluating
the health effects of exposure to diesel exhaust, in June 2012 the
World Health Organization's International Agency for Research on Cancer
(IARC), a recognized international authority on the carcinogenic
potential of chemicals and other agents, evaluated the full range of
cancer-related health effects data for diesel engine exhaust. IARC
concluded that diesel exhaust should be regarded as ``carcinogenic to
humans.'' \2254\ This designation was an update from its 1988
evaluation that considered the evidence to be indicative of a
``probable human carcinogen.''
---------------------------------------------------------------------------
\2254\ IARC (International Agency for Research on Cancer)
(2013). Diesel and gasoline engine exhausts and some nitroarenes.
IARC Monographs Volume 105. Available at http://monographs.iarc.fr/ENG/Monographs/vol105/index.php.
---------------------------------------------------------------------------
(c) Current Concentrations
Because DPM is part of overall ambient PM and cannot be easily
distinguished from overall PM, the agencies do not have direct
measurements of DPM in the ambient air. DPM concentrations are
estimated using ambient air quality modeling based on DPM emission
inventories. DPM emission inventories are computed as the exhaust PM
emissions from mobile sources combusting diesel or residual oil fuel.
DPM concentrations were recently estimated as part of the 2014 NATA.
Areas with high concentrations are clustered in the Northeast, Great
Lake States, California, and the Gulf Coast States and are also
distributed throughout the rest of the U.S.
(7) Air Toxics
(a) Background
Light-duty vehicle emissions contribute to ambient levels of air
toxics that are known or suspected human or animal carcinogens, or that
have noncancer health effects. The population experiences an elevated
risk of cancer and other noncancer health effects from exposure to the
class of pollutants known collectively as ``air toxics.'' \2255\ These
compounds include, but are not limited to, benzene, 1,3-
[[Page 24865]]
butadiene, formaldehyde, acetaldehyde, acrolein, polycyclic organic
matter, and naphthalene. These compounds were identified as national or
regional risk drivers or contributors in the 2014 or past National-
scale Air Toxics Assessment and have significant inventory
contributions from mobile sources.2256 2257
---------------------------------------------------------------------------
\2255\ U.S. EPA (2015). Summary of Results for the 2011
National-Scale Assessment. http://www3.epa.gov/sites/production/files/2015-12/documents/2011-nata-summary-results.pdf.
\2256\ U.S EPA (2018) Technical Support Document EPA's 2014
National Air Toxics Assessment. Available at https://www.epa.gov/national-air-toxics-assessment/2014-nata-assessment-results.
\2257\ U.S. EPA (2015). 2011 National Air Toxics Assessment.
http://www3.epa.gov/national-air-toxics-assessment/2011-national-air-toxics-assessment.
---------------------------------------------------------------------------
(b) Benzene
EPA's Integrated Risk Information System (IRIS) database lists
benzene as a known human carcinogen (causing leukemia) by all routes of
exposure, and concludes that exposure is associated with additional
health effects, including genetic changes in both humans and animals
and increased proliferation of bone marrow cells in
mice.2258 2259 2260 EPA states in its IRIS database that
data indicate a causal relationship between benzene exposure and acute
lymphocytic leukemia and suggest a relationship between benzene
exposure and chronic non-lymphocytic leukemia and chronic lymphocytic
leukemia. EPA's IRIS documentation for benzene also lists a range of
2.2 x 10-6 to 7.8 x10-6 per [micro]g/m\3\ as the unit risk estimate
(URE) for benzene.2261 2262 The International Agency for
Research on Cancer (IARC) has determined that benzene is a human
carcinogen and the U.S. Department of Health and Human Services (DHHS)
has characterized benzene as a known human
carcinogen.2263 2264
---------------------------------------------------------------------------
\2258\ U.S. EPA. (2000). Integrated Risk Information System File
for Benzene. This material is available electronically at: http://www3.epa.gov/iris/subst/0276.htm.
\2259\ International Agency for Research on Cancer, IARC
monographs on the evaluation of carcinogenic risk of chemicals to
humans, Volume 29, some industrial chemicals and dyestuffs,
International Agency for Research on Cancer, World Health
Organization, Lyon, France 1982.
\2260\ Irons, R.D.; Stillman, W.S.; Colagiovanni, D.B.; Henry,
V.A. (1992). Synergistic action of the benzene metabolite
hydroquinone on myelopoietic stimulating activity of granulocyte/
macrophage colony-stimulating factor in vitro, Proc. Natl. Acad.
Sci. 89:3691-3695.
\2261\ A unit risk estimate is defined as the increase in the
lifetime risk of an individual who is exposed for a lifetime to 1
[micro]g/m\3\ benzene in air.
\2262\ U.S. EPA (2000). Integrated Risk Information System File
for Benzene. This material is available electronically at: http://www3.epa.gov/iris/subst/0276.htm.
\2263\ International Agency for Research on Cancer (IARC, 2018.
Monographs on the evaluation of carcinogenic risks to humans, volume
120. World Health Organization--Lyon France. Available at http://publications.iarc.fr/Book-And-ReportSeries/Iarc-Monographs-On-The-ldentification-Of-Carcinogenic-Hazards-To-Humans/Benzene-2018.
\2264\ NTP (National Toxicology Program). 2016. Report on
Carcinogens, Fourteenth Edition.; Research Triangle Park, NC: U.S.
Department of Health and Human Services Public Health Service.
Available at https://ntp.niehs.nih.gov/go/roc.
---------------------------------------------------------------------------
A number of adverse noncancer health effects including blood
disorders, such as pre- leukemia and aplastic anemia, have also been
associated with long-term exposure to benzene. The most sensitive
noncancer effect observed in humans, based on current data, is the
depression of the absolute lymphocyte count in blood. EPA's inhalation
reference concentration (RfC) for benzene is 30 [micro]g/m\3\. The RfC
is based on suppressed absolute lymphocyte counts seen in humans under
occupational exposure conditions. In addition, recent work, including
studies sponsored by the Health Effects Institute, provides evidence
that biochemical responses are occurring at lower levels of benzene
exposure than previously known.2265 2266 2267 2268 EPA's
IRIS program has not yet evaluated these new data. EPA does not
currently have an acute reference concentration for benzene. The Agency
for Toxic Substances and Disease Registry (ATSDR) Minimal Risk Level
(MRL) for acute exposure to benzene is 29 [micro]g/m\3\ for 1-14 days
exposure.
---------------------------------------------------------------------------
\2265\ Qu, O.; Shore, R.; Li, G.; Jin, X.; Chen, C.L.; Cohen,
B.; Melikian, A.; Eastmond, D.; Rappaport, S.; Li, H.; Rupa, D.;
Suramaya, R.; Songnian, W.; Huifant, Y.; Meng, M.; Winnik, M.; Kwok,
E.; Li, Y.; Mu, R.; Xu, B.; Zhang, X.; Li, K. (2003). HEI Report
115, Validation & Evaluation of Biomarkers in Workers Exposed to
Benzene in China.
\2266\ Qu, Q., R. Shore, G. Li, X. Jin, L.C. Chen, B. Cohen, et
al. (2002). Hematological changes among Chinese workers with a broad
range of benzene exposures. Am. J. Industr. Med. 42: 275-285.
\2267\ Lan, Qing, Zhang, L., Li, G., Vermeulen, R., et al.
(2004). Hematotoxically in Workers Exposed to Low Levels of Benzene.
Science 306: 1774-1776.
\2268\ Turtletaub, K.W. and Mani, C. (2003). Benzene metabolism
in rodents at doses relevant to human exposure from Urban Air.
Research Reports Health Effect Inst. Report No.113.
---------------------------------------------------------------------------
(c) 1,3-Butadiene
EPA has characterized 1,3-butadiene as carcinogenic to humans by
inhalation.2269 2270 The IARC has determined that 1,3-
butadiene is a human carcinogen and the U.S. DHHS has characterized
1,3-butadiene as a known human
carcinogen.2271 2272 2273 2274 There are numerous studies
consistently demonstrating that 1,3-butadiene is metabolized into
genotoxic metabolites by experimental animals and humans. The specific
mechanisms of 1,3-butadiene-induced carcinogenesis are unknown;
however, the scientific evidence strongly suggests that the
carcinogenic effects are mediated by genotoxic metabolites. Animal data
suggest that females may be more sensitive than males for cancer
effects associated with 1,3-butadiene exposure; there are insufficient
data in humans from which to draw conclusions about sensitive
subpopulations. The URE for 1,3-butadiene is 3 x 10-5 per
[micro]g/m\3\.\2275\ 1,3-butadiene also causes a variety of
reproductive and developmental effects in mice; no human data on these
effects are available. The most sensitive effect was ovarian atrophy
observed in a lifetime bioassay of female mice.\2276\ Based on this
critical effect and the benchmark concentration methodology, an RfC for
chronic health effects was calculated at 0.9 ppb (approximately 2
[micro]g/m\3\).
---------------------------------------------------------------------------
\2269\ U.S. EPA (2002). Health Assessment of 1,3-Butadiene.
Office of Research and Development, National Center for
Environmental Assessment, Washington Office, Washington, DC. Report
No. EPA600-P-98-001F. This document is available electronically at
http://www3.epa.gov/iris/supdocs/buta-sup.pdf.
\2270\ U.S. EPA (2002). ``Full IRIS Summary for 1,3-butadiene
(CASRN 106-99-0)'' Environmental Protection Agency, Integrated Risk
Information System (IRIS), Research and Development, National Center
for Environmental Assessment, Washington, DC. Available at http://www3.epa.gov/iris/subst/0139.htm.
\2271\ International Agency for Research on Cancer (IARC)
(1999). Monographs on the evaluation of carcinogenic risk of
chemicals to humans, Volume 71, Re-evaluation of some organic
chemicals, hydrazine and hydrogen peroxide World Health
Organization, Lyon, France.
\2272\ International Agency for Research on Cancer (IARC).
(2012). Monographs on the evaluation of carcinogenic risk of
chemicals to humans, Volume 100F chemical agents and related
occupations, World Health Organization, Lyon, France.
\2273\ International Agency for Research on Cancer (IARC).
(2008). Monographs on the evaluation of carcinogenic risk of
chemicals to humans, 1,3-Butadiene, Ethylene Oxide and Vinyl Halides
(Vinyl Fluoride, Vinyl Chloride and Vinyl Bromide) Volume 97, World
Health Organization, Lyon, France.
\2274\ NTP (National Toxicology Program). 201 6. Report on
Carcinogens, Fourteenth Edition.; Research Triangle Park NC: U.S.
Department of Health and Human Services Public Health Service.
Available at https://ntp.niehs.nih.gov/go/rocl4.
\2275\ U.S. EPA (2002). ``Full IRIS Summary for 1,3-butadiene
(CASRN 106-99-0)'' Environmental Protection Agency, Integrated Risk
Information System (IRIS), Research and Development, National Center
for Environmental Assessment, Washington, DC http://www3.epa.gov/iris/subst/0139.htm.
\2276\ Bevan, C.; Stadler, J.C.; Elliot, G.S.; et al. (1996).
Subchronic toxicity of 4-vinylcyclohexene in rats and mice by
inhalation. Fundam. Appl. Toxicol. 32:1-10.
---------------------------------------------------------------------------
(d) Formaldehyde
In 1991, EPA concluded that formaldehyde is a carcinogen based on
nasal tumors in animal bioassays.\2277\ An Inhalation URE for cancer
and a Reference Dose for oral noncancer
[[Page 24866]]
effects were developed by the agency and posted on the IRIS database.
Since that time, the National Toxicology Program (NTP) and
International Agency for Research on Cancer (IARC) have concluded that
formaldehyde is a known human carcinogen.2278 2279 2280
---------------------------------------------------------------------------
\2277\ EPA Integrated Risk Information System. Formaldehyde
(CASRN 50-00-0) http://www3.epa.gov/iris/subst/0419/htm.
\2278\ NTP (National Toxicology Program). 2016. Report on
Carcinogens. Fourteenth Edition.; Research Triangle Park, NC: U.S.
Department of Health and Human Services. Public Health Service.
Available at https://ntp.niehs.nih.gov/go/roc 14.
\2279\ IARC Monographs on the Evaluation of Carcinogenic Risks
to Humans Volume 100F (2012): Formaldehyde.
\2280\ IARC Monographs on the Evaluation of Carcinogenic Risks
to Humans Volume 88 (2006): Formaldehyde, 2- Butoxyethanol and 1 -
tert-Butoxypropan-2-ol.
---------------------------------------------------------------------------
The conclusions by IARC and NTP reflect the results of
epidemiologic research published since 1991 in combination with
previous animal, human and mechanistic evidence. Research conducted by
the National Cancer Institute reported an increased risk of
nasopharyngeal cancer and specific lymph hematopoietic malignancies
among workers exposed to formaldehyde.2281 2282 2283 A
National Institute of Occupational Safety and Health study of garment
workers also reported increased risk of death due to leukemia among
workers exposed to formaldehyde.\2284\ Extended follow-up of a cohort
of British chemical workers did not report evidence of an increase in
nasopharyngeal or lymph hematopoietic cancers, but a continuing
statistically significant excess in lung cancers was reported.\2285\
Finally, a study of embalmers reported formaldehyde exposures to be
associated with an increased risk of myeloid leukemia but not brain
cancer.\2286\
---------------------------------------------------------------------------
\2281\ Hauptmann, M.; Lubin, J.H.; Stewart, P.A.; Hayes, R.B.;
Blair, A. 2003. Mortality from lymphohematopoietic malignancies
among workers in formaldehyde industries. Journal of the National
Cancer Institute 95, pp. 1615-23.
\2282\ Hauptmann, M.; Lubin, J.H.; Stewart, P.A.; Hayes, R.B.;
Blair, A. 2004. Mortality from solid cancers among workers in
formaldehyde industries. American Journal of Epidemiology 159: 1117-
30.
\2283\ Beane Freeman, L.E.; Blair, A.; Lubin, J.H.; Stewart,
P.A.; Hayes, R.B.; Hoover, R.N.; Hauptmann, M. 2009. Mortality from
lymph hematopoietic malignancies among workers in formaldehyde
industries: The National Cancer Institute cohort. J. National Cancer
Inst. 101: 751-61.
\2284\ Pinkerton, L.E. 2004. Mortality among a cohort of garment
workers exposed to formaldehyde: an update. Occup. Environ. Med. 61:
193-200.
\2285\ Coggon, D, EC Harris, J Poole, KT Palmer. 2003. Extended
follow-up of a cohort of British chemical workers exposed to
formaldehyde. J National Cancer Inst. 95:1608-15.
\2286\ Hauptmann, M.; Stewart P.A.; Lubin J.H.; Beane Freeman,
L.E.; Hornung, R.W.; Herrick, R.F.; Hoover, R.N.; Fraumeni, J.F.;
Hayes, R.B. 2009. Mortality from lymph hematopoietic malignancies
and brain cancer among embalmers exposed to formaldehyde. Journal of
the National Cancer Institute 101:1696-1708.
---------------------------------------------------------------------------
Health effects of formaldehyde in addition to cancer were reviewed
by the Agency for Toxics Substances and Disease Registry in 1999,\2287\
supplemented in 2010,\2288\ and by the World Health Organization.\2289\
These organizations reviewed the scientific literature concerning
health effects linked to formaldehyde exposure to evaluate hazards and
dose response relationships and defined exposure concentrations for
minimal risk levels (MRLs). The health endpoints reviewed included
sensory irritation of eyes and respiratory tract, reduced pulmonary
function, nasal histopathology, and immune system effects. In addition,
research on reproductive and developmental effects and neurological
effects were discussed along with several studies that suggest that
formaldehyde may increase the risk of asthma--particularly in the
young.
---------------------------------------------------------------------------
\2287\ ATSDR (1999). Toxicological Profile for Formaldehyde,
U.S. Department of Health and Human Services (HHS), July 1999.
\2288\ ATSDR (2010). Addendum to the Toxicological Profile for
Formaldehyde. U.S. Department of Health and Human Services (HHS),
October 2010.
\2289\ IPCS (2002). Concise International Chemical Assessment
Document 40. Formaldehyde. World Health Organization.
---------------------------------------------------------------------------
EPA released a draft Toxicological Review of Formaldehyde--
Inhalation Assessment through the IRIS program for peer review by the
National Research Council (NRC) and public comment in June 2010.\2290\
The draft assessment reviewed more recent research from animal and
human studies on cancer and other health effects. The NRC released
their review report in April 2011.\2291\ EPA is currently developing a
revised draft assessment in response to this review.
---------------------------------------------------------------------------
\2290\ EPA (2010). Toxicological Review of Formaldehyde (CAS No.
50-00-0)-Inhalation Assessment: In Support of Summary Information on
the Integrated Risk Information System (IRIS). External Review
Draft. EPA/635/R-10/002A. U.S. Environmental Protection Agency,
Washington DC. Available at http://cfpub.epa.gov/ncea/irs_drats/recordisplay.cfm?deid=223614.
\2291\ NRC (National Research Council) (2011). Review of the
Environmental Protection Agency's Draft IRIS Assessment of
Formaldehyde. Washington DC: National Academies Press. http://books.nap.edu/openbook.php?record_id=13142.
---------------------------------------------------------------------------
(e) Acetaldehyde
Acetaldehyde is classified in EPA's IRIS database as a probable
human carcinogen, based on nasal tumors in rats, and is considered
toxic by the inhalation, oral, and intravenous routes.\2292\ The URE in
IRIS for acetaldehyde is 2.2 x 10-6 per [micro]g/m\3\.\2293\
Acetaldehyde is reasonably anticipated to be a human carcinogen by the
U.S. DHHS in the 13th Report on Carcinogens and is classified as
possibly carcinogenic to humans (Group 2B) by the
IARC.2294 2295 Acetaldehyde is currently listed on the IRIS
Program Multi-Year Agenda for reassessment within the next few years.
---------------------------------------------------------------------------
\2292\ U.S. EPA (1991). Integrated Risk Information System File
of Acetaldehyde. Research and Development, National Center for
Environmental Assessment, Washington, DC. This material is available
electronically at http://www3.epa.gov/iris/subst/0290.htm.
\2293\ U.S. EPA (1991). Integrated Risk Information System File
of Acetaldehyde. This material is available electronically at http://www3.epa.gov/iris/subst/0290.htm.
\2294\ NTP (National Toxicology Program) 2016. Report on
Carcinogens Fourteenth Edition, Research Triangle Park, NC: U.S.
Department of Health and Human Services. Public Health Service.
Available at https://ntp.niehs.nih.gov/go/roc14.
\2295\ International Agency for Research on Cancer (IARC)
(1999). Re-evaluation of some organic chemicals, hydrazine, and
hydrogen peroxide. IARC Monographs on the Evaluation of Carcinogenic
Risk of Chemical to Humans, Vol 71. Lyon, France.
---------------------------------------------------------------------------
The primary noncancer effects of exposure to acetaldehyde vapors
include irritation of the eyes, skin, and respiratory tract.\2296\ In
short-term (4 week) rat studies, degeneration of olfactory epithelium
was observed at various concentration levels of acetaldehyde
exposure.2297 2298 Data from these studies were used by EPA
to develop an inhalation reference concentration of 9 [micro]g/m\3\.
Some asthmatics have been shown to be a sensitive subpopulation to
decrements in functional expiratory volume (FEV1 test) and
bronchoconstriction upon acetaldehyde inhalation.\2299\
---------------------------------------------------------------------------
\2296\ U.S. EPA (1991). Integrated Risk Information System File
of Acetaldehyde. This material is available electronically at http://www3.epa.gov/iris/subst/0290.htm.
\2297\ U.S. EPA. (2003). Integrated Risk Information System File
of Acrolein. Research and Development, National Center for
Environmental Assessment, Washington, DC. This material is available
electronically at http://www3.epa.gov/iris/subst/0364.htm.
\2298\ Appleman, L.M., R.A. Woutersen, and V.J. Feron. (1982).
Inhalation toxicity of acetaldehyde in rats. I. Acute and subacute
studies. Toxicology. 23: 293-297.
\2299\ Myou, S.; Fujimura, M.; Nishi K.; Ohka, T.; and Matsuda,
T. (1993) Aerosolized acetaldehyde induces histamine-mediated
bronchoconstriction in asthmatics. Am. Rev. Respir. Dis. 148(4 Pt
1): 940-943.
---------------------------------------------------------------------------
(f) Acrolein
EPA most recently evaluated the toxicological and health effects
literature related to acrolein in 2003 and concluded that the human
carcinogenic potential of acrolein could not be determined because the
available data were inadequate. No information was available on the
carcinogenic effects of acrolein in humans and the animal data provided
inadequate evidence of
[[Page 24867]]
carcinogenicity.\2300\ The IARC determined in 1995 that acrolein was
not classifiable as to its carcinogenicity in humans.\2301\
---------------------------------------------------------------------------
\2300\ U.S. EPA (2003). Integrated Risk Information System File
of Acrolein. Research and Development, National Center for
Environmental Assessment, Washington, DC. This material is available
at http://www3.epa.gov/iris/subst/0364.htm.
\2301\ International Agency for Research on Cancer (1995).
Monographs on the evaluation of carcinogenic risk of chemicals to
humans, Volume 63. Dry cleaning, some chlorinated solvents and other
industrial chemicals, World Health Organization, Lyon, France.
---------------------------------------------------------------------------
Lesions to the lungs and upper respiratory tract of rats, rabbits,
and hamsters have been observed after sub-chronic exposure to
acrolein.\2302\ The agency has developed an RfC for acrolein of 0.02
[micro]g/m\3\ and an RfD of 0.5 [micro]g/kg-day.\2303\
---------------------------------------------------------------------------
\2302\ U.S. EPA (2003). Integrated Risk Information System File
of Acrolein. Office of Research and Development, National Center for
Environmental Assessment, Washington, DC. This material is available
at http://www3.epa.gov/iris/subst/0364.htm.
\2303\ U.S. EPA (2003). Integrated Risk Information System File
of Acrolein. Office of Research and Development, National Center for
Environmental Assessment, Washington, DC. This material is available
at http://www3.epa.gov/iris/subst/0364.htm.
---------------------------------------------------------------------------
Acrolein is extremely acrid and irritating to humans when inhaled,
with acute exposure resulting in upper respiratory tract irritation,
mucus hypersecretion and congestion. The intense irritancy of this
carbonyl has been demonstrated during controlled tests in human
subjects, who suffer intolerable eye and nasal mucosal sensory
reactions within minutes of exposure.\2304\ These data and additional
studies regarding acute effects of human exposure to acrolein are
summarized in EPA's 2003 Toxicological Review of Acrolein.\2305\
Studies in humans indicate that levels as low as 0.09 ppm (0.21 mg/
m\3\) for five minutes may elicit subjective complaints of eye
irritation with increasing concentrations leading to more extensive
eye, nose and respiratory symptoms. Acute exposures in animal studies
report bronchial hyper-responsiveness. Based on animal data (more
pronounced respiratory irritancy in mice with allergic airway disease
in comparison to non-diseased mice) \2306\ and demonstration of similar
effects in humans (e.g., reduction in respiratory rate), individuals
with compromised respiratory function (e.g., emphysema, asthma) are
expected to be at increased risk of developing adverse responses to
strong respiratory irritants such as acrolein. EPA does not currently
have an acute reference concentration for acrolein. The available
health effect reference values for acrolein have been summarized by EPA
and include an ATSDR MRL for acute exposure to acrolein of 7 [micro]g/
m\3\ for 1-14 days' exposure; and Reference Exposure Level (REL) values
from the California Office of Environmental Health Hazard Assessment
(OEHHA) for one-hour and 8-hour exposures of 2.5 [micro]g/m\3\ and 0.7
[micro]g/m\3\, respectively.\2307\
---------------------------------------------------------------------------
\2304\ U.S. EPA (2003). Toxicological review of acrolein in
support of summary information on Integrated Risk Information System
(IRIS) National Center for Environmental Assessment, Washington, DC.
EPA/635/R-03/003. p. 10. Available online at: http://www3.epa.gov/ncea/iris/toxreviews/0364tr.pdf.
\2305\ U.S. EPA (2003). Toxicological review of acrolein in
support of summary information on Integrated Risk Information System
(IRIS) National Center for Environmental Assessment, Washington, DC.
EPA/635/R-03/003. Available online at: http://www3.epa.gov/ncea/iris/toxreviews/0364tr.pdf.
\2306\ Morris JB, Symanowicz PT, Olsen JE, et al. (2003).
Immediate sensory nerve-mediated respiratory responses to irritants
in healthy and allergic airway-diseased mice. J Appl Physiol
94(4):1563-71.
\2307\ U.S. EPA (2009). Graphical Arrays of Chemical-Specific
Health Effect Reference Values for Inhalation Exposures (Final
Report). U.S. Environmental Protection Agency, Washington, DC, EPA/
600/R-09/061, 2009. Available at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=211003.
---------------------------------------------------------------------------
(g) Polycyclic Organic Matter
The term polycyclic organic matter (POM) defines a broad class of
compounds that includes the polycyclic aromatic hydrocarbon compounds
(PAHs). One of these compounds, naphthalene, is discussed separately
below. POM compounds are formed primarily from combustion and are
present in the atmosphere in gas and particulate form. Cancer is the
major concern from exposure to POM. Epidemiologic studies have reported
an increase in lung cancer in humans exposed to diesel exhaust, coke
oven emissions, roofing tar emissions, and cigarette smoke; all of
these mixtures contain POM compounds.2308 2309 Animal
studies have reported respiratory tract tumors from inhalation exposure
to benzo[a]pyrene and alimentary tract and liver tumors from oral
exposure to benzo[a]pyrene.\2310\ In 1997 EPA classified seven PAHs
(benzo[a]pyrene, benz[a]anthracene, chrysene, benzo[b]fluoranthene,
benzo[k]fluoranthene, dibenz[a,h]anthracene, and indeno[1,2,3-
cd]pyrene) as Group B2, probable human carcinogens.\2311\ Since that
time, studies have found that maternal exposures to PAHs in a
population of pregnant women were associated with several adverse birth
outcomes, including low birth weight and reduced length at birth, as
well as impaired cognitive development in preschool children (3 years
of age).2312 2313 These and similar studies are being
evaluated as a part of the ongoing IRIS reassessment of health effects
associated with exposure to benzo[a]pyrene.
---------------------------------------------------------------------------
\2308\ Agency for Toxic Substances and Disease Registry (ATSDR).
(1995). Toxicological profile for Polycyclic Aromatic Hydrocarbons
(PAHs). Atlanta, GA: U.S. Department of Health and Human Services,
Public Health Service. Available electronically at http://www.atsdr.cdc.gov/ToxProfiles/TP.asp?id=122&tid=25.
\2309\ U.S. EPA (2002). Health Assessment Document for Diesel
Engine Exhaust. EPA/600/8-90/057F Office of Research and
Development, Washington DC. http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060.
\2310\ International Agency for Research on Cancer (IARC).
(2012). Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals for Humans, Chemical Agents and Related Occupations. Vol.
100F. Lyon, France.
\2311\ U.S. EPA (1997). Integrated Risk Information System File
of indeno (1,2,3-cd) pyrene. Research and Development, National
Center for Environmental Assessment, Washington, DC. This material
is available electronically at http://www3.epa.gov/ncea/iris/subst/0457.htm.
\2312\ Perera, F.P.; Rauh, V.; Tsai, W-Y.; et al. (2002). Effect
of transplacental exposure to environmental pollutants on birth
outcomes in a multiethnic population. Environ Health Perspect. 111:
201-05.
\2313\ Perera, F.P.; Rauh, V.; Whyatt, R.M.; Tsai, W.Y.; Tang,
D.; Diaz, D.; Hoepner, L.; Barr, D.; Tu, Y.H.; Camann, D.; Kinney,
P. (2006). Effect of prenatal exposure to airborne polycyclic
aromatic hydrocarbons on neurodevelopment in the first 3 years of
life among inner-city children. Environ Health Perspect 114: 1287-
92.
---------------------------------------------------------------------------
(h) Naphthalene
Naphthalene is found in small quantities in gasoline and diesel
fuels. Naphthalene emissions have been measured in larger quantities in
both gasoline and diesel exhaust compared with evaporative emissions
from mobile sources, indicating it is primarily a product of
combustion. Acute (short-term) exposure of humans to naphthalene by
inhalation, ingestion, or dermal contact is associated with hemolytic
anemia and damage to the liver and the nervous system.\2314\ Chronic
(long term) exposure of workers and rodents to naphthalene has been
reported to cause cataracts and retinal damage.\2315\ The National
Toxicology
[[Page 24868]]
Program listed naphthalene as ``reasonably anticipated to be a human
carcinogen'' in 2004 on the basis of bioassays reporting clear evidence
of carcinogenicity in rats and some evidence of carcinogenicity in
mice.\2316\ California EPA has released a new risk assessment for
naphthalene, and the IARC has reevaluated naphthalene and re-classified
it as Group 2B: Possibly carcinogenic to humans.\2317\
---------------------------------------------------------------------------
\2314\ U.S. EPA (1998). Toxicological Review of Naphthalene
(Reassessment of the Inhalation Cancer Risk), Environmental
Protection Agency, Integrated Risk Information System, Research and
Development, National Center for Environmental Assessment,
Washington, DC. This material is available electronically at http://www3.epa.gov/iris/subst/0436.htm.
\2315\ U.S. EPA (1998). Toxicological Review of Naphthalene
(Reassessment of the Inhalation Cancer Risk), Environmental
Protection Agency, Integrated Risk Information System, Research and
Development, National Center for Environmental Assessment,
Washington, DC. This material is available electronically at http://www3.epa.gov/iris/subst/0436.htm.
\2316\ NTP (National Toxicology Program), 2016. Report on
Carcinogens Fourteenth Edition, Research Triangle Park NC: U.S.
Department of Health and Human Services, Public Health Service.
Available at https://ntp.niehs.nih.gov/go/roc14.
\2317\ International Agency for Research on Cancer (IARC).
(2002). Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals for Humans. Vol. 82. Lyon, France.
---------------------------------------------------------------------------
Naphthalene also causes a number of chronic non-cancer effects in
animals, including abnormal cell changes and growth in respiratory and
nasal tissues.\2318\ The current EPA IRIS assessment includes noncancer
data on hyperplasia and metaplasia in nasal tissue that form the basis
of the inhalation RfC of 3 [micro]g/m\3\.\2319\ The ATSDR MRL for acute
exposure to naphthalene is 0.6 mg/kg/day.
---------------------------------------------------------------------------
\2318\ U.S. EPA (1998). Toxicological Review of Naphthalene,
Environmental Protection Agency, Integrated Risk Information System,
Research and Development, National Center for Environmental
Assessment, Washington, DC. This material is available
electronically at http://www3.epa.gov/iris/subst/0436.htm.
\2319\ U.S. EPA (1998). Toxicological Review of Naphthalene.
Environmental Protection Agency, Integrated Risk Information System
(IRIS), Research and Development, National Center for Environmental
Assessment, Washington, DC. Available at http://www3.epa.gov/iris/subst/0436.htm.
---------------------------------------------------------------------------
(i) Other Air Toxics
In addition to the compounds described above, other compounds in
gaseous hydrocarbon and PM emissions from motor vehicles will be
affected by this action. Mobile source air toxic compounds that will
potentially be impacted include ethylbenzene, propionaldehyde, toluene,
and xylene. Information regarding the health effects of these compounds
can be found in EPA's IRIS database.\2320\
---------------------------------------------------------------------------
\2320\ U.S. EPA Integrated Risk Information System (IRIS)
database is available at: www3.epa.gov/iris.
---------------------------------------------------------------------------
(j) Current Concentrations
The most recent available data indicate that the majority of
Americans continue to be exposed to ambient concentrations of air
toxics at levels which have the potential to cause adverse health
effects. The levels of air toxics to which people are exposed vary
depending on where people live and work and the kinds of activities in
which they engage, as discussed in detail in EPA's most recent Mobile
Source Air Toxics Rule. According to the National Air Toxic Assessment
(NATA) for 2014, mobile sources were responsible for 51 percent of
outdoor anthropogenic toxic emissions and were the largest contributor
to cancer and noncancer risk from directly emitted pollutants. Mobile
sources are also significant contributors to precursor emissions which
react to form air toxics. Formaldehyde is the largest contributor to
cancer risk of all 71 pollutants quantitatively assessed in the 2014
NATA. Mobile sources were responsible for more than 30 percent of
primary anthropogenic emissions of this pollutant in 2014 and also
contribute to formaldehyde precursor emissions. Benzene is also a large
contributor to cancer risk, and mobile sources account for
approximately 54 percent of ambient exposure. Over the years, EPA has
implemented a number of mobile source and fuel controls which have
resulted in VOC reductions, which also reduced formaldehyde, benzene
and other air toxic emissions.
(k) Exposure and Health Effects Associated With Traffic
Locations in close proximity to major roadways generally have
elevated concentrations of many air pollutants emitted from motor
vehicles. Hundreds of such studies have been published in peer-reviewed
journals, concluding that concentrations of CO, NO, NO2,
benzene, aldehydes, particulate matter, black carbon, and many other
compounds are elevated in ambient air within approximately 300-600
meters (approximately 1,000-2,000 feet) of major roadways. Highest
concentrations of most pollutants emitted directly by motor vehicles
are found at locations within 50 meters (approximately 165 feet) of the
edge of a roadway's traffic lanes.
A large-scale review of air quality measurements in the vicinity of
major roadways between 1978 and 2008 concluded that the pollutants with
the steepest concentration gradients in vicinities of roadways were CO,
ultrafine particles, metals, elemental carbon (EC), NO, NOX,
and several VOCs.\2321\ These pollutants showed a large reduction in
concentrations within 100 meters downwind of the roadway. Pollutants
that showed more gradual reductions with distance from roadways
included benzene, NO2, PM2.5, and
PM10. In the review article, results varied based on the
method of statistical analysis used to determine the trend.
---------------------------------------------------------------------------
\2321\ Karner, A.A.; Eisinger, D.S.; Niemeier, D.A. (2010).
Near-roadway air quality: synthesizing the findings from real-world
data. Environ Sci. Technol. 44: pp. 5334-44.
---------------------------------------------------------------------------
For pollutants with relatively high background concentrations
relative to near-road concentrations, detecting concentration gradients
can be difficult. For example, many aldehydes have high background
concentrations as a result of photochemical breakdown of precursors
from many different organic compounds. This can make detection of
gradients around roadways and other primary emission sources difficult.
However, several studies have measured aldehydes in multiple weather
conditions and found higher concentrations of many carbonyls downwind
of roadways.2322 2323 These findings suggest a substantial
roadway source of these carbonyls.
---------------------------------------------------------------------------
\2322\ Liu, W.; Zhang, J.; Kwon, J.l; et l. (2006).
Concentrations and source characteristics of airborne carbonyl
comlbs measured outside urban residences. J Air Waste Manage Assoc.
56: 1196-1204.
\2323\ Cahill, T.M.; Charles, M.J.; Seaman, V.Y. (2010).
Development and application of a sensitive method to determine
concentrations of acrolein and other carbonyls in ambient air.
Health Effects Institute Research Report 149. Available at http://dx.doi.org.
---------------------------------------------------------------------------
In the past 15 years, many studies have been published with results
reporting that populations who live, work, or go to school near high-
traffic roadways experience higher rates of numerous adverse health
effects, compared to populations far away from major roads.\2324\ In
addition, numerous studies have found adverse health effects associated
with spending time in traffic, such as commuting or walking along high-
traffic roadways.2325 2326 2327 2328 The health outcomes
with the strongest evidence linking them with traffic-associated air
pollutants are respiratory effects, particularly in asthmatic children,
and cardiovascular effects.
---------------------------------------------------------------------------
\2324\ In the widely-used PubMed database of health
publications, between January 1, 1990 and August 18, 2011, 605
publications contained the keywords ``traffic, pollution,
epidemiology,'' with approximately half the studies published after
2007.
\2325\ Laden, F.; Hart, J.E.; Smith, T.J.; Davis, M.E.;
Garshick, E. (2007) Cause-specific mortality in the unionized U.S.
trucking industry. Environmental Health Perspect 115:1192-96.
\2326\ Peters, A.; von Klot, S.; Heier, M.; Trentinaglia, I.;
H[ouml]rmann, A.; Wichmann, H.E.; L[ouml]wel, H. (2004) Exposure to
traffic and the onset of myocardial infarction. New England J Med
351: 1721-30.
\2327\ Zanobetti, A.; Stone, P.H.; Spelzer, F.E.; Schwartz,
J.D.; Coull, B.A.; Suh, H.H.; Nearling, B.D.; Mittleman, M.A.;
Verrier, R.L.; Gold, D.R. (2009) T-wave alternans, air pollution and
traffic in high-risk subjects. Am J Cardiol 104: 665-670.
\2328\ Dubowsky Adar, S.; Adamkiewicz, G.; Gold, D.R.; Schwartz,
J.; Coull, B.A.; Suh, H. (2007) Ambient and microenvironmental
particles and exhaled nitric oxide before and after a group bus
trip. Environ Health Perspect 115: 507-512.
---------------------------------------------------------------------------
[[Page 24869]]
Numerous reviews of this body of health literature have been
published as well. In 2010, an expert panel of the Health Effects
Institute (HEI) published a review of hundreds of exposure,
epidemiology, and toxicology studies.\2329\ The panel rated how the
evidence for each type of health outcome supported a conclusion of a
causal association with traffic-associated air pollution as either
``sufficient,'' ``suggestive but not sufficient,'' or ``inadequate and
insufficient.'' The panel categorized evidence of a causal association
for exacerbation of childhood asthma as ``sufficient.'' The panel
categorized evidence of a causal association for new onset asthma as
between ``sufficient'' and ``suggestive but not sufficient.''
``Suggestive of a causal association'' was how the panel categorized
evidence linking traffic-associated air pollutants with exacerbation of
adult respiratory symptoms and lung function decrement. It categorized
as ``inadequate and insufficient'' evidence of a causal relationship
between traffic-related air pollution and health care utilization for
respiratory problems, new onset adult asthma, chronic obstructive
pulmonary disease (COPD), nonasthmatic respiratory allergy, and cancer
in adults and children. Other literature reviews have been published
with conclusions generally similar to the HEI
panel's.2330 2331 2332 2333 However, in 2014, researchers
from the U.S. Centers for Disease Control and Prevention (CDC)
published a systematic review and meta-analysis of studies evaluating
the risk of childhood leukemia associated with traffic exposure and
reported positive associations between ``postnatal'' proximity to
traffic and leukemia risks, but no such association for ``prenatal''
exposures.\2334\
---------------------------------------------------------------------------
\2329\ Health Effects Institute Panel on the Health Effects of
Traffic-Related Air Pollution (2010). Traffic-related air pollution:
a critical review of the literature on emissions, exposure, and
health effects. HEI Special Report 17. Available at http://www.healtheffects.org.
\2330\ Boothe, V.L.; Shendell, D.G. (2008). Potential health
effects associated with residential proximity to freeways and
primary roads: review of scientific literature, 1999-2006. J Environ
Health 70: 33-41.
\2331\ Salam, M.T.; Islam, T.; Gilliland, F.D. (2008). Recent
evidence for adverse effects of residential proximity to traffic
sources on asthma. Curr Opin Pulm Med 14: 3-8.
\2332\ Sun, X.; Zhang, S.; Ma, X. (2014) No association between
traffic density and risk of childhood leukemia: a meta-analysis.
Asia Pac J Cancer Prev 15: 5229-32.
\2333\ Raaschou-Nielsen, O.; Reynolds, P. (2006). Air pollution
and childhood cancer: a review of the epidemiological literature.
Int J Cancer 118: 2920-9.
\2334\ Boothe, VL.; Boehmer, T.K.; Wendel, A.M.; Yip, F.Y.
(2014) Residential traffic exposure and childhood leukemia: a
systematic review and meta-analysis. Am J Prev Med 46: 413-422.
---------------------------------------------------------------------------
Health outcomes with few publications suggest the possibility of
other effects still lacking sufficient evidence to draw definitive
conclusions. Among these outcomes with a small number of positive
studies are neurological impacts (e.g., autism and reduced cognitive
function) and reproductive outcomes (e.g., preterm birth, low birth
weight).2335 2336 2337 2338.
---------------------------------------------------------------------------
\2335\ Volk, H.E.; Hertz-Picciotto, I.; Delwiche, L.; et al.
(2011). Residential proximity to freeways and autism in the CHARGE
study. Environ Health Perspect 119: 873-77.
\2336\ Franco-Suglia, S.; Gryparis, A.; Wright, R.O.; et al.
(2007). Association of black carbon with cognition among children in
a prospective birth cohort study. Am J Epidemiol. doi: 10.1093/aje/
kwm308. Available at http://dx.doi.org.
\2337\ Power, M.C.; Weisskopf, M.G.; Alexeef, SE; et al. (2011).
Traffic-related air pollution and cognitive function in a cohort of
older men. Environ Health Perspect 2011: 682-687.
\2338\ Wu, J.; Wilhelm, M.; Chung, J.; et al. (2011). Comparing
exposure assessment methods for traffic-related air pollution in and
adverse pregnancy outcome study. Environ Res 111: 685-6692.
---------------------------------------------------------------------------
In addition to health outcomes, particularly cardiopulmonary
effects, conclusions of numerous studies suggest mechanisms by which
traffic-related air pollution affects health. Numerous studies indicate
that near-roadway exposures may increase systemic inflammation,
affecting organ systems, including blood vessels and
lungs.2339 2340 2341 2342 Long-term exposures in near-road
environments have been associated with inflammation-associated
conditions, such as atherosclerosis and
asthma.2343 2344 2345
---------------------------------------------------------------------------
\2339\ Riediker, M. (2007). Cardiovascular effects of fine
particulate matter components in highway patrol officers. Inhal
Toxicol 19: 99-105. doi: 10.1080/08958370701495238 Available at
http://dx.doi.org.
\2340\ Alexeef, SE; Coull, B.A.; Gryparis, A.; et al. (2011).
Medium-term exposure to traffic-related air pollution and markers of
inflammation and endothelial function. Environ Health Perspect 119:
481-486. doi:10.1289/ehp.1002560 Available at http://dx.doi.org.
\2341\ Eckel. S.P.; Berhane, K.; Salam, M.T.; et al. (2011).
Traffic-related pollution exposure and exhaled nitric oxide in the
Children's Health Study. Environ Health Perspect (IN PRESS).
doi:10.1289/ehp.1103516. Available at http://dx.doi.org.
\2342\ Zhang, J.; McCreanor, J.E.; Cullinan, P.; et al. (2009).
Health effects of real-world exposure diesel exhaust in persons with
asthma. Res Rep Health Effects Inst 138. Available at http://www.healtheffects.org.
\2343\ Adar, S.D.; Klein, R.; Klein, E.K.; et al. (2010). Air
pollution and the microvasculatory: a cross-sectional assessment of
in vivo retinal images in the population-based Multi-Ethnic Study of
Atherosclerosis. PLoS Med 7(11): E1000372. doi:10.1371/
journal.pmed.1000372. Available at http://dx.doi.org.
\2344\ Kan, H.; Heiss, G.; Rose, K.M.; et al. (2008).
Prospective analysis of traffic exposure as a risk factor for
incident coronary heart disease: the Atherosclerosis Risk in
Communities (ARIC) study. Environ Health Perspect 116: 1463-1468.
doi:10.1289/ehp.11290. Available at http://dx.doi.org.
\2345\ McConnell, R.; Islam, T.; Shankardass, K.; et al. (2010).
Childhood incident asthma and traffic-related air pollution at home
and school. Environ Health Perspect 1021-26.
---------------------------------------------------------------------------
Several studies suggest that some factors may increase
susceptibility to the effects of traffic-associated air pollution.
Several studies have found stronger respiratory associations in
children experiencing chronic social stress, such as in violent
neighborhoods or in homes with high family
stress.2346 2347 2348
---------------------------------------------------------------------------
\2346\ Islam, T.; Urban, R.; Gauderman, W.J.; et al. (2011).
Parental stress increases the detrimental effect of traffic exposure
on children's lung function. Am J Respir Crit Care Med (In press).
\2347\ Clougherty, J.E.; Levy, J.I.; Kubzansky, L.D.; et al.
(2007). Synergistic effects of traffic-related air pollution and
exposure to violence on urban asthma etiology. Environ Health
Perspect 115: 1140-46.
\2348\ Chen, E.; Schrier, H.M.; Strunk, R.C.; et al. (2008).
Chronic traffic-related air pollution and stress interact to predict
biologic and clinical outcomes in asthma. Environ Health Perspect
116: 970-5.
---------------------------------------------------------------------------
The risks associated with residence, workplace, or schools near
major roads are of potentially high public health significance due to
the large population in such locations. According to the 2009 American
Housing Survey, over 22 million homes (17.0 percent of all U.S. housing
units) were located within 300 feet of an airport, railroad, or highway
with four or more lanes. This corresponds to a population of more than
50 million U.S. residents in close proximity to high-traffic roadways
or other transportation sources. Based on 2010 Census data, a 2013
publication estimated that 19 percent of the U.S. population (over 59
million people) lived within 500 meters of roads with at least 25,000
annual average daily traffic (AADT), while about 3.2 percent of the
population lived within 100 meters (about 300 feet) of such
roads.\2349\ Another 2013 study estimated that 3.7 percent of the U.S.
population (about 11.3 million people) lived within 150 meters (about
500 feet) of interstate highways or other freeways and
expressways.\2350\ On average, populations near major roads have higher
fractions of minority residents and lower socioeconomic status.
Furthermore, on average, Americans spend more than an hour traveling
each day, bringing nearly all residents into a
[[Page 24870]]
high-exposure microenvironment for part of the day.
---------------------------------------------------------------------------
\2349\ Rowangould, G.M. (2013). A census of the U.S. near-
roadway population: public health and environmental justice
considerations. Transportation Research Part D 25: 59-67.
\2350\ Boehmer, T.K.; Foster, S.L.; Henry, J.R.; Woghiren-
Akinnifesi, E.L.; Yip, F.Y. (2013) Residential proximity to major
highways--United States, 2010. Morbidity and Mortality Weekly Report
62(3); 46-50.
---------------------------------------------------------------------------
In light of these concerns, EPA has required through the NAAQS
process that air quality monitors be placed near high-traffic roadways
for determining concentrations of CO, NO2, and
PM2.5 (in addition to those existing monitors located in
neighborhoods and other locations farther away from pollution sources).
Near-roadway monitors for NO2 began operation between 2014
and 2017 in Core Based Statistical Areas (CBSAs) with population of at
least 500,000. Monitors for CO and PM2.5 began operation
between 2015 and 2017. These monitors will further the understanding of
exposure in these locations.
EPA and DOT continue to research near-road air quality, including
the types of pollutants found in high concentrations near major roads
and health problems associated with the mixture of pollutants near
roads.
(8) Environmental Effects of Non-GHG Pollutants
(a) Visibility
Visibility can be defined as the degree to which the atmosphere is
transparent to visible light.\2351\ Visibility impairment is caused by
light scattering and absorption by suspended particles and gases.
Visibility is important because it has direct significance to people's
enjoyment of daily activities in all parts of the country. Individuals
value good visibility for the well-being it provides them directly,
where they live and work, and in places where they enjoy recreational
opportunities. Visibility is also highly valued in significant natural
areas, such as national parks and wilderness areas, and special
emphasis is given to protecting visibility in these areas. For more
information on visibility see the final 2019 p.m.
ISA.2352 2353
---------------------------------------------------------------------------
\2351\ National Research Council, (1993). Protecting Visibility
in National Parks and Wilderness Areas. National Academy of Sciences
Committee on Haze in National Parks and Wilderness Areas. National
Academy Press, Washington, DC. Available at http://www.nap.edu/books/0309048443/html/.
\2352\ U.S. EPA. Integrated Science Assessment (ISA) for
Particulate Matter (Final Report 2019). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-19/188, 2019.
\2353\ There is an ongoing review of the ISA for Oxides of
Nitrogen Oxides of Sulfur, and Particulate Matter (Ecological
Criteria), Available at https://wwwepa.gov/isa/integrated-science-assessment-isa-oxides-nitrogen-oxides-sulfur-andparticulate-matter.
---------------------------------------------------------------------------
EPA is working to address visibility impairment. Reductions in air
pollution from implementation of various programs associated with the
Clean Air Act Amendments of 1990 (CAAA) provisions have resulted in
substantial improvements in visibility and will continue to do so in
the future. Because trends in haze are closely associated with trends
in particulate sulfate and nitrate due to the relationship between
their concentration and light extinction, visibility trends have
improved as emissions of SO2 and NOX have
decreased over time due to air pollution regulations such as the Acid
Rain Program.\2354\
---------------------------------------------------------------------------
\2354\ U.S. EPA (2009). Final Report: Integrated Science
Assessment for Particulate Matter. U.S. Environmental Protection
Agency, Washington, DC, EPA/600/R-08/139F, 2009.
---------------------------------------------------------------------------
In the Clean Air Act Amendments of 1977, Congress recognized
visibility's value to society by establishing a national goal to
protect national parks and wilderness areas from visibility impairment
caused by manmade pollution.\2355\ In 1999, EPA finalized the regional
haze program to protect the visibility in Mandatory Class I Federal
areas.\2356\ There are 156 national parks, forests and wilderness areas
categorized as Mandatory Class I Federal areas.\2357\ These areas are
defined in CAA Section 162 as those national parks exceeding 6,000
acres, wilderness areas and memorial parks exceeding 5,000 acres, and
all international parks which were in existence on August 7, 1977.
---------------------------------------------------------------------------
\2355\ See Section 169(a) of the Clean Air Act.
\2356\ 64 FR 35714 (July 1, 1999).
\2357\ 62 FR 38680-81 (July 18, 1997).
---------------------------------------------------------------------------
EPA has also concluded that PM2.5 causes adverse effects
on visibility in other areas that are not targeted by the Regional Haze
Rule, such as urban areas, depending on PM2.5 concentrations
and other factors such as dry chemical composition and relative
humidity (i.e., an indicator of the water composition of the
particles). EPA revised the PM2.5 standards in December 2012
and established a target level of protection that is expected to be met
through attainment of the existing secondary standards for
PM2.5.
(b) Plant and Ecosystem Effects of Ozone
The welfare effects of ozone include effects on ecosystems, which
can be observed across a variety of scales, i.e. subcellular, cellular,
leaf, whole plant, population and ecosystem. Ozone can produce both
acute and chronic injury in sensitive species depending on the
concentration level and the duration of the exposure.\2358\ In those
sensitive species,\2359\ effects from repeated exposure to ozone
throughout the growing season of the plant can tend to accumulate, so
that even relatively low concentrations experienced for a longer
duration have the potential to create chronic stress on
vegetation.\2360\ Ozone damage to sensitive species includes impaired
photosynthesis and visible injury to leaves. The impairment of
photosynthesis, the process by which the plant makes carbohydrates (its
source of energy and food), can lead to reduced crop yields, timber
production, and plant productivity and growth. Impaired photosynthesis
can also lead to a reduction in root growth and carbohydrate storage
below ground, resulting in other, more subtle plant and ecosystems
impacts.\2361\ These latter impacts include increased susceptibility of
plants to insect attack, disease, harsh weather, interspecies
competition and overall decreased plant vigor. The adverse effects of
ozone on areas with sensitive species could potentially lead to species
shifts and loss from the affected ecosystems,\2362\ resulting in a loss
or reduction in associated ecosystem goods and services. Additionally,
visible ozone injury to leaves can result in a loss of aesthetic value
in areas of special scenic significance like national parks and
wilderness areas and reduced use of sensitive ornamentals in
landscaping.\2363\
---------------------------------------------------------------------------
\2358\ 73 FR 16486 (March 27, 2008).
\2359\ 73 FR 16491 (March 27, 2008). Only a small percentage of
all the plant species growing within the U.S. (over 43,000 species
have been catalogued in the USDA PLANTS database) have been studied
with respect to ozone sensitivity.
\2360\ The concentration at which ozone levels overwhelm a
plant's ability to detoxify or compensate for oxidant exposure
varies. Thus, whether a plant is classified as sensitive or tolerant
depends in part on the exposure levels being considered. Chapter 9,
Section 9.3.4 of U.S. EPA, 2013 Integrated Science Assessment for
Ozone and Related Photochemical Oxidants. Office of Research and
Development/National Center for Environmental Assessment. U.S.
Environmental Protection Agency. EPA 600/R-10/076F.
\2361\ 73 FR 16492 (March 27, 2008).
\2362\ 73 FR 16493-94 (March 27, 2008). Ozone impacts could be
occurring in areas where plant species sensitive to ozone have not
yet been studied or identified.
\2363\ 73 FR 16490-97 (March 27, 2008).
---------------------------------------------------------------------------
The most recent Integrated Science Assessment (ISA) for Ozone
presents more detailed information on how ozone affects vegetation and
ecosystems.2364 2365 The ISA concludes that ambient
concentrations of ozone are associated with a number of adverse welfare
effects and characterizes the
[[Page 24871]]
weight of evidence for different effects associated with ozone.\2366\
The ISA concludes that visible foliar injury effects on some
vegetation, reduced vegetation growth, reduced productivity in
terrestrial ecosystems, reduced yield and quality of some agricultural
crops, and alteration of below-ground biogeochemical cycles are
causally associated with exposure to ozone. It also concludes that
reduced carbon sequestration in terrestrial ecosystems, alteration of
terrestrial ecosystem water cycling, and alteration of terrestrial
community composition are likely to be causally associated with
exposure to ozone.
---------------------------------------------------------------------------
\2364\ U.S. EPA. Integrated Science Assessment of Ozone and
Related Photochemical Oxidants (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-10/076F, 2013. The ISA
is available at http://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=247492#Download.
\2365\ There is an ongoing review of the ozone NAAQS, EPA
intends to finalize an updated Integrated Science Assessment in
early 2020 Available at (https://www.epa.gov naaqs/ozone-o3-
standards-integrated-science-assessments-currentreview).
\2366\ The Ozone ISA evaluates the evidence associated with
different ozone related health and welfare effects, assigning one of
five ``weight of evidence'' determinations: causal relationship,
likely to be a causal relationship, suggestive of a causal
relationship, inadequate to infer a causal relationship, and not
likely to be a causal relationship. For more information on these
levels of evidence, please refer to Table II of the ISA.
---------------------------------------------------------------------------
(c) Atmospheric Deposition
Wet and dry deposition of ambient particulate matter delivers a
complex mixture of metals (e.g., mercury, zinc, lead, nickel, aluminum,
and cadmium), organic compounds (e.g., polycyclic organic matter,
dioxins, and furans) and inorganic compounds (e.g., nitrate, sulfate)
to terrestrial and aquatic ecosystems. The chemical form of the
compounds deposited depends on a variety of factors including ambient
conditions (e.g., temperature, humidity, oxidant levels) and the
sources of the material. Chemical and physical transformations of the
compounds occur in the atmosphere as well as the media onto which they
deposit. These transformations in turn influence the fate,
bioavailability and potential toxicity of these compounds.
Adverse impacts to human health and the environment can occur when
particulate matter is deposited to soils, water, and biota.\2367\
Deposition of heavy metals or other toxics may lead to the human
ingestion of contaminated fish, impairment of drinking water, damage to
terrestrial, freshwater and marine ecosystem components, and limits to
recreational uses. Atmospheric deposition has been identified as a key
component of the environmental and human health hazard posed by several
pollutants including mercury, dioxin and PCBs.\2368\
---------------------------------------------------------------------------
\2367\ U.S. EPA. Integrated Science Assessment for Particulate
Matter (Final Report). U.S. Environmental Protection Agency,
Washington, DC, EPA/600/R-08/139F, 2009.
\2368\ U.S. EPA (2000). Deposition of Air Pollutants to the
Great Waters: Third Report to Congress. Office of Air Quality
Planning and Standards. EPA-453/R-00-0005.
---------------------------------------------------------------------------
The ecological effects of acidifying deposition and nutrient
enrichment are detailed in the Integrated Science Assessment for Oxides
of Nitrogen and Sulfur-Ecological Criteria.2369 2370
Atmospheric deposition of nitrogen and sulfur contributes to
acidification, altering biogeochemistry and affecting animal and plant
life in terrestrial and aquatic ecosystems across the United States.
The sensitivity of terrestrial and aquatic ecosystems to acidification
from nitrogen and sulfur deposition is predominantly governed by
geology. Prolonged exposure to excess nitrogen and sulfur deposition in
sensitive areas acidifies lakes, rivers and soils. Increased acidity in
surface waters creates inhospitable conditions for biota and affects
the abundance and biodiversity of fishes, zooplankton and
macroinvertebrates and ecosystem function. Over time, acidifying
deposition also removes essential nutrients from forest soils,
depleting the capacity of soils to neutralize future acid loadings and
negatively affecting forest sustainability. Major effects in forests
include a decline in sensitive tree species, such as red spruce (Picea
rubens) and sugar maple (Acer saccharum). In addition to the role
nitrogen deposition plays in acidification, nitrogen deposition also
leads to nutrient enrichment and altered biogeochemical cycling. In
aquatic systems increased nitrogen can alter species assemblages and
cause eutrophication. In terrestrial systems nitrogen loading can lead
to loss of nitrogen-sensitive lichen species, decreased biodiversity of
grasslands, meadows and other sensitive habitats, and increased
potential for invasive species.
---------------------------------------------------------------------------
\2369\ NOX and SOX secondary ISA2369 U.S.
EPA. Integrated Science Assessment (ISA) for Oxides of Nitrogen and
Sulfur Ecological Criteria (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/082F, 2008.
\2370\ There is an ongoing review of the ISA for Oxides and
Nitrogen, Oxides of Sulfur, and Particulate Matter (Ecological
Criteria), Available at https://www.epa.gov/isa/integrated-science-assessment-isa-oxides-nitrogen-oxides-sulfur-and-particulate-matter.
---------------------------------------------------------------------------
Building materials including metals, stones, cements, and paints
undergo natural weathering processes from exposure to environmental
elements (e.g., wind, moisture, temperature fluctuations, sunlight,
etc.). Pollution can worsen and accelerate these effects. Deposition of
PM is associated with both physical damage (materials damage effects)
and impaired aesthetic qualities (soiling effects). Wet and dry
deposition of PM can physically affect materials, adding to the effects
of natural weathering processes, by potentially promoting or
accelerating the corrosion of metals, by degrading paints and by
deteriorating building materials such as stone, concrete and
marble.\2371\ The effects of PM are exacerbated by the presence of
acidic gases and can be additive or synergistic due to the complex
mixture of pollutants in the air and surface characteristics of the
material. Acidic deposition has been shown to have an effect on
materials including zinc/galvanized steel and other metal, carbonate
stone (as monuments and building facings), and surface coatings
(paints).\2372\ The effects on historic buildings and outdoor works of
art are of particular concern because of the uniqueness and
irreplaceability of many of these objects. In addition to aesthetic and
functional effects on metals, stone and glass, altered energy
efficiency of photovoltaic panels by PM deposition is also becoming an
important consideration for impacts of air pollutants on materials.
---------------------------------------------------------------------------
\2371\ U.S. EPA. Integrated Science Assessment (ISA) for
Particulate Matter (Final Report, 2019). U.S Environmental
Protection Agency, Washington, DC, EPA/600/R-l9/188, 2019.
\2372\ Irving, P.M., e.d. 1991. Acid Deposition: State of
Science and Technology, Volume III, Terrestrial, Materials, Health,
and Visibility Effects, The U.S. National Acid Precipitation
Assessment Program, Chapter 24, pp. 24-76.
---------------------------------------------------------------------------
(d) Environmental Effects of Air Toxics
Emissions from producing, transporting and combusting fuel
contribute to ambient levels of pollutants that contribute to adverse
effects on vegetation. Volatile organic compounds, some of which are
considered air toxics, have long been suspected to play a role in
vegetation damage.\2373\ In laboratory experiments, a wide range of
tolerance to VOCs has been observed.\2374\ Decreases in harvested seed
pod weight have been reported for the more sensitive plants, and some
studies have reported effects on seed germination, flowering and fruit
ripening. Effects of individual VOCs or their role in conjunction with
other stressors (e.g., acidification, drought, temperature extremes)
have not been well studied. In a recent study of a mixture of VOCs
including ethanol and toluene on herbaceous plants, significant effects
on seed production, leaf water content and photosynthetic
[[Page 24872]]
efficiency were reported for some plant species.\2375\
---------------------------------------------------------------------------
\2373\ U.S. EPA (1991). Effects of organic chemicals in the
atmosphere on terrestrial plants. EPA/600/3-91/001.
\2374\ Cape JN, ID Leith, J Binnie, J Content, M Donkin, M
Skewes, DN Price AR Brown, AD Sharpe. (2003). Effects of VOCs on
herbaceous plants in an open-top chamber experiment. Environ.
Pollut. 124:341-343.
\2375\ Cape JN, ID Leith, J Binnie, J Content, M Donkin, M
Skewes, DN Price AR Brown, AD Sharpe. (2003). Effects of VOCs on
herbaceous plants in an open-top chamber experiment. Environ.
Pollut. 124:341-343.
---------------------------------------------------------------------------
Research suggests an adverse impact of vehicle exhaust on plants,
which has in some cases been attributed to aromatic compounds and in
other cases to nitrogen oxides.2376 2377 2378 The impacts of
VOCs on plant reproduction may have long-term implications for
biodiversity and survival of native species near major roadways. Most
of the studies of the impacts of VOCs on vegetation have focused on
short-term exposure and few studies have focused on long-term effects
of VOCs on vegetation and the potential for metabolites of these
compounds to affect herbivores or insects.
---------------------------------------------------------------------------
\2376\ Viskari E.-L. (2000). Epicuticular wax of Norway spruce
needles as indicator of traffic pollutant deposition. Water, Air,
and Soil Pollut. 121:327-337.
\2377\ Ugrekhelidze D, F Korte, G Kvesitadze (1997). Uptake and
transformation of benzene and toluene by plant leaves. Ecotox.
Environ. Safety 37:24-29.
\2378\ Kammerbauer H, H Selinger, R Rommelt, A Ziegler-Jons, D
Knoppik, B Hock. (1987). Toxic components of motor vehicle emissions
for the spruce Picea abies. Environ. Pollut. 48: 235-43.
---------------------------------------------------------------------------
(c) How the Agencies Estimated Impacts on Emissions
The rule implements an emissions inventory methodology for
estimating impacts. Vehicle emissions inventories are often described
as three-legged stools, comprised of activity (i.e., miles traveled,
hours operated, or gallons of gasoline burned), population (or number
of vehicles), and emission factors. An emissions factor is a
representative value that attempts to relate the quantity of a
pollutant released to the atmosphere with an activity associated with
the release of that pollutant.\2379\ Depending on the vehicle activity
available, emission factors may be on a distance-, time-, or fuel-
basis. For example, an emissions inventory for a light-duty fleet could
simply be the vehicle miles traveled multiplied by the appropriate per-
mile emission factor for a chosen pollutant.
---------------------------------------------------------------------------
\2379\ USEPA, Basics Information of Air Emissions Factors and
Quantification, https://www.epa.gov/air-emissions-factors-and-quantification/basic-information-air-emissions-factors-and-quantification.
---------------------------------------------------------------------------
As described in Section VI.A, Overview of Methods, the agencies
used specific models to develop inputs to the CAFE model, such as fuel
prices and emission factors. The CAFE model estimates how manufacturers
might respond to a given regulatory scenario (CAFE/CO2
standards) and fuel prices, and what impact that response will have on
emissions. As mentioned above, the agencies have used DOT's CAFE model
to estimate impacts of the CAFE and CO2 standards
promulgated today. Details of the analysis are presented below and in
the accompanying RIA, EIS, and model documentation. To estimate the
response on emissions, several steps are involved. The estimation of
emissions involves accounting for vehicular fuel type (e.g., gasoline,
diesel, electric) and fuel economy (accounting for the estimated gap,
discussed below, between ``laboratory'' and actual on-road fuel
economy), vehicular turnover and travel demand, fuel properties (carbon
content), and upstream process emissions. Like other models, the CAFE
model includes procedures to estimate annual rates at which new
vehicles are used and subsequently scrapped. Together, these procedures
result in, for each vehicle model in each model year, estimates of the
number remaining in service in each calendar year, as well as the
annual mileage accumulation (i.e. VMT) in each calendar year.
Quantities of emissions derive from this vehicle operation.
For every vehicle model in the market file, the model estimates the
VMT per vehicle (using the assumed VMT schedule, the vehicle fuel
economy, fuel price, and the rebound assumption). Those miles are
multiplied by the number off each vehicle model/configuration remaining
in service in any given calendar year. Fuel consumption is the product
of miles driven and fuel economy, which can be tracked by model year
cohort in the model. Carbon dioxide emissions from vehicle tailpipes
are the simple product of gallons consumed and the carbon content of
each gallon. As discussed in the CAFE model overview, the simulated
application of technology results in estimates of the cost, fuel type,
fuel economy, and fuel share applicable to each vehicle model in each
model year. Together with quantities of travel, and with estimates of
the ``gap'' between ``laboratory'' and ``on-road'' fuel economy, these
enable calculation of quantities of fuel consumed in each year during
the useful life of each vehicle model produced in each model year. The
model calculates emissions of CO2, CH4, and
N2O, criteria pollutants, and air toxics, reporting
emissions both from vehicle tailpipes and from upstream processes
(e.g., petroleum refining) involving in producing and supplying fuels.
In order to calculate calendar year fuel consumption, the model
needs to account for the inherited on-road fleet in addition to the
model year cohorts affected by this rule. Using the VMT of the average
passenger car and light truck from each cohort, the model computes the
fuel consumption of each model year class of vehicles for its age in a
given CY. The sum across all ages (and thus, model year cohorts) in a
given CY provides estimated CY fuel consumption.
For this rule, vehicle tailpipe (downstream) and upstream emission
inventories were developed separately. In addition to the tailpipe
emissions of carbon dioxide, each gallon of gasoline produced for
consumption by the on-road fleet has associated ``upstream'' emissions
that occur in the extraction, transportation, refining, and
distribution of the fuel. The tailpipe inventories apply per-mile
emission factors from the Motor Vehicle Emission Simulator (MOVES) and
the upstream inventories apply per-gallon of fuel consumed emission
factors from the Argonne National Laboratory's Greenhouse gases,
Regulated Emissions, and Energy use in Transportation (GREET) Model.
The model accounts for upstream emissions and reports them accordingly.
More detailed descriptions of emission data sources and calculations
are provided in the following section.
The agencies received several comments on estimation of criteria
pollutant impacts in the NPRM. As discussed elsewhere in this preamble,
EDF modified aspects of the CAFE model as part of their comments to the
agencies. Specifically in regards to criteria pollutant emissions, EDF
made several alternative assumptions, including assertions that
criteria pollutant impacts were not as negligible as the agencies
claimed, and that fatalities due to criteria pollutant emissions would
be higher than the agencies showed in the NPRM. The agencies declined
to adopt EDF's suggested changes to the model and inputs, but did make
the changes discussed in this section that refined the agencies'
accounting of criteria pollutant emissions and explicitly modeled
criteria pollutant fatalities, as discussed below.
Also discussed elsewhere in this preamble, some commenters
expressed that the agencies' analysis (by implication, their modeling)
should account for some States' mandates that manufacturers sell
minimum quantities of ``Zero Emission Vehicles'' (ZEVs).\2380\ These
commenters stressed the
[[Page 24873]]
importance of the ZEV mandate in relation to maintaining air quality
requirements and reducing effects of climate change.
---------------------------------------------------------------------------
\2380\ CBD et al., NHTSA-2018-0067-12123; States and Cities,
NHTSA-2018-0067-11735; SCAQMD, NHTSA-2018-0067-11813.
---------------------------------------------------------------------------
The reference case analysis for today's rule, like that for the
proposal, does not simulate compliance with ZEV mandates,\2381\ because
such mandates are subject to preemption under EPCA and are therefore
not enforceable. As discussed in the One National Program Action,
California and other states remain free to revise their overall average
emissions standards to further reduce ozone forming emissions and seek
a waiver of Clean Air Act preemption from EPA, as described above,
while not violating NHTSA's preemption authority. These States and
local governments would continue to be allowed to take other actions so
long as those are not related to fuel economy and are consistent with
any other relevant Federal law.
---------------------------------------------------------------------------
\2381\ The NPRM version of the model included experimental
capabilities to account for mandates and credits for the sale of
ZEVs, but the agencies did not utilize those capabilities for the
NPRM for the same reasons discussed above.
---------------------------------------------------------------------------
(1) Activity Levels
As discussed in Section VI.A, for each vehicle model/configuration
in each model year during 2017-2050, the CAFE model estimates and
records the fuel type (e.g., gasoline, electricity), fuel economy, and
number of units sold in the U.S. The model also makes use of an
aggregated representation of vehicles sold in the U.S. during 1978-
2016. The model estimates the numbers of each cohort of vehicles
remaining in service in each calendar year, and the amount of driving
accumulated by each such cohort in each calendar year. The CAFE model
estimates annual vehicle-miles of travel (VMT) for each individual car
and light truck model produced in each model year at each age of their
lifetimes, which extend for a maximum of 40 years. Since a vehicle's
age is equal to the current calendar year minus the model year in which
it was originally produced, the age span of each vehicle model's
lifetime corresponds to a sequence of 40 calendar years beginning in
the calendar year corresponding to the model year it was
produced.\2382\ These estimates reflect the gradual decline in the
fraction of each car and light truck model's original model year
production volume that is expected to remain in service during each
year of its lifetime, as well as the well-documented decline in their
typical use as they age. Using this relationship, the CAFE model
calculates total VMT for the entire fleet of cars and light trucks in
service during each calendar year spanned by the agencies' analysis.
---------------------------------------------------------------------------
\2382\ In practice, many vehicle models bearing a given model
year designation become available for sale in the preceding calendar
year, and their sales can extend through the following calendar year
as well. However, the CAFE model does not attempt to distinguish
between model years and calendar years; vehicles bearing a model
year designation are assumed to be produced and sold in that same
calendar year.
---------------------------------------------------------------------------
Based on these estimates, the model also calculates quantities of
each type of fuel or energy, including gasoline, diesel, and
electricity, consumed in each calendar year. By combining these with
estimates of each model's fuel or energy efficiency, the model also
estimates the quantity and energy content of each type of fuel consumed
by cars and light trucks at each age, or viewed another way, during
each calendar year of their lifetimes. As with the accounting of VMT,
these estimates of annual fuel or energy consumption for each vehicle
model and model year combination are combined to calculate the total
volume of each type of fuel or energy consumed during each calendar
year, as well as its aggregate energy content.
The procedures the CAFE model uses to estimate annual VMT for
individual car and light truck models produced during each model year
over their lifetimes and to combine these into estimates of annual
fleet-wide travel during each future calendar year, together with the
sources of its estimates of their survival rates and average use at
each age, are described in detail in Section VI.D.1 of this final rule.
The data and procedures it employs to convert these estimates of VMT to
fuel and energy consumption by individual model, and to aggregate the
results to calculate total consumption and energy content of each fuel
type during future calendar years, are also described in detail in that
same section.
The model documentation accompanying today's notice describes these
procedures in detail.\2383\ The quantities of travel and fuel
consumption estimated for the cross section of model years and calendar
years constitutes a set of ``activity levels'' based on which the model
calculates emissions. The model does so by multiplying activity levels
by emission factors. As indicated in the previous section, the
resulting estimates of vehicle use (VMT), fuel consumption, and fuel
energy content are combined with emission factors drawn from various
sources to estimate emissions of GHGs, criteria air pollutant, and
airborne toxic compound that occur throughout the fuel supply and
distribution process, as well as during vehicle operation, storage, and
refueling. Emission factors measure the mass of each GHG or criteria
pollutant emitted per vehicle-mile of travel, gallon of fuel consumed,
or unit of fuel energy content. The following section identifies the
sources of these emission factors and explains in detail how the CAFE
model applies them to its estimates of vehicle travel, fuel use, and
fuel energy consumption to estimate total annual emissions of each GHG,
criteria pollutant, and airborne toxic.
---------------------------------------------------------------------------
\2383\ CAFE model documentation is available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
(2) What emission factors did the agencies apply?
(a) Tailpipe (Downstream) Emission Factors
In a full fuel cycle analysis, emissions that occur from the
fueling pump to vehicle wheels are usually referred to as tailpipe or
simply downstream emissions. Today's rule primarily impacts
CO2 emissions. The agencies have calculated tailpipe
CO2 emissions based on fuel consumption and fuel properties
(i.e., fuel density and carbon content) that result in gram per gallon
emission factors. For all other exhaust constituents (except sulfur
dioxide, discussed below), the agencies have calculated emissions by
applying per-mile emission factors to quantities of travel (i.e., VMT).
This rulemaking's tailpipe emission factors are from EPA's Motor
Vehicle Emission Simulator (MOVES), which serves as the federal
regulatory model for mobile-source emission inventories, with a few
notable exceptions. In particular, light-duty gasoline and diesel
tailpipe emission factors for the following criteria pollutants,
greenhouse gases (other than CO2), and air toxics are drawn
from MOVES2014a: \2384\
---------------------------------------------------------------------------
\2384\ For the emission factors informing the Final EIS,
updating to MOVES 2014b would have produced values identical to
those based on MOVES 2014a.
Criteria pollutants
[cir] Carbon monoxide (CO),
[cir] Volatile organic compounds (VOC),
[cir] Nitrogen oxides (NOX), and
[cir] Fine particulate matter (PM2.5)
Greenhouse gases
[cir] Methane (CH4), and
[cir] Nitrous oxide (N2O)
Air toxics
[cir] Acetaldehyde,
[cir] Acrolein,
[cir] Benzene,
[cir] Butadiene,
[cir] Formaldehyde,
[cir] Diesel particulate matter (DPM10), and
[[Page 24874]]
[cir] Methyl tert-butyl ether (MTBE)
These MOVES-based emission factors are specified separately for
gasoline and diesel vehicles, by model year (ranging from MY 1975 to
2050), and by vehicle age (ranging from zero to 39 years old). The
structure of criteria pollutant emission standards is such that these
factors do not vary with fuel economy unless a change in fuel type
(e.g., from gasoline to electricity) is involved.
Since tailpipe sulfur dioxide (SO2) emissions are
dependent on the sulfur content of the fuel, a single SO2
emission factor in grams per million British thermal units (MMBTU) of
fuel consumed is applied respectively for gasoline, diesel, and ethanol
(E85) across all model years after MY 2017 based on a longitudinal
analysis in MOVES.
As previously mentioned, EDF submitted supplemental comments on
SO2 emissions, stating that ``SO2 emissions
should be proportional to fuel consumption'' and ``that the tailpipe
SO2 emissions by calendar year from the Volpe Model do not
change proportionally to the changes in fuel consumption across various
CO2 control scenarios.'' \2385\ The version of the model
supporting the 2012 final rule calculated tailpipe SO2
emissions on a gram per gallon basis. Supporting the ensuing rulemaking
regarding heavy-duty pickups and vans, and the 2016 draft TAR, EPA
staff provided SO2 emission factors specified on a gram per
mile basis. DOT modified the model in order to apply these
SO2 emission factors as provided by EPA. The CAFE Model
documentation released with the NPRM clearly describes how the agencies
calculated emissions in the model. Although the version of model
applied for the NPRM did not change this approach to calculating
tailpipe SO2 emissions, the agencies agree that
SO2 emissions should be proportional to fuel consumption,
and DOT has revised the model accordingly. For SO2
emissions, the inputs to the model include the number of grams of
SO2 emitted by a vehicle per gallon of fuel consumed by the
vehicle.
---------------------------------------------------------------------------
\2385\ EDF, NHTSA-2018-0067-12363.
---------------------------------------------------------------------------
The agencies also received comments on the use of MOVES. Most
notably, the National Farmers Union stated ``Concerns have been raised
regarding the models used by EPA to determine emissions from fuels.
Third-party reviews have shown that MOVES2014 may be inadequate as a
tool for estimating the exhaust emissions of gasoline blends containing
more than 10 percent ethanol. The model's results for mid-level ethanol
blends have been shown to be inconsistent with other results from the
scientific literature for both exhaust emissions and evaporative
emissions, including results from real-world emissions testing.''
\2386\ The agencies considered comments on the use of MOVES and ethanol
blends and notes that MOVES may be unreliable for fuel blends over E10;
however, MOVES is not designed to model mid-level ethanol blends.
MOVES2014 is designed to model ethanol volumes up to 15 percent (E0 to
E15), and it can also model E85 (ethanol volumes of 70 to 85 percent),
but MOVES2014 is not designed to model intermediate fuel blends.
Moreover, the agencies did not explicitly consider blends above E10 as
part of the analysis, but rather ethanol blending is considered in
relation to how to achieve a higher octane level and a higher anti-
known index.
---------------------------------------------------------------------------
\2386\ National Farmers Union, NHTSA-2018-0067-11972.
---------------------------------------------------------------------------
The Pennsylvania Department of Environmental Protection stated that
there may be a significant State-specific rebound effect in
Pennsylvania given Pennsylvania's regional role in natural gas and
petroleum processing and refining. According to this commenter, the
proposed rule does not adequately take into account significant local,
State, and regional air quality impacts because it dilutes the
emissions impact of the rule across the entire Nation. The Center for
Biological Diversity, the Consumer Federation of America, and other
commenters expressed concern that the proposed rule would increase
criteria pollutants in areas with large minority populations,
especially those in areas near oil refineries.
Results of these tailpipe emissions calculations are summarized
below in Section VII and in the FRIA accompanying today's notice, and
presented in greater detail in the accompanying Final EIS.
(b) Upstream Emission Factors
Fuel cycle emissions occurring between the extraction well and the
fueling pump are often called upstream emissions. This rule has drawn
upstream emission factors exclusively from the Greenhouse gases,
Regulated Emissions, and Energy use in Transportation (GREET) model,
developed by the U.S. Department of Energy's Argonne National
Laboratory. The upstream gasoline, diesel, and electricity emission
factors for criteria pollutants--namely, CO, VOC, NOX,
PM2.5, and SO2--and greenhouse gases--namely,
CO2, CH4, and N2O--have been updated
with GREET 2018 data. The upstream emission factors for the air toxics
mentioned above were unchanged from the proposal. For the final rule,
upstream emission factors cover the following analysis years, 2017,
2020, 2025, 2030, 2035, 2040, 2045, and 2050, and four distinct
upstream processes:
Petroleum Extraction,
Petroleum Transportation,
Petroleum Refining, and
Fuel Transportation, Storage, and Distribution (TS&D).
These upstream emission factors for each fuel type and analysis
year were generated by a process using emission factor values found in
the GREET 2018 spreadsheet tool and adjustment factors where
appropriate. Emission factors for the petroleum extraction process are
the aggregation of different crude feedstock--such as crude oil, oil
sands, and shale oil--emission factors multiplied by their associated
adjustments for transportation to refineries losses, storage losses,
and energy share by crude feedstock. Emission factors for the petroleum
transportation process are emissions by crude feedstock sources--such
as crude oil fields, surface and in-situ mining, and shale reserves--
and multiplied the associated energy shares. Emission factors for the
petroleum refining are the sum of the crude input, combustion, and non-
combustion products multiplied by the transportation of blended fuel
loss factors. The refining emission factors applies a non-ethanol
energy content adjustment for gasoline, blended at E10. Diesel does not
have any such ethanol content adjustment. Emission factors for the Fuel
TS&D process are based on the blended fuel transportation and
distribution emissions as well as an energy content factor for both the
petroleum and ethanol portions of the fuels. Again, diesel does not
have an ethanol adjustment.
The aggregated upstream emission factors used in the rule are
aggregated across the four processes for each fuel type and analysis
year. The aggregated upstream emission factor in the sum of the fuel
TS&D emission factor, the petroleum refining emission factor multiplied
by the share of fuel savings leading to reduced domestic refining, the
pair of petroleum extraction and transportation emission factors
multiplied by both the share of fuel savings and the share of reduced
domestic refining from domestic crude. The upstream adjustments are
replicated from the proposal.
Finally, the upstream emission factors for electricity are also
updated with GREET 2018 data. Upstream electricity emissions factors
are derived from
[[Page 24875]]
electricity for transportation use feedstock and fuel emissions by
analysis year. As the analysis supporting the proposal noted, there are
three possible supply ``pathways'' for fuel consumed by the U.S. light-
duty vehicle fleet:
1. Importing fuel that has been refined overseas into the U.S.
2. Refining fuel in the U.S. from crude petroleum produced overseas
and imported into the U.S.
3. Refining fuel in the U.S. from crude petroleum produced in the
U.S.\2387\
---------------------------------------------------------------------------
\2387\ The proposal assumed that all fuel refined outside the
U.S. and then imported into the U.S. would be refined from petroleum
that was also produced outside the U.S. Although some of it could be
refined from crude petroleum produced in the U.S. and exported, the
analysis assumed that the fraction supplied via this pathway is
negligible.
---------------------------------------------------------------------------
The distribution of fuel consumed within the U.S. that is supplied
via each of these pathways has important implications for domestic
``upstream'' emissions, because each pathway produces domestic
emissions arising from a different combination of activities that occur
within the U.S. For example, pathway 1 involves domestic emissions that
occur during crude petroleum extraction, transportation of crude oil
from production or nearby temporary storage facilities to domestic
refineries, refining of crude petroleum to produce transportation
fuels, and storage and distribution of refined fuels.\2388\ In
contrast, pathway 2 generates domestic emissions during transportation
of crude petroleum from U.S. coastal ports to domestic refineries, as
well as from fuel refining, storage, and distribution, while pathway 3
produces domestic emissions only from storage and distribution of
refined fuel.
---------------------------------------------------------------------------
\2388\ By longstanding EPA convention, emissions that occur when
vehicles are being refueled at retail stations or vehicle storage
depots (such as buses) are ascribed to vehicle use, rather than to
fuel supply.
---------------------------------------------------------------------------
The analysis supporting the proposal made two central assumptions
in estimating upstream emissions from fuel supply. First, 50 percent of
any change in domestic fuel consumption by cars and light trucks
operating on petroleum-based liquid fuels (gasoline and diesel) would
be reflected in changes in imports of refined fuel, while the remaining
50 percent would be reflected in changes in the volume of those fuels
refined domestically. Second, 90 percent of any change in the volume of
fuel refined domestically was assumed to be reflected in changes in the
volume of crude petroleum imported into the U.S, with the remaining 10
percent reflected in changes in the volume of crude petroleum produced
within the U.S. The agencies developed these assumptions to analyze the
environmental impacts of alternative CAFE and CO2 standards
for model years 2012-2016, and have continued to rely in their analyses
supporting subsequent rules.
To illustrate the effect of these assumptions, for each increase in
domestic fuel consumption of 100 gallons, 50 additional gallons would
be supplied via pathway 1 (refined outside the U.S. and imported in
already-refined form). Additional fuel supplied via pathway 2 (U.S.
domestic refining of imported crude oil) would account for 90 percent
of the remaining 50 gallons of increased consumption, or 45 gallons.
Finally, the remaining 5 gallons of increased fuel consumed within the
U.S. would be supplied via pathway 3 (domestic refining of crude oil
produced within the U.S.). This same breakdown was applied to changes
in fuel consumption estimated to occur throughout the analysis period
used for the proposal, which extended from 2017 through 2050.
The agencies estimated the resulting changes in upstream emissions
of criteria air pollutants and airborne toxics occurring within the
U.S. by applying emission factors for the appropriate stages of the
fuel supply chain (petroleum extraction, petroleum transportation to
refineries, fuel refining, and fuel storage and distribution) to the
changes in the total energy content of fuel supplied by each pathway,
and summed the results.\2389\ The energy content of fuel rather than
its volume was used as the basis for estimating emissions, because
emission factors are typically expressed in mass per unit of fuel
energy supplied--for example, grams per million Btu--rather than per
unit volume of fuel supplied.
---------------------------------------------------------------------------
\2389\ Increases in upstream GHG emissions were calculated from
the increase in U.S. domestic fuel consumption, without regard to
whether they occurred within the U.S.
---------------------------------------------------------------------------
In the proposal, the agencies made no explicit assumptions about
the future mix of electric generating capacity that would be used to
supply increased electricity consumed by BEVs and PHEVs. Instead, the
agencies implicitly relied on the assumptions about future evolution of
the nationwide mix of generation sources that were reflected in the
U.S. average emission factors for electricity produced to power
transportation vehicles, including cars and light trucks, which as
described previously were drawn from the most recent version of Argonne
National Laboratory's GREET model that was available at the time of the
proposal. These assumptions were consistent with those made by EIA in
its AEO 2017 Reference case analysis and publications.\2390\
---------------------------------------------------------------------------
\2390\ https://greet.es.anl.gov/publication-greet-2017-summary.
---------------------------------------------------------------------------
While the agencies' use of these assumptions to estimate upstream
emissions did not prompt widespread comments on their analyses in
support of previous CAFE rulemakings, the more recent proposal did draw
a large number of comments focusing on those same assumptions. Most
commenters asserted that the entirety of any increase in consumption of
petroleum-based fuels by cars and light trucks resulting from the
proposal would be met via increased domestic refining, primarily from
crude petroleum produced in the U.S., and would thus generate
additional upstream emissions within the U.S. throughout the fuel
supply process. Even some commenters who argued elsewhere that the U.S.
would continue to be a large-scale importer of petroleum asserted that
the entire increase in fuel consumption resulting from the proposal
would be refined from additional domestically-produced petroleum.\2391\
---------------------------------------------------------------------------
\2391\ For example, IPI notes that AEO 2019 shows the U.S. will
continue to import crude petroleum through 2050, and will remain a
net importer as measured by the energy content rather than the
volume of U.S. petroleum exports and imports; see IPI, NHTSA-2018-
0067-12213. Similarly, EDF argued that because U.S. petroleum
imports have been declining and gasoline imports are currently low,
the best assumption was that the entire increase in gasoline
consumption resulting from the proposal would be supplied from
increased domestic refining of U.S.-produced crude petroleum; see
EDF, NHTSA-2018-0067-12108.
---------------------------------------------------------------------------
As a consequence, most commenters argued that the agencies'
analysis of the proposal significantly underestimated the increases in
upstream emissions that were likely to result, with some also asserting
that the increases in emissions of criteria air pollutants would cause
potentially serious degradation of air quality in the areas surrounding
U.S. refineries. For example, EDF stated, ``NHTSA assumed that 50% of
all the gasoline saved by more stringent CAFE and CO2
standards would have been imported (i.e., refined overseas). . . . It
is difficult to see how this could be the case when the nation is
producing enough crude oil to be a net exporter. It is also difficult
to see how this could be the case when gasoline consumption is
decreasing and sufficient domestic refining capacity exists to fulfill
today's demand, let alone decreased demand in the future. . . .
Assuming that 100% of the differences in gasoline consumption between
control scenarios will be refined in the U.S. appears to be much more
consistent with the available data. Likewise, it seems reasonable to
assume
[[Page 24876]]
that differences in the crude oil requirements of the various scenarios
will also affect domestic production more so than imports.'' \2392\
---------------------------------------------------------------------------
\2392\ EDF, NHTSA-2018-0067-12108, p. 53. Others making similar
assertions include IPI, NHTSA-2018-0067-12213, p. 5.
---------------------------------------------------------------------------
However, one commenter did agree with the agencies' assessment of
the proposal's likely impact on U.S. petroleum imports, noting that
``Through 2050, there will only be a small increase in domestic oil
production due to increased demand, well under 1%. . . . The vast
majority (88% through 2050) of the additional petroleum that will be
required to fuel light-duty vehicles in the proposed case will be
imported. This assessment is not too far off of a single comment in the
NPRM, `Using NEMS, it was estimated that 50% of increased gasoline
consumption would be supplied by increased domestic refining and that
90% of this additional refining would use imported crude petroleum.' ''
\2393\
---------------------------------------------------------------------------
\2393\ David Gohlke, EPA-HQ-OAR-2018-0283-5082, p. 1.
---------------------------------------------------------------------------
The agencies note that there seems to be considerable confusion
among commenters about the agencies' assumptions regarding import
shares, and what they are attempting to measure. The agencies'
assumptions are intended to measure the effects of changes in
consumption of petroleum-derived transportation fuels by cars and light
trucks that are attributable to this final rule on changes in U.S.
production and imports of crude petroleum, in domestic refining of
crude petroleum to produce transportation fuels, and in the volume of
refined fuel distributed for domestic consumption. While recent data on
U.S. fuel consumption, domestic production and imports of crude
petroleum, and imports of refined petroleum products may be useful in
estimating these desired measures, they are not themselves measures of
the marginal impacts of changes in fuel consumption on the volumes of
fuel supplied via each of the supply pathways described previously.
Instead, the agencies rely on two types of information to estimate
the current and likely future values of the desired measures. First,
they examine recent changes in domestic consumption of petroleum-based
motor fuels--particularly gasoline, since it is the primary fuel used
by vehicles that are subject to CAFE and CO2 standards--and
compare them to the accompanying changes in the three gasoline supply
pathways, namely domestic petroleum production, U.S. imports of crude
petroleum, and U.S. imports of refined gasoline (or components that are
blended domestically to produce gasoline). Second, the agencies examine
differences in forecasts of U.S. petroleum production, fuel refining,
and imports of refined fuel under alternative future scenarios that
were included in AEO 2018 whose projections of domestic fuel
consumption differ in ways that include alternative CAFE standards.
While this latter approach would ideally compare scenarios that differ
only in their assumptions about the stringency of CAFE and
CO2 standards but are otherwise strictly comparable, such
idealized comparisons are rarely possible because other factors almost
always differ as well between the alternative scenarios being compared.
(i) Assumptions Used To Analyze Impacts of the Final Rule on Petroleum
Imports and Emissions
In response to comments, the agencies conducted a detailed
examination of recent changes in U.S. fuel consumption, domestic fuel
refining, and U.S. imports and exports of crude petroleum as well as
refined fuel (primarily gasoline). This included comparing changes in
these variables at both the national aggregate level and for three
separate regions of the U.S. In addition, they examined differences in
the forecast values of these variables under alternative assumptions
about fuel economy standards, although as indicated above these
comparisons are complicated by the fact that factors other than CAFE
and CO2 standards also differ between these alternative
scenarios.
The agencies also identified a fourth ``pathway'' to supply the
increase in U.S. gasoline consumption anticipated to result from this
final rule. The U.S. is now a net exporter of refined gasoline (and
products that are blended to produce gasoline), and the volume of U.S
gasoline exports is likely to increase for at least the next two
decades. This introduces the possibility that some--and perhaps all--of
the anticipated increase in domestic gasoline consumption will be met
simply by redirecting U.S. gasoline exports to serve domestic
consumption. This additional source of supply would result in no
increase in domestic refining activity, and thus no increase in
emissions from refining of petroleum-based transportation fuels.\2394\
---------------------------------------------------------------------------
\2394\ Increased domestic emissions would only occur in this
case to the extent that domestic distribution of gasoline entailed
higher emissions than transporting it to U.S. coastal ports for
export.
---------------------------------------------------------------------------
Throughout most of the past half-century, the nation has been a
large net importer of crude petroleum, taking its price as determined
in world markets and importing the volumes necessary to meet the
difference between U.S. demand for refined petroleum products and
domestic supplies. Throughout this period, the U.S. has also been
largely self-sufficient in refining, meaning that the gap between
domestic demand for refined products and the volumes refined from crude
petroleum extracted within the U.S. was primarily met by domestic
refining of imported crude petroleum, with only marginal volumes of
gasoline and other products imported or exported. U.S. refinery
capacity and output generally increased over this period in proportion
to growth in domestic consumption of fuel and other products refined
from petroleum.
In the past decade, however, this situation has changed
dramatically. U.S. production of crude petroleum has more than doubled
since 2008, making the nation one of the world's largest producers,
while net imports of crude oil and refined products have declined by
nearly 80 percent.\2395\ Domestic gasoline consumption declined by more
than 6 percent between 2007 and 2012, and recovered to its 2007 levels
only as recently as 2016, remaining near or slightly below its 2016
level since then.\2396\ As a consequence, the U.S. shifted from being a
net importer of refined petroleum products to a net exporter in 2011,
and has become a net exporter of gasoline and ``blending stock'' since
2016.\2397\
---------------------------------------------------------------------------
\2395\ These and other petroleum statistics cited here were
calculated from data available at EIA, Petroleum and Other Liquids,
2019, https://www.eia.gov/petroleum/data.php. U.S. production of
crude petroleum rose from 1.83 billion barrels in 2008 to 4.01
billion barrels in 2018, or by 119%, During that same period, net
U.S. imports of crude petroleum and refined products declined from
4.07 billion to 0.85 billion barrels, or by 79%. Net U.S. imports
are the difference between the nation's total (or gross) imports
from elsewhere in the world and the volumes it exports to other
nations.
\2396\ U.S. gasoline consumption declined from 3.39 billion
barrels in 2007 to 3.18 billion barrels in 2012, or by 6.2 percent,
rose to 3.41 billion barrels in 2016, and remained near that level
through 2018.
\2397\ In 2010, U.S. net imports of refined petroleum products
were 98 million barrels, but by 2011 U.S. net exports were 160
million barrels. U.S. net exports of refined products then increased
steadily through 2018, reaching 1.23 billion barrels in that year.
In 2015, U.S. net imports of gasoline and blending components
totaled 19 million barrels, but by 2016, U.S. net exports were 20
million barrels, and grew to 93 million barrels in 2018. Another
recent change in petroleum markets has been the increasing
production and trade in gasoline blendstock in domestic and
international petroleum trade. While in earlier periods refineries
normally produced finished gasoline and shipped it to local storage
terminals for distribution and retailing, in recent years,
refineries have increasingly shifted to producing standardized
gasoline blendstocks, such as Reformulated Blendstock for Oxygenate
Blending (or ``RBOB''), which are then shipped and blended with
ethanol or other additives to make finished gasoline that meets
local regulatory requirements or customer specifications. Although
this process has clear cost and operational advantages, particularly
with extensive geographic and seasonal variation in gasoline
formulations, it complicates the tabulation and comparison of
petroleum statistics. In both EIA and most international trade
statistics, finished gasoline and blendstocks are treated as
separate products, and as reported in EIA statistics, large volumes
of finished gasoline are now produced from blendstocks by local
``blenders,'' rather than by more centralized ``refiners.'' In
addition, the volume of refinery production of gasoline and
blendstock is now systematically lower than consumption of finished
gasoline, because up to 10 percent of the volume of gasoline sold at
retail can be made up of ethanol that is blended into gasoline after
it leaves the refinery.
---------------------------------------------------------------------------
[[Page 24877]]
Over the past decade, increased availability of crude petroleum and
other refinery feedstocks in combination with declining gasoline
consumption has presented U.S. refiners with a choice between
continuing to produce gasoline at or near their capacity while boosting
exports, or cutting back on refinery output. U.S. refiners elected not
to cut back on their production of gasoline; instead, they actually
increased the volume they refined. U.S. production of finished gasoline
increased by 9 percent between 2007 and 2018.
The excess of gasoline production resulting from increased refinery
capacity and stable consumption has partly displaced previous gasoline
and blendstock imports, with the remainder taking the form of increased
U.S. exports. Thus, as Figure VI-92 below shows, the nation now has a
capacity to produce gasoline that considerably exceeds its current
domestic consumption. This surplus of gasoline appears likely to
increase in coming few years, as EIA's Annual Energy Outlook 2019
reference case (EIA, 2019) anticipates that domestic gasoline
consumption will continue to decline until nearly 2040. Therefore, the
U.S. seems likely to remain a net exporter of gasoline through the next
three decades.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.513
Although EIA's Annual Energy Outlook does not include separate
forecasts of gasoline exports and imports, that same agency's Short
Term Energy Outlook projects that U.S. gasoline exports will continue
to rise through 2020 (EIA, 2019).\2398\ Combined with EIA's reference
case forecast in the AEO 2019, the forecasts of declining U.S. gasoline
consumption and rising net exports of refined petroleum products
suggest that the United States will remain a growing net exporter of
refined petroleum products--including gasoline--through nearly 2040. In
turn, this suggests that any increase in domestic gasoline consumption
resulting from this final rule is likely to
[[Page 24878]]
low anticipated growth in U.S. exports, rather than prompting growth in
domestic refining and associated upstream emissions.
---------------------------------------------------------------------------
\2398\ AEO does not forecast gasoline refining, imports, or
exports separately, instead reporting them as part of total refined
petroleum products.
---------------------------------------------------------------------------
Regional patterns of U.S. gasoline consumption, refining, and trade
also suggests that redirecting U.S. gasoline exports to domestic
markets is likely to be an important source of additional supply to
meet any increase in U.S. consumption stemming from this final rule.
The nation's East Coast (which comprises the Energy Information
Administration's Production and Distribution District 1, or PADD 1)
currently accounts for about 32 percent of U.S. gasoline consumption,
but has historically produced significantly less than gasoline than it
consumes. As Figure VI-93 below shows, the gap between consumption and
local supply within PADD1 has recently narrowed, as gasoline production
along the East Coast has increased rapidly in recent years, while
shipments into the region from the remainder of the U.S. and foreign
imports (which come mostly from Canada) declined. In June 2019,
however, press reports suggested that that one of the largest East
Coast refineries (Philadelphia Energy Solutions, which represents some
28 percent of East Coast refining capacity) would be closed.\2399\ At
the same time, construction of new refineries continues to be hindered
by the density of population concentrations and commercial development
along the nation's East Coast, casting doubt on the potential for
continued increases in local gasoline refining and supply within PADD
1.
---------------------------------------------------------------------------
\2399\ Seba, E. (2019, July 5). Philadelphia refinery closing
reverses two years of U.S. capacity gains. Retrieved September 19,
2019, from Reuters: https://www.reuters.com/article/us-usa-refinery-blast-capacity/philadelphia-refinery-closing-reverses-two-years-of-u-s-capacity-gains-idUSKCN1U0283.
[GRAPHIC] [TIFF OMITTED] TR30AP20.514
BILLING CODE 4910-59-C
As a consequence, it seems likely that at least in the near term,
any increase in gasoline consumption along the Nation's East Coast in
response to this rule would be supplied primarily by Gulf Coast
refineries or increased foreign imports, rather than from increased
production in East Coast refineries. Pipelines available to transport
refined petroleum products from Gulf Coast refineries to the East Coast
may also face capacity limitations, in which case most of any increase
in gasoline consumption there would need to be met by increased imports
from abroad. Over the longer term, however, it is possible that
increases in East Coast gasoline consumption could be met partly by
expanded refining activity within the region.
The West Coast, which includes Nevada and Arizona (EIA's PADD 5),
currently accounts for 168 percent of
[[Page 24879]]
U.S. gasoline consumption. Almost all of the gasoline consumed in that
region is also refined within it, although small volumes are shipped
into Arizona from neighboring PADDs by pipeline, and small volumes are
also exported to Latin America by tanker. The West Coast is relatively
isolated from other U.S. sources of refined gasoline by long
transportation distances and limited pipeline capacity, while import
terminals for crude petroleum are relatively numerous, and it therefore
appears more likely that marginal increases in gasoline consumption
from the rule will be met from increases in local (i.e., within-PADD)
refining. Figure VI-94 shows that this has been the case in recent
decades, as growth in gasoline production within PADD 5 throughout that
period has closely paralleled growth in local consumption, while net
exports have remained minimal.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.515
The central region of the United States (PADDs 2-4) accounts for
the remaining 52 percent of current U.S. gasoline consumption, while
producing about three-quarters of the nation's gasoline and blendstock.
Although as Figure VI-95 shows the central region was a minor net
exporter of gasoline as recently as 2007, it now exports some 800,000
barrels per day of gasoline and blendstock, and has accounted for
virtually all of the recent growth in U.S. exports of these two
categories of refined products. Recent press reports indicate that
firms are currently making significant new investments to add refining
capacity on the Gulf Coast to process the growing supply of U.S. shale
oil (Douglas, 2019), and with the projected future decline in U.S.
consumption, any additional gasoline refined there is likely to
increase U.S. exports. Thus, future increases in gasoline consumption
in the central region of the U.S. of the magnitude likely to result
from adopting these final standards is expected to be met by diverting
gasoline exports to domestic consumption, even in the absence of
additional refinery investments.
[[Page 24880]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.516
BILLING CODE 4910-59-C
Table VI-278 below compares recent changes in gasoline consumption
and various sources of supply for these three U.S. regions during the
recent period (2012-18) when gasoline consumption has generally
increased. As it shows, recent increases in consumption along the U.S.
East Coast have been supplied by increased production within the
region. As noted previously, however, it appears likely that production
capacity there will contract significantly in the near term, and that
future increases in consumption will need to be met from foreign
imports or shipments from other U.S. regions. As the table also shows,
recent increases in gasoline production in the Midwest and Gulf Coast
region have been adequate to supply increased consumption within the
region as well as major increases in foreign exports and shipments to
other U.S. regions. Finally, increased consumption on the Nation's West
Coast appears to have been met via a combination of increased
production within the region and drawdowns of previously accumulated
inventories (not shown in the table).
At the national level, where net shipments among regions
necessarily cancel one another (resulting in the zero entry for Net
Receipts from Other PADDS shown in the table), recent increases in
production have been sufficient to meet increased domestic consumption,
while simultaneously enabling a major increase in exports. This
suggests that from the nationwide aggregate perspective, incremental
increases in domestic gasoline consumption resulting from this rule
could be met by a reduction in U.S. exports of domestically-refined
gasoline to other nations, accompanied by increases in shipments from
the Midwest and Gulf Coast regions to the nation's East and West
Coasts.
[[Page 24881]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.517
To summarize, based on changes in the various sources of supply
that have accompanied recent changes in consumption within different
regions of the U.S., the agencies anticipate that:
Most of any marginal increases in U.S. gasoline
consumption resulting from this rule that occur on the East Coast of
the U.S. is likely to be met in the near term by increased transfers
from other regions of the U.S. or higher foreign imports, and possibly
by expanded refining activity in the longer term;
Most of any marginal increases in U.S. gasoline
consumption resulting from this rule that occur on the West Coast is
likely to be supplied by increased gasoline refining within that
region; and
Most or all of any marginal increase in U.S. gasoline
consumption resulting from this rule that occurs in the Central region
is likely to be supplied by redirecting foreign exports to supply
markets within that region.
With these expectations and acknowledging the uncertainty
surrounding them, the agencies have concluded that assuming 50 percent
of any increase in U.S. gasoline consumption will lead to increased
domestic refining activity--and thus to increases in domestic refinery
emissions--continues to be reasonable, and perhaps even overstates the
expected increase in domestic refinery emissions. In particular, the
agencies find that assuming 50 percent is more reasonable than assuming
that either none or 100 percent of any change in gasoline consumption
will be translated into changes in domestic gasoline refining. Thus,
the agencies have elected to continue to employ the 50 percent
assumption in their central analysis, and to examine the sensitivity of
its results to varying this fraction over the entire possible range,
from zero to 100 percent.
(ii) Changes in Crude Oil Supply to Domestic Refineries
The agencies also re-evaluated their assumption that 90 percent of
the increase in crude petroleum refined in the U.S. to produce
additional gasoline consumed as a result of this rule would be imported
from abroad (thus resulting in increased emissions for its storage at
import terminals, and transportation to domestic refineries), while the
remaining 10 percent would be produced domestically (thus resulting in
emissions from its extraction, local storage, and transportation to
U.S. refineries). As discussed in more detail below, the agencies
conclude that domestic petroleum production responds primarily to
technological innovations, investments in exploration and development
of new domestic sources of oil, and variation in the world price of
petroleum, rather than to U.S. demand for refined products such as
gasoline. As a consequence, they conclude that any increase in gasoline
consumption attributable to this final rule is unlikely by itself to
have a significant effect on domestic petroleum production, and that
their previous assumption continues to be reasonable.
U.S. oil production is primarily a function of development
opportunities identified during prior exploration programs, innovations
in the technological for drilling and extracting crude petroleum,
producer's expectations regarding future world petroleum prices, and
the U.S. tax and regulatory situations surrounding petroleum
exploration and production. Crude oil is a fungible, non-perishable
commodity, and can usually be transported among local oil markets
around the globe at some cost. As a consequence, the price of oil in a
U.S. domestic market such as Texas is highly correlated with its price
in markets located in Northern Europe, the Far East, and the Middle
East.
In contrast, U.S. gasoline consumption depends on a broad array of
factors that overlap only partially with the determinants of U.S. crude
petroleum production. These include domestic economic growth and its
consequences for transportation demand, current and future vehicle fuel
economy, gasoline prices, excise and sales taxes levied on gasoline,
technological and cultural changes, vehicle prices, and the evolution
of transportation systems and the built environment.
As a consequence, changes in U.S. consumption and supply of
petroleum products will primarily be reflected in changes the
destinations of domestically produced and imported crude petroleum,
rather than in changes in their production volumes. To the extent that
changes in U.S. gasoline demand for lead to changes in the volume
refined domestically (the subject of the previous analysis), increased
refining activity is thus likely to be reflected in a shift in U.S.
imports or exports of crude oil, rather than in a change in U.S.
production of crude oil. Instead, any effect of this rule on U.S. crude
oil production would arise primarily from the impact of increased
domestic gasoline demand on global oil prices, which will be limited by
the fact that U.S. gasoline demand accounts for a relatively small
share of total global demand for petroleum products, and by
[[Page 24882]]
the response of global supply to any upward pressure on prices. Thus,
any effect of this rule on U.S. petroleum production is likely to be
extremely modest.\2400\
---------------------------------------------------------------------------
\2400\ U.S. gasoline consumption currently accounts for about 9%
of total global demand for refined petroleum products, and the AEO
2019 reference case projects that this will decline to 6% by the
year 2035, and remain at that level through 2050. These figures are
calculated from AEO 2019 Reference Case, Tables 11 and 21, available
at https://www.eia.gov/outlooks/aeo/tables_ref.php.
---------------------------------------------------------------------------
Localized and temporary changes in domestic production might arise
in response to capacity limitations or transportation bottlenecks
associated with particular regions or refineries, which could
temporarily create markets for higher-priced crude oil. However, these
situations would normally be localized and prevail for only a limited
time. At the same time, the effects of any change in domestic petroleum
consumption on world oil prices would be attenuated, because as
indicated previously the impact of increased domestic consumption would
be felt on prices and volumes supplied in the much larger global
petroleum market, rather than confined to the smaller U.S. market. Any
resulting changes in global oil prices and petroleum production would
inevitably be small when viewed on a world scale, and likely to prompt
only minimal responses in U.S. petroleum supply.
As one indication of the likely minimal impacts of higher U.S.
gasoline consumption on U.S. production of crude petroleum, EIA's
Annual Energy Outlook 2018 included a side case called ``No New
Efficiency Requirements,'' which included a freeze on U.S. fuel economy
standards beginning in 2020. Although this scenario does not correspond
exactly to either the agencies' earlier proposal or this final rule,
comparing its results to those from the AEO 2018 reference case
illustrates the insensitivity of domestic crude oil production to
increases in gasoline consumption, as represented in EIA's National
Energy Modeling System (NEMS).
Figure VI-96 below presents such a comparison, showing historical
trends is U.S. gasoline consumption and petroleum production, and
comparing their projected future trends in the AEO 2018 Reference Case
and No New Efficiency Requirements alternative. As the figure
illustrates, the large increase in U.S. gasoline consumption under the
latter scenario relative to the Reference Case is accompanied by an
almost indiscernible change in U.S. crude petroleum production, for
exactly the reasons described above.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.518
[[Page 24883]]
BILLING CODE 4910-59-C
The agencies conclude that in the context of the current global
petroleum market, increases in U.S. gasoline demand on the scale likely
to result from this final rule are unlikely to produce changes in the
market that prompt a significant increase in domestic petroleum
production. Instead, they are likely to affect mainly the destinations
and uses of crude petroleum--including refining gasoline within the
U.S.--that is already being supplied to the global market. As a
consequence, the agencies have elected to retain our previous
assumption that any increase in domestic gasoline refining that occurs
as a consequence of adopting this final rule is unlikely by itself to
lead to a significant increase in domestic crude oil production or in
the associated upstream emissions. Specifically, the agencies continue
to assume that 10 percent of any increase in domestic gasoline refining
would utilize increased U.S. production of crude petroleum.
The agencies chose to model upstream emissions in order to generate
full fuel cycle emissions--using GREET for the upstream component and
MOVES for the downstream component--because each alternative has
varying levels of fuel consumption, and the specific gallons of
gasoline, diesel, E85, and other fuels evaluated in today's rule will
lead to different tailpipe and upstream emission outcomes.
While it may be fair to characterize MOVES and GREET as partial
equilibrium models rather than general equilibrium models, the agencies
did not make any modifications to the MOVES or GREET emission factors
themselves. Changes in emission results were initiated through changes
in fleet composition or activity, especially changes in vehicle miles
travelled as well as vehicle sales and population. Other changes were
made to average vehicle mass and road load coefficients such as
aerodynamic drag and rolling resistance corresponding to the various
regulatory alternatives. Each alternative consists of a package of
technology changes, so a particular technology change was not modeled
alone and would need to be evaluated separately to quantify incremental
changes. Please consult the FRIA for quantified impacts for the
technology packages laid out by alternative.
d) How Did the Agencies Estimate and Value Health Impacts From Changes
in Air Quality
The agencies' analyses estimates changes in the population-wide
incidence of selected health impacts, as well as changes in the
aggregate monetary value of those health impacts that may occur from
the changes in emissions of criteria air pollutants projected to result
from this final rule and the alternative that were considered. As with
other estimated impacts of the final rule and alternatives, these
changes are measured from a baseline that is represented by the
adoption of the augural CAFE standards and the extension of EPA's
updated CO2 estimates, providing a more precise accounting
of physical impacts and costs and benefits of the standards, and also
directly responds to comments, as discussed below.\2401\
---------------------------------------------------------------------------
\2401\ See EPA, Office of Air and Radiation, Office of Air
Quality Planning and Standards, Technical Support Document,
Estimating the Benefit per Ton of Reducing PM2.5
Precursors from 17 Sectors, February 2018, available at https://www.epa.gov/sites/production/files/2018-02/documents/sourceapportionmentbpttsd_2018.pdf.
---------------------------------------------------------------------------
Many commenters expressed concern over the health impacts from
increased GHG emissions and criteria pollutants. The American Lung
Association et al. stated ``Today, nearly 40 percent of Americans--more
than 124 million--live in communities in nonattainment for ozone and
particulate matter, with many residents impacted more severely by local
pollution sources, including near-road pollution. . . . Near-road
pollution has been found to increase asthma attacks in children,
cardiovascular health impacts, impaired lung function and premature
death. . . . Reducing VOC emissions will help reduce the burden of
these carcinogens on many communities, especially those living or
working near these roadways.'' \2402\ As discussed in this Section, the
agencies agree with these statements and have considered health effects
as part of the analysis for today's rule. The Institute for Policy
Integrity stated ``the agencies fixate on alleged on-road fatality
effects while arbitrarily ignoring the mortalities, morbidities, and
other welfare effects associated with emissions.'' \2403\ As described
in this Section, in the analysis for this rule, the agencies estimate
both air quality-related fatalities and their costs, in addition to the
agencies' analysis on vehicle-related fatalities. Many public
commenters also expressed concern for health issues associated with
increased pollutants and emissions over what was anticipated by the
agencies' 2012 analysis. The agencies carefully considered these
comments and provided additional analysis to consider health impacts,
as described below.
---------------------------------------------------------------------------
\2402\ American Lung Association et al., NHTSA-2018-0067-11765.
\2403\ Institute for Policy Integrity, NHTSA-2018-0067-12213.
---------------------------------------------------------------------------
The estimated health impacts reflect the nationwide baseline level
of emissions of each pollutant, an assumed geographic distribution of
increased emissions, the resulting changes in concentrations of
criteria pollutants at various locations nationwide (some of which
reflect accumulations of emissions, while others are chemical by-
products formed in atmospheric reactions), increased exposure of the
U.S. population to unhealthful concentrations of each pollutant, and
the consequences of increased exposure for the aggregate frequency of
each health impact. The agencies' analysis assumes that the increases
in upstream and vehicle emissions are distributed in proportion to
current emissions associated with fuel supply and vehicle use. This is
consistent with the way EPA estimates health impacts and health damage
costs for the refining and on-road mobile sources sectors, since those
are estimated by assuming an increase in emissions from those sectors
that is distributed in proportion to current emissions from each one,
and estimating the resulting changes in accumulations of air
pollutants, population exposure, health impacts, and associated
monetary value. The accompanying estimates of per-ton damage costs
apply unit values to the increased frequency of each health effect,
representing the dollar costs or estimated willingness-to-pay to avoid
its occurrence, and combine the results to estimate total damage costs.
EPA analysts utilize a large volume of underlying data, a number of
intermediate calculations, and many simplifying assumptions to develop
these estimates of health impacts and health damage costs per ton of
additional emissions, and discussing these in detail is well beyond the
scope of this rule. These underlying data, assumptions, and
calculations are described in detail in the document that reports the
values used for the agencies' analysis.\2404\ EPA quantifies health
impacts and damage costs for emissions from 17 separate sectors of U.S.
economic activity, and reports values for increases in premature
mortality and the combined costs of damages from premature mortality
and various other health impacts per ton of PM2.5, nitrate,
[[Page 24884]]
and sulfate emissions.\2405\ These values include high and low
estimates of both premature mortality and health damage costs, which
primarily reflect alternative published estimates of the premature
mortality impact of PM2.5 emissions.\2406\ Alternative
values are also reported for 3 percent and 7 percent discount rates;
discounting affects the values because of the delay (or ``latency
period'') between exposure to air pollution and the development of some
health impacts, most notably premature deaths.
---------------------------------------------------------------------------
\2404\ See EPA, Office of Air and Radiation, Office of Air
Quality Planning and Standards, Technical Support Document,
Estimating the Benefit per Ton of Reducing PM2.5
Precursors from 17 Sectors, February 2018, available at https://www.epa.gov/sites/production/files/2018-02/documents/sourceapportionmentbpttsd_2018.pdf.
\2405\ Premature mortality includes deaths that are estimated to
occur before the normally expected life span of persons with
specified demographic characteristics.
\2406\ Estimated willingness to pay to avoid premature death
accounts for 98% of the total health damage costs included in these
estimates; see EPA, p. 10.
---------------------------------------------------------------------------
The agencies' analysis uses those values for the petroleum refining
sector (sector 15) to represent impacts resulting from emissions that
occur during the fuel production and distribution process (upstream
emissions), and those for the on-road mobile source sector (sector 13)
to represent the impacts of emissions resulting from car and light
truck use. The agencies apply EPA's estimates of per-ton increases in
premature mortality and health damage costs for these sectors to their
estimates of changes in nationwide total emissions of PM2.5,
nitrogen oxides (NOx), and sulfur dioxide (SO2) from the
fuel supply process and from car and light truck use.
Table VI-279 and Table VI-280 below report values the agencies used
in the estimates of premature mortality impacts and total health damage
costs per ton of emissions to analyze the consequences of this final
rule. The results for this analysis are provided in Section VII of this
rule. The dollar values reported in the tables below differ slightly
from those reported in the underlying source, because they have been
adjusted from the 2015$ used in that source to the 2018 dollars used
throughout this analysis. Values for intervening years were
interpolated from those shown in the tables, and values for the year
2030 shown in the tables were assumed to prevail for years beyond 2030.
The agencies' central analysis of the rule uses averages of the low and
high values shown in each table, while the low and high values
themselves are used in the sensitivity analyses described in Section
VII of this rule.
BILLING CODE 4910-59-P
[[Page 24885]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.519
[[Page 24886]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.520
BILLING CODE 4910-59-C
The valuation of premature mortality effects rely on the results of
``benefits per ton'' approach (BPT). This approach is a reduced form
approach, which is
[[Page 24887]]
less complex than full-scale air quality modeling, requiring less
agency resources and time. Based on EPA's work to examine reduced form
approach, the BPT may yield estimates of PM2.5--benefits for
the mobile sector that are as much as 10 percent greater than those
estimated when using full air quality modeling.
The EPA is currently working on a systematic comparison of results
from its BPT technique and other reduced-form techniques with results
from full-form photochemical modelling. While this analysis employed
photochemical modeling simulations, we acknowledge that the Agency has
elsewhere applied reduced-form techniques. The summary report from the
``Reduced Form Tool Evaluation Project'', which has not yet been peer
reviewed, is available on EPA's website at https://www.epa.gov/benmap/reduced-form-evaluation-project-report. Under the scenarios examined in
that report, EPA's BPT approach in the 2012 rule (which was based off a
2005 inventory) may yield estimates of PM2.5--benefits for
the mobile sector that are as much as 10 percent greater than those
estimated when using full air quality modeling. The estimate increases
to 30 percent greater for the electricity sector. The EPA continues to
work to develop refined reduced-form approaches for estimating
PM2.5 benefits.
In addition, considerable uncertainty surrounds many of the
assumptions and other inputs used in the agencies' analysis of economic
and environmental impacts likely to result from adopting the final
standards, rather than ratifying the augural standards. Perhaps most
notably, because fuel prices are inherently volatile and forecasts of
their future level depend critically on developments in the often
unstable and politicized global oil market, those forecasts are
inherently uncertain, as evidenced by the fact that actual gasoline
prices are well below those the agencies relied on in their 2012
analysis of CAFE and CO2 standards for model years 2017-25.
While the agencies' current analysis updates those projections to
reflect EIA's 2019 Annual Energy Outlook, which now anticipates that
future prices will remain well below those the agencies projected in
their 2012 analysis, it remains possible that EIA's current forecast
will continue to overestimate actual future prices (of course, EIA's
current forecast could also prove to be too low, although the recent
record suggests a larger risk that the opposite will be the case).
Further, gasoline prices are only one of a number of assumptions about
which the agencies have reason to be uncertain; others include the fuel
economy and other features of car and light truck models that
manufacturers will offer during future model years, how buyers will
respond to changes in the features of competing models in the face of
future fuel prices and economic conditions, and how much they (and
subsequent owners) will ultimately drive the models they purchase over
their lifetimes. Uncertainty about all of these factors is reflected in
similar risks that the agencies' projections of changes in vehicle use
and fuel consumption under the final standards will prove to be in
error. Finally, uncertainty about the agencies' companion projections
of those standards' impacts on PM emissions and premature mortality is
compounded by the currently unknown effects of future control
technologies and regulations on actual refinery and vehicle emissions,
as well as by the sources of potential error in estimating the effects
of changes in emissions on premature mortality discussed above.
Although it may seem that the agencies' estimates of increases in
premature mortality resulting from the final standards are more likely
to be too high than too low, it is extremely difficult to anticipate
whether this is actually the case.
Separately, the DEIS and FEIS accompanying this rule describe that
the BPT estimates are subject to several assumptions and uncertainties
that make it difficult to draw conclusions about the estimated monetary
values.\2407\ Non-exhaustively, these reasons include that estimates do
not reflect local variability in population density, meteorology,
exposure, baseline health incidence rates, or other local factors that
might lead to an overestimate or underestimate of the actual benefits
of controlling fine particulates, and that the health impact studies
include several sources of uncertainties, including: Within-study
variability (the precision with which a given study estimates the
relationship between air quality changes and health impacts), across-
study variation (different published studies of the same pollutant/
health effect relationship typically do not report identical findings,
and in some cases the differences are substantial), the application of
concentration-response functions nationwide (does not account for any
relationship between region and health impact to the extent that there
is such a relationship), and extrapolation of impact functions across
population (the agencies assumed that certain health impact functions
applied to age ranges broader than those considered in the original
epidemiological study).
---------------------------------------------------------------------------
\2407\ See DEIS and FEIS at Chapter 4, Air Quality--Health
Impacts.
---------------------------------------------------------------------------
Full-scale photochemical modeling provides the needed spatial and
temporal detail to more precisely estimate changes in ambient levels of
these pollutants and their associated impacts on human health and
welfare. This modeling provides insight into the uncertainties
associated with the use of benefit-per-ton estimates. The agencies
conducted a photochemical modeling analysis for the Final EIS using the
same methods as in the previous CAFE Final EISs 2408 2409
and the HD Fuel Efficiency Standards Phases 1 and 2 Final
EISs.2410 2411 The air quality modeling and health effects
analysis focused on ozone and fine particulate matter equal to or less
than 2.5 microns in diameter (PM2.5). As indicated in the
Draft EIS, the agencies performed photochemical air quality modeling
based on the inputs and emissions forecasts used in the Draft EIS.
Consistent with prior rulemakings and as described in the scoping
notice, to accommodate the substantial time required to complete the
air quality modeling analysis, NHTSA proposed to initiate air quality
modeling before the inputs and emissions forecasts for the Final EIS
were finalized.\2412\ NHTSA received no public comments in response to
the scoping notice addressing this analytical approach, and the agency
proceeded accordingly. Therefore, NHTSA used the inputs and emissions
forecasts for the Proposed Action and alternatives as stated in the
Draft EIS for the analysis in this final rulemaking. For additional
[[Page 24888]]
information on the scoping notice and comments received, see Section X.
---------------------------------------------------------------------------
\2408\ NHTSA (2010). Final Environmental Impact Statement,
Corporate Average Fuel Economy Standards, Passenger Cars and Light
Trucks, Model Years 2012-2016. Washington, DC, National Highway
Traffic Safety Administration.
\2409\ NHTSA (2012). Final Environmental Impact Statement,
Corporate Average Fuel Economy Standards Passenger Cars and Light
Trucks, Model Years 2017-2025, Docket No. NHTSA-2011-0056. July
2012. Available at: https://one.nhtsa.gov/Laws-&-Regulations/CAFE-%E2%80%93-Fuel-Economy/Environmental-Impact-Statement-for-CAFE-Standards,-2017%E2%80%93202.
\2410\ NHTSA (2011). Final Environmental Impact Statement,
Medium and Heavy-Duty Fuel Efficiency Improvement Program.
Washington, DC, National Highway Traffic Safety Administration.
\2411\ NHTSA (2016). Phase 2 Fuel Efficiency Standards for
Medium- and Heavy-Duty Engines and Vehicles. Final Environmental
Impact Statement. Available at: https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/mdhd2-final-eis.pdf.
\2412\ NHTSA, ``Notice of Intent to Prepare an Environmental
Impact Statement for Model Year 2022-2025 Corporate Average Fuel
Economy Standards,'' 82 FR 34740, 34743 fn. 15 (Jul. 26, 2017).
---------------------------------------------------------------------------
Some stakeholders submitted comments about the agencies' use of
underlying NPRM modeling to conduct the photochemical modeling; for
example, NCDEQ recognized the agencies statement that there was not
sufficient time to collect the modeling, but stated that they
``strongly believe that the inputs and results should be readily
available for public comment before the EIS and rulemaking are
finalized.'' \2413\ Those comments are addressed in Section X and in
the FEIS accompanying this rule. As part of EDF's alternative
examination of the CAFE model and inputs, EDF utilized the same EPA
benefit-per-ton method the agencies utilized for the final rule
(discussed further below) to estimate health effects due to criteria
pollutant emissions, concluding that the proposal would increase
premature mortality due to increases in particulate matter emissions.
EDF stated that these results indicated that the potential impacts of
the rule are large, and accordingly, ``NHTSA and EPA must conduct
detailed and thorough emission, photochemical and health effects
modeling to quantify the effect of this or any other proposal to relax
the CAFE and CO2 standards and increase upstream
emissions.'' \2414\
---------------------------------------------------------------------------
\2413\ North Carolina Department of Environmental Quality,
NHTSA-2018-0067-12025.
\2414\ Environmental Defense Fund, NHTSA-2018-0067-12108.
---------------------------------------------------------------------------
The agencies estimated air quality changes and health-related
benefits at the national scale based on a detailed analysis of air
quality and health effects throughout the contiguous 48 states.
Different regions of the country could experience either a net increase
or a net decrease in emissions because of the rule, depending on the
relative magnitude of the changes in emissions from decreased fuel
economy, decreased vehicle use, and increased fuel production and
distribution under each alternative. The EIS air quality analysis
addresses regional differences using grid-based air quality modeling
and analysis techniques, which account for local and regional
differences in emissions and many of the other factors (such as
meteorology and atmospheric processes) that affect air quality and the
resulting health effects at any given location. This air quality
modeling analysis is intended as a screening application of both the
Community Multiscale Air Quality (CMAQ) model and the Environmental
Benefits Mapping and Analysis Program (BenMAP) tool for the purposes of
quantifying and comparing the air quality and health-related benefits.
To examine and quantify the air quality and health-related benefits
associated with implementing the final CAFE standards for MY 2021-2026
light-duty vehicles, the agencies performed a national-scale
photochemical air quality modeling and health benefit assessment with
the following key steps:
Preparing emission inventories.
Modeling air quality.
Assessing air quality-related health impacts.
The following widely used tools were used for the air quality and
health effects assessment:
Sparse-Matrix Operator Kernel Emissions (SMOKE) processing
tool (version 3.7) to prepare model-ready emissions.
Community Multiscale Air Quality (CMAQ) model (version
5.2.1) to quantify air quality changes for the different fuel economy
alternatives.
Environmental Benefits Mapping and Analysis Program--
Community Edition (BenMAP-CE) tool (version 1.4) to assess the health-
related impacts of the simulated changes in air quality.
The national-scale modeling analysis employed the standard CMAQ
continental modeling domain. The horizontal resolution of the grid for
this modeling domain is 36 kilometers (22.4 miles). Air quality and
health-related impacts were calculated for each grid cell in the entire
contiguous United States (48 states). Although the modeling domain does
not include all 50 states, nearly all of the affected emissions and
population are included in the domain; therefore, the results are
expected to represent those for a national-scale analysis. The agencies
applied the CMAQ model for an annual simulation period using
meteorological inputs for a base year of 2011.
The agencies performed modeling for 2035 (although the emission
inputs represented a variety of different projection years, including
2030, 2035, and 2040, based on best available data). As in the Draft
EIS, the agencies chose 2035 for analysis of the various fuel economy
alternatives because a large proportion of vehicles in operation are
expected to meet the level of the standards set forth by 2035. EPA
provided up-to-date, projected, national-scale emissions data for 2040
for motor vehicles and for 2030 for all other sources. The emissions
were processed for the 36-kilometer (22.4-mile) resolution modeling
domain using SMOKE. The resulting model-ready inventories contain
emissions for all criteria pollutants (as required for photochemical
modeling) for multiple source categories (sectors), including on-road
mobile sources, non-road mobile sources (e.g., construction equipment,
locomotives, ships, and aircraft), electric generating unit (EGU) point
sources, non-EGU point sources, area sources, and biogenic sources.
Following preparation of baseline emissions inventories, the
baseline emissions for the light-duty vehicle portion of the on-road
mobile emissions and the relevant upstream categories were replaced
with data reflecting the alternatives analyzed in the Draft EIS. As
discussed above, NHTSA calculated national estimates of on-road
emissions for these vehicle classes for 2035, including both downstream
and upstream emissions.
The agencies then applied CMAQ, using the emissions specific to
each alternative. The simulated difference in air quality between the
Draft EIS No Action Alternative and each action alternative represents
the change in air quality associated with that alternative. Following
the application of CMAQ, the agencies processed the CMAQ outputs for
input to the BenMAP-CE health effects analysis tool, and used BenMAP-CE
to estimate the health impacts and monetized health-related benefits
associated with the changes in air quality simulated by CMAQ for each
of the action alternatives. The BenMAP-CE tool includes health impact
functions, which relate a change in the concentration of a pollutant
with a change in the incidence of a health endpoint. BenMAP-CE also
calculates the economic value of health impacts. For this study, the
health effects analysis considered the effects of ozone and
PM2.5. The PM2.5 analysis includes sulfate and
nitrate particulates (secondary PM2.5) formed from emissions
of SO2 (sulfur dioxide) and NOX, respectively.
BenMAP-CE does not estimate health impacts associated with changes in
directly emitted sulfur dioxide (SO2), carbon monoxide (CO),
and other emissions. Health effects were calculated at the 36-kilometer
scale (grid cell size) and aggregated nationally to determine overall
impact.
Figure VI-97 shows the components of the air quality modeling and
health-related benefits analysis. Note that both the emissions and
meteorological inputs are used by SMOKE.
[[Page 24889]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.521
Discussion of the photochemical modeling results is presented in
the FEIS accompanying this final rule.
E. Compliance Example Walk-Through
To illustrate the CAFE model's simulation of a manufacturer's
potential response to fuel prices and new standards, the NPRM provided
an example of how the preliminary version of the model showed, on a
year-by-year basis, how GM could potentially respond to a set of CAFE
standards, starting from MY 2016 (the latest year for which the
agencies were able to develop a full and detailed characterization of
the fleet of vehicles produced for sale in the U.S. at the time of
publishing the NPRM). Although no analysis that does not rely heavily
on a manufacturer's confidential product planning information can, with
high fidelity, predict what that manufacturer will do, the CAFE model,
by realistically reflecting product planning considerations in a
detailed year-by-year context, can describe a course that manufacturer
could realistically take. Indeed, when manufacturers provide
information to the agencies, they often emphasize year-by-year plans.
Although such information is typically considered confidential business
information (CBI), public comments by the Alliance illustrate the
concept for a hypothetical manufacturer. Although the illustration
includes credit carry-back (aka borrowing) that most manufacturers have
a history of avoiding, the illustration clearly demonstrates that the
Alliance views product planning as a year-by-year exercise:
BILLING CODE 4910-59-P
[[Page 24890]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.522
BILLING CODE 4910-59-C
Like the peer reviewers who examined the model's simulation of
technology application and compliance, automakers have been widely
supportive of the CAFE model's approach of year-by-year analysis
informed by product planning realities. For example, Toyota commented,
``The preamble correctly notes that manufacturers try to keep costs
down by applying most major changes mainly during vehicle redesigns and
more modest changes during product refresh, and that redesign cycles
for vehicle models can range from six to ten years, and eight to ten-
years for powertrains. . . This appreciation for standard business
practice enables the modeling to capture more accurately the way
vehicles share engines, transmissions, and platforms. There are now
more realistic limits placed on the number of engines and transmissions
in a powertrain portfolio which better recognizes manufacturers must
manage limited engineering resources and control supplier, production,
and service costs.'' \2416\
---------------------------------------------------------------------------
\2415\ NHTSA-2018-0067-12073, at 28.
\2416\ NHTSA-2018-0067-12098, at 6.
---------------------------------------------------------------------------
The CAFE model's year-by-year approach to estimating manufacturers'
potential responses to standards and fuel prices is consistent with
EPCA/EISA's requirement that CAFE standards be set at the maximum
feasible levels for each fleet (passenger car and light truck) in each
model year. Some commenters correctly observe that the CAA (which
provides no direction regarding tailpipe CO2 emissions
standards) does not require such a year-by-year determination, but
suggest, further, that EPA should refrain from making use of year-by-
year analysis. In particular, CBD et. al. commented as follows:
Furthermore, the Volpe model and association [sic] tools are not
designed in accordance with EPA's independent statutory authority
under Clean Air Act Section 202. The Volpe and OMEGA models have an
overarching difference in their architecture--one where the Volpe
modeling approach is designed to match NHTSA's statutory authority,
but not EPA's. The EPCA requirements drive the design of the Volpe
model, in that it performs a year-by-year analysis in order to
demonstrate that NHTSA is meeting its EPCA obligations. As a result,
the Volpe model attempts to simulate for each manufacturer, by year,
their refresh and redesign cadence across their vehicle platforms
and then predict a manufacturer's technology deployment decision-
making process for each platform. But under the Clean Air Act, EPA
is not required to demonstrate that standards are set at the maximum
feasible level year-by-year, as EPCA explicitly requires for
NHTSA.\2417\
---------------------------------------------------------------------------
\2417\ NHTSA-2018-0067-12000, Appendix A, at 24-25.
Although CBD is correct that the CAA does not require a year-by-
year determination or year-by-year analysis, CBD wrongly claims that
the CAFE model's modeling approach is not ``in accordance'' with the
CAA. CBD's claim is analogous to saying ``just say you want to drive
across the country; don't bother looking at a map.'' As the NPRM
demonstrated, the CAFE model can be used to simulate compliance with
CO2 standards. That the model follows a year-by-year
approach to doing so simply means that it takes greater pains to
describe realistic pathways forward from a known model year.
Manufacturers are by no means the only stakeholders to recognize that
product planning is actually a year-by-year process. Supporting its
comments on the agencies' proposal, CARB provided a study by Roush
Industries, focusing on a potential design pathway for the Toyota
RAV4.\2418\ While this report, which was cited by CARB in its comments,
asserted the agencies' modeling underestimated fuel consumption
benefits and overestimated costs, Roush, like the Alliance, clearly
interpreted the question of realism as a
[[Page 24891]]
year-by-year question, as illustrated by the following chart in Roush's
report:
---------------------------------------------------------------------------
\2418\ Rogers, G., ``Technical Review of: The Safer Affordable
Fuel-Efficient (SAFE) Vehicles Rule for Model Years 2021-2026
Passenger Cars and Light Trucks, Final Report.'' Roush Industries.
October 25, 2018. See CARB, NHTSA-2018-0067-11984.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.523
BILLING CODE 4910-59-C
While a year-by-year representation is essential to the estimation
of pathways that individual manufacturers could realistically take to
apply technologies to specific vehicle models, the CAFE model also
accounts for a range of other important engineering and product
planning considerations. For example, among specific vehicle models,
engines and transmissions are often shared, and a given vehicle design
platform may encompass a range of different specific vehicle models.
This means not every configuration of every vehicle model can be as
optimized for fuel economy as if each could be considered in isolation.
This isn't to say that such optimization is technologically impossible,
but rather to say that the resources involved in such optimization
would be financially impracticable. Moreover, CAFE and CO2
standards apply to fleets, not specific products. This means, for
example, that if a given engine is shared among both passenger cars and
light trucks, changes made to that engine in response to one fleet's
standard will impact products in the other fleet. Consistent with the
fact that CAFE and CO2 compliance applies to fleets on a
year-by-year basis, the CAFE model explicitly accounts for sharing
among specific model/configurations when simulating year-by-year
compliance. The Roush report's authors ``have not performed a complete
fleet-compliance simulation.'' \2420\ Therefore, even notwithstanding
differences in estimates of redesign schedules and technology efficacy
and costs, Roush's analysis of the RAV4 is highly idealized. As
discussed below, together with inputs based on Toyota's actual MY 2017
production, the CAFE model represents the RAV4 as encompassing multiple
configurations, spanning both the passenger car and light truck
regulatory classes, all on a common vehicle platform that includes
several other vehicle models, and some RAV4s sharing engines with some
Camrys. Compared to estimating the potential to apply technology to a
handful of specific model/configurations in isolation, analysis that
accounts for manufacturers' actual production considerations produces
more realistic results.
---------------------------------------------------------------------------
\2419\ Rogers, G., ``Technical Review of: The Safer Affordable
Fuel-Efficient (SAFE) Vehicles Rule for Model Years 2021-2026
Passenger Cars and Light Trucks, Final Report,'' at 26. Roush
Industries. October 25, 2018. See CARB, NHTSA-2018-0067-11984.
\2420\ Ibid. at 6.
---------------------------------------------------------------------------
Nothing about the CAA discourages realism in regulatory analysis,
and even if the CAA did so, the CAFE model can easily be run for
isolated model years, or run in a manner that otherwise ignores
practical limits on development and manufacturing complexity.\2421\ EPA
elected to use the CAFE model as designed because doing so produces a
more realistic basis to estimate regulatory impacts. EPA considers its
use of the CAFE model entirely consistent with all CAA and other
statutory and other requirements governing the agency's development of
[[Page 24892]]
motor vehicle CO2 emissions standards which, unlike criteria
pollutant standards, are specified on a year-by-year basis, and
inherently involve the entirety of manufacturers' vehicles and fleets.
---------------------------------------------------------------------------
\2421\ Idealized simulation of compliance with a hypothetically
isolated model year could be accomplished by, when running the
model, setting the various ``start'' and ``end'' years to the same
value. Sharing of engines and transmission among different model/
configurations could be ignored by, in the CAFE model's ``market''
input file, assigning each engine, transmission, and vehicle
platform to a single model/configuration (e.g., such that each of
the six versions of the RAV4 is on its own vehicle platform, and
uses a dedicated engine and transmission).
---------------------------------------------------------------------------
Of course, like any other model, the CAFE model used for the NPRM
had room for improvement. As discussed above, the agencies have
responded to public comments by making changes to some aspects of the
CAFE model itself. Only a few such changes, all of which are discussed
above in greater detail, impact the CAFE model's simulation of
manufacturers' application of fuel-saving technologies. Among these,
three are especially important: First, the model now uses a more
``open'' application of its technology ``decision trees.'' While the
primary objective of this change is to make the model's cost accounting
more transparent (by recasting costs as absolute rather than
incremental), it also makes the model somewhat more likely to identify
and apply any highly cost-effective yet comparatively ``advanced''
combinations of technology. Second, the model introduces a ``cost per
credit'' metric for comparing available opportunities to add specific
technologies to specific vehicles.\2422\ As discussed above and in the
summary of the sensitivity analysis conducted for today's notice,
changing from the NPRM's ``effective cost'' metric to this new ``cost
per credit'' metric leads the model to, at least for the combination of
inputs in today's central analysis, more frequently select less costly
technology pathways than more costly pathways, at least when simulating
compliance with CO2 standards. Third, the CAFE model can now
extend its explicit simulation of manufacturers' technology application
well into the future. Today's analysis extends this explicit simulation
through model year 2050. Because today's reference case input estimates
include continued increases in fuel prices alongside continued
(``learning''-related) reductions in technology costs, extending the
explicit simulation shows manufacturers making significant voluntary
improvement in the longer term (e.g., after MY 2035), even if CAFE and
CO2 remain unchanged.
---------------------------------------------------------------------------
\2422\ Notable comments on this metric appear at NHTSA-2018-
0067-12039, Appendix, pp. 28-34, and at NHTSA-2018-0067-12108,
Appendix B, pp. 66-70.
---------------------------------------------------------------------------
The agencies have also revised most of the inputs to the CAFE
model, both to respond to comments and to better reflect an ever-
changing world. Sections appearing above discuss changes to model
inputs, such as the analysis fleet, technology-related inputs, and fuel
prices. Many of these changes are important to the model's simulated
application of fuel-saving technology. Updating the analysis fleet from
a MY 2016 to a MY 2017 basis ensures that fuel economy and
CO2 improvements manufacturers actually realized by adding
technologies between those model years is accounted for, and ensures
that changes in product offerings and production volumes between those
model years are also accounted for. With this update, the agencies also
more fully accounted for compliance credits accumulated prior to the
MYs represented explicitly in today's analysis. Some manufacturers have
accumulated large volumes of such credits, and are able to apply those
credits well past MY 2016, and to trade them to other manufacturers.
Updated vehicle simulations correct errors and make use of additional
engine performance estimates (i.e., engine efficiency ``maps''), and
cost estimates for some technologies reflect additional data and
consideration of comments. Also, fuel prices in the forecast used for
today's analysis are somewhat higher than those used for the NPRM; by
itself, this change makes the model tend to show larger and more
widespread voluntary fuel economy increases and accompanying
CO2 emissions reductions, although this increased tendency
is countered by the impact of changing to the ``cost per credit''
metric.
The following example will illustrate the model's behavior when
simulating compliance with CO2 standards. While the example
focuses on the baseline CO2 standards and on a specific
manufacturer (Toyota), and highlights a specific vehicle model (the
Toyota RAV4), results for other scenarios, manufacturers, and vehicle
models reflect application of the same logic. Because this example
begins with the MY 2017 fleet, and does not make use of manufacturers'
product plans (which the agencies have historically treated as
confidential business information, today's analysis cannot and does not
fully reflect manufacturers' actual product design decisions, even in
the short term. Nevertheless, the analysis yields a realistic and
detailed characterization of a path each manufacturer could take in
response to a given set of standards and other input estimates (e.g.,
of technology costs and fuel prices).
As discussed above, the model considers all models and model/
configurations produced for sale in the U.S. by a given manufacturer.
The Toyota Camry and Tundra are examples of specific Toyota passenger
car and light truck models, Toyota produces a range of configurations
(e.g., with different engines) of each of these vehicle models, and
inputs to the CAFE model ensure that each such configuration is
accounted for. CAFE model output files show the progressive application
of technology to each model/configuration over time under each
regulatory alternative. Here, focusing on different versions of one
model, the RAV4, illustrates the process and results.
The RAV4 is one of the vehicle models included in a vehicle
platform that also includes the Camry, Corolla, Prius, Lexus CT 200h,
Lexus NX 200t, and Lexus NX 300h. As mentioned above, the CAFE model
reflects the agencies' assumption that significant changes to vehicle
structures or materials will most practicably be applied throughout a
vehicle platform as models within the platform are redesigned. Within
this platform, the CAFE model identifies the Corolla LE, at more than
180,000 units produced in MY 2017, as the most likely ``leader'' for
such changes. Inputs to today's analysis also show that most of the
RAV4s produced for the U.S. in MY 2017 shared a 2.5L naturally
aspirated 4-cylinder gasoline engine with many Camrys. The CAFE model
identifies the Camry as the leader for new versions of that engine. The
same inputs show many RAV4s shared a 6-speed automatic transmission
with a range of other vehicle models, including the Avalon, Camry,
Lexus ES 350, Highlander, Lexus NX 200t, and the CAFE model identifies
the Camry as the most likely leader for changes to this transmission.
Model inputs also show other RAV4s shared a different 6-speed automatic
transmission with the Lexus NX 200t, and the CAFE model identifies the
RAV4 as the most likely leader for changes to this transmission.
Finally, the MY2017 RAV4 also included two ``strong'' (power split)
hybrid-electric versions (SE and XLE). Although these shared an engine
with other Toyota hybrids (Avalon, Camry, Lexus ES 300h and NX 300h),
the CAFE model reflects the agencies' assumption that it could be
practicable to ``split off'' plug-in (or fuel cell) configurations
rather than necessarily replace all strong hybrids sharing an engine
with PHEVs, BEVs, or FCVs.
Inputs for today's analysis have Toyota redesigning the RAV4 every
five years, starting with MY 2019, and freshening the model 2-3 years
after each redesign. Given this design cycle, and all the other inputs
to today's analysis, the CAFE model shows that under the baseline
CO2 standards,
[[Page 24893]]
Toyota could potentially make changes to the RAV4 summarized in the
table that follows. The first changes occur in 2019, with Toyota
improving aerodynamics of the hybrid RAV4s, and with the conventional
RAV4s inheriting a new high compression ratio (HCR) engine introduced
with the MY 2018 redesign of the Camry, and also adding 8-speed
automatic (A8) transmissions,\2423\ improved accessories (IACC), and
tires with reduced rolling resistance (ROLL20). With the MY 2024
redesign, all versions of the RAV4 receive further aerodynamic
improvements (AERO20) and ``Level 1'' mass reduction, engine friction
reduction (EFR) is applied to the HCR engine the non-hybrid versions
share with the Camry, and secondary axle disconnect (SAX) is applied to
the non-hybrid versions of the RAV4. With the MY 2027 freshening,
Toyota applies low-drag brakes to all the RAV4s. The MY 2029 redesign
does not make any powertrain changes, but applies more significant mass
reduction (MR3) to all RAV4s. In MY 2039, Toyota replaces the hybrid
RAV4 SE and XLE with 200-mile (BEV200) and 300-mile (BEV300) electric
vehicle, respectively.
---------------------------------------------------------------------------
\2423\ While it is not necessary for the compliance simulation
to produce real predictions of manufacturer product designs, only
plausible ones, these changes to the RAV4 did in fact occur during
the 2019 redesign.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.524
[[Page 24894]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.525
This progressive application of technology to the RAV4 produces a
series of emission reductions shown in the following table (and, though
not shown, corresponding fuel economy improvements). The table also
shows the progression of CO2 targets for these vehicles,
reflecting the fact that targets are higher for the hybrid and
conventional AWD versions of the RAV4, classified as light trucks, than
for the FWD RAV4s classified as passenger cars. Also notably, the
conventional RAV4s never achieve their respective CO2
emissions targets. This merely reflects the fact that credits for
reducing A/C refrigerant leakage apply at the fleet level rather than
on a per-vehicle basis and, in any event, Toyota can respond by
improving CO2 levels enough among enough other vehicle
models that Toyota's overall average CO2 levels comply with
Toyota's overall requirements, taking into account the potential
application of compliance credits.
[[Page 24895]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.526
These CO2 values could be converted to equivalent fuel
economy levels by multiplying their reciprocals by 8887 grams per
gallon (e.g., 8887 g/gal x 1/(144 g/mi) = 62 mpg), differences in
compliance provisions are such that results would be offset from actual
fuel economy levels under CAFE standards. When simulating compliance
with CAFE or CO2 standards, the CAFE model reports both fuel
economy and CO2 targets and achieved levels, even when the
model is ``enforcing'' compliance with only one of these sets of
standards. When simulating
[[Page 24896]]
compliance with baseline CO2 standards, results for the
example discussed here show the following fuel economy targets and
achieved levels for the RAV4.
[GRAPHIC] [TIFF OMITTED] TR30AP20.527
[[Page 24897]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.528
The progressive application of technology also produces increases
(and some eventual decreases) in costs. For each RAV4 configuration,
the following table shows costs beyond MY 2017 technology, in 2018
dollars. The conventional RAV4s incur a significant cost increase in MY
2019, primarily for the new HCR engine inherited from the Camry. Costs
continue to increase through MY 2029 as additional technology
accumulates, with another significant increase for MR4 in MY 2029.
After MY 2029, technology costs for conventional RAV4s gradually
decline through MY 2050, in response to ongoing learning. In MY 2039,
the BEV200 RAV4 is less expensive than the HEV RAV4 it replaces,
leading this version's cost to drop by about $500 between MY 2033 and
MY 2034, and with learning, to fall quickly well below this version's
MY 2017 cost. Conversely, the BEV300 RAV4 introduced in MY 2039 is
about $950 more expensive than the MY 2038 hybrid RAV4 it replaces, and
even with learning, the BEV300 remains more expensive through MY 2050
than the hybrid RAV4. These BEVs are not needed for compliance; the
model shows Toyota could introduce them because, if battery costs
continue to decline while gasoline prices continue to increase, BEVs
could eventually become attractive on an economic basis.
[[Page 24898]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.529
[[Page 24899]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.530
BILLING CODE 4910-59-C
As mentioned above, by making sufficient improvements to other
vehicle models, Toyota could refrain from making the conventional RAV4s
meet their CO2 emissions targets. More broadly, Toyota can
also use compliance credits to cover compliance gaps. The CAFE model
accounts for the potential to transfer compliance credits between the
passenger car (PC) and light truck (LT) fleets. The model also accounts
for the potential to apply credits from prior model years (i.e.,
credits that have been ``banked'' or, equivalently, ``carried
forward''), including compliance credits earned prior to MY 2017. These
aspects of the model interact with the model's accounting for multiyear
planning--that is, the potential that a manufacturer, depending on its
product design cadence and on the progression of standards, might apply
``extra'' technology in some model years in order to facilitate
compliance in later model years. For example, if a manufacturer is only
redesigning 15% of its fleet volume in MY 2025, that manufacturer might
be best off--even setting aside credit banking--applying some ``extra''
technology (i.e., technology that leads to overcompliance) as part of
vehicle redesigns planned for MYs 2018-2024, and carrying that
technology forward into MY 2025 when there are fewer opportunities
available to reduce CO2 emissions in new models. As shown in
Figure VI-100, in Toyota's case, the model shows that Toyota could
offset its light truck compliance gaps during MY 2017-2019 by applying
compliance credits earned for light trucks prior to MY 2017. The graph
also shows Toyota applying extra technology to its passenger car fleet
during MYs 2018-2024 in order to comply with the MY 2025 passenger car
standard, but also to carry forward compliance credits and use those
credits to offset large compliance gaps for Toyota's light truck fleet
during MYs 2023-2027. After MY 2025, the model shows the effects of
some technology continuing to be inherited (especially during MYs 2026-
2030) from prior MYs, of Toyota continuing to make voluntary
improvements where economically attractive (like the MY 2039 RAV4 EV
mentioned above), and of Toyota continuing to transfer compliance
credits from the passenger car to the light truck fleet.\2424\
---------------------------------------------------------------------------
\2424\ While the fleets (PC and LT) are shown separately for
compliance purposes in this example, the ability to utilize credits
from either fleet toward total model year compliance (in the current
year, without caps or limits) means that the fleets for a
manufacturer comply jointly in each model year.
---------------------------------------------------------------------------
[[Page 24900]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.531
As the above figure shows, credit banking and transfers play an
important role in Toyota's simulated response to the standards. If
exercised in a manner that sets aside credit banking, the CAFE model
shows Toyota increasing its application of fuel-saving technologies
through MY 2025, and carrying those improvements forward, such that
Toyota's overall average CO2 emission rate is 16 g/mi lower
in MY 2025 when credit banking is not accounted for, as illustrated by
the next chart appearing below. Though not shown here, accounting for
credit banking also impacts the simulation other OEMs' compliance
pathways, because inputs to today's analysis assume that Toyota would
likely not need to use all of its pre-2017 compliance credits before
these credits expire in 2021, and that Toyota could therefore sell
those older credits other manufacturers (e.g., FCA, VW). By accounting
for credit banking, the CAFE model thereby avoids considerable
potential understatement of future CO2 emissions from light-
duty vehicles.
[[Page 24901]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.532
As indicated by the following chart, a failure to account for
credit banking would also increase Toyota's modeled per-vehicle costs
by nearly $1,000 in MY 2025. By accounting for credit banking, the CAFE
model thus avoids considerable potential overstatement of compliance
costs. Though not shown here, accounting for credit banking while also
applying inputs that reflect Toyota's ability to sell older credits to
some other OEMs also enables the CAFE model to avoid overstatement of
compliance costs for those OEMs.
[[Page 24902]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.533
While the model's simulation of manufacturers' potential responses
to CAFE standards applies the same inputs and analytical methods, it
does so accounting for several important statutory and regulatory
differences between CO2 standards and CAFE standards, and
for specific statutory direction regarding how CAFE standards are to be
considered for purposes of setting standards at the maximum feasible
levels in each model year. EPCA places specific limits on the amount of
credit that can be transferred between fleets, and requires that
domestic passenger cars meet minimum standards without applying
credits. EPCA also requires that the determination of maximum feasible
stringency set aside the potential to apply compliance credits or
introduce new alternative fuel vehicles (include BEVs and FCVs, but not
including plug-in HEVs) during the model years under consideration.
Especially with standards that continue to become more stringent,
applying these statutory constraints to the analysis leads the model to
tend to show greater overcompliance with standards in earlier model
years, because even setting aside the potential to carry forward or
transfer credits, Toyota is likely to find it more practicable to apply
some ``extra'' technology when redesigning vehicles during MYs 2017-
2024 than to attempt to address MY 2025 standards by working with only
vehicles scheduled to be redesigned in MY 2025. The model also tends to
show greater overcompliance in later model years, because some of that
extra technology from years leading up to the last year of stringency
increases takes time to carry forward to ensuing model years. These
aspects of the CAFE ``standard setting'' analysis are evident in the
model's solution for Toyota, shown in the following figure. With the
use of credits set aside after MY 2020, Toyota overcomplies with light
truck standards during MYs 2018-2023 in order to carry technology
forward into MY 2025. Although Toyota only marginally overcomplies with
MY 2025 standards, the inheritance of technology during MYs 2026-2029
contributes to increased overcompliance (which is to be expected given
the degree of platform and powertrain sharing between the fleets).
Continued increases in overcompliance after 2030 arise due to cost
learning effects (especially for batteries) and increased fuel prices.
[[Page 24903]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.534
VII. What Does the Analysis Show, and What Does It Mean?
A. Impacts of the Standards--Final and Alternatives
New CAFE and CO2 standards will have a range of impacts.
EPCA/EISA and NEPA require DOT to consider such impacts when making
decisions about new CAFE standards, and the CAA requires EPA to do so
when making decisions about new emissions standards. Like past
rulemakings, today's announcement is supported by the analysis of many
potential impacts of new standards. Today's rulemaking finalizes new
standards through model year 2026. While the CAFE model explicitly
estimates manufacturers responses to standards through model year 2050
and the associated impacts through calendar year 2089, today's
rulemaking presents estimates of impacts on model years through MY
2029, including impacts through these vehicles' full useful lives
(i.e., for MY 2029 vehicles, through 2068). Today's rulemaking also
presents estimates of overall impacts in each calendar year through
2050, accounting for all model years through 2050. The agencies of
course do not know today what will actually come to pass decades from
now under the new final standards or under any of alternatives under
consideration. The analysis is intended less as a forecast, than as an
assessment--reflecting the best judgments regarding many different
factors--of impacts that could occur.\2425\ As discussed below, the
analysis was conducted using several defined alternatives to explore
the sensitivity of this assessment to a variety of potential changes in
key analytical inputs (e.g., fuel prices).
---------------------------------------------------------------------------
\2425\ ``Prediction is very difficult, especially if it's about
the future.'' Attributed to Niels Bohr, Nobel Laureate in Physics.
---------------------------------------------------------------------------
This section summarizes various impacts of the final standards and
other regulatory alternatives defined above. The no-action alternative
provides the baseline relative to which all impacts are shown. Because
the final standards (and the other alternatives considered), being of a
``deregulatory'' nature, are less stringent than the no-action
alternative, all impacts are directionally opposite to impacts reported
in recent CAFE and CO2 rulemakings. For example, while past
rulemakings reported positive values for fuel consumption avoided under
new standards, today's rulemaking reports negative values, as fuel
consumption is expected be somewhat greater under today's new final
standards than under standards defining the baseline no-action
alternative. Reported negative values for avoided fuel consumption
could also be properly interpreted as simply ``additional fuel
consumption.'' Similarly, reported negative values for costs could be
properly interpreted as ``avoided costs'' or ``benefits,'' and reported
negative values for benefits could be properly interpreted as ``forgone
benefits'' or ``costs.'' However, today's rulemaking retains reporting
conventions consistent with past rulemakings, anticipating that,
compared to other options, doing so will facilitate review by most
stakeholders.
Today's analysis presents results for individual model years in two
different ways. The first way is similar to past rulemakings and shows
how
[[Page 24904]]
manufacturers could respond in each model year under the new final
standards and each alternative covering MYs 2021/2022-2026. The second,
expanding on the information provided in past rulemakings, evaluates
incremental impacts of new standards for each model year, in turn. In
past rulemaking analyses, NHTSA modeled year-by-year impacts under the
aggregation of standards applied in all model years, and EPA modeled
manufacturers' hypothetical compliance with a single model years'
standards in that model year. Especially considering multiyear planning
effects, neither approach provides a clear basis to attribute impacts
to specific standards first introduced in each of a series of model
years. For example, of the technology manufacturers applied in MY 2017,
some would have been applied even under the MY 2014 standards, and some
were likely applied to position manufacturers toward compliance with
(including credit banking to be used toward) MY 2018 standards.
Therefore, of the impacts attributable to the model year 2017 fleet,
only a portion can be properly attributed to the MY 2017 standards, and
the impacts of the MY 2017 standards involve fleets leading up and
extending well beyond MY 2016. Considering this, the final standards
were examined on an incremental basis, modeling each new model year's
standards over the entire span of included model years, using those
results as a baseline relative to which to measure impacts attributable
to the next model year's standards. For example, incremental costs
attributable to the new standards for MY 2023 are calculated as
follows:
COSTNew final,MY 2023 = (COSTNew final\through\MY 2023-COSTNo-
Action\through\MY 2023)-(COSTNew final\through\MY 2022-COSTNo-
Action\through\MY 2022)
where
COSTNew final,MY 2023: Incremental technology cost during MYs 2018-
2029 and attributable to the new final standards for MY 2023.
COSTNew final\through\MY 2022: Technology cost for MYs 2018-2029
under new final standards through MY 2022.
COSTNew final\through\MY 2023: Technology cost for MYs 2018-2029
under new final standards through MY 2023.
COSTNo-Action\through\MY 2022: Technology cost for MYs 2018-2029
under no-action alternative standards through MY 2022.
COSTNo-Action\through\MY 2023: Technology cost for MYs 2018-2029
under no-action alternative standards through MY 2023.
Furthermore, today's analysis includes impacts on new vehicle sales
volumes and the use (i.e., survival) of vehicles of all model years,
such that standards introduced in a model year produce impacts
attributable to vehicles having been in operation for some time. For
example, as modeled here, standards for MY 2021 will impact the prices
of new vehicles starting in MY 2017, and those price impacts will
affect the survival of all vehicles still in operation in calendar
years 2018 and beyond (e.g., MY 2021 standards impact the operation of
MY 2007 vehicles in calendar year 2027). Therefore, while past
rulemaking analyses focused largely on impacts over the useful lives of
the explicitly modeled fleets, much of today's analysis considers all
model years through 2029, as operated over their entire useful lives.
For some impacts, such as on technology penetration rates, average
vehicle prices, and average vehicle ownership costs, the focus was on
the useful life of the MY 2029 fleet, as the simulation of
manufacturers' technology application and credit use (when included in
the analysis) continues to evolve after model year 2026, stabilizing by
model year 2029.
Responding to comments recommending that the agencies present
impacts on a calendar year basis, today's rulemaking does so, with the
presented results extending through calendar year 2050, the last
calendar year that includes an on-road fleet with all vehicle vintages
represented.
Effects were evaluated from four perspectives: The social
perspective, the manufacturer perspective, the private perspective, and
the physical perspective. The social perspective focuses on economic
benefits and costs, setting aside economic transfers such as fuel taxes
but including economic externalities such as the social cost of
CO2 emissions. The manufacturer perspective focuses on
average requirements and levels of performance (i.e., average fuel
economy level and CO2 emission rates), compliance costs, and
degrees of technology application. The private perspective focuses on
costs of vehicle purchase and ownership, including outlays for fuel
(and fuel taxes). The physical perspective focuses on national-scale
highway travel, fuel consumption, highway fatalities, and carbon
dioxide and criteria pollutant emissions.
This analysis does not explicitly identify ``co-benefits,'' as such
a concept would include all benefits other than cost savings to vehicle
buyers. Instead, it distinguishes between private benefits--which
include economic impacts on vehicle manufacturers, buyers of new cars
and light trucks, and owners (or users) of used cars and light trucks--
and external benefits, which represent indirect benefits (or costs) to
the remainder of the U.S. economy that stem from the final rule's
effects on the behavior of vehicle manufacturers, buyers, and users. In
this accounting framework, changes in fuel use and safety impacts
resulting from the final rule's effects on the number of used vehicles
in use represent an important component of its private benefits and
costs, despite the fact that previous analyses have failed to recognize
these effects. The agencies' presentation of private costs and benefits
clearly distinguishes between those that would be experienced by owners
and users of cars and light trucks produced during previous model years
and those that would be experienced by buyers and users of new cars and
light trucks subject to the final standards. Moreover, it clearly
separates these into benefits related to fuel consumption and those
related to safety consequences of vehicle use. This is more meaningful
and informative than simply identifying all impacts other than changes
in fuel savings to buyers of new vehicles as ``co-benefits.''
For the social perspective, the following effects for model years
through 2029 as operated through calendar year 2068 are summarized:
Technology Costs: Incremental cost, as expected to be paid
by vehicle purchasers, of fuel-saving technology beyond that added
under the no-action alternative.
Hybrid Vehicle Welfare Loss: Loss of value to vehicle
owners resulting from incremental increases in the numbers of strong
and plug-in hybrid electric vehicles (strong HEVs or SHEVs, and PHEVs)
and/or battery electric vehicles (BEVs), beyond increases occurring
under the no-action alternative.\2426\ The loss of value is a function
of the factors that lead to different valuations for conventional and
electric versions of similar-size vehicles (e.g., differences in:
Travel range, recharging time versus refueling time, performance, and
comfort).
---------------------------------------------------------------------------
\2426\ Through MY 2029, the ``standard setting'' analysis of
CAFE standards sets aside the potential that manufacturers might by
introduce new BEV (or FCV) vehicle models, but allows that the
numbers of such vehicles produced might increase or decrease along
with overall U.S. sales of new passenger cars and light trucks, and
allows that additional BEV or FCV vehicle models might be intruded
after MY 2029.
---------------------------------------------------------------------------
Pre-tax Fuel Savings: Incremental savings, beyond those
achieved under the no-action alternative, in outlays for fuel
purchases, setting aside fuel taxes.
Mobility Benefit: Value of incremental travel, beyond that
[[Page 24905]]
occurring under the no-action alternative.
Lost New Vehicle Consumer Surplus: Value of incremental
savings to new vehicle buyers due to cheaper vehicle prices.
Implicit Opportunity Cost: \2427\ Value of other vehicle
attributes forwent to apply technology to meet the standards.
---------------------------------------------------------------------------
\2427\ This value is set to ``0'' for the central analysis.
---------------------------------------------------------------------------
Refueling Benefit: Value of incremental reduction,
compared to the no-action alternative, of time spent refueling
vehicles.
Non-Rebound Fatality Costs: Social value of additional
fatalities, beyond those occurring under the no-action alternative,
setting aside any additional travel attributable to the rebound effect.
Rebound Fatality Costs: Social value of additional
fatalities attributable to the rebound effect, beyond those occurring
under the no-action alternative.
Benefits Offsetting Rebound Fatality Costs: Assumed
further value, offsetting rebound fatality costs internalized by
drivers, of additional travel attributed to the rebound effect.
Non-Rebound Non-Fatal Crash Costs: Social value of
additional crash-related losses (other than fatalities), beyond those
occurring under the no-action alternative, setting aside any additional
travel attributable to the rebound effect.
Rebound Non-Fatal Crash Costs: Social value of additional
crash-related losses (other than fatalities) attributable to the
rebound effect, beyond those occurring under the no-action alternative.
Benefits Offsetting Rebound Non-Fatal Crash Costs: Assumed
further value, offsetting rebound non-fatal crash costs internalized by
drivers, of additional travel attributed to the rebound effect.
Additional Congestion and Noise (Costs): Value of
additional congestion and noise resulting from incremental travel,
beyond that occurring under the no-action alternative.
Energy Security Benefit: Value of avoided economic
exposure to petroleum price ``shocks,'' the avoided exposure resulting
from incremental reduction of fuel consumption beyond that occurring
under the no-action alternative.
Avoided CO2 Damages (Benefits): Social value of
incremental reduction of CO2 emissions, compared to
emissions occurring under the no-action alternative.
Other Avoided Pollutant Damages (Benefits): Social value
of incremental reduction of criteria pollutant emissions, compared to
emissions occurring under the no-action alternative.
Total Costs: Sum of incremental technology costs, hybrid
vehicle welfare loss, fatality costs, non-fatal crash costs, and
additional congestion and noise costs.
Total Benefits: Sum of pretax fuel savings, mobility
benefits, refueling benefits, Benefits Offsetting Rebound Fatality
Costs, Benefits Offsetting Rebound Non-Fatal Crash Costs, energy
security benefits, and benefits from reducing emissions of
CO2, the CO2 equivalent of other associated
gases, and criteria pollutants.
Net Benefits: Total benefits minus total costs.
Retrievable Electrification Costs: The portion of HEV,
PHEV, and BEV technology costs which can be passed onto consumers,
using the willingness to pay analysis described above.
Electrification Tax Credits: Estimates of the portion of
HEV, PHEV, and BEV technology costs which are covered by Federal or
State tax incentives.
Irretrievable Electrification Costs: The portion of HEV,
PHEV, and BEV technology costs OEM's must either absorb as a profit
loss, or cross-subsidize with the prices of internal combustion engine
(ICE) vehicles.
Total Electrification Costs: Total incremental technology
costs attributable to HEV, PHEV, or BEV vehicles.
For the manufacturer perspective, the following effects for the
aggregation of model years 2017-2029 are summarized:
Average Required Fuel Economy: Average of manufacturers'
CAFE requirements for indicated fleet(s) and model year(s).
Percent Change in Stringency from Baseline: Percentage
difference between averages of fuel economy requirements under no-
action and indicated alternatives.
Average Required Fuel Economy: Industry-wide average of
fuel economy levels achieved by indicated fleet(s) in indicated model
year(s).
Percent Change in Stringency from Baseline: Percentage
difference between averages of fuel economy levels achieved under no-
action and indicated alternatives.
Total Technology Costs ($b): Cost of fuel-saving
technology beyond that applied under no-action alternative.
Total Civil Penalties ($b): Cost of civil penalties (for
the CAFE program) beyond those levied under no-action alternative.
Total Regulatory Costs ($b): Sum of technology costs and
civil penalties.
Sales Change (millions): Change in number of vehicles
produced for sale in U.S., relative to the number estimated to be
produced under the no-action alternative.
Revenue Change ($b): Change in total revenues from vehicle
sales, relative to total revenues occurring under the no-action
alternative.
Curb Weight Reduction: Reduction of average curb weight,
relative to MY 2017.
Technology Penetration Rates: MY 2030 average technology
penetration rate for indicated ten technologies (three engine
technologies, advanced transmissions, and six degrees of
electrification).
Average Required CO2: Average of manufacturers'
CO2 requirements for indicated fleet(s) and model year(s).
Percent Change in Stringency from Baseline: Percentage
difference between averages of CO2 requirements under no-
action and indicated alternatives.
Average Achieved CO2: Average of manufacturers'
CO2 emission rates for indicated fleet(s) and model year(s).
For the private perspective, the following effects for the MY 2030
fleet are summarized:
Average Price Increase: Average increase in vehicle price,
relative to the average occurring under the no-action alternative.
Implicit Opportunity Cost: The lost benefit of vehicle
attributes that consumers prefer, which are sacrificed by manufacturers
to comply with the standards.
Hybrid Vehicle Welfare Loss (Costs): Average loss of value
to vehicle owners resulting from incremental increases in the numbers
of strong HEVs, PHEVs) and/or BEVs, beyond increases occurring under
the no-action alternative. The loss of value is a function of the
factors that lead to different valuations for conventional and electric
versions of similar-size vehicles (e.g., differences in: Travel range,
recharging time versus refueling time, performance, and comfort).
Ownership Costs: Average increase in some other costs of
vehicle ownership (taxes, fees, financing), beyond increase occurring
under the no-action alternative.
Lost Consumer Surplus: Value of incremental savings to new
vehicle buyers due to cheaper vehicle prices.
Fuel Savings: Average of fuel outlays (including taxes)
avoided over a vehicle's expected useful lives, compared to outlays
occurring under the no-action alternative.
[[Page 24906]]
Mobility Benefit: Average incremental value of additional
travel over average vehicles' useful lives, compared to travel
occurring under the no-action alternative.
Refueling Benefit: Average incremental value of avoided
time spent refueling over average vehicles' useful lives, compared to
time spent refueling under the no-action alternative.
Total Costs: Sum of average price increase, welfare loss,
and ownership costs.
Total Benefits: Sum of fuel savings, the mobility benefit,
and the refueling benefit.
Net Benefits: Total benefits minus total costs.
For the physical perspective, the following effects for model years
through 2029 as operated through calendar year 2068 are summarized:
Fuel Consumption, with rebound (billion gallons):
Reduction of fuel consumption, relative to the no-action alternative,
and including the rebound effect.
Fuel Consumption, without rebound (billion gallons):
Reduction of fuel consumption, relative to the no-action alternative,
and excluding the rebound effect.
Greenhouse Gases: Includes carbon dioxide
(CO2), methane (CH4), and nitrous oxide
(N2O), and values are reported separately for vehicles
(tailpipe) and upstream processes (combining fuel production,
distribution, and delivery) and shown as reductions in carbon dioxide
or its equivalent relative to the no-action alternative.
Criteria Pollutants: Includes carbon monoxide (CO),
volatile organic compounds (VOC), nitrogen oxides (NOX),
sulfur dioxide (SO2) and particulate matter (PM), and values
are shown as reductions relative to the no-action alternative.
Fuel Consumption: Aggregates all fuels, with electricity,
hydrogen, and compressed natural gas (CNG) included on a gasoline-
equivalent-gallon (GEG) basis, and values are shown as reductions
relative to the no-action alternative.
VMT, with rebound (billion miles): Increase in highway
travel (as vehicle miles traveled), relative to the no-action
alternative, and including the rebound effect.
VMT, without rebound (billion miles): Increase in highway
travel (as vehicle miles traveled), relative to the no-action
alternative, and excluding the rebound effect.
Fatalities, with rebound: Increase in highway fatalities,
relative to the no-action alternative, and including the rebound
effect.
Fatalities, without rebound: Increase in highway
fatalities, relative to the no-action alternative, and excluding the
rebound effect.
Health Effects: Increase in the occurrence of a variety of
health effects of criteria pollutant emissions, relative to the no-
action alternative, and reported separately for tailpipe and upstream
emissions.
Below, this section tabulates results for each of these four
perspectives and does so separately for the new final CAFE and
CO2 standards. More detailed results are presented in the
FRIA accompanying today's rulemaking, and additional and more detailed
analysis of environmental impacts for CAFE regulatory alternatives is
provided in the corresponding Final Environmental Impact Statement
(FEIS). Underlying CAFE model output files are available (along with
input files, model, source code, and documentation) on NHTSA's
website.\2428\ Summarizing and tabulating results for presentation here
involved considerable ``off model'' calculations (e.g., to combine
results for selected model years and calendar years, and to combine
various components of social and private costs and benefits); tools
Volpe Center staff used to perform these calculations are also
available on NHTSA's website.\2429\
---------------------------------------------------------------------------
\2428\ Compliance and Effects Modeling System, National Highway
Traffic Safety Administration, https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
\2429\ These tools, available at the same location, are scripts
executed using R, a free software environment for statistical
computing. R is available through https://www.r-project.org/.
---------------------------------------------------------------------------
While the National Environmental Policy Act (NEPA) requires NHTSA
to prepare an EIS documenting estimating environmental impacts of the
regulatory alternatives under consideration in CAFE rulemakings, NEPA
does not require EPA to do so for EPA rulemakings. With CO2
standards for each regulatory alternative being harmonized as practical
with corresponding CAFE standards, environmental impacts of
CO2 standards should be directionally identical and similar
in magnitude to those of CAFE standards. Nevertheless, in this section,
following the series of tables below, today's announcement provides a
more detailed analysis of estimated impacts of the new final CAFE and
CO2 standards. Results presented herein for the CAFE
standards differ slightly from those presented in the FEIS; while, as
discussed above, EPCA/EISA requires that the Secretary determine the
maximum feasible levels of CAFE standards in manner that, as presented
here, sets aside the potential use of CAFE credits or application of
alternative fuels toward compliance with new standards, NEPA does not
impose such constraints on any analysis presented in corresponding
FEISs, and the FEIS presents results of an ``unconstrained'' analysis
that considers manufacturers' potential application of alternative
fuels and use of CAFE credits.
In terms of all estimated impacts, including estimated costs and
benefits, the results of today's analysis are different for CAFE and
CO2 standards. Differences arise because, even when the
mathematical functions defining fuel economy and CO2 targets
are ``harmonized,'' surrounding regulatory provisions may not be. For
example, while both CAFE and CO2 standards allow credits to
be transferred between fleets and traded between manufacturers, EPCA/
EISA places explicit and specific limits on the use of such credits,
such as by requiring that each domestic passenger car fleet meet a
minimum CAFE standard (as discussed above). The CAA provides no
specific direction regarding CO2 standards, and while EPA
has adopted many regulatory provisions harmonized with specific EPCA/
EISA provisions (e.g., separate standards for passenger cars and light
trucks), EPA has not adopted all such provisions. For example, EPA has
not adopted the EPCA/EISA provisions limiting transfers between
regulated fleet or requiring separate compliance by domestic and
imported passenger car fleets. Such differences introduce variance
between impacts estimated under CAFE standards and under CO2
standards. Also, as mentioned above, Congress has required that new
CAFE standards be considered in a manner that sets aside the potential
use of CAFE credits and the potential additional application of
alternative fuel vehicles (such as electric vehicles) during the model
years under consideration. Congress has provided no corresponding
direction regarding the analysis of potential CO2 standards,
and today's analysis does consider these potential responses to
CO2 standards.
Tables in the remaining section summarize these estimated impacts
for each alternative, considering the same measures as shown above for
the final standards. For the final standards, social costs and
benefits, private costs and benefits, and environmental and energy
impacts were evaluated, and were done so separately for CAFE and
CO2 standards defining each regulatory alternative. Also,
for the final standards, the compliance-related private costs and
[[Page 24907]]
benefits were evaluated separately for domestic and imported passenger
cars under CAFE standards but not under CO2 standards
because EPCA/EISA's requirement for separate compliance applies only to
CAFE standards.
Both the final standards and, all other alternatives involve
standards less stringent than the no-action alternative. Therefore, as
discussed above, incremental benefits and costs for each alternative
are negative--in other words, each alternative involves forgone
benefits and avoided costs. Environmental and energy impacts are
correspondingly negative, involving forgone avoided CO2
emissions and forgone avoided fuel consumption. For consistency with
past rulemakings, these are reported as negative values rather than as
additional CO2 emissions and additional fuel consumption.
Like the NPRM and PRIA (and past rulemakings), today's rulemaking
and FRIA emphasize a ``model year'' perspective when reporting impacts.
That is, for enough model years (here, through MY 2029) to extend
beyond those when the estimated use of ``banked'' credits is reasonably
likely to be sufficient to show the average manufacturer not achieving
required CAFE or CO2 levels, the presentation of results
mainly considers the lifetime impacts attributable to vehicles produced
in these model years. Because standards are actually enforced on a
model year basis, this perspective aligns well with the consideration
of impacts on manufacturers and new vehicle buyers. However, impacts on
national energy consumption and the natural environment will involve
all vehicles on the road in future years, including those produced
after MY 2029. Therefore, similar to the approach followed in recent
and past EISs (and today's FEIS), today's rulemaking also presents
impacts on a ``calendar year'' basis--that is, summarizing overall
impacts (i.e., including those attributable to vehicles produced after
MY 2029) in each calendar year through 2050. As discussed in below, the
model year and calendar year perspectives draw on the same CAFE model
outputs, but differ in the scope of those outputs included in
summarized information.
As discussed above, more detailed results are available in the FRIA
and FEIS accompanying today's rulemaking, as well as in underlying
model output files posted on NHTSA's website.
1. Average Required Fuel Economy and CO2 Standard for PCs,
LTs, and Combined
The model fully represents the required CAFE and CO2
levels for every manufacturer and every fleet. The standard for each
manufacturer is based on the harmonic average of footprint targets (by
volume) within a fleet, just as the standards prescribe. Unlike earlier
versions of the CAFE model, the current version further disaggregates
passenger cars into domestic and imported classes (which manufacturers
report to NHTSA and EPA as part of their CAFE compliance submissions).
This allows the CAFE model to more accurately estimate the requirement
on the two passenger car fleets, represent the domestic passenger car
floor (which must be exceeded by every manufacturer's domestic fleet,
without the use of credits, but with the possibility of civil penalty
payment), and allows it to enforce the transfer cap limit that exists
between domestic and imported passenger cars, all for purposes of the
CAFE program.
In calculating the achieved CAFE level, the model uses the
prescribed harmonic average of fuel economy ratings within a vehicle
fleet. Under an ``unconstrained'' analysis, or in a model year for
which standards are already final, it is possible for a manufacturer's
CAFE to fall below its required level without generating penalties
because the model will apply expiring or transferred credits to
deficits if it is strategically appropriate to do so. Consistent with
current EPA regulations, the model applies simple (not harmonic)
production-weighted averaging to calculate average CO2
levels.
While the CAFE and CO2 standards themselves are, as
discussed in Section VI, inputs to the agencies' analysis, because the
standards are attribute-based standards specified separately for
passenger car and light truck fleets and applicable to average fuel
economy and CO2 levels, average requirements under these
standards are analytical results, not analytical inputs. Also, because
EPCA requires NHTSA to determine in advance minimum requirements that
will be applicable to manufacturers' fleets of domestic passenger cars,
these, too, are analytical results. The remainder of this section
presents these results.
a) Passenger Car Requirements
As discussed in Section V, the final standards are different from
the preferred alternative identified in the proposal.
We do not know yet with certainty what CAFE and CO2
levels will ultimately be required of individual manufacturers, because
those levels will depend on the mix of vehicles that each manufacturer
produces for sale in future model years. Based on the market forecast
of future sales used to examine the final standards, the agencies
currently estimate that the target functions shown above would result
in the following average required fuel economy and CO2
emissions levels for all manufacturers during MYs 2021-2026:
[[Page 24908]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.535
We emphasize again that the values in these tables are estimates,
and not necessarily the ultimate levels with which each of these
manufacturers will have to comply, for the reasons described
above.\2430\
---------------------------------------------------------------------------
\2430\ MY2017 values reflect the agencies' analysis, which uses
an analysis fleet developed using MY2017 compliance data as of
summer 2019. The analysis does not reflect subsequent updates and
corrections to manufacturers' MY2017 compliance data.
---------------------------------------------------------------------------
b) Light Truck Requirements
Again, while the agencies do not know yet with certainty what CAFE
and CO2 levels will ultimately be required of individual
manufacturers, because those levels will depend on the mix of vehicles
that each manufacturer produces for sale in future model years, based
on the market forecast of future sales used to examine today's proposed
standards, the agencies currently estimate that the target functions
shown above would result in the following average required fuel economy
and CO2 emissions levels for individual manufacturers during
MYs 2021-2026.
[GRAPHIC] [TIFF OMITTED] TR30AP20.536
[[Page 24909]]
We emphasize again the values in these tables are estimates and not
necessarily the ultimate levels with which each of these manufacturers
will have to comply for reasons described above.\2431\
---------------------------------------------------------------------------
\2431\ MY2017 values reflect the agencies' analysis, which uses
an analysis fleet developed using MY2017 compliance data as of
summer, 2019. The analysis does not reflect subsequent updates and
corrections to manufacturers' MY2017 compliance data.
---------------------------------------------------------------------------
c) Average of PassengerCcar and Light Truck Requirements
Overall average requirements will depend, further, on the relative
shares of passenger cars and light trucks in the new vehicle fleet. The
agencies' analysis estimates future shifts in these shares as vehicles'
average prices and fuel economy levels change, and as fuel prices also
change. Resultant estimates of overall average requirements are as
follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.537
(d) Estimated Average Requirements for Specific Manufacturers
Overall average requirements (e.g., reflecting both passenger car
and light truck fleets) applicable to each manufacturer will depend on
the mix (i.e., footprint distribution) of vehicles produced in each
model year, and relative production shares of passenger cars and light
trucks. Tables appearing below summarize estimated requirements through
model year 2029. Estimates for specific fleets (e.g., domestic
passenger cars, imported passenger cars, light trucks) are available in
CAFE model output files accompanying today's rulemaking, as are
estimates for each MYs 2030-2050.\2432\
---------------------------------------------------------------------------
\2432\ The model and all inputs and outputs supporting today's
notice are available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 24910]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.538
[[Page 24911]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.539
[[Page 24912]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.540
[[Page 24913]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.541
[[Page 24914]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.542
[[Page 24915]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.543
[[Page 24916]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.544
[[Page 24917]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.545
[[Page 24918]]
BILLING CODE 4910-59-C
2. Impacts on Vehicle Manufacturers
As mentioned above, impacts are presented from two different
perspectives for today's final rule. From either perspective, overall
impacts are the same. The first perspective, taken above in VII.A,
examines overall impacts of the standards--i.e., the entire series of
year-by-year standards--on each model year. The second perspective,
presented here, provides a clearer characterization of the incremental
impacts attributable to standards introduced in each successive model
year. For example, the new final standards for MY 2023 are likely to
impact manufacturers' application of technology in model years prior to
MY 2023, as well as model years after MY 2023. By conducting analysis
that successively introduces standards for each MY, in turn, isolates
the incremental impacts attributable to new standards introduced in
each MY, considering the entire span of MYs 1975-2029 and calendar
years 2016-2069 included in the analysis that only considers the full
series of successive MYs' standards. Tables appearing below summarize
results as aggregated across these model and calendar years. Underlying
model output files \2433\ report physical impacts and specific
monetized costs and benefits attributable to each model year in each
calendar (thus providing information needed to, for example,
differentiate between impacts attributable to the MY 1975-2017 and MY
2018-2029 cohorts). The FRIA presents costs and benefits for individual
model years (with MY's 1975-2017 in a single bucket) for the final
standards.
---------------------------------------------------------------------------
\2433\ Available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
a) Industry Average Technology Penetration Rates
The CAFE model tracks and reports technology application and
penetration rates for each manufacturer, regulatory class, and model
year, calculated as the volume of vehicles with a given technology
divided by the total volume. The ``application rate'' accounts only for
those technologies applied by the model during the compliance
simulation, while the ``penetration rate'' accounts for the total
percentage of a technology present in a given fleet, whether applied by
the CAFE model or already present at the start of the simulation.
In addition to the aggregate representation of technology
penetration, the model also tracks each individual vehicle model on
which it has operated. Accordingly, the CAFE model produces a record
for every model year and every alternative that identifies with which
technologies the vehicle started the simulation and which technologies
the same vehicle had at the conclusion of each model year. Interested
parties may use these outputs to assess how the compliance simulation
modified any vehicle that was offered for sale in MY 2017 in response
to a given regulatory alternative.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.546
[[Page 24919]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.547
[GRAPHIC] [TIFF OMITTED] TR30AP20.548
[[Page 24920]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.549
[GRAPHIC] [TIFF OMITTED] TR30AP20.550
[[Page 24921]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.551
[GRAPHIC] [TIFF OMITTED] TR30AP20.552
[[Page 24922]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.553
[GRAPHIC] [TIFF OMITTED] TR30AP20.554
[[Page 24923]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.555
[GRAPHIC] [TIFF OMITTED] TR30AP20.556
[GRAPHIC] [TIFF OMITTED] TR30AP20.557
[[Page 24924]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.558
[GRAPHIC] [TIFF OMITTED] TR30AP20.559
[[Page 24925]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.560
[[Page 24926]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.561
[[Page 24927]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.562
[[Page 24928]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.563
[[Page 24929]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.564
[[Page 24930]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.565
[[Page 24931]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.566
[[Page 24932]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.567
[[Page 24933]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.568
[[Page 24934]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.569
[[Page 24935]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.570
[[Page 24936]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.571
[[Page 24937]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.572
[[Page 24938]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.573
[[Page 24939]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.574
[[Page 24940]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.575
[[Page 24941]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.576
[[Page 24942]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.577
[[Page 24943]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.578
[[Page 24944]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.579
[[Page 24945]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.580
[[Page 24946]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.581
[[Page 24947]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.582
[[Page 24948]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.583
[[Page 24949]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.584
[[Page 24950]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.585
[[Page 24951]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.586
[[Page 24952]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.587
[[Page 24953]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.588
[[Page 24954]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.589
[[Page 24955]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.590
[[Page 24956]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.591
[[Page 24957]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.592
[[Page 24958]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.593
[[Page 24959]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.594
[[Page 24960]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.585
[[Page 24961]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.596
[[Page 24962]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.597
[[Page 24963]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.598
[[Page 24964]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.599
[[Page 24965]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.600
[[Page 24966]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.601
[[Page 24967]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.602
[[Page 24968]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.603
[[Page 24969]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.604
[[Page 24970]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.605
[[Page 24971]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.606
[[Page 24972]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.607
[[Page 24973]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.608
[[Page 24974]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.609
[[Page 24975]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.610
[[Page 24976]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.611
[[Page 24977]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.612
[[Page 24978]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.613
[[Page 24979]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.614
[[Page 24980]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.615
[[Page 24981]]
BILLING CODE 4910-59-C
b) Technology Costs
For each technology that the model adds to a given vehicle, it
accumulates cost. The technology costs are defined incrementally and
vary both over time and by technology class, where the same technology
may cost more to apply to larger vehicles as it involves more raw
materials or requires different specifications to preserve some
performance attributes. While learning-by-doing can bring down cost,
and should reasonably be implemented in the CAFE model as a rate of
cost reduction that is applied to the cumulative volume of a given
technology produced by either a single manufacturer or the industry as
a whole, in practice this notion is implemented as a function of time,
rather than production volume. Thus, depending upon where a given
technology starts along its learning curve, it may appear to be cost-
effective in later years where it was not in earlier years. As the
model carries forward technologies that it has already applied to
future model years, it similarly adjusts the costs of those
technologies based on their individual learning rates.
BILLING CODE 4910-59-P
[[Page 24982]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.616
[[Page 24983]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.617
[[Page 24984]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.618
[[Page 24985]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.619
[[Page 24986]]
BILLING CODE 4910-59-C
c) Civil Penalties
The other costs that manufacturers incur as a result of CAFE
standards are civil penalties resulting from non-compliance with CAFE
standards. The CAFE model accumulates costs of $5.50 per 1/10-MPG under
the standard, multiplied by the number of vehicles produced in that
fleet, in that model year. The model reports as the full ``regulatory
cost,'' the sum of total technology cost and total fines by the
manufacturer, fleet, and model year. As mentioned above, the relevant
EPCA/EISA provisions do not also appear in the CAA, so this option and
these costs apply only to simulated compliance with CAFE standards.
d) Average Prices, Sales, and Revenue Changes
In all previous versions of the CAFE model, the total number of
vehicles sold in any model year, in fact the number of each individual
vehicle model sold in each year, has been a static input that did not
vary in response to price increases induced by CAFE standards, nor
changes in fuel prices, or any other input to the model. The only way
to alter sales, was to update the entire forecast in the market input
file. However, in the 2012 final rule, the agencies included a dynamic
fleet share model that was based on a module in the Energy Information
Administration's NEMS model. This fleet share model did not change the
size of the new vehicle fleet in any year, but it did change the share
of new vehicles that were classified as passenger cars (or light
trucks). That capability was not included in the central analysis but
was included in the uncertainty analysis, which looked at the baseline
and final standards in the context of thousands of possible future
states of the world. As some of those futures contained extreme cases
of fuel prices, it was important to ensure consistent modeling
responses within that context. For example, at a gasoline price of $7/
gallon, it would be unrealistic to expect the new vehicle market's
light truck share to be the same as the future where gasoline cost $2/
gallon. The current model has slightly modified, and fully integrated,
the dynamic fleet share model. Every regulatory alternative and
sensitivity case considered for this analysis reflects a dynamically
responsive fleet mix in the new vehicle market.
While the dynamic fleet share model adjusts unit sales across body
styles (cars, SUVs, and trucks), it does not modify the total number of
new vehicles sold in a given year. The CAFE model now includes a
separate function to account for changes in the total number of new
vehicles sold in a given year (regardless of regulatory class or body
style), in response to certain macroeconomic inputs and changes in the
average new vehicle price. The price impact is modest relative to the
influence of the macroeconomic factors in the model. The combination of
these two models modify the total number of new vehicles, the share of
passenger cars and light trucks, and, as a consequence, the number of
each given model sold by a given manufacturer. However, these two
factors are insufficient to cause large changes to the composition of
any of a manufacturer's fleets. In order to change significantly the
mix of models produced within a given fleet, the CAFE model would
require a way to trade off the production of one vehicle versus another
both within a manufacturer's fleet and across the industry. While the
agencies have experimented with fully-integrated consumer choice
models, their performance has yet to satisfy the requirements of a
rulemaking analysis.
Above, Section VI discusses at length the sales model the agencies
have applied in the analysis supporting today's rulemaking.
BILLING CODE 4910-59-P
[[Page 24987]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.620
[[Page 24988]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.621
[[Page 24989]]
BILLING CODE 4910-59-C
e) Labor
As discussed in Section VI the analysis includes estimates of
impacts on U.S. auto industry labor, considering the combined impact of
changes in sales volumes and changes in outlays for additional fuel-
saving technology. Note: This analysis does not consider the
possibility that potential new jobs and plants attributable to
increased stringency will not be located in the United States, or that
increased stringency will not lead to the relocation of current jobs or
plants to foreign countries. Compared to the no-action alternative
(i.e., the baseline standards), the new final standards (alternative 1)
and other regulatory alternatives under consideration all involve
reduced regulatory costs expected to lead to reduced average vehicle
prices and, in turn, increased sales. While the increased sales
slightly increase estimated U.S. auto sector labor hours, because
producing and selling more vehicles uses additional U.S. labor, the
reduced outlays for fuel-saving technology slightly reduce estimated
U.S. auto sector labor hours, because manufacturing, integrating, and
selling less technology means using less labor to do so. Of course,
this is technology that may not otherwise be produced or deployed were
it not for regulatory mandate, and the additional costs of this
technology would be borne by a reduced number of consumers given
reduction in sales in response to increased prices. Today's analysis
shows the negative impact of reduced mandatory technology outlays
outweighing the positive impact of increased sales. However, both of
these underlying factors are subject to uncertainty. For example, if
fuel-saving technology that would have been applied under the baseline
standards is more likely to have come from foreign suppliers than
estimated here, less of the forgone labor to manufacture that
technology would have been U.S. labor. Also, if sales would be more
positively impacted by reduced vehicle prices than estimated here,
correspondingly positive impacts on U.S. auto sector labor could be
magnified. Alternatively, if manufacturers are able to deploy
technology to improve vehicle attributes that new car buyers prefer to
fuel economy improvements, both technology spending and vehicle sales
would correspondingly increase.
The labor utilization analysis was focused on automotive labor
because adjacent labor utilization factors and consumer spending
factors for other goods and services are uncertain and difficult to
predict. How direct labor changes may affect the macro economy and
possibly change employment in adjacent industries were not considered.
For instance, possible labor changes in vehicle maintenance and repair
were not considered, nor were changes in labor at retail gas stations
considered. Possible labor changes due to raw material production, such
as production of aluminum, steel, copper, and lithium were not
considered, nor were possible labor impacts due to changes in
production of oil and gas, ethanol, and electricity considered. Effects
of how consumers could spend money saved due to improved fuel economy
were not analyzed, nor were effects of how consumers would pay for more
expensive fuel savings technologies at the time of purchase analyzed;
either could affect consumption of other goods and services, and hence
affect labor in other industries. The effects of increased usage of
car-sharing, ride-sharing, and automated vehicles were not analyzed.
How changes in labor from any industry could affect gross domestic
product and possibly affect other industries as a result were not
estimated.
Also, no assumptions were made about full-employment or not full-
employment and the availability of human resources to fill positions.
When the economy is at full employment, a fuel economy regulation is
unlikely to have much impact on net overall U.S. labor utilization;
instead, labor would primarily be shifted from one sector to another.
These shifts in employment impose an opportunity cost on society,
approximated by the wages of the employees, as regulation diverts
workers from other activities in the economy. In this situation, any
effects on net employment are likely to be transitory as workers change
jobs (e.g., some workers may need to be retrained or require time to
search for new jobs, while shortages in some sectors or regions could
bid up wages to attract workers). On the other hand, if a regulation
comes into effect during a period of high unemployment, a change in
labor demand due to regulation may affect net overall U.S. employment
because the labor market is not in equilibrium. Schmalansee and Stavins
point out that net positive employment effects are possible in the near
term when the economy is at less than full employment due to the
potential hiring of idle labor resources by the regulated sector to
meet new requirements (e.g., to install new equipment) and new economic
activity in sectors related to the regulated sector. In the longer run,
the net effect on employment is more difficult to predict and will
depend on the way in which the related industries respond to the
regulatory requirements. For that reason, this analysis does not
include multiplier effects but instead focuses on labor impacts in the
most directly affected industries. Those sectors are likely to face the
most concentrated labor impacts.
The tables presented below summarize these results for the final
standards and other regulatory alternatives considered. While values
are reported as thousands of person-years, changes in labor utilization
would not necessarily involve the same number of changes in actual
jobs, as auto industry employers may use a range of strategies (e.g.,
shift changes, overtime) beyond simply adding or eliminating jobs.
(1) CAFE Standards
[[Page 24990]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.622
[GRAPHIC] [TIFF OMITTED] TR30AP20.623
(2) CO2 Standards
[[Page 24991]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.624
3. Impacts to Vehicle Buyers
BILLING CODE 4910-59-P
[[Page 24992]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.625
[[Page 24993]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.626
[[Page 24994]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.627
[[Page 24995]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.628
[[Page 24996]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.629
[[Page 24997]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.630
[[Page 24998]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.631
[[Page 24999]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.632
[[Page 25000]]
BILLING CODE 4910-59-C
a) Average Price Increase
BILLING CODE 4910-59-P
[[Page 25001]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.633
[[Page 25002]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.634
[[Page 25003]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.635
[[Page 25004]]
BILLING CODE 4910-59-C
4. Impacts to Society
As the CAFE model simulates manufacturer compliance with regulatory
alternatives, it estimates and tracks a number of consequences that
generate social costs. The most obvious cost associated with the
program is the cost of additional fuel economy improving/CO2
emissions reducing technology that is added to new vehicles as a result
of the rule. However, the model does not inherently draw a distinction
between costs and benefits. For example, the model tracks fuel
consumption and the dollar value of fuel consumed. This is the cost of
travel under a given alternative (including the baseline). The ``cost''
or ``benefit'' associated with the value of fuel consumed is determined
by the reference point against which each alternative is considered.
The CAFE model reports absolute values for the amount of money spent on
fuel in the baseline, then reports the amount spent on fuel in the
alternatives relative to the baseline. If the baseline standard were
fixed at the current level, and an alternative achieved significantly
greater mpg by 2025, the total expenditures on fuel in the alternative
would be lower, creating a fuel savings ``benefit.'' This analysis uses
a baseline that is more stringent than each alternative considered, so
the incremental fuel expenditures are greater for the alternatives than
for the baseline.
Other social costs and benefits emerge as the result of physical
phenomena, like tailpipe emissions or highway fatalities, which are the
result of changes in the composition and use of the on-road fleet. The
social costs associated with those quantities represent an economic
estimate of the social damages associated with the changes in each
quantity. The model tracks and reports each of these quantities by:
Model year and vehicle age (the combination of which can be used to
produce calendar year totals), regulatory class, fuel type, and social
discount rate.
The full list of potential costs and benefits is presented in Table
VII-90 as well as the population of vehicles that determines the size
of the factor (either new vehicles or all registered vehicles) and the
mechanism that determines the size of the effect (whether driven by the
number of miles driven, the number of gallons consumed, or the number
of vehicles produced).
BILLING CODE 4910-59-P
[[Page 25005]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.636
[[Page 25006]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.637
[[Page 25007]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.638
[[Page 25008]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.639
[[Page 25009]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.640
[[Page 25010]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.641
[[Page 25011]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.642
[[Page 25012]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.643
[[Page 25013]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.644
[[Page 25014]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.645
[[Page 25015]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.646
[[Page 25016]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.647
[[Page 25017]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.648
[[Page 25018]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.649
[[Page 25019]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.650
[[Page 25020]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.651
[[Page 25021]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.652
[[Page 25022]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.653
[[Page 25023]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.654
The above tables summarizing estimated benefits and costs of the
regulatory alternatives considered here exclude results of the implicit
opportunity cost calculations discussed above and in Section
VI.D.1.b)(8)
[[Page 25024]]
Implicit Opportunity Cost. The following four tables show corresponding
benefits and costs when results of these calculations are included:
[[Page 25025]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.655
[[Page 25026]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.656
[[Page 25027]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.657
[[Page 25028]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.658
[[Page 25029]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.659
[[Page 25030]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.660
[[Page 25031]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.661
[[Page 25032]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.662
a) Impacts on Total Fleet Size, Usage, and Safety
[[Page 25033]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.663
[[Page 25034]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.664
[[Page 25035]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.665
[[Page 25036]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.666
[[Page 25037]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.667
[[Page 25038]]
BILLING CODE 4910-59-C
(1) Total Fleet Size and VMT
The CAFE model carries a complete representation of the registered
vehicle population in each calendar year, starting with an aggregated
version of the most recent available data about the registered
population for the first year of the simulation. In this analysis, the
first model year considered is MY 2017, and the registered vehicle
population enters the model as it appeared at the end of calendar year
2016. The initial vehicle population is stratified by age (or model
year cohort) and regulatory class--to which the CAFE model assigns
average fuel economies based on the reported regulatory class industry
average compliance value in each model year (and class). Once the
simulation begins, new vehicles are added to the population from the
market data file and age throughout their useful lives during the
simulation, with some fraction of them being retired (or scrapped)
along the way. For example, in calendar year 2018, the new vehicles
(age zero) are MY 2018 vehicles (added by the CAFE model simulation and
represented at the same level of detail used to simulate compliance),
the age one vehicles are MY 2017 vehicles (added by the CAFE model
simulation), and the age two vehicles are MY 2016 vehicles (inherited
from the registered vehicle population and carried through the analysis
with less granularity). This national registered fleet is used to
calculate annual fuel consumption, vehicle miles traveled (VMT),
pollutant emissions, and safety impacts under each regulatory
alternative.
In support of prior CAFE rulemakings, the CAFE model accounted for
new travel that results from fuel economy improvements that reduce the
cost of driving. The magnitude of the increase in travel demand is
determined by the rebound effect. In both previous versions and the
current version of the CAFE model, the amount of travel demanded by the
existing fleet of vehicles is also responsive to the rebound effect
(representing the price elasticity of demand for travel)--increasing
when fuel prices decrease relative to the fuel price when the VMT on
which our mileage accumulation schedules were built was observed. Since
the fuel economy of those vehicles is already fixed, only the fuel
price influences their travel demand relative to the mileage
accumulation schedule and so is identical for all regulatory
alternatives.
While the average mileage accumulation per vehicle by age is not
influenced by the rebound effect in a way that differs by regulatory
alternative, three other factors influence total VMT in the model in a
way that produces different total mileage accumulation by regulatory
alternative. The first factor is the total industry sales response: New
vehicles are both driven more than older vehicles and are more fuel
efficient (thus producing more rebound miles). To the extent that more
(or fewer) of these new models enter the vehicle fleet in each model
year, total VMT will increase (or decrease) as a result. The second
factor is the dynamic fleet share model. The fleet share influences not
only the fuel economy distribution of the fleet, as light trucks are
less efficient than passenger cars on average, but the total miles are
influenced by fact that light trucks are driven more than passenger
cars as well. Both of the first two factors can magnify the influence
of the rebound effect on vehicles that go through the compliance
simulation (MY 2017-2050) in the manner discussed above. The third
factor influencing total annual VMT is the scrappage model. By
modifying the retirement rates of on-road vehicles under each
regulatory alternative, the scrappage model either increases or
decreases the lifetime miles that accrue to vehicles in a given model
year cohort.
In addition to dynamically modifying the total number of new
vehicles sold, a dynamic model of vehicle retirement, or scrappage, has
also been implemented. The model implements the scrappage response by
defining the instantaneous scrappage rate at any age using two
functions. For ages less than 30, instantaneous scrappage is defined as
a function of vehicle age, new vehicle price, fuel prices, cost per
mile of driving (the ratio of fuel price and fuel economy), and GDP
growth rate. For ages greater than 30, the instantaneous scrappage rate
is a simple exponential function of age. While the scrappage response
does not affect manufacturer compliance calculations, it impacts the
lifetime mileage accumulation (and thus fuel savings) of all vehicles.
Previous CAFE analyses have focused exclusively on new vehicles,
tracing the fuel consumption and social costs of these vehicles
throughout their useful lives; the scrappage effect also impacts the
registered vehicle fleet that exists when a set of standards is
implemented.
For a given calendar year, the retirement rates of the registered
vehicle population are governed by the scrappage model. To the extent
that a given set of CAFE or CO2 standards accelerates or
decelerates the retirement of vehicles, fuel consumption and social
costs may change. The CAFE model accounts for those costs and benefits,
as well as tracking all of the standard benefits and costs associated
with the lifetimes of new vehicles produced under the rule. For more
detail about the derivation of the scrappage functions, see Section VI.
(2) Fuel Consumption
For every vehicle model in the market file, the model estimates the
VMT per vehicle (using the assumed VMT schedule, the vehicle fuel
economy, fuel price, and the rebound assumption). Those miles are
multiplied by the volume for each vehicle. Fuel consumption is the
product of miles driven and fuel economy, which can be tracked by model
year cohort in the model. Carbon dioxide emissions from vehicle
tailpipes are the simple product of gallons consumed and the carbon
content of each gallon.
In order to calculate calendar year fuel consumption, the model
needs to account for the inherited on-road fleet in addition to the
model year cohorts affected by this new final rule. Using the VMT of
the average passenger car and light truck from each cohort, the model
computes the fuel consumption of each model year class of vehicles for
its age in a given CY. The sum across all ages (and thus, model year
cohorts) in a given CY provides estimated CY fuel consumption.
Because the model produces an estimate of the aggregate number of
gallons sold in each CY, it is possible to calculate both the total
expenditures on motor fuel and the total contribution to the Highway
Trust Fund (HTF) that result from that fuel consumption. The Federal
fuel excise tax is levied on every gallon of gasoline and diesel sold
in the U.S., with diesel facing a higher per-gallon tax rate. The model
uses a national perspective, where the State taxes present in the input
files represent an estimated average fuel tax across all U.S. States.
Accordingly, while the CAFE model cannot reasonably estimate potential
losses to State fuel tax revenue from increasingly the fuel economy of
new vehicles, it can do so for the HTF.
In addition to the tailpipe emissions of carbon dioxide, each
gallon of gasoline produced for consumption by the on-road fleet has
associated ``upstream'' emissions that occur in the extraction,
transportation, refining, and distribution of the fuel. The model
accounts for these emissions as well (on a per-gallon basis) and
reports them accordingly.
(3) Safety
Earlier versions of the CAFE model accounted for the safety impacts
associated with reducing vehicle mass
[[Page 25039]]
in order to improve fuel economy. In particular, NHTSA's safety
analysis estimated the additional fatalities that would occur as a
result of new vehicles getting lighter, then interacting with the on-
road vehicle population. In general, taking mass out of the heaviest
new vehicles improved safety outcomes, while taking mass from the
lightest new vehicles resulted in a greater number of expected highway
fatalities. However, the change in fatalities did not adequately
account for changes in exposure that occur as a result of increased
demand for travel as vehicles become cheaper to operate. The current
version of the model resolves that limitation and addresses additional
sources of fatalities that can result from the implementation of CAFE
or CO2 standards. These are discussed in greater detail in
Section VI.
The agencies have observed that older vehicles in the population
are responsible for a disproportionate number of fatalities, both by
number of registrations and by number of miles driven. Accordingly, any
factor that causes the population of vehicles to turn over more slowly
will induce additional fatalities--as those older vehicles continue to
be driven, rather than being retired and replaced with newer (even if
not brand new) vehicle models. The scrappage effect, which delays (or
accelerates) the retirement of registered vehicles, impacts the number
of fatalities through this mechanism--importantly affecting not just
new vehicles sold from model years 2017-2050 but existing vehicles that
are already part of the on-road fleet. Similarly, to the extent that a
CAFE or CO2 alternative reduces new vehicle sales, it can
slow the transition from older vehicles to newer vehicles, reducing the
share of total vehicle miles that are driven by newer, more
technologically advanced vehicles. Furthermore, newer vehicles are
equipped with technologies that make driving safer not only safer for
the occupants of newer vehicles, but also pedestrians, cyclists, and
even occupants of other vehicles. Accounting for the change in vehicle
miles traveled that occurs when vehicles become cheaper to operate
leads to a number of fatalities that can be attributed to the rebound
effect, independent of any changes to new vehicle mass, price, or
longevity.
The CAFE model estimates fatalities by combining the effects
discussed above. In particular, the model estimates the fatality rate
per billion miles VMT for each model year vehicle in the population
(the newest of which are the new vehicles produced that model year).
This estimate is independent of regulatory class and varies only by
year (and not vehicle age). The estimated fatality rate is then
multiplied by the estimated VMT (in billions of miles) for each vehicle
in the population and the product of the change in curb weight and the
relevant safety coefficient, as in the equation below.
[GRAPHIC] [TIFF OMITTED] TR30AP20.668
For the vehicles in the historical fleet, meaning all those
vehicles that are already part of the registered vehicle population in
CY 2017, only the model year effect that determines the
``FatalityEstimate'' is relevant. However, each vehicle that is
simulated explicitly by the CAFE model, and is eligible to receive mass
reduction technologies, must also consider the change between its curb
weight and the threshold weights that are used to define safety
classes. For vehicles above the threshold, reducing vehicle mass can
have a smaller negative impact on fatalities (or even reduce
fatalities, in the case of the heaviest light trucks). The
``ChangePer100Lbs'' depends upon this difference. The sum of all
estimated fatalities for each model year vehicle in the on-road fleet
determines the reported fatalities, which can be summarized by either
model year or calendar year.
b) Environmental Impacts
Today's final rule directly involves the fuel economy and average
CO2 emissions of light-duty vehicles, and the final rule is
expected directly and significantly to impact national fuel consumption
and CO2 emissions. Fuel economy and CO2 emissions
are closely related, so that it is expected the impacts on national
fuel consumption and national CO2 emissions will track in
virtual lockstep with each other.
Today's final rule does not directly involve pollutants such as
carbon monoxide, smog-forming pollutants (nitrogen oxides and unburned
hydrocarbons), fine particulate matter, or ``air toxics'' (e.g.,
formaldehyde, acetaldehyde, benzene). While today's final rule is
expected to impact such emissions indirectly (by reducing travel demand
and accelerating fleet turnover to newer and cleaner vehicles on one
hand while, on the other, increasing activity at refineries and in the
fuel distribution system), it is expected that these impacts will be
much smaller than impacts on fuel use and CO2 emissions
because standards for these other pollutants are independent of those
for CO2 emissions.
Following decades of successful regulation of criteria pollutants
and air toxics, modern vehicles are already vastly cleaner than in the
past, and it is expected that new vehicles will continue to improve.
For example, the following chart shows trends in new vehicles' emission
rates \2434\ for volatile organic compounds (VOCs) and nitrogen oxides
(NOX)--the two motor vehicle criteria pollutants that
contribute to the formation of smog.
---------------------------------------------------------------------------
\2434\ The emission rate is the rate at which a vehicle emits a
given pollutant into the atmosphere. Tailpipe emission rates are
expressed on a gram per mile basis. For example, driving 15,000
miles in a year, a vehicle with a 0.4 g/mi NOX emission
rate would emit 6,000 grams of NOX.
---------------------------------------------------------------------------
[[Page 25040]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.669
Because new vehicles are so much cleaner than older models, it is
expected that under any of the alternatives considered here for fuel
economy and CO2 standards, emissions of smog-forming
pollutants would continue to decline nearly identically over the next
two decades. The following chart shows estimated total fuel
consumption, CO2 emissions, and smog-forming emissions under
the baseline and new final standards (CAFE standards--trends for
CO2 standards would be very similar), normalized to 2017
levels in order to allow the three to be shown together on a single
chart:
[[Page 25041]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.670
The following table summarizes relative differences between the
baseline/augural and final standards:
[GRAPHIC] [TIFF OMITTED] TR30AP20.671
As indicated, the agencies' analysis indicates that through 2050,
increases in annual light-duty fuel consumption and CO2
emissions would remain below 10 percent, and increases in annual light-
duty emissions of smog-forming pollutants would remain below 2.5
percent.
As the analysis affirms, while fuel economy and CO2
emissions are two sides (or, arguably, the same side) of the same coin,
fuel economy and CO2 are only incidentally related to
pollutants such as smog, and any positive or negative impacts of
today's rulemaking on these other air quality problems would most
likely be far too small to observe.
[[Page 25042]]
The remainder of this section summarizes the impacts on fuel
consumption and emissions for both the new final CAFE standards and the
new final CO2 standards.
(1) Understanding Energy and Environmental Impacts
Today's rulemaking and accompanying FRIA and FEIS all examine a
range of physical impacts. These impacts reflect the combined effect of
a range of different factors, some of which are independent of one
another, and some of which interact. The scope and nature of this set
of factors is such that, even among knowledgeable experts, intuition is
often uninformative or even misleading.
On one hand, it is reasonable to be confident that the more CAFE
and CO2 standards are relaxed, the more national-scale fuel
consumption and CO2 emissions will increase, because the
standards apply directly to the average rates at which new vehicle
consume fuel and, in turn, emit CO2. While other factors--
including some that work against this expectation--are involved, these
other factors are insufficient to belie this basic expectation that
less stringent standards will lead to increased fuel consumption and
CO2 emissions.
On the other hand, while it is intuitive to expect that the
increased fuel consumption should lead to some additional emissions to
produce and distribute fuel, those processes are expected to become
cleaner over time, and refineries may respond by reducing exports of
petroleum products rather than increasing overall activity. Although
many believe that more fuel-efficient vehicles are, by definition,
``cleaner,'' most pollutants impacting air quality are regulated on an
average per-mile basis, such that vehicles' ``cleanliness'' is
effectively independent from vehicles' fuel economy.\2435\ However,
because emissions standards relevant to air quality are so much more
stringent than in the past, and because some emission control
technologies (e.g., catalytic converters) tend to deteriorate as
vehicles age, average emission rates of vehicles are very dependent on
when those vehicles were produced and how old they are. This means that
total vehicular emissions of pollutants impacting air quality depend
not directly on fuel economy, but rather on the amount of highway
travel (since emissions are regulated on a per-mile basis) and on how
that travel is distributed among older and newer vehicles. The agencies
estimate that relaxing CAFE and CO2 standards will, by
decreasing the price and fuel economy levels of vehicles produced after
MY 2017, lead to changes in the quantities of new vehicles produced and
sold in the U.S., as well as changes in fleet mix (i.e., the relative
shares of passenger cars and light trucks, which are subject to
different emissions standards), and changes in the rates at which older
vehicles are removed from service (i.e., scrapped). Is it reasonable to
expect that less stringent standards will necessarily accelerate the
turnover to newer, cleaner vehicles? Does that depend on fuel prices?
Yet another factor involves the prevalence of electric vehicles, which
emit no air pollutants directly, but do use electricity. How might that
electricity be generated in the future? Also, does it necessarily
follow that less stringent CAFE and CO2 standards will
reduce the sale of battery electric vehicles (BEVs) in the long term?
Could less stringent standards increase long-term BEV sales if
manufacturers are able to make early investments in BEV research and
development, or wait for the costs of BEV systems to decline, rather
than making larger nearer-term commitments to, say, very advanced
engine technologies? With air quality depending on how emissions of
various pollutants are impacted (and sometimes in different ways) by
these factors, there is scant basis for a priori expectations regarding
the direction, much less the magnitude of air quality impacts under the
various regulatory alternatives.
---------------------------------------------------------------------------
\2435\ For example, in 42 U.S.C. 7521(g), the 1990 Clean Air Act
Amendments defined specific numerical standards for passenger car
and light truck CO, NMHC (i.e., VOC), and NOx emission rates, and
defined them on a gram per mile basis, such that the 3-cylinder 1993
Geo Metro and the 12-cylinder 1993 Ferrari 512 were both regulated
to 0.4 grams per mile of NOx, even though the Metro's average fuel
economy rating, at 47 mpg, was more than four times greater than the
Ferrari's 11 mpg rating.
---------------------------------------------------------------------------
Although, like any other model, the CAFE model involves many
uncertainties and does not account for every possible factor or
interaction, the model does enable the agencies to estimate emissions
impacts accounting for the factors mentioned above, and specific
results can be understood through careful examination of model inputs,
outputs, and methods. To illustrate this, the agencies consider
estimated emissions of nitrogen oxides (NOX), a class of
pollutants that contribute to the formation of ground-level ozone
(i.e., smog) that is harmful to public health and welfare. The agencies
apply the same ``unconstrained'' modeling approach as underlies the
FEIS. Graphing estimated annual tailpipe, upstream, and combined total
NOX emissions from passenger cars and light trucks shows
emissions declining significantly over time, with results from the
various action alternatives (focusing here on the least stringent,
preferred, and most stringent alternatives, and applying the same
vertical scale to all three charts) being virtually indistinguishable
from the no-action alternative:
BILLING CODE 4910-59-P
[[Page 25043]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.672
[[Page 25044]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.673
[[Page 25045]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.674
BILLING CODE 4910-59-C
Closer examination, though, reveals that although differences are
very small on a relative scale, they do exhibit definitive trends.
Reducing stringency causes total annual tailpipe NOX
emissions to decline initially, as scrappage of older higher-emitting
vehicles is accelerated and sales of new vehicles increase slightly
relative to augural standards. Over time, both of these trends are
impacted by steadily increasing fuel prices, but more important,
reducing stringency causes the market to shift somewhat more slowly to
electric vehicles than under the augural standards. Because electric
vehicles emit no NOX directly, the impact on NOX
emissions of this dampening of electric vehicle sales eventually
outweighs the other impacts, such that by approximately 2035, less
stringent standards begin increasing annual tailpipe NOx emissions
rather than decreasing these emissions (relative to the augural
standards):
[[Page 25046]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.675
On the other hand, at least through 2050, less stringent standards
show increased upstream NOX emissions. These increases
continue to build through the late 2030s, as total fuel consumption
under the less stringent standards continues to increase relative to
levels under the augural standards. However, by 2040, these increases
are steadily shrinking, due to the same delayed shift to electric
vehicles:
[[Page 25047]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.676
Model outputs indicate that on a per-mile basis, upstream
NOX emissions beyond 2030 are 2-24 percent greater for
electricity than for gasoline, varying over time and between regulatory
alternatives. (Although the agencies have applied the same upstream
emission factors to all regulatory alternatives, comparative per-mile
upstream emissions also depend on comparative vehicle efficiency.) This
means that, although a shift to electrification reduces tailpipe
emissions, it also tends to increase net upstream emissions.
Taken together, these changes in tailpipe emissions produce very
slight decreases in overall annual NOX emissions through
about 2026 under each regulatory alternative. Beyond 2026, the
regulatory action alternatives all produce increased overall annual
NOX emissions relative to the augural standards, although
for the most stringent regulatory alternative considered here, these
increases plateau after about 2040:
[[Page 25048]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.677
Still, although trends and differences between regulatory
alternatives are clear on the scale of the last three of the above
charts, the preceding three charts place these emissions changes in
context, and show that they are barely discernable. For example, the
largest increase shown in the last of the above charts is about 0.015
million tons, in 2050, when total emissions are 0.33-0.35 million tons,
down from about 1.5 million tons in 2017. In other words, the largest
increase in overall annual NOX emissions is only about 1
percent of recent annual NOX emissions attributable to
passenger cars and light trucks.
The FEIS accompanying today's rulemaking presents tailpipe,
upstream, and total emissions for a range of pollutants, and presents
results of photochemical modeling to estimate corresponding changes in
air quality, as well as results of calculations to estimate resultant
health impacts. As indicated by the following chart, at least for the
final standards, VOC and PM emissions follow overall trends broadly
similar to those followed by NOX emissions, although,
relative to recent (2017) total emissions attributable to passenger
cars and light trucks, changes in VOC and PM emissions are not as small
as changes in NOX emissions. Under the final standards,
combined tailpipe and upstream CO emissions are very slightly lower
than under the augural standards through the early 2030s, after which
these emissions changes begin increasing at rates similar to those for
VOC, NOX, and PM. CO2 emissions changes exhibit
the expected trend mentioned above, with combined tailpipe and upstream
emissions steadily increasing under the final standards. However, the
final standards lead combined tailpipe and upstream SO2
emissions to decrease relative to the augural standards, and as a share
of 2017 emissions, these decreases grow from about 2 percent in 2035 to
about 10 percent in 2050:
[[Page 25049]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.678
As indicated by the following chart, changes in tailpipe
SO2 emission follow trends nearly identical to those
followed by changes in CO2 emissions, because both result
directly from the quantity and composition (sulfur and carbon per
gallon, respectively) of fuel consumed:
[[Page 25050]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.679
This means that the decreases in overall SO2 emissions
must be attributable to decreases in upstream SO2 emissions.
The following chart shows SO2 emissions decreases becoming
steadily larger after the mid-2030s, suggesting that, as discussed
above, delaying the shift to electric vehicles leads to delays in
emissions from electricity generation, and for some pollutants (notably
below, SO2 and CO2), these emissions from
electricity generation are large enough to reverse trends in overall
emissions changes. For SO2, this reflects, among other
things, the fact that, in order to enable catalytic converters to
operate more efficiently, gasoline in sulfur is now limited to an
average of 10 parts per million.\2436\
---------------------------------------------------------------------------
\2436\ See https://www.epa.gov/regulations-emissions-vehicles-and-engines/final-rule-control-air-pollution-motor-vehicles-tier-3.
---------------------------------------------------------------------------
[[Page 25051]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.680
Again, the FEIS accompanying today's rulemaking further explores
changes in emissions; the purpose of this discussion is not to
duplicate material appearing in the FEIS, but rather to discuss some of
the underlying factors and how they can lead to some of the trends
reported in the FEIS.
Unlike the FEIS, today's rulemaking and accompanying FRIA largely
examine impacts on a ``model year basis.'' As discussed below, while a
calendar year basis involves considering impacts in one or a series of
calendar years, a model year basis involves considering impacts over
the useful lives of vehicles produced in one or over a series of model
years. A calendar year approach answers the question ``what do we
estimate will happen in, for example, 2035?,'' and a model year
approach answers the question ``what impacts do we estimate will be
attributable to vehicles produced in 2025?'' The calendar approach does
not extend beyond 2050, the last year in which the analysis includes a
complete on-road fleet. On the other hand, while it accounts for model
year 2050 vehicles' fuel consumption and emissions through 2089, the
model year approach as implemented here does not extend beyond model
year 2029.
These are differences in temporal perspective that, for some types
of impacts, lead to differences in reported trends. For example,
returning to tailpipe NOX emissions, Figure VII-6 (using the
calendar year perspective) shows that relaxing the stringency of CAFE
standards leads annual tailpipe NOX emissions to increase
starting around 2035, but leads these emissions to decrease in the
nearer term. As discussed above, this shift can be attributed to the
less stringent standards leading to a delayed shift toward electric
vehicles. Because the model year perspective as implemented here
extends through 2029, it largely sets aside this shift to electric
vehicles, even for the ``unconstrained'' modeling underlying the FEIS
(modeling which, unlike the ``standard setting'' type of analysis
required by EPCA, considers that, even during 2018-2029, additional
electric vehicles might be produced in response to standards).
Consequently, unlike the calendar year perspective as applied beyond
2035, the model year perspective that extends through MY 2029 always
shows tailpipe NOX emissions decreasing as the stringency of
CAFE standards is relaxed relative to the augural standards.
In addition to this difference in temporal perspective, the FEIS,
relative to the rulemaking and FRIA, applies a perspective that is
different in terms of how manufacturers could respond to standards. The
``unconstrained'' modeling underlying the FEIS allows for the potential
that manufacturers might apply CAFE compliance credits or introduce
additional electric vehicles in any model year. This is intended to
reflect how manufacturers might respond to standards in the real world.
However, EPCA requires that, for purposes of determining the maximum
feasible standards, NHTSA set aside the potential that manufacturers
might apply credits or increase electric vehicle offerings in the model
years under consideration. Therefore, for CAFE, the preamble and FRIA
use modeling that sets aside the potential use of credits
[[Page 25052]]
and the potential introduction of new electric vehicles through 2029
(although, since standards prior to MY 2021 are not subject to
reconsideration, this modeling does consider the potential use of
credits through MY 2020). As indicated by the following chart,
especially prior to model year 2030, this leads to significant
differences in EV market penetration between the two types of analyses:
[GRAPHIC] [TIFF OMITTED] TR30AP20.681
Over time, these differences in EV sales lead to significant
differences in the steadily accumulating share of overall highway
travel powered with electricity:
[[Page 25053]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.682
For most pollutants, the fact that EVs do not emit air pollutants
outweighs the fact that combustion-based power plants do. As discussed
above, sulfur content in gasoline is so low that the opposite is the
case for net SO2 emissions.
A complete quantitative analysis of differences between calendar
year-based emissions trends shown in the FEIS and model year-based
emissions trends shown in the rulemaking and FRIA would involve
examination of all of the factors mentioned above. However, considering
the temporal difference in perspective between the two types of
analyses, and considering the differences in the timing and pace of the
estimated transition to electric vehicles, differences in emissions
trends are inevitable.
(2) CO2 Damages
Section V discusses, among other things, the need of the Nation to
conserve energy, providing context for the estimated impacts on
national-scale fuel consumption summarized below. Corresponding to
these changes in fuel consumption, the agencies estimate that today's
final rule will impact CO2 emissions. CO2 is one
of several gases that absorb infrared radiation, thereby trapping heat
and potentially making the planet warmer. The most important such gases
directly emitted by human activities include carbon dioxide
(CO2), methane (CH4), nitrous oxide
(N2O), and several fluorine-containing halogenated
substances. Although CO2, CH4, and N2O
occur naturally in the atmosphere, human activities have changed their
atmospheric concentrations. From the pre-industrial era (i.e., ending
about 1750) to 2016, concentrations of these gases have increased
globally by 44, 163, and 22%, respectively.\2437\ The FEIS accompanying
today's rulemaking discusses the potential impacts of the emission of
such gases at greater length, and also summaries analysis quantifying
some of these impacts (e.g., average temperatures) for each of the
considered regulatory alternatives.
---------------------------------------------------------------------------
\2437\ Impacts and U.S. emissions of CO2 are
discussed at greater length in EPA's 2018 ``Inventory of U.S.
Greenhouse Gas Emissions and Sinks,'' EPA 430-R-18-003 (Apr. 12,
2018), available at https://www.epa.gov/sites/production/files/2018-01/documents/2018_complete_report.pdf.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 25054]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.683
[[Page 25055]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.684
BILLING CODE 4910-59-C
[[Page 25056]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.685
(3) Other Pollutant Damages--Criteria and Toxic Pollutants
The CAFE model uses the entire on-road fleet, calculated VMT
(discussed above), and emissions factors (which are an input to the
CAFE model, specified by model year and age) to calculate tailpipe
emissions associated with a given alternative. Just as it does for
additional CO2 emissions associated with upstream emissions
from fuel production, the model captures criteria pollutants that occur
during other parts of the fuel life cycle. While this is typically a
function of the number of gallons of gasoline consumed (and miles
driven, for tailpipe criteria pollutant emissions), the CAFE model also
estimates electricity consumption and the associated upstream emissions
(resource extraction and generation, based on U.S. grid mix).
(a) Emissions Increases
[[Page 25057]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.686
BILLING CODE 4910-59-P
[[Page 25058]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.687
[[Page 25059]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.688
[[Page 25060]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.689
[[Page 25061]]
BILLING CODE 4910-69-C
(b) Air Quality Impacts of Other Pollutants
Although this final rule focuses on standards for fuel economy and
CO2, it will also have an impact on criteria and air toxic
pollutant emissions, although as discussed above, it is expected that
incremental impacts on criteria and air toxic pollutant emissions would
be too small to observe under any of the regulatory alternatives under
consideration. Nevertheless, the following sections detail the criteria
pollutant and air toxic inventory impacts of this final rule; the
methodology used to calculate those impacts; the health and
environmental effects associated with the criteria and toxic air
pollutants that are being impacted by this final rule; the potential
impact of this final rule on concentrations of criteria and air toxic
pollutants in the ambient air; and other unquantified health and
environmental effects.
Today's analysis reflects the combined result of several underlying
impacts, all discussed above. CAFE and CO2 standards are
estimated to impacts new vehicle prices, fuel economy levels, and
CO2 emission rates. These changes are estimated to impact
the size and composition of the new vehicle fleet and to impact the
retention of older vehicles (i.e., vehicle survival and scrappage) that
tend to have higher criteria and toxic pollutant emission rates. Along
with the rebound effect, these lead to changes in the overall amount of
highway travel and the distribution among different vehicles in the on-
road fleet. Vehicular emissions depend on the overall amount of highway
travel and the distribution of that travel among different vehicles,
and emissions from ``upstream'' processes (e.g., petroleum refining,
electricity generation) depend on the total consumption of different
types of fuels for light-duty vehicles.
(i) Impacts
As discussed above, in addition to affecting fuel consumption and
emissions of carbon dioxide or its equivalent, this rule would also
influence other pollutants, i.e., ``criteria'' air pollutants and their
precursors, and air toxics. The final rule would affect emissions of
carbon monoxide (CO), fine particulate matter (PM2.5),
sulfur dioxide (SOX), volatile organic compounds (VOC),
nitrogen oxides (NOX), benzene, 1,3-butadiene, formaldehyde,
acetaldehyde, and acrolein. Consistent with the evaluation conducted
for the Environmental Impact Statement accompanying today's rule, the
agency analyzed criteria air pollutant impacts in 2025, 2035, and 2050
(as a representation of future program impacts). Estimates of these
other emission impacts are shown by pollutant in Table VII-124 through
Table VII-127 and are broken down by the two drivers of these changes:
a) ``downstream'' emission changes, reflecting the estimated effects of
VMT rebound (discussed in Section VIII of the FRIA), changes in vehicle
fleet age, changes in vehicle emission standards, and changes in fuel
consumption; and b) ``upstream'' emission increases because of
increased refining and distribution of motor vehicle gasoline relative
to the baseline. Program impacts on criteria and toxics emissions are
discussed below.
As discussed above, these changes in total annual criteria
pollutant emissions attributable to passenger cars and light trucks
reflect trends in both vehicular and upstream emissions, and these
trends can either be mutually reinforcing or mutually offsetting,
depending on the pollutant and year. Above, Figure VII-9 places these
total changes in emissions in context, showing that, except for
SO2, these changes in criteria pollutant emissions are very
small. For SO2 emissions, changes are also very small
through the late 2030s, after which reduced upstream emissions cause
net emission reductions to exceed 10 percent of 2017 emissions by 2050.
BILLING CODE 4910-59-P
[[Page 25062]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.690
[[Page 25063]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.691
[[Page 25064]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.692
[[Page 25065]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.693
BILLING CODE 4910-59-C
As shown in Table VII-128 through Table VII-131, it is estimated
that the new final program would result in small changes for air toxic
emissions
[[Page 25066]]
compared to total U.S. inventories across all sectors. These changes
also reflect the changing balance between vehicular and upstream
emissions.
BILLING CODE 4910-59-P
[[Page 25067]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.694
[[Page 25068]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.695
[[Page 25069]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.696
[[Page 25070]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.697
[[Page 25071]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.698
BILLING CODE 4910-59-C
Changes in emissions of other pollutants due to these rules will
impact air quality. Information on current air quality and the results
of our air quality
[[Page 25072]]
modeling of the projected impacts of these rules are summarized in the
following section.
(ii) Other Unquantified Health and Environmental Effects
In the proposal, the agencies sought comment on whether there are
any other health and environmental impacts associated with advancements
in technologies that should be considered. For example, the use of
technologies and other strategies to reduce fuel consumption and/or
CO2 emissions could have effects on a vehicle's life-cycle
impacts (e.g., materials usage, manufacturing, end of life disposal),
beyond the issues regarding fuel production and distribution (upstream)
CO2 emissions discussed in Section VI.D.2. The agencies
sought comment on any studies or research in this area that should be
considered in the future to assess a fuller range of health and
environmental impacts from the light-duty vehicle fleet shifting to
different technologies and/or materials. At this point, the agencies
find there is insufficient information about the lifecycle impacts of
the myriad of available technologies, materials, and cradle-to-grave
pathways to conduct the type of detailed assessments that would be
needed in a regulatory context, especially considering the
characterization of specific vehicles in the analysis fleet and the
characterization of specific technology options.
(c) Health Effects of Other Pollutants
This section presents results of the analysis showing health
effects associated with exposure to some of the criteria and air toxic
pollutants impacted by the new final vehicle standards. As discussed
above, the health impacts presented here are subject to a number of
uncertainties, some of which arise from the less complex benefits-per-
ton approach relied on in this analysis, and some of which arise from
the uncertainty surrounding many of the assumptions and other inputs
relied on in the agencies' analysis. As the agencies conclude above,
although it may seem that the agencies' estimates of increases in
premature mortality resulting from the final standards are more likely
to be too high than too low, it is extremely difficult to anticipate
whether this is actually the case.
BILLING CODE 4910-59-P
[[Page 25073]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.699
[[Page 25074]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.700
[[Page 25075]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.701
[[Page 25076]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.702
[[Page 25077]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.703
[[Page 25078]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.704
[[Page 25079]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.705
[[Page 25080]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.706
[[Page 25081]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.707
[[Page 25082]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.708
[[Page 25083]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.709
[[Page 25084]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.710
[[Page 25085]]
BILLING CODE 4910-59-C
B. Impacts on Calendar Year Basis
As with the NPRM, the agencies' analysis primarily examines
regulatory impacts on a model year basis, accounting for the physical
impacts and monetized costs and benefits attributable to vehicles
produced prior to model year 2030 and occurring throughout these
vehicles' useful lives. EDF submitted comments arguing that the
agencies should examine impacts on a calendar year basis, as discussed
above in VI.A.\2438\ CAFE analysis has historically examined effects of
the standards on a model year basis, because CAFE (and CO2)
standards are enforced on a model year basis, and manufacturers'
responses to these standards (i.e., their costs), which are the direct
effects of the standards, occur on a model year basis. On the other
hand, overall impacts on national energy consumption and the
environment result from the evolution and operation of the overall on-
road fleet, and this motivates consideration of results on a calendar
year basis. As also discussed in VI.A., the agencies have expanded the
presentation of results in today's rulemaking and FRIA by presenting
some impacts for each of CYs 2017-2050 and, to enable doing so, have
extended the analysis to cover model years through 2050.
---------------------------------------------------------------------------
\2438\ EDF, NHTSA-2018-0067-12108, Appendix A at 9, et seq., and
Appendix B at 11-14.
---------------------------------------------------------------------------
For this analysis, the CAFE model reports impacts for each model
year through 2050, and, to capture the entire useful lives of these
vehicles, for each of calendar years 2017-2089.\2439\ One way to
illustrate the model's outputs is to consider three cohorts of model
years: MYs 1978-2017 (MYs to which the analysis applies no additional
fuel-saving technology), MYs 2018-2029 (MYs included in both the ``MY
basis'' and ``CY basis'' approaches), and MYs 2030-2050 (MYs included
only the ``CY basis'' approach). On a calendar year basis, impacts of
the final standards on annual CO2 emissions (impacts on fuel
consumption would follow essentially the same trends) may be attributed
to these cohorts as follows:
---------------------------------------------------------------------------
\2439\ As for the NPRM, DOT has made the model and all inputs
and outputs for today's analysis available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system. The model documentation available at the same location
explains, among other things, the structure and contents of each
type of input and output file. The
``annual_societal_effects_report.csv ``and
``annual_societal_costs_report.csv'' reports contain, respectively,
estimates of physical impacts and monetized costs and benefits
attributable to each model year in each calendar years. Other output
file types contain corresponding aggregations either all calendar
years, or across all model years.
[GRAPHIC] [TIFF OMITTED] TR30AP20.711
BILLING CODE 4910-59-P
Here, the large lower area of the chart shows annual CO2
emissions estimated to occur under the baseline/augural CAFE standards,
through calendar year
[[Page 25086]]
2089, which is the last year any MY 2050 vehicles are estimated still
to be on the road. The steady declines through 2050 reflect turnover to
more efficient vehicles produced under either regulatory alternative,
and the steep decline after 2050 reflects vehicles included in the
analysis being removed from service. Of the increased annual emissions
under the final standards, the black area shows the portion
attributable to vehicles produced during MYs 2018-2029, and the topmost
area shows the portion attributable to vehicles produced during MYs
2030-2050. The final standards are estimated to reduce emissions from
vehicles produced during MYs 1978-2017 by accelerating scrappage of
these vehicles, but these changes are too small to be visible in this
chart.
The bulk of the reporting of results here and in the FRIA examines
impacts over the useful lives of vehicles produced prior to MY 2030. In
terms of the above chart, this means excluding the topmost area,
producing the following:
[GRAPHIC] [TIFF OMITTED] TR30AP20.712
On the other hand, calendar year accounting, as considered for this
analysis, includes all model years included in the analysis (i.e.,
through MY 2050), and examines impacts in all calendar years for which
a full on-road fleet is simulated. In terms of the first of the above
charts, this means ``cutting off'' results at calendar year 2050:
[[Page 25087]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.713
Here, the horizontal axis extends through 2089 to make clear that
this calendar year accounting involves excluding emissions impacts over
most of the useful lives of the latest model years included in the
analysis. On a scale covering just those calendar years included in the
calendar year analysis, the same chart appears as follows:
[[Page 25088]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.714
Viewed on the same calendar year basis, technology costs appear as
follows, with differences between costs under the baseline/augural
standards and under the final standards shown as amounts by which the
former exceed the latter (e.g., in 2025, the final standards are
estimated to avoid about $19 billion in technology costs that would
have been incurred under the baseline/augural standards):
[[Page 25089]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.715
Present value analysis considered involves discounting all
estimated future costs and benefits to 2019. At a 7 percent discount
rate, the undiscounted technology costs shown above correspond to
discounted costs shown in the following chart:
[[Page 25090]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.716
Without discounting, therefore, the final standards avoid $457
billion in technology costs through 2050, each additional year of
analysis after 2036 adding about $14 billion to that total. At a 7
percent discount rate, the final standards still avoid $183 billion in
technology costs, while incremental amounts attributable to each
additional year of analysis are (of course) lower than the undiscounted
amounts--declining to about $5 billion during 2035-2036 and, by 2045,
about $2 billion.
For each of the regulatory alternatives considered here, the
following tables summarize results of such aggregations for each
reported category of monetized costs and benefits. The first three
tables focus on the final CAFE standards, presenting total amounts
through 2050 at 3 percent and 7 percent discount rates. The second
three tables show results for corresponding CO2 standards.
[[Page 25091]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.717
[[Page 25092]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.718
[[Page 25093]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.719
[[Page 25094]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.720
[[Page 25095]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.721
[[Page 25096]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.722
[[Page 25097]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.723
[[Page 25098]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.724
As illustrated above, the model year analysis answers the question
``what impacts do we think might eventually be attributable to vehicles
produced before 2030?,'' and the calendar year analysis answers the
question ``what do
[[Page 25099]]
we think might happen between now and 2050?'' Again, CAFE and
CO2 standards are enforced on a model year basis, and the
agencies accordingly simulate manufacturers' responses to these
standards--and estimate manufacturers' corresponding costs--on a model
year basis. This motivates consideration of results on a model year
basis. On the other hand, overall impacts on national energy
consumption and the environment result from the evolution and operation
of the overall on-road fleet, and this motivates consideration of
results on a calendar year basis.
These different perspectives produce results that, without careful
consideration, appear to conflict. The model year perspective as
applied through MY 2029 shows less stringent standards producing
environmental benefits (compared to the augural standards) attributable
to the aggregate of vehicles produced prior to MY 2030. While the
calendar year perspective also shows similar trends prior to (calendar
year) 2035, with the estimated transition to electric vehicles
accelerating over time, the calendar year perspective shows less
stringent standards mostly increasing emissions (SO2 being
an exception) relative to the augural standards.
Still, some important aspects of estimated social benefits and
costs are common to both the model year and calendar year perspectives.
For each of the regulatory action alternatives, the magnitude of total
incremental benefits (relative to the baseline augural standards) is
similar to the magnitude of total incremental costs. This stands in
marked contrast to the agencies' 2012 rulemaking announcing the augural
standards, and finding of estimated benefits that were 3-4 times larger
than costs.\2440\ Under today's analysis, estimated benefits and costs
are instead of similar magnitude, with estimated net benefits, by
comparison, small enough to be even directionally uncertain, such that
an alternative estimated to produce small positive net benefits under
one perspective and applying a 7 percent discount rate might be
estimated to produce small negative net benefits under the other
perspective and/or applying a 3 percent discount rate. While the
agencies obviously must consider benefits, costs, and net benefits, our
decisions are based on wider considerations. Consistent with the
agencies' 2012 final rule, today's final rule finds--from both the
model year and calendar year perspectives--that forgone fuel savings
(forgone because today's final rule involves relaxing rather than
increasing the stringency of CAFE and CO2 standards) account
for the bulk of estimated forgone social benefits. These are private
benefits, which raises a significant question of whether there is a
meaningful market failure that needs to be addressed by more stringent
regulation.
---------------------------------------------------------------------------
\2440\ 77 FR at 62629 (Oct. 15, 2012).
---------------------------------------------------------------------------
Section VI contains an extensive discussion and analysis of the
existence and nature of various market failures related to fuel economy
standards. These potential market failures include the well-established
externalities of environmentally harmful emissions, congestion, and
safety; as well the debatable and hypothetical market failures related
to the ``energy paradox.'' The energy paradox refers to an observation
that some consumers appear voluntarily to forgo investments in energy
conservation even when those initial investments appear to repay
themselves--in the form of savings in energy costs--over the relatively
near term. Section VI.D.1 discussion casts doubt on the theoretical
underpinnings that the energy paradox represents a market failure,
discusses recent research that suggests the extent consumers are
undervaluing fuel economy has been overstated, and suggests the
analysis supporting claims of an energy paradox overlooks the
opportunity costs of other vehicle attributes that consumers and
manufacturers trade off with fuel efficiency technology. As stated in
Section VI, while the agencies have reservations about the extent to
which a market failure capable of driving very large net private
financial harm to consumers exists, the agencies do not take a position
on the existence of an energy paradox in this rulemaking.
The primary analysis shows that the CAFE final rule would generate
$12.9 billion in total social net benefits using a 7 percent discount
rate, but without the large net private loss of $26.4 billion, the net
social benefits would equal the external net benefits, or $39.3
billion. Therefore, given significant questions about whether
government action to impose restrictions in private markets could
improve net social benefits absent a market failure, if no market
failure exists to motivate the $26.4 billion in private losses to
consumers, the net benefits of these final standards would be $39.3
billion. The CY analysis produces similar results, though the estimated
private losses are exacerbated relative to the external gains. The CY
analysis shows the CAFE final rule would generate -$6 billion in total
net social benefits using a 7 percent discount rate, but without the
large net private loss of $65 billion, the net social benefits would
equal the external net benefits of $59 billion.
One commenter suggested that the agencies should elect to use CY
accounting in the primary analysis because the MY accounting approach
resulted in an inconsistent accounting of costs and benefits owing to
the scrappage effect. While the CY accounting approach does reduce non-
rebound safety benefits from $9 billion to $8 billion (combined fatal
and non-fatal benefits), the total external net benefits of the rule
actually increase by $20 billion using the CY approach. This result is
driven primarily by a significant increase in congestion cost savings
from less rebound driving, from $44 billion to $69 billion. Any changes
in the net benefits in the opposite direction using CY accounting
result from increased net private costs to consumers own financial
wellbeing from allowing more consumer choice. These increased net
private costs occur because the CY analysis captures model years far
into the future, which are more uncertain and not subject to today's
CAFE final rule. Therefore, the agencies see little evidence that the
inconsistency suggested by the commenter is important, or that the
primary conclusions of the analysis are meaningfully influenced by it.
Sensitivity Analysis
As discussed at the beginning of this section, results presented
today reflect the agencies' best judgments regarding many different
factors. Based on analyses in past rulemakings, the agencies recognize
that some analytical inputs are especially uncertain, some are likely
to exert considerable influence over specific types of estimated
impacts, and some are likely to do so for the bulk of the analysis. To
explore the sensitivity of estimated impacts to changes in model
inputs, analysis was conducted using alternative values for a range of
different inputs. Results of this sensitivity analysis are summarized
in the Final Regulatory Impact Analysis (FRIA) accompanying today's
rulemaking, and detailed model inputs and outputs are available on
NHTSA's website.\2441\ The following table lists the cases included in
the sensitivity analysis.
---------------------------------------------------------------------------
\2441\ The CAFE model and all inputs and outputs supporting
today's rulemaking are available at https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 25100]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.725
[[Page 25101]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.726
[[Page 25102]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.727
BILLING CODE 4910-59-C
VIII. How do the final standards fulfill the agencies' statutory
obligations?
A. How Does the technical assessment support the final CO2
standards as compared to the alternatives that EPA has considered?
1. Introduction
Title II of the Clean Air Act provides for comprehensive regulation
of mobile sources, authorizing EPA to regulate emissions of air
pollutants from all mobile source categories. Under Section 202(a) and
relevant case law, as discussed below, EPA considers such issues as
technology emission reduction effectiveness, its cost (both per
vehicle, per manufacturer, and per consumer), the lead time necessary
to implement the technology, and based on this the feasibility of
potential standards; the impacts of potential standards on emissions
reductions of both GHGs and non-GHGs; the impacts of standards on oil
conservation and energy security; the impacts of standards on fuel
savings by consumers; the impacts of standards on the auto industry;
other energy impacts; as well as other relevant factors such as impacts
on safety.
EPA is afforded considerable discretion under section 202(a) when
assessing issues of technical feasibility and availability of lead time
and in weighing these factors. In light of its consideration of the
relevant factors, EPA has concluded, for the reasons discussed below,
that the previous standards (which increase stringency at a rate of
about 5% per year) are not appropriate, and the best action is to
revise the standards to increase stringency by 1.5% per year. Beginning
in 2009, EPA and NHTSA have worked together jointly to establish fuel
economy and tailpipe CO2 emission standards for light duty
vehicles. The first rulemaking, finalized in 2010, established
standards for the 2012 through 2016 model years. Shortly thereafter, in
2012, the agencies established standards for the 2017 through 2025
model years--but given the limitation in EPCA that only allows for
standards to be set five years at a time, the 2022-2025 model year
standards were only final for EPA's tailpipe CO2 emissions
regulation. This rapid period of rulemaking to establish standards over
a decade in advance may have marked a departure for NHTSA, but it
followed EPA's longstanding
[[Page 25103]]
approach when regulating vehicular criteria pollutant emissions to
provide a significant period of time for the industry to develop
technologies to achieve standards.
While EPA had decades of experience regulating light duty vehicle
emissions, it did not previously have experience regulating tailpipe
CO2 emissions. And regulating CO2 emissions is
quite different from regulating criteria pollutant emissions. With
criteria pollutants, technological emission controls exist primarily in
the form of engine controls and catalytic conversion. Today's emission
controls for criteria pollutants have only a de minimis effect on
performance or functionality of the vehicle.
Controlling tailpipe CO2 emissions for an internal
combustion engine requires controlling the amount of energy used to
propel the vehicle. All else being equal, better performance (in
acceleration or passing speed) requires more energy. Similarly,
vehicles with more storage capacity tend to be larger, and moving an
object with larger mass requires more energy than objects with smaller
mass. Vehicles with greater towing performance likewise require more
energy. Maintaining utility and performance requires sophisticated and
expensive technological solutions, such as reducing mass through
advanced materials, changing engine combustion cycles, increasing
compression ratios, or turbo-charging the engine. Consumers often can
feel the difference in vehicle performance as a result of these
controls, and as will be discussed herein.
As discussed when issuing the 2012 Final Rule, the economic and
market assumptions underlying the standards the agencies finalized were
crucial, and long-term projections are inherently uncertain. Upon
review of those assumptions, such as the price of gas and the sales mix
of pick-up trucks and sport-utility vehicles as compared to passenger
cars, the agencies have now concluded that many of these assumptions
have not proven to be accurate and therefore have been updated. Given
the uncertainty about the 2012 assumptions at the time of that
rulemaking, the agencies incorporated a mid-term evaluation process for
EPA's 2022-2025 model year standards that would be ``collaborative,
robust and transparent,'' and ``based on information available at the
time of the mid-term evaluation and an updated assessment of all the
factors considered in setting the standards and the impacts of those
factors on the manufacturers' ability to comply.'' \2442\
---------------------------------------------------------------------------
\2442\ 77 FR at 62633.
---------------------------------------------------------------------------
While that process was expected to take place throughout 2017, and
a final determination issued in the Spring of 2018, this process was
expedited. On July 27, 2016, the agencies published a Federal Register
notice making the public aware of the availability of a draft Technical
Assessment Report, with comments due at the end of September 2016. On
December 6, 2016, EPA published a notice in the Federal Register making
the public aware of its proposed Final Determination and extensive
Technical Support Document to keep the standards set in 2012 in place
through the 2025 model year without change. The public was given until
December 30, 2016 to comment on the proposed determination. Less than
two weeks later, on January 12, 2017, EPA finalized its determination.
Industry commenters stated that the 2017 Final Determination ``is
the product of egregious procedural and substantive defects and EPA
should withdraw it,'' that EPA had ``fail[ed] to provide an adequate
period for meaningful notice and comment,'' that EPA had
``acknowledg[ed] that the Proposed Determination adjusted a number of
EPA assumptions in response to commenters who pointed out errors at
earlier stages'' while stating that ``there was no need for more time
because [it] did not include much new material,'' and that ``EPA [had]
underestimated the burden [of the standards],'' ``EPA [made] cursory
assertions that downplayed the impact of its mandate on auto sales and
employment,'' and ``EPA refused to consider many of the [industry's]
technical concerns even when supported by an outside consultant,
asserted [industry] provided insufficient data, and then refused
further meetings for clarification.'' \2443\
---------------------------------------------------------------------------
\2443\ Alliance letter to Administrator Pruitt, Feb. 21, 2017,
available at https://autoalliance.org/wp-content/uploads/2017/02/Letter-to-EPA-Admin.-Pruitt-Feb.-21-2016-Signed.pdf.
---------------------------------------------------------------------------
In light of commenters' concerns about EPA's 2017 final
determination, in March 2017, EPA announced its intent to reconsider
the final determination in order to allow additional opportunity to
hear from the public, and additional consultation and coordination with
NHTSA in support of a national harmonized program. In August 2017, EPA
published a notice in the Federal Register requesting comment on its
reconsideration of the initial determination, and held a public hearing
on the matter in September 2017. Then, in April 2018, EPA issued a
revised final determination finding that the 2022-2025 model year GHG
standards set in 2012 were not appropriate and a rulemaking should be
initiated to revise the standards, as appropriate.
In this proceeding, in order to determine what standards are
appropriate, EPA and NHTSA sought comment on a wide range of potential
standards--ranging from holding the 2020 standards flat through the
2026 model year to retaining the standards finalized in 2012. Similar
to the 2012 rulemaking, EPA considered a number of different
alternatives--ranging from the standards finalized in 2012, to holding
the 2020 MY standards flat through MY 2026. As in 2012, the manner in
which different factors are weighed can yield very different result--
more stringent standards would improve CO2 emissions, reduce
energy consumption, and save consumers fuel. Less stringent standards
would reduce technology costs for manufacturers and save consumers in
upfront purchase prices, enabling the fleet to turnover more quickly.
While weighing these factors, EPA has considered compliance results
that have been observed throughout the fleet. While the agencies have
seen extraordinary reductions in tailpipe CO2 emissions
since EPA has begun regulation in this area, manufacturers are
increasingly falling short of meeting their performance targets, and
are increasingly using acquired or earned credits to comply with
requirements. For the 2016 model year, the overall fleet failed, for
the first time in regulation history, to meet emission targets--
achieving 272 grams per mile, when the standard was 263 grams per
mile.\2444\ The 2016 model year saw only five major manufacturers
perform at or better than their CO2 footprint standards--
Honda, Hyundai, Mazda, Nissan, and Subaru. For the 2017 model year,
only three major manufacturers--BMW, Honda, and Subaru--performed
better than their CO2 standards, and the total fleet
underperformed compared to the standards--achieving 263 grams per mile,
when the fleetwide standard was 258 grams per mile.\2445\ The emissions
averaging, credit banking and trading system was established to allow
[[Page 25104]]
manufacturers greater flexibility and lead time to address technical
feasibility and cost without sacrificing effectiveness of the
standards, but widespread reliance upon credits across the industry may
raise concerns about compliance in future years, particularly since the
more significant increases in stringency in the 2012 rulemaking have
yet to be effective. Taken together, the agencies now believe this
information supports the conclusion that the lead time EPA estimated
would be sufficient to achieve compliance with the previous standards
for MYs 2021-26, was not sufficient.
---------------------------------------------------------------------------
\2444\ EPA Greenhouse Gas Emission Standards for Light-Duty
Vehicles: Manufacturer Performance Report for the 2016 Model Year.
EPA-420-R-18-002 (January 2018).
\2445\ 2018 EPA Automotive Trends Report: Greenhouse Gas
Emissions, Fuel Economy, and Technology since 1975, available at:
https://www.epa.gov/automotive-trends/download-automotive-trends-report.
---------------------------------------------------------------------------
In this action, EPA is reducing the rate of stringency increases
from those adopted in the 2012 rulemaking in part to ensure that the
standards remain reasonable and appropriate. As in 2012, EPA is
deciding against selecting alternatives that are more stringent or less
stringent than appropriate. The final rule analysis projects that the
1.5 percent alternative would result in less significant shortfalls
compared to more stringent alternatives, which will ease compliance
burdens while nonetheless pushing the market beyond what it would
demand in the absence of standards or what would be achieved with less
stringent standards. The standards finalized today will result in
continuing improvements compared to the 2020 model year, and are best
viewed in the context of the larger rulemaking, as shown in the chart
below:
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TR30AP20.728
BILLING CODE 4910-59-C
2. Basis for the CO2 Standards Under Section 202(a) of the
Clean Air Act
Title II of the Clean Air Act (CAA) provides for comprehensive
regulation of mobile sources, authorizing EPA to regulate emissions of
air pollutants from all mobile source categories. This rule implements
a specific provision from Title II, section 202(a).\2446\ Section
202(a)(1) states that ``[t]he Administrator shall by regulation
prescribe (and from time to time revise) . . . standards applicable to
the emission of any air pollutant from any class or classes of new
motor vehicles or new motor vehicle engines, which in his judgment
cause, or contribute to, air pollution which may reasonably be
anticipated to endanger public health or welfare.'' If EPA makes the
appropriate endangerment and cause or contribute findings, then section
202(a) directs EPA to issue standards applicable to emissions of those
pollutants.
---------------------------------------------------------------------------
\2446\ 42 U.S.C. 7521(a).
---------------------------------------------------------------------------
[[Page 25105]]
Any standards under CAA section 202(a)(1) ``shall be applicable to
such vehicles and engines for their useful life.'' Emission standards
set by the EPA under section 202(a)(1) are technology-based, as the
levels chosen must be premised on a finding of technological
feasibility. Thus, standards promulgated under section 202(a) are to
take effect only after ``such period as the Administrator finds
necessary to permit the development and application of the requisite
technology, giving appropriate consideration to the cost of compliance
within such period.'' \2447\ EPA must consider costs to those entities
which are directly subject to the standards.\2448\ Thus, ``the
[s]ection 202(a)(2) reference to compliance costs encompasses only the
cost to the motor-vehicle industry to come into compliance with the new
emission standards.'' \2449\ EPA is afforded considerable discretion
under section 202(a) when assessing issues of technical feasibility and
availability of lead time to implement new technology. Such
determinations are ``subject to the restraints of reasonableness,''
which ``does not open the door to `crystal ball' inquiry.'' \2450\ In
developing such technology-based standards, EPA has the discretion to
consider different standards for appropriate groupings of vehicles
(``class or classes of new motor vehicles''), or a single standard for
a larger grouping of motor vehicles.\2451\
---------------------------------------------------------------------------
\2447\ CAA section 202 (a)(2); see also NRDC v. EPA, 655 F.2d
318, 322 (DC Cir. 1981).
\2448\ Motor & Equipment Mfrs. Ass'n Inc. v. EPA, 627 F. 2d
1095, 1118 (DC Cir. 1979).
\2449\ Coalition for Responsible Regulation, 684 F.3d at 128;
see also id. at 126-27 (rejecting arguments that EPA was required to
consider or should have considered costs to other entities, such as
stationary sources, which are not directly subject to the emission
standards).
\2450\ NRDC, 655 F.2d at 328 (quoting International Harvester
Co. v. Ruckelshaus, 478 F.2d 615, 629 (DC Cir. 1973)).
\2451\ NRDC, 655 F.2d at 338.
---------------------------------------------------------------------------
Although standards under CAA section 202(a)(1) are technology-
based, they are not based exclusively on technological capability. EPA
has the discretion, and in some instances has been specifically
directed by Congress, to consider and weigh various factors along with
technological feasibility, such as the cost of compliance, \2452\ lead
time necessary for compliance, \2453\ safety,\2454\ other impacts on
consumers,\2455\ and energy impacts associated with use of the
technology.\2456\
---------------------------------------------------------------------------
\2452\ See section 202(a)(2).
\2453\ Id.
\2454\ See NRDC, 655 F.2d at 336 n. 31.
\2455\ Since its earliest Title II regulations, EPA has
considered the safety of pollution control technologies. See 45 FR
14496, 14503 (March 5, 1980). (``EPA would not require a particulate
control technology that was known to involve serious safety
problems. If during the development of the trap-oxidizer safety
problems are discovered, EPA would reconsider the control
requirements implemented by this rulemaking.'').
\2456\ See George E. Warren Corp. v. EPA, 159 F.3d 616, 623-624
(DC Cir. 1998) (ordinarily permissible for EPA to consider factors
not specifically enumerated in the CAA).
---------------------------------------------------------------------------
Unlike standards set under provisions such as section 202(a)(3) and
section 213(a)(3), EPA is not required to set technology-forcing
standards when such standards would not be appropriate. EPA has
interpreted a similar statutory provision, CAA section 231,\2457\ as
follows:
\2457\ Section 231(a)(2)(A) of the CAA provides: ``The
Administrator shall, from time to time, issue proposed emission
standards applicable to the emission of any air pollutant from any
class or classes of aircraft engines which in his judgment causes,
or contributes to, air pollution which may reasonably be anticipated
to endanger public health or welfare.'' Section 231(a)(3) provides
in part: ``Within 90 days after the issuance of such proposed
regulations, he shall issue such regulations with such modifications
as he deems appropriate. Such regulations may be revised from time
to time.'' Sectiion 231(b) provides: ``Any regulation prescribed
under this section (and any revision thereof) shall take effect
after such period as the Administrator finds necessary (after
consultation with the Secretary of Transportation) to permit the
development and application of the requisite technology, giving
appropriate consideration to the cost of compliance within such
period.''
---------------------------------------------------------------------------
While the statutory language of section 231 is not identical to
other provisions in title II of the CAA that direct EPA to establish
technology-based standards for various types of engines, EPA
interprets its authority under section 231 to be somewhat similar to
those provisions that require us to identify a reasonable balance of
specified emissions reduction, cost, safety, noise, and other
factors. See, e.g., Husqvarna AB v. EPA, 254 F.3d 195 (D.C. Cir.
2001) (upholding EPA's promulgation of technology-based standards
for small non-road engines under section 213(a)(3) of the CAA).
However, EPA is not compelled under section 231 to obtain the
``greatest degree of emission reduction achievable'' as per sections
213 and 202 of the CAA, and so EPA does not interpret the Act as
requiring the agency to give subordinate status to factors such as
cost, safety, and noise in determining what standards are reasonable
for aircraft engines. Rather, EPA has greater flexibility under
section 231 in determining what standard is most reasonable for
aircraft engines, and is not required to achieve a ``technology
forcing'' result.\2458\
---------------------------------------------------------------------------
\2458\ 70 FR 69664, 69676 (Nov. 17, 2005).
This interpretation was upheld as reasonable in NACAA v. EPA.\2459\
CAA section 202(a), as with section 231, does not specify the degree of
weight to apply to each factor, and EPA accordingly interprets its
authority under section 202(a) similarly to its interpretation of
section 231 as set forth above: EPA has discretion in choosing an
appropriate balance among the statutory factors.\2460\
---------------------------------------------------------------------------
\2459\ 489 F.3d 1221, 1230 (DC Cir. 2007).
\2460\ See Sierra Club v. EPA, 325 F.3d 374, 378 (D.C. Cir.
2003) (even where a provision is technology-forcing, the provision
``does not resolve how the Administrator should weigh all [the
statutory] factors in the process of finding the 'greatest emission
reduction achievable'''); see also Husqvarna AB v. EPA, 254 F. 3d
195, 200 (D.C. Cir. 2001) (great discretion to balance statutory
factors in considering level of technology-based standard, and
statutory requirement ``[to give] appropriate consideration to the
cost of applying . . . technology'' does not mandate a specific
method of cost analysis); Hercules Inc. v. EPA, 598 F. 2d 91, 106-07
(D.C. Cir. 1978) (``In reviewing a numerical standard, we must ask
whether the agency's numbers are within a `zone of reasonableness,'
not whether its numbers are precisely right''); Permian Basin Area
Rate Cases, 390 U.S. 747, 797 (1968) (same); Federal Power
Commission v. Conway Corp., 426 U.S. 271, 278 (1976) (same); Exxon
Mobil Gas Marketing Co. v. FERC, 297 F. 3d 1071, 1084 (D.C. Cir.
2002) (same).
---------------------------------------------------------------------------
As noted above, EPA has found that the elevated concentrations of
greenhouse gases in the atmosphere may reasonably be anticipated to
endanger public health and welfare.\2461\ EPA defined the ``air
pollution'' referred to in CAA section 202(a) to be the combined mix of
six long-lived and directly emitted GHGs: carbon dioxide
(CO2), methane (CH4), nitrous oxide
(N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs),
and sulfur hexafluoride (SF6). The EPA further found under
CAA section 202(a) that emissions of the single air pollutant defined
as the aggregate group of these same six greenhouse gases from new
motor vehicles and new motor vehicle engines contribute to air
pollution. As a result of these findings, section 202(a) requires EPA
to issue standards applicable to emissions of that air pollutant. New
motor vehicles and engines emit CO2, CH4,
N2O, and HFC. EPA has established standards and other
provisions that control motor vehicle emissions of CO2,
HFCs, N2O, and CH4. EPA has not set any standards
for PFCs or SF6 as they are not emitted by motor vehicles.
---------------------------------------------------------------------------
\2461\ 74 FR 66496 (Dec. 15, 2009).
---------------------------------------------------------------------------
3. EPA's Conclusion That the Final CO2 Standards Are
Appropriate and Reasonable
In this section, EPA discusses the factors, data and analysis the
Administrator has considered in the selection of the EPA's revised
CO2 emission standards for MYs 2021 and later and the
comments received on EPA's consideration of these factors (see further
discussion below on EPA's summary and analysis of comments).
As discussed in Section VIII.A.1 above, the primary purpose of
Title II of the Clean Air Act is the protection of public health and
welfare, and GHG
[[Page 25106]]
emissions from light-duty vehicles have been found by EPA to endanger
public health and welfare.\2462\ The goal of the light-duty vehicle GHG
standards is to reduce these emissions which cause or contribute to air
pollution which may reasonably be anticipated to endanger public health
or welfare, while taking into account other factors as discussed above.
---------------------------------------------------------------------------
\2462\ Id.
---------------------------------------------------------------------------
CAA section 202(a)(2) states when setting emission standards for
new motor vehicles, the standards ``shall take effect after such period
as the Administrator finds necessary to permit the development and
application of the requisite technology, giving appropriate
consideration to the cost of compliance within such period.'' 42 U.S.C.
7521(a)(2). That is, when establishing emission standards, the
Administrator must consider both the lead time necessary for the
development of technology that can be used to achieve the emission
standards and the resulting costs of compliance on those entities that
are directly subject to the standards. In previous rulemakings,
including the rulemaking that established the current standards, EPA
considered lead time-related elements, including comparative per-
vehicle cost increases by manufacturer for both cars and trucks,
comparative penetration rates of advanced technologies by manufacturers
for both cars and trucks, and lead time concerns about increasing
technology penetration rates for these advanced technologies beyond
current levels. EPA also considered comparative industry-wide costs and
differences between alternatives, framed in terms of total costs and
percentage differences between alternatives. These elements are
discussed in detail throughout the analysis. As mentioned previously,
however, the performance of the fleet in recent years indicates that
the lead time deemed as adequate in the 2012 rulemaking was not
sufficient.
EPA is not limited to consideration of the factors specified in CAA
section 202(a)(2) when establishing standards for light-duty vehicles.
In addition to feasibility and cost of compliance, EPA may (and
historically has) considered such factors as safety, energy use and
security, degree of reduction of both GHG and non-GHG pollutants,
technology cost-effectiveness, and costs and other impacts on
consumers.
EPA also considers relevant case law. Critical to this series of
joint rulemakings with NHTSA, the Court in Massachusetts v. EPA,\2463\
recognized EPA's argument that ``it cannot regulate carbon dioxide
emissions from motor vehicles'' without ``tighten[ing] mileage
standards . . . .''--a task assigned to DOT. The Court found that
``[t]he two obligations may overlap, but there is no reason to think
the two agencies cannot both administer their obligations and yet avoid
inconsistency.'' \2464\ Accordingly, the agencies have worked closely
together in setting standards, and many of the factors that NHTSA
considers to set maximum feasible standards overlap with factors that
EPA considers under the Clean Air Act. Just as EPA considers energy use
and security, NHTSA considers these factors when evaluating the need of
the nation to conserve energy, as required by EPCA. Just as EPA
considers technological feasibility, the cost of compliance,
technological cost-effectiveness and cost and other impacts upon
consumers, NHTSA considers these factors when weighing the
technological feasibility and economic practicability of potential
standards. EPA and NHTSA both consider implications of the rulemaking
on CO2 emissions as well as criteria pollutant emissions.
And, NHTSA's role as a safety regulator inherently leads to the
consideration of safety implications when establishing standards. The
balancing of competing factors by both EPA and NHTSA are consistent
with each agency's statutory authority and recognize the overlapping
obligations the Supreme Court pointed to in directing collaboration.
---------------------------------------------------------------------------
\2463\ 549 U.S. 497, 531 (2007).
\2464\ Id. at 532.
---------------------------------------------------------------------------
As discussed in prior rulemakings setting GHG standards,\2465\ EPA
may establish technology-forcing standards under section 202(a), but it
must provide a rationale for concluding that the industry can develop
the needed technology in the available time. However, EPA is not
required to set technology-forcing standards under section 202(a).
Rather, because section 202(a), unlike the text of section 202(a)(3)
and section 213(a)(3),\2466\ does not specify that standards shall
obtain ``the greatest degree of emission reduction achievable,'' EPA
retains considerable discretion under section 202(a) in deciding how to
weigh the various factors, consistent with the language and purpose of
the Clean Air Act, to determine what standards are appropriate.
---------------------------------------------------------------------------
\2465\ See, e.g., 77 FR 62624, 62673 (Oct. 15, 2012), EPA and
NHTSA final rule for 2017 and later model year light-duty GHG
emissions and CAFE standards.
\2466\ Section 202(a)(3) provides that regulations applicable to
emissions of certain specified pollutants from heavy-duty vehicles
or engines ``shall contain standards which reflect the greatest
degree of emission reduction achievable through the application of
technology which the Administrator determines will be available . .
. giving appropriate consideration to cost, energy, and safety
factors associated with the application of such technology.'' 42
U.S.C. 7521(a)(3). Section 213(a)(3) contains a similar provision
for new nonroad engines and new nonroad vehicles (other than
locomotives or engines used in locomotives). 42 U.S.C. 7547(a)(3).
---------------------------------------------------------------------------
The proposed rule presented an analysis of alternatives, in support
of the Administrator's consideration of a range of alternative
CO2 standards as potential revisions of the existing
standards for model years 2021 and later, from the previous standards
(representing an increase in stringency of approximately 5 percent per
year from MY 2021 through MY 2025) to several less stringent
alternatives. These alternatives ranged from a zero percent increase in
stringency to a stringency increase for passenger cars of 2 percent per
year and for light trucks of 3 percent per year, in addition to the
baseline alternative consisting of the previous standards.\2467\ The
analysis supported the range of alternative standards based on factors
relevant to the EPA's exercise of its section 202(a) authority, such as
emissions reductions of GHGs and other air pollutants, the necessary
technology and associated lead-time, the costs of compliance for
automakers, the impact on consumers with respect to cost and vehicle
choice, and effects on safety. The proposed rule identified the
alternative composed of a zero percent increase in stringency as the
preferred alternative.
---------------------------------------------------------------------------
\2467\ 83 FR 42990, Table I-4 (August 24, 2018).
---------------------------------------------------------------------------
EPA received numerous public comments on the range of stringency
alternatives in the proposed rule and the Administrator's consideration
of various factors in determining appropriate GHG standards under
section 202(a) of the CAA. Below EPA responds to comments on these
issues. EPA notes that many comments concerned the technical foundation
and analysis upon which EPA was basing its regulatory decisions, such
as the modeling of emission control technologies and costs, the safety
analysis, and consumer issues. Comments specific to these analyses are
discussed elsewhere in this preamble. The section below addresses
comments specifically addressing EPA's considerations in finalizing
appropriate CO2 emissions standards under the CAA.
EPA's conclusion, after consideration of the factors described
below, public comments, and other information in the administrative
record for this action is that holding CO2 emissions
standards for MY 2020 flat through MY 2026 is not appropriate or
reasonable. EPA
[[Page 25107]]
concludes steady stringency increases year over year are warranted, but
that the MY 2021-2026 standards first established in 2012 are not
appropriate taking into account lead time and the various factors
described below. Accordingly, the Administrator has concluded that 1.5
percent annual increases in stringency from the MY 2020 standards
through MY 2026 (Alternative 3 of this final rule analysis) \2468\ are
reasonable and appropriate.
---------------------------------------------------------------------------
\2468\ The numbered Alternatives presented in the SAFE proposed
rule (see Table I-4 at 83 FR 42990, August 24, 2018) were in some
cases defined differently than those presented in this final rule
(see Section V). Unless otherwise stated, the Alternatives described
in this section refer to those presented in this final rule.
---------------------------------------------------------------------------
a) Consideration of the Development and Application of Technology To
Reduce CO2 Emissions
When EPA establishes emission standards under CAA section 202, it
considers both what technologies are currently available and what
technologies under development may become available. For today's final
rule, EPA considered the analysis of the potential penetration into the
future vehicle fleet of a wide range of technologies that both reduce
CO2 and improve fuel economy (see FRIA Chapter X). The
majority of these technologies have already been developed, have been
commercialized, and are in-use on vehicles today. These technologies
include, but are not limited to, engine and transmission technologies,
vehicle mass reduction technologies, technologies to reduce aerodynamic
drag, and a range of electrification technologies. The electrification
technologies include 12-volt stop-start systems, 48-volt mild hybrids,
strong hybrid systems, plug-in hybrid electric vehicles, and dedicated
electric vehicles.
This consideration is especially important given current
projections about relatively lower fuel prices than what was projected
in 2012. In that rulemaking, EPA expressed concern that some
alternatives may require too much advanced technologies (including
electrification) in light of uncertain consumer acceptance of added
costs, as well as the technologies themselves.\2469\ There, EPA
concluded that more stringent increases in technology penetration rates
raise serious concerns about the ability and likelihood that
manufacturers can smoothly implement additional technologies to meet
requirements.\2470\
---------------------------------------------------------------------------
\2469\ 77 FR 62879.
\2470\ See 77 FR at 62875, discussion about certain alternatives
may require too much electrification and ``may well be overly
aggressive in the face of uncertain consumer acceptance of both the
added costs and the technologies themselves. EPA continues to
believe these technology penetration rates are inappropriate given
the concerns just voiced.'' At 62877, ``This increase in tech
penetration rates raises serious concerns about the ability and
likelihood manufacturers can smoothly implement. . . .''
---------------------------------------------------------------------------
As shown in Section VII of this preamble and in FRIA Section VII,
the projected penetration of technologies varies across the
Alternatives considered for this final rule. In general, the baseline
alternative consisting of the previous EPA standards as finalized in
2012 was projected to result in the highest penetration of advanced
technologies into the vehicle fleet, in particular mild hybrids at 7.1
percent penetration and strong hybrids at 9 percent penetration by MY
2030. By contrast, the revised final standards adopted today (1.5
percent per year stringency improvement from MY 2021 through MY 2026)
are projected to result in a significantly lower level of mild and
strong hybrids used to meet the standards, at 1.6 percent mild hybrids
and 2.2 percent strong hybrids by MY 2030. Further, the final rule
analysis indicates that the previous CO2 standards would
have led to a projected 5.7 percent penetration of dedicated electric
vehicles (EV), with 0.4 percent penetration of plug-in hybrid electric
vehicles (PHEV); the revised final standards reduce this projected
level to 3.7 percent EV penetration (with 0.2 percent PHEV
penetration), which again is more in line with what the EPA believes is
a more appropriate projected level of market penetration.
The technology penetration rates in the analysis for the final rule
are changed since EPA's prior analysis. These changes in the estimated
penetrations in this rulemaking are due to changes in the model that
are meant to reflect consumer response to the standards, as well as
changes to estimates for technology costs and effectiveness. In the
2017 Final Determination on Model Year 2022-2025 standards, where EPA
found there was available and effective technology to meet the MY 2022-
2025 standards, the technology was available at reasonable cost to the
vehicle manufacturers and consumers, there was adequate lead time, and
the standards were feasible and practicable. EPA also found that the
previous MY 2022-2025 standards could be met largely through advanced
gasoline vehicle technologies, with low levels of electrified
vehicles.\2471\ The levels of electrified vehicle technologies
projected in this final rule to meet the baseline Alternative (the
previous GHG standards) differ slightly from those projected in the
2017 Final Determination. In this final rule, EPA projects a combined
strong and mild hybrid penetration of 16 percent (compared to 20
percent in the 2017 Final Determination), with the share of mild
hybrids somewhat lower (7 percent compared to 18 percent in the 2017
Final Determination) and the share of strong hybrids higher (9 percent
compared to 2 percent in in the 2017 Final Determination). EPA projects
a total level of plug-in vehicles of 6 percent, similar to the 5
percent total projected in the 2017 Final Determination, but with a
slightly different mix of plug-in hybrid electric vehicles (0.4 percent
compared to 2 percent in the 2017 Final Determination) and dedicated
electric vehicles (5.7 percent compared to 3 percent in the 2017 Final
Determination).
---------------------------------------------------------------------------
\2471\ ``Final Determination on the Appropriateness of the Model
Year 2022-2025 Light-Duty Vehicle Greenhouse Gas Emissions Standards
under the Midterm Evaluation,'' EPA-420-R-17-001, January 2017. See
Table ES-1, page 4-5, and Section II (i), (ii), and (iii), pages 28-
24. Hereafter ``2017 Final Determination.''
---------------------------------------------------------------------------
Another aspect of the analysis that EPA considered related to
technology development and application is manufacturers' projected
level of over-compliance under the alternatives considered for the
final rule. Under the least stringent Alternatives (Alternative 1, zero
percent stringency improvement, and Alternative 2, 0.5 percent per year
stringency improvement), manufacturers overall are projected to over-
comply with those levels of stringency. For example, under Alternative
1, manufacturers are projected to achieve a CO2 level of 206
g/mi in MY 2029, 16 g/mi below (more stringent than) the required
target level of 222 g/mi. Similarly, for Alternative 2, manufacturers
are projected to achieve a CO2 level of 205 g/mi in MY 2029,
10 g/mi below the required target level of 215 g/mi. Thus, the industry
is projected to considerably over-comply with the Alternative 1 and 2
standards. Under the final standards, the projected level of over-
compliance is much narrower, only 4 g/mi (198 g/mi by MY 2029 compared
to a 202 g/mi target), and for other alternatives that are more
stringent than the final standards, that gap is similar or even more
narrow as shown in Table VII-7. This is an indication that the
standards in Alternatives 1 and 2 may not represent
[[Page 25108]]
an appropriate level of stringency when compared to the pace at which
manufacturers would be applying technologies. While some level of over-
compliance is expected so that manufacturers retain a reasonable
compliance margin, Alternatives 1 and 2 would, based on the final rule
analysis, result in manufacturers retaining a compliance margin more
than 2-3 times that of the other alternatives. The Administrator has
rejected those lower stringency Alternatives in part for this reason
and believes that the final standards (Alternative 3, 1.5 percent per
year stringency improvement) represent an appropriate margin of
compliance that can be attained given the projected pace of
manufacturers' application of technologies.
EPA received several comments regarding its consideration of the
development and application of GHG reducing technologies. The
California Air Resources Board (CARB) commented that, despite what they
characterize as evidence of widely available technology, EPA has
proposed to promulgate emission standards that are less stringent than
existing standards and that would lead to increased emissions of GHGs.
The New York State Department of Environmental Conservation commented
that the proposal did not ``appropriately value, or consider,
technology advancement and innovation by OEMs and automotive parts
suppliers'' and noted the role of technology innovation in reducing
technology costs. EPA notes that the agencies specifically considered
technology cost-savings attributable to experience with technology--in
other words, the analysis provides that technology costs reduce over
time.
The Center for Biological Diversity (CBD) et al. commented that
since technologies exist today that can achieve the current standards,
reducing the standards to the level proposed in the NPRM is contrary to
the objectives of the Clean Air Act. These parties further commented
that EPA failed to make a proposed finding that additional lead-time is
necessary, as they argue is required by Section 202(a)(2). The Green
Energy Institute at Lewis and Clark Law School and others similarly
commented that EPA lacks a reasonable justification for extending the
phase-in period for the current standards because compliant
technologies currently exist and are already commercially available.
The Attorney General of California and others commented that EPA
acknowledges that most or all technology necessary to meet the current
standards is available, and does not provide evidence to support how
additional lead time is ``necessary to permit the development and
application of the requisite technology.''
In response to the public comments, and as EPA indicated in the
proposal and in the 2012 Final Rule establishing the previous
standards, the technologies projected to be used to meet the GHG
standards, including the alternatives in the proposal as well as the
final standards, are currently available and in production. If the
appropriateness of the standards were based solely on an assessment of
technology availability, and lead time considerations were limited to
the development of such technology, EPA might consider more stringent
CO2 standards to be potentially appropriate. But this is not
the sole or predominant factor to be weighed. In 2012, EPA had to
balance this issue as well. As in 2012, manufacturers today are capable
of building vehicles that can meet the standards that any of the
regulatory alternatives evaluated in the final rule would require.
However, greater uncertainty about consumer acceptance of those
technologies (as compared to what EPA believed was likely in 2012)
means that providing more lead time is appropriate.\2472\
---------------------------------------------------------------------------
\2472\ See 77 FR at 62871 (``As stated above, EPA's analysis
indicates that there is a technology pathway for all manufacturers
to build vehicles that would meet their final standards as well as
the alternative standards. The differences between the final
standards and these analyzed alternatives lie in the per-vehicle
costs and the associated technology penetration rates.'').
---------------------------------------------------------------------------
As in 2012, EPA disagrees with commenters that a finding that
necessary technology is available is, by itself, determinative of the
appropriate emission standard under CAA section 202(a). As described in
the proposed rule and in this section of the final rule, the
Administrator weighs technology availability and lead time along with
several other factors, including costs, emissions impacts, safety, and
consumer impacts in determining the appropriate standards under section
202(a) of the CAA.
Under this analysis, given the factors discussed later in this
Section, the previous standards would yield technology penetration
rates for advanced technologies beyond what is appropriate and
reasonable. By contrast, the final standards are projected to result in
more modest penetration rates for advanced technologies that
nonetheless will achieve an increased level of technology penetration
compared to the standards applicable for MY 2020. For example, the
final rule analysis projects that dynamic cylinder deactivation
penetration for MY 2030 would be 39.2 percent under the previous
standards for, but 34.4 percent under today's final standards.
Similarly, turbocharged engine penetration would be a projected 48
percent by MY 2030 under the previous standards, compared to 36.4
percent under the final standards. In addition, mild hybrids are
projected to change from 7.1 percent to 1.6 percent, strong hybrids
from 9 percent to 2.2 percent, and dedicated electric vehicles from 5.7
percent to 3.7 percent (all for MY 2030) under the final standards
instead of the previous standards. The Administrator believes that the
level of technology development and application for the final standards
is an appropriate balance, in light of the relevant factors considered
as a whole, as discussed below.
(b) Consideration of the Cost of Compliance
EPA is required to consider costs of compliance when setting
standards under section 202(a). The standards finalized today would
reduce required technology costs for the industry by an estimated $108
billion for the vehicles produced from MY 2017 through MY 2029 (at 3
percent discount rate, see Section VII) compared to the EPA standards
established in 2012. While less-stringent increases would result in
additional technology cost savings ($129 billion and $126 billion for
Alternatives 1 and 2, respectively), technology cost savings are only
one element that EPA considers.
In addition to capital cost savings, the final standards would
reduce the per-vehicle costs by $1,250 per vehicle in MY 2030, compared
to the standards set in 2012, as shown in Table VII-77. While less-
stringent increases would result in greater per-vehicle technology
cost-savings, cost-savings alone do not dictate the appropriate
standards. For example, Alternatives 1 and 2 would save manufacturers
$1,218 and $1,181 in per-vehicle costs in MY 2030 compared to the
previously issued standards. Alternatives more stringent than the final
standards would be more burdensome to manufacturers, with Alternatives
4 through 8 ranging from a cost savings to manufacturers of $927 to
$351 per-vehicle compared to the previous standards.
The costs to comply projected in this final rule are higher than
those previously projected by EPA in the 2017 Final Determination: In
2017 EPA projected that the per-vehicle cost to meet the MY 2025
standards would be $875 on average, with a range of $800 to $1,115
considering a range of
[[Page 25109]]
sensitivities (in 2015 dollars).\2473\ The costs to the auto industry
for complying with the previous MY 2022-2025 standards projected in the
2017 Final Determination were $24 billion to $33 billion (in 2015$ at 7
percent and 3 percent discount rates, respectively).\2474\ Again, EPA
notes that the values in this final rule analysis and the values in the
2017 Final Determination have different points of reference making them
not directly comparable, as discussed above.
---------------------------------------------------------------------------
\2473\ See 2017 Final Determination Table ES-1, page 4-5, and
II(v), page 24-26.
\2474\ Id. at Table ES-4, page 7.
---------------------------------------------------------------------------
Several public comments addressed EPA's consideration of costs of
compliance in setting the revised standards. The Alliance of Automobile
Manufacturers (Alliance) commented that the proposal's cost estimates
for the current MY 2021 and later standards differed from what EPA
projected in 2012 when setting those standards. The Alliance argued
that that those changes in the expected costs of the previously issued
standards provide significant reasoned support for EPA's view that the
existing standards should be reduced.
The Association of Global Automakers (Global Automakers) commented
on the importance of lead time for technology investment. While it
agreed that the existing standards are too stringent, it stated that
vehicle manufacturers and suppliers have invested $76 billion in
manufacturing facilities, and that much of that was for improvement in
CO2 emission reductions and fuel economy improvements. At
least some of that investment, according to Global Automakers, was made
to meet the standards set in 2012. Global Automakers expressed concern
with an abrupt halt to gradual fuel economy improvements, as such an
approach could result in stranded capital investments for automakers
and suppliers.
CBD and others disagreed with EPA's conclusion that the cost of
broader adoption of technologies is unreasonable in light of other
factors considered by EPA. CBD and others claimed that the Clean Air
Act narrowly allows for consideration of cost only as a question of
whether costs of compliance make it infeasible for manufacturers to
meet standards within the relevant period. They argue that this
consideration relates to lead time, and not to a broader consideration
of costs. They assert that broader compliance cost considerations apply
only to the motor vehicle industry. They also claim that compliance
costs to meet the standards set in 2012 for the 2017-2025 model years
are not challenging to the industry.
These commenters also state that the costs to industry to meet the
standards are not high enough to require reducing standards, to permit
development and application of the required technology. They claim that
the only burden that Congress intended to impose as a constraint on
emission reduction requirements are costs that are ``so severe as to
preclude the deployment of required technology during the relevant
period.''
The New York State Department of Environmental Conservation
commented on the role of technology innovation in considering
technology feasibility, while acknowledging that the feasibility
analysis allows for consideration of numerous factors argues that since
technology exists today to meet the standards for MY 2026, no lead time
is necessary. It further states that EPA did not appropriately balance
or consider in the proposal future technological advancements and OEM
innovation that will further constrain the costs of new technology.
In response to the Alliance's comment that the projected compliance
costs have changed significantly from EPA's 2012 rule, EPA agrees.
Indeed, this is a significant factor in EPA's conclusion that the
previous standards were too stringent. EPA notes that the projected
difference between the cost to comply with the previous standards and
the costs to comply with the standards established today is lower in
this final rule analysis as compared to the projected difference
between the proposal's preferred alternative and the previous
standards. EPA concludes that the final standards nevertheless result
in significant reductions in required technology costs for auto
manufacturers compared to the previous standards.
EPA also considered the Global Automakers' concern that freezing
the standards from MY 2021-2026 as proposed could result in stranded
capital for the auto industry and automotive suppliers who have
invested significantly in meeting the previous standards. The standards
EPA is finalizing today, unlike the proposed preferred alternative,
will require the gradual increase in CO2 improvements across
the fleet, at a rate of 1.5 percent per year stringency improvement,
thus supporting investments in GHG-reducing technologies, at a pace
that EPA believes is more reasonable than that of the previous
standards.
EPA disagrees with CBD et al.'s comments that the agency's
consideration of costs is inappropriate or not supported by the record.
EPA disagrees that Congress intended section 202(a)(2)'s requirement to
give ``appropriate consideration to the cost of compliance within such
period'' to mean that the agency ``only consider compliance costs if
they are so severe as to preclude deployment of the requisite
technology during the period.'' EPA does not interpret the Clean Air
Act as limiting EPA's consideration of costs to manufacturers only to
the question of whether such costs are so high that a manufacturer
could not afford to deploy the technology in question for a given model
year--that would be tantamount to suggesting that EPA must always set a
standard to achieve ``the greatest degree of emission reduction
achievable through the application of technology,'' which as discussed
above is not EPA's approach to setting standards such as these under
section 202(a). And this is particularly important when setting
CO2 standards, which, as described above, have a significant
impact on vehicle utility and performance that differs from other
standards established under Section 202. As discussed above, Congress
specified such technology-forcing standards elsewhere in section 202
and could have done so here (or otherwise specified that standards
shall take effect ``as soon as practicable'' while taking into
consideration costs and other factors)--but did not do so. Section
202(a) prevents EPA from implementing standards sooner than feasible,
taking into account lead time considerations and the cost of
compliance, but does not require standards be implemented as soon as
feasible or at the limit of feasibility, taking into account the cost
of compliance. EPA notes that it received numerous comments on the
analysis underlying the proposed rule, and the analysis for this final
rule in fact was changed from the proposal in consideration of these
comments, as discussed in Section VI.B. Nevertheless, the projected
costs to comply with the previous MY 2021-2026 standards remain
significant as discussed above, and EPA has considered these costs
along with other factors under the CAA in determining the final
standards, as discussed in Section VIII.A.3.h) below.
(c) Consideration of Costs to Consumers
In this section EPA considers the cost impacts on consumers. First,
the initial up-front costs to consumers are discussed, then the costs
associated with fuel expenditures, and finally the total ownership
costs to consumers over the life of the vehicles.
[[Page 25110]]
In addition to the $1,250 per-vehicle technology costs to the
automotive industry described above, which EPA expects could, and
likely would, be passed on to consumers, the analysis estimates other
per-vehicle costs that could be borne by consumers, specifically costs
attributed to changes in financing, insurance, taxes, and other fees,
as shown in Section VII. Considering these additional costs, EPA's
final standards (Alternative 3) would result in reduced costs to
consumers of $1,385 in MY 2029 (at a 3 percent discount rate) compared
to EPA's previously issued standards. While alternatives lower in
stringency than the final standards would save consumers more (i.e.,
Alternatives 1 and 2 would save consumers $1,665 and $1,637,
respectively, in MY 2029 at 3 percent discount rate), while
alternatives more stringent than the final standards would save
consumers less (i.e., Alternatives 4 through 7 would save consumers a
range of from $1,329 to $620, for MY 2029 at 3 percent discount rate),
this is only one of the factors EPA considers in setting standards. On
balance, EPA believes that further increases in stringency, compared to
the proposal, are appropriate and reasonable.
Compared to the previously issued CO2 standards, the
standards finalized today will result in increased fuel consumption and
associated expenditures for consumers. The analysis detailed in the
Final RIA and summarized in Section VII of this preamble projects the
increased fuel consumption for owners of the vehicle over the projected
life of the vehicle, up to 39 years, as compared to the previously
issued standards as the baseline. For example, as shown in Table VII-84
(at a 3 percent discount rate), consumers will spend $1,461 more in
fuel costs over the vehicle lifetime, which the analysis assumes can be
up to 39 years,\2475\ under today's final standards (Alternative 3)
compared to the previously issued standards.
---------------------------------------------------------------------------
\2475\ For further information of on the modeled distribution of
registrations by age see, e.g., Table VI-238--Registrations, Total
VMT, and Proportions of Total VMT by Vehicle Age (in Section
VII.D.2.b).2.(d)) which shows the distribution of registrations by
vehicle age.
---------------------------------------------------------------------------
EPA notes that, when comparing lifetime fuel savings for all owners
of a vehicle to the upfront additional ownership costs--generally borne
by the initial purchaser, a net reduction in benefits of $175 is seen
under the final standards. That said, as noted by several commenters,
consumers keep vehicles for a much shorter period of time prior to
trading the vehicle in for another or selling the vehicle.\2476\ CFA,
for instance mentioned that consumers retain vehicles for more than
five years, and a group of State Comptrollers and Treasurers referred
to an IHS Markit report that the average length of time a consumer
keeps a new car is approximately 6.6 years. Accordingly, such a
simplistic comparative approach would anticipate that a consumer
account for fuel savings over a much longer period of time than would
be rational. Further, it is important to note that consumers are
informed of estimated average annual fuel costs for the vehicle, as
well as a comparison of the difference between five years'-worth of
fuel costs or savings compared to an average new vehicle on the
Monroney label that must be posted on every new vehicle offered for
sale.
---------------------------------------------------------------------------
\2476\ It should be noted, however, that, all else being equal,
improved fuel economy can improve resale value of a vehicle. That
said, it is not at all clear that consumers generally anticipate
potential future incremental trade-in value attributable to improved
fuel economy when making a decision as to which new vehicle to
purchase.
---------------------------------------------------------------------------
In the 2017 Final Determination, EPA projected that the previous MY
2022-2025 standards compared to the MY 2021 standards would provide
fuel savings of $52 billion to $92 billion and total net benefits of
$59 billion to $98 billion (in 2015 dollars and at 7 percent and 3
percent discount rates, respectively, and based on AEO2016 reference
case fuel prices). The up-front vehicle costs to consumers were
projected to be approximately $926 per vehicle, including the vehicle
technology costs, taxes and insurance.\2477\ EPA projected that
consumers would realize net savings of $1,650 over the lifetime of a
new MY 2025 vehicle (net of increased lifetime costs and lifetime fuel
savings).\2478\ Under the final standards, vehicle sales are expected
to increase by 2.2 million vehicles over MY 2017-2029 compared to
projected sales under the previous standards. EPA views this projection
of vehicle sales increases resulting from the final standards as
important in facilitating the turnover of the fleet to newer, safer
vehicles, all of which will be subject to increasingly stringent
criteria pollutant emission requirements as federal Tier 3 emission
standards continue to phase in from MY 2017 through MY 2025.
Below the major comments are summarized regarding EPA's
consideration of the impact of the revised standards on consumers.
Securing America's Future Energy (SAFE) commented that vehicle prices
are influenced by many factors beyond the GHG standards, and that costs
to improve fuel economy make up only a portion of the vehicle price.
SAFE notes that fuel savings from efficient vehicles offsets increase
ownership costs. SAFE further claims, without support, that standards
``do not have a major role in creating higher vehicle prices, or in
suppressing sales.'' Accordingly, SAFE argues that pausing fuel economy
increases, as proposed in the NPRM, is not justified. SAFE suggests
that fuel savings impacts should be discussed along with technology
cost increases.
CBD and others commented that EPA's consideration of consumer
costs, including finance and insurance costs, cannot outweigh its
public health mandate. Such commenters noted that some of the options
analyzed in the notice showed that fuel savings of the lifetime of the
vehicle outweighed upfront vehicle price increases, and that not
choosing such an alternative is not justified. CBD then goes on to
argue that the analysis inflates technology costs and undercounts fuel
savings.
The California Attorney General and others claim that EPA's
consideration of the potential increased costs for consumers related to
maintenance, financing, insurance, taxes, and other fees is
unjustified, unlawful, and contrary to its prior position that
compliance cost considerations include only costs to the motor-vehicle
industry.
EPA notes that fuel efficiency and GHG standards affect labor and
materials costs, technology add-ons, and sales mix, and expects the
estimated cost decrease from these final standards to have a positive
effect on the auto market and vehicle buyers. As described in the
notice and throughout this preamble, EPA disagrees that standards have
no major impact on increasing prices or suppressing sales. Fuel-saving
technology adds costs, and as prices increase, fewer consumers can
afford to buy new cars--either because they cannot afford a new car, or
because they decide to purchase an older vehicle, or because they
decide to keep their existing vehicle. EPA also notes that both the
notice and this preamble discusses fuel savings from the various
alternatives analyzed. Some commenters suggest EPA calculate and
consider fuel savings, spread over the lifetime of the vehicle up to 39
years and experienced by multiple owners--compared to the upfront
vehicle costs, which are generally paid for by the original purchaser
either in cash or through additional finance costs over a much shorter
period of time. This approach, which would yield a projected $175 in
additional costs (additional lifetime outlays for fuel minus avoided
upfront vehicle costs)
[[Page 25111]]
over the multi-owner, lifetime of a vehicle beyond the initial
ownership savings, distorts the comparison. Instead, EPA concludes that
the upfront vehicle technology costs (and associated financing costs)
are a more important factor. In other words, a consumer is more likely
to buy a new vehicle at a lower up-front price even if that vehicle
will incur a more-than offsetting level of fuel costs over its lifetime
that will be borne by the first and all subsequent owners of the
vehicle.\2479\ By reducing upfront costs, more consumers will be able
to afford new vehicles, which will result in a quicker fleet turnover
to safer, more efficient vehicles that emit lower amounts of criteria
pollutants than the existing fleet. In fact, the agencies project that
the revised standards will result in 2.2 million additional new
vehicles sold--all of which would meet the latest safety standards and
be subject to the phase-in of the Tier 3 criteria pollutant emission
standards.
---------------------------------------------------------------------------
\2479\ For further discussion regarding consumers valuation of
fuel economy, see preamble section VI.D.1.b).(2) (sales), preamble
section VI.D.1.b).(8), and Final Regulatory Impact Analysis section
III.C.
---------------------------------------------------------------------------
With respect to the comments that consideration of costs to
consumers is contrary to CAA section 202(a)(2), EPA disagrees. As
discussed above, section 202(a)(2) requires EPA to consider the cost of
compliance, which EPA has done, and it allows EPA to consider other
costs, including costs to consumers, which EPA also have done, in this
rule and past rules setting standards under section 202(a). The statute
sets some minimum requirements for EPA's consideration, but permits a
wider range of concerns to be considered, including public health and
welfare but also safety, costs to consumers, and other factors
discussed herein. As discussed above, and below, EPA has considered the
effects of a range of potential standards across this entire set of
factors. The agency is permitted to take all of these factors into
account, and that is what it has done in selecting the final standards.
d) Consideration of GHG Emissions and Other Air Pollutant Emissions
As discussed above, the purpose of GHG standards established under
CAA section 202 is to reduce GHG emissions, which EPA has found to
endanger public health and welfare, in an appropriate manner that takes
into account other factors as directed by Congress and in the
reasonable exercise of EPA's discretion under the statute. Today's
final standards are projected to increase CO2 emissions
compared to the previously issued standards, by a total of 867 million
metric tons (MMT) over the lifetime of MY 1977 through MY 2029 vehicles
(see Section VII of this preamble)--i.e., by 2.9% of the amount
projected to be attributable to passeners cars and light trucks under
the baseline/augural standards. Of this CO2 emissions
increase, 731 MMT would come from tailpipe emissions, and an additional
136 MMT from upstream sources, both being nearly 3% greater than
projected to occur under the baseline/augural standards. The analysis
projects that Alternatives more stringent than the final standards
would result in smaller increases in CO2 emissions. Also
compared to the baseline/augural standards, and also over the lifetime
of MY 1977-2029 vehicles, Alternatives 4 through 7 are projected to
increase CO2 emissions by 826 MMT (2.8%) to 361 MMT (1.2%).
Alternatives less stringent than the final standards would increase
CO2 emissions by a greater amount, 1,074 MMT (3.5%) and
1,044 MMT (3.6%), for Alternatives 1 and 2 respectively.\2480\
---------------------------------------------------------------------------
\2480\ This preamble and the FRIA document estimate annual GHG
emissions from light-duty vehicles under the baseline CO2
standards, the final standards, and the standards defined by each of
the other regulatory alternatives considered. For the final rule
issued in 2012, EPA estimated changes in atmospheric CO2,
global temperature, and sea level rise using GCAM and MAGICC with
outputs from its OMEGA model. Because the agencies are now using the
same model and inputs, outputs from NHTSA's EIS (that used more
recent versions of GCAM and MAGICC) were analyzed. Today's analysis
estimates that annual GHG emissions from light-duty vehicles under
the CO2 standards and corresponding CAFE standards, which
are very similar. Especially considering the uncertainties involved
in estimating future climate impacts, the very similar estimates of
future GHG emissions under CO2 standards and
corresponding CAFE standards means that climate impacts presented in
NHTSA's EIS represent well the climate impacts of the CO2
standards.
---------------------------------------------------------------------------
In addition to GHG emissions, EPA has considered the change in
criteria air pollutant emissions impacts due to the revised
CO2 standards. EPA has considered both tailpipe emissions
and upstream emissions associated with increased fuel consumption.
Unlike with CO2 emissions, which EPA found to be a long-
lived greenhouse gas well-mixed throughout the global atmosphere,
criteria pollutant emissions contribute primarily to local and regional
air pollution. Generally, tailpipe emissions for volatile organic
compounds (VOC), nitrogen oxides (NOX), and particulate
matter (PM) decrease under the final standards compared to the previous
standards, leading to improvements in human health in areas where air
quality improves. Upstream emissions attributable to refining and
transportation of the additional fuel needed under less stringent
standards increase under the final standards, leading to adverse
impacts on public health in locations where air quality worsens. The
additional upstream emissions generally exceed the reduced tailpipe
emissions, leading to net increases in these pollutants and net
increases in adverse health effects. Under the model year analysis
(changes in pollutants summed over the lifetimes of MY 1977-2029
vehicles for calendar year 2017 and later), and relative to total
emissions projected to be attributable to passenger cars and light
trucks under the baseline/augural standards, these increases range from
0.1% (for NOX) to 0.7% (for SO2 and PM). On the
other hand, projected net emissions of carbon monoxide (CO) are 0.4%
lower under the final standards than under the baseline/augural
standards, and emissions of air toxics (e.g., benzene) are 0.1-0.4%
lower under the final standards, varying among different toxic
compounds.
In addition to evaluating emissions impacts under the model year
analysis described above, EPA has considered the emissions impacts
under a calendar year analysis, which provides information over a
longer time horizon about the interactions between all vehicle model
years on the road in any given calendar year--that is, considering the
effects of the revised MY 2021 and later standards on fleet turnover
and utilization from calendar year 2017 out to 2050. Both the model
year analysis and the calendar year analysis provide relevant
information about the impacts of EPA's standards. When viewed from the
calendar year analysis perspective that extends through 2050, the
emissions impacts of the revised MY 2021 and later standards compared
to the baseline/augural standards vary over time, with cumulative
differences generally being greater in magnitude than under the model
year analysis: EPA's analysis shows cumulative VOC emissions through
2050 under the final standards increasing by a total of nearly 575
thousand tons (1.9%) relative to the cumulative amount projected to
accrue through 2050 under the baseline/augural standards. On the same
basis, estimated NOX and PM emissions increase by about 173
thousand tons (0.8%) and 16.5 thousand tons (1.7%), respectively. On
the other hand, also on the same basis, estimated CO and SO2
emissions decrease by about 278 thousand tons (0.1%) and 38 thousand
tons (0.8%), respectively.
As shown in the NHTSA Final Environmental Impact Statement (FEIS),
[[Page 25112]]
NHTSA's analysis indicates small air quality improvements in some areas
and small decrements in others which could help or hinder individual
areas' efforts to attain the NAAQS in the future.
EPA has also considered the health effects of air pollution
associated with today's final standards. As discussed above, it is the
cumulative contribution of the lower projected vehicle tailpipe
emissions with the higher projected upstream emissions (primarily from
the production and distribution of gasoline) which impact air quality.
As noted above and presented in detail elsewhere in this preamble and
the Final RIA, vehicle emissions are generally reduced due to the SAFE
final rule.
Due largely to the projected increase in upstream emissions
resulting from the increased production and transportation of gasoline
resulting from the standards finalized today compared to the previous
EPA standards, the Final Rule analysis projects increases in premature
deaths, asthma exacerbation, respiratory symptoms, non-fatal heart
attacks, and a wide range of other health impacts. While these health
impacts are presented in detail elsewhere in this preamble and in the
Final RIA, two factors suggest that the forgone premature mortality
benefits are overstated. First, in the last year, EPA has completed
analysis that demonstrated the likelihood that the air quality modeling
approach used here (i.e., benefits per ton) overestimates foregone PM
premature mortality benefits. Second, the 2012 rulemaking significantly
overestimated gasoline price projections in its baseline, predicting
lower fuel consumption, thus overestimating the premature mortality
benefits in that rule. While gasoline price projections in this
rulemaking have been updated to reflect recent data, the potential for
this kind of unanticipated fluctuation in gasoline prices remains, thus
estimates of fuel consumption and the correlated foregone premature
mortality benefits may not capture actual market outcomes.
The valuation of premature mortality effects rely on the results of
``benefits per ton'' approach (BPT). This approach is a reduced form
approach, which is less complex than full-scale air quality modeling,
requiring less agency resources and time. Based on EPA's work to
examine reduced form approach, the BPT may yield estimates of
PM2.5-benefits for the mobile sector that are as much as 10
percent greater than those estimated when using full air quality
modeling.
The EPA is currently working on a systematic comparison of results
from its BPT technique and other reduced-form techniques with results
from full-form photochemical modelling. While this analysis employed
photochemical modeling simulations, we acknowledge that the Agency has
elsewhere applied reduced-form techniques. The summary report from the
``Reduced Form Tool Evaluation Project'', which has not yet been peer
reviewed, is available on EPA's website at https://www.epa.gov/benmap/reduced-form-evaluation-project-report. Under the scenarios examined in
that report, EPA's BPT approach in the 2012 rule (which was based off a
2005 inventory) may yield estimates of PM2.5-benefits for
the mobile sector that are as much as 10 percent greater than those
estimated when using full air quality modeling. The estimate increases
to 30 percent greater for the electricity sector. The EPA continues to
work to develop refined reduced-form approaches for estimating
PM2.5 benefits.
Also, in this regulation, a key projection that influences the
estimation about car purchase and driving behavior is the gasoline
price projection. From 2008 through 2018, the average monthly gasoline
price ranged from less $1/gallon to $4/gallon.\2481\ The gasoline price
level and the volatility of price changes are major drivers of car
purchasing behavior thereby gasoline consumption and the resulting
criteria pollutant emissions. If gasoline prices are lower than
projected in an analysis, consumers are more likely to purchase less
fuel efficient cars, resulting in more emissions and vice versa.
---------------------------------------------------------------------------
\2481\ https://www.eia.gov/energyexplained/gasoline/price-fluctuations.php.
---------------------------------------------------------------------------
With a lower fuel price projection and an expectation that new
vehicle buyers respond to fuel prices, the 2012 rule would have shown
much smaller fuel savings attributable to the more stringent standards.
Projected fuel prices are considerably lower today than in 2012. The
agencies now understand new vehicle buyers to be at least somewhat
responsive to fuel prices, and the agencies have therefore updated
corresponding model inputs to produce an analysis the agencies consider
to be more realistic.
The first of these assumptions, fuel prices, was simply an artifact
of the timing of the rule. Following recent periodic spikes in the
national average gasoline price and continued volatility after the
great recession, the fuel price forecast then produced by EIA (as part
of AEO 2011) showed a steady march toward historically high, sustained
gasoline prices in the United States. However, the actual series of
fuel prices has skewed much lower. As it has turned out, the observed
fuel price in the years between the 2012 final rule and this rule has
frequently been lower than the ``Low Oil Price'' sensitivity case in
the 2011 AEO, even when adjusted for inflation. The discrepancy in fuel
prices is important to the discussion of differences between the
current rule and the 2012 final rule, because that discrepancy leads in
turn to differences in analytical outputs and thus to differences in
what the agencies consider in assessing what levels of standards are
reasonable, appropriate, and/or maximum feasible. Long-term predictions
are challenging and the fuel price projections in the 2012 rule were
within the range of conventional wisdom at the time. However, it does
suggest that fuel economy and tailpipe CO2 regulations set
almost two decades into the future are vulnerable to surprises, in some
ways, and reinforces the value of being able to adjust course when
critical assumptions are proven inaccurate. This value was codified in
regulation when EPA bound itself to the mid-term evaluation process as
part of the 2012 final rule.\2482\
---------------------------------------------------------------------------
\2482\ See 40 CFR 86-1818-12(h).
---------------------------------------------------------------------------
Because of these uncertainties surrounding air quality modeling of
premature mortality effects, the projections of foregone PM premature
mortality benefits are uncertain and may be over-stated. Fluctuations
in gasoline prices contribute to this uncertainty, making it difficult
to accurately project gasoline consumption and its related premature
mortality benefits.
The analysis projects that the air pollution emission increases
associated with the revised standards will lead to an increase of 440
to 1,000 premature deaths--deaths that occur before the normally
expected life span--0.5% more than the number of such deaths projected
to occur under the baseline/augural standards and over the lifetime of
the MY 1977-MY 2029 vehicles. In addition, a wide range of health
impacts are projected to increase by 0.4-0.6% under the final standards
compared to occurrences projected to occur the standards established in
2012, as summarized in Table VII-132 et seq.
When quantified using the calendar year (CY) analysis perspective
(CYs 2018-2050), under the revised final standards (compared to the
previous standards), premature mortality is expected to increase from
460 to 1,010 deaths (i.e., by 0.4%), upper and lower respiratory
symptoms are expected to increase by 22,000 cases (0.4%), asthma
exacerbations are projected to increase by 16,000 cases (0.4%), acute
bronchitis
[[Page 25113]]
cases are projected by increase by 720 (0.4%), non-fatal heart attacks
are projected to increase by 450 (0.4%), hospital admissions for
cardiovascular and respiratory issues are projected to increase by 225
(0.4%) cases, and emergency room visits for respiratory issues are
projected to increase by 260 (0.4%). In addition, these additional
health impacts are expected to result in an additional 61,000 work loss
days (0.3% of the number projected under the baseline/augural
standards) and 355,000 minor restricted activity days (0.4% more than
under that baseline/augural standards) for the public. Compared to the
baseline/augural standards, the agencies estimate that the final
standards rule will increase by 0.3-0.4% each of the various health
impacts accumulated through 2050 (e.g., premature deaths, upper and
lower respiratory symptoms, asthma exacerbations, acute bronchitis
cases, hospital admissions for cardiovascular and respiratory issues,
emergency room visits for respiratory issues).
In the 2017 Final Determination, EPA projected GHG emissions
reductions of 540 million metric tons over the lifetimes of MY 2022-
2025 vehicles.\2483\ EPA also projected criteria pollutant emission
reductions for CY2040 of 97,000 tons of VOC, 24,000 tons of
NOX, 3,600 tons of PM2.5, and 15,000 tons of
SO2.\2484\ EPA projected that these emissions reductions
would result in positive health benefits through CY2050.\2485\ In this
final rule, the revised final standards compared to the previous
standards are projected to result in an increase in emissions and
health incidences, as discussed above, resulting in $5 billion or $3
billion (in 2018 $, and reflecting, respectively, either a 7 percent or
3 percent discount rate) in foregone public health benefits (see Table
VII-103 and Table VII-104).
---------------------------------------------------------------------------
\2483\ 2017 Final Determination at Table ES-3, page 6, and
Section II (iv), page 24.
\2484\ 2016 Proposed Determination at Appendix C, Table C.54,
page A-163.
\2485\ Id. at Table C.87, page A-183.
---------------------------------------------------------------------------
In public comments on these topics, the Attorney General of
California and others commented that, in adopting the previous
standards, EPA focused on obtaining significant CO2 emission
reductions, but now proposed to increase emissions relative to the
previous standards without sufficient justification. They claim that
EPA offered no justification of acknowledgement of a change in
position, stating that none of the alternatives further the goal of
CO2 emission reductions. They argue that EPA justifies its
proposal on the limited impact of the rule on global climate change,
and that failing to seek incremental improvements is contrary to the
EPA's duties under the Clean Air Act.
The United States Conference of Catholic Bishops commented that
considering public safety of any set of standards requires giving
significant weight to the effect of air pollution, and that the
proposal failed to promote public health and safety.
The Chesapeake Bay Foundation (CBF) claims that the proposal would
have significant health consequences that disproportionately impact
minority and low-income communities in the Chesapeake Bay. They discuss
general impacts of climate change CBF argues that criteria pollutant
health impacts of the proposal, should be more heavily weighed against
safety impacts of the rule.
The State of Washington commented that the agencies did not analyze
public health effects from increased criteria pollutant emissions
arising from increased petroleum consumption or environmental justice
concerns. They claim that the NPRM's discussion of the negligible
impact of the rulemaking on global climate change is ``deeply
concerning.''
As noted above, EPA agrees that the purpose of Title II emission
standards is to protect the public health and welfare from air
pollution, and in establishing emission standards, the agency is
cognizant of the importance of this goal. At the same time, EPA
balances multiple factors in determining what standards are reasonable
and appropriate. And, contrary to some commenters' views, unlike other
provisions in Title II, section 202(a) does not require the
Administrator to set standards which result in the greatest degree of
emissions control achievable. Thus, in setting these standards, the
Administrator has taken into consideration other factors discussed
above and below, including not only technological feasibility, lead-
time, and the cost of compliance, but also potential impacts of vehicle
emission standards on safety and other impacts on consumers.
Several commenters claimed that the agencies did not analyze health
impacts of the various alternatives, but this is not accurate. First,
the notice and PRIA included this information in monetized terms to
facilitate the balancing of various factors. Further, NHTSA conducted a
comprehensive Draft Environmental Impact Statement, which discussed
these effects in detail. For this final rule, these health impacts have
been separately itemized, as summarized above. Other commenters claimed
that the agencies did not sufficiently consider environmental justice
elements in the proposal. This, too, is inaccurate, as discussed
elsewhere in this preamble.
In response to comments of the California Attorney General and
others, that the Clean Air Act cannot allow for increases in a
regulated emission, EPA notes that the 2012 Final Rule specifically
called for a Mid Term Evaluation process that envisioned the potential
for an adjustment of the standards in case the stringency increases
established in 2012 were no longer reasonable and appropriate. As
discussed above, the increases in stringency of the standards for MY
2021-2025 are, on balance, not reasonable and appropriate based on a
consideration of the factors described in this preamble. EPA now
recognizes based on updated information and analysis that industry
should be provided additional lead time to meet the later model years
of standards set in the 2012 rule, and, as discussed in this preamble,
industry is having unanticipated difficulties complying with earlier
years of the standards, with fleetwide performance failing to meet
CO2 emission targets in MY 2016 and MY 2017. That is not to
say that CO2 and criteria pollutant emissions are not
significant factors in this rulemaking. Indeed, they are weighed
heavily along with other important factors considered by EPA, which has
led to increasing stringency on a 1.5 percent annual basis for the
2021-2026 model years. Importantly, the agencies project that the
revised standards will result in an additional 2 million new vehicles
sold before 2030 compared to under the baseline/augural standards. This
means that an additional 2 million vehicles will be produced during the
phase-in of the Tier 3 emission standards, which implement more
stringent tailpipe standards for criteria pollutants, displacing
greater numbers of higher-emitting older vehicles and providing
significant health benefits. As discussed, when finalizing the Tier 3
standards in 2014, ``[t]he final Tier 3 vehicle and fuel standards
together will reduce dramatically emissions of NOX, VOC,
PM2.5, and air toxics.'' \2486\
---------------------------------------------------------------------------
\2486\ 79 FR 23425.
---------------------------------------------------------------------------
Although GHG emissions reductions would be lessened under the
standards finalized today compared to the previously issued EPA
standards, in light of this assessment indicating higher vehicle costs
and associated impacts on consumers, EPA believes that, on balance, the
final standards
[[Page 25114]]
(Alternative 3) are justified and appropriate.
(e) Consideration of Consumer Choice
EPA believes that consumer demand is an important consideration in
setting CO2 emission standards, because one of EPA's goals
in setting the standards has been and continues to be to allow
manufacturers to provide, and consumers to purchase, vehicles with
varying attributes and functionality rather than to shift demand to
certain vehicle types or sizes. Societal and economic trends play a
role in this area as well--if fuel prices are relatively high, demand
for fuel-efficient vehicles increase and, as a result, compliance with
standards is easier to achieve. If fuel prices are relatively low--as
they are now and are projected to be in the mid-term--consumer demand
for fuel-efficiency is less strong, making it harder for manufacturers
to comply with the standard. While manufacturer difficulty in complying
due to lack of consumer demand may not be the deciding factor in
determining the appropriate levels of stringency for standards, it is
relevant to understanding lead time difficulties, which EPA is required
to consider under Section 202(a)(2).
As discussed previously, the EPA CO2 standards are based
on vehicle footprint, and in general smaller footprint vehicles have
individual CO2 targets that are lower (more stringent) than
larger footprint vehicles. The passenger car fleet has footprint curves
that are distinct from the light-truck fleet. One of EPA's goals in
designing the footprint-based standards, in considering the shape,
slope, and stringency of the footprint standard curves, and in adopting
various compliance flexibilities (e.g., emissions averaging, banking,
and trading, air-conditioning credits, off-cycle credits) was to
maintain consumer choice. The EPA standards are designed to require
reductions of CO2 emissions over time from the vehicle fleet
as a whole, but also to provide sufficient flexibility to the
automotive manufacturers so that firms can produce vehicles that serve
the needs of their customers. The past several model years in the
marketplace show that, while this approach reduces the impact of
increased fuel economy on consumer choice, it does not adequately
account for changes in consumer preference. As a result, as discussed
throughout this preamble, manufactures are struggling to meet
CO2 emission standards based upon their fleet performance.
In fact, the 2017 model year saw that only three major manufacturers
had fleets that met the standards. One reason behind these challenges
is that, while the footprint-based attribute standards account for
vehicle length and width, they do not account for vehicle height or
weight. And, since many crossovers sold today are classified as
passenger cars and not light trucks, the additional weight of such
vehicles to provide for requisite ride height puts pressure on
CO2 emission compliance for automaker passenger car fleets.
Similarly, large SUVs are subject to the same footprint-based standards
as lighter trucks, putting pressure on CO2 emission standard
compliance. For the 2017 model year, 12 percent of the fleet consisted
of car-based SUVs, and 32 percent of the fleet consisted of truck-based
SUVs.\2487\ Taller and heavier vehicles, including crossovers and SUVs,
are more popular today than was expected at the time the standards were
set. While automobile manufacturers have continued to offer a broad
range of vehicles (e.g., full-size pick-up trucks with high towing
capabilities, minivans, cross-over vehicles, SUVs, and passenger cars;
vehicles with off-road capabilities; luxury/premium vehicles,
supercars, performance vehicles, entry level vehicles, etc.) despite
continuing required increases in fuel economy stringency, this has
largely been possible because of well-stocked over-compliance credit
banks from when standards were less stringent and the ability to
acquire credits from other manufacturers. As mentioned earlier, the
agencies have concerns whether this is sustainable. Automotive
companies have been able to reduce their fleet-wide CO2
emissions while continuing to produce and sell the many diverse
products that serve the needs of consumers in the market. The agencies
recognize that automotive customers are diverse, that automotive
companies do not all compete for the same segments of the market, and
that increasing stringency in the standards can be expected to have
different effects not only on certain vehicle segments but also on
certain manufacturers that have developed market strategies around
those vehicle segments. Taking into consideration this diversity of the
automotive customer base, and of the strategies which have developed to
meet specific segments, EPA concludes that the previous standards are
not reasonable or appropriate.
---------------------------------------------------------------------------
\2487\ 2018 EPA Automotive Trends Report: Greenhouse Gas
Emissions, Fuel Economy, and Technology since 1975, available at:
https://www.epa.gov/automotive-trends/download-automotive-trends-report.
---------------------------------------------------------------------------
In the initial determination, EPA assessed several factors related
to consumer choice, including the costs to consumers of new vehicles
and fuel savings to consumers, as described above under Section
VII.A.2.c). In 2017, EPA found that the previous standards would
increase the upfront costs of vehicles but overall would have positive
net benefits because lifetime fuel savings outweighed the lifetime
vehicle costs for consumers. As discussed above, the costs of
technology to comply with the standards are generally borne by the
initial purchaser, with understanding of fuel cost implication given
statutorily required disclosures. In contrast, the fuel savings are
realized by many subsequent owners over the vehicles' lifetime, which
this analysis assumes can be up to 39 years. New vehicle purchasers are
not likely to place as much weight on fuel savings that will be
realized by subsequent owners. Accordingly, EPA is placing greater
weight on the up-front vehicle cost savings to consumers in light of
the goal of accelerating the turnover of the motor vehicle fleet to
safer cars that emit fewer criteria pollutants.
EPA received many comments regarding the agency's consideration of
consumer choice in determining appropriate standards under section
202(a) of the CAA. The Alliance commented that EPA's concerns regarding
consumer choice are well founded, stating ``in the years since 2012
(and in part due to the unexpected decrease in fuel prices), consumers
have demonstrated less interest in high-efficiency/low-emission
vehicles than EPA and NHTSA projected in issuing the 2012 Final Rule.
As such, compliance with the existing standards would require a
substantially greater variance than EPA expected from the vehicle fleet
that consumers would otherwise choose.''
Global Automakers agreed that consumer acceptance is an important
factor, but does not justify holding standards flat through the 2026
model year. Global Automakers further commented that ``[f]uel economy
remains a factor in vehicle purchase decisions, though perhaps not a
dominant one.''
CBD and others commented that the Clean Air Act does not allow EPA
to reduce stringency based upon consumer choice factors. They point to
the diversity of the vehicle fleet and argue that EPA's consideration
of projected tech levels and associated costs as ``speculative'' and
not grounded in fact.
U.S. Congressman Mark DeSaulnier claimed that the justification for
the proposal appeared to be consumer
[[Page 25115]]
willingness to buy new vehicles. He claimed that absent any standards
whatsoever, automakers could produce more vehicles that consumers would
want to purchase. He stated that the standards require all vehicles to
become more efficient and that EPA has an overly simplistic
understanding of American consumers, who, according to him, are ``wary
of the price tag'' when shopping, but, nonetheless, ``overwhelmingly
want more efficient vehicles, and they want to reduce the health burden
of air pollution.''
The Institute for Policy Integrity (IPI) claims, without support,
that as fuel efficiency technology is introduced and becomes
widespread, consumer attitudes will change and will start focusing on
such technology. IPI also claims that manufacturers can change consumer
preference through advertising. IPI implies that manufactures play a
larger role in shaping consumer options of their needs that consumers
do themselves. IPI also comments that academic literature relating to
demand- and supply-side obstacles to fuel economy indicates that the
proposal's justification runs counter to available evidence.
The University of California Berkeley Environmental Law Clinic
(Berkeley) argued against EPA's consideration of consumer choice in
setting standards, claiming that low-income households bear exposure to
operating costs, fuel price fluctuations, and environmental impacts.
Berkeley also claimed that EPA's purported list of features consumers
may favor over fuel economy is not supported by evidence, and, in any
event, should be categorized into lists of ``needs'' versus ``wants.''
Consumer choice is a complex consideration when setting standards.
As Congressman DeSaulnier correctly notes, EPA cannot disregard its
consideration of public health and welfare based upon the agency-
projected whims of consumers. At the same time, the willingness of
consumers to pay for fuel economy improvements, which as described
above affects vehicle performance and utility in a manner
distinguishable from criteria pollutant emissions, has a direct effect
upon the ability of manufacturers to sell their product. And as
consumers demand vehicles with increased ride height (which, all else
being equal, increases CO2 emissions), establishing
standards that account for this--but still require manufacturers to
focus on improving emission performance, is reasonable and appropriate.
In response to Global Automakers' comment that consumers do not
heavily focus on fuel economy in making purchase decisions, EPA agrees,
but notes that this is a consumer's choice, as federal law requires
that consumers are made aware of fuel economy impacts, pursuant to 49
U.S.C. 32908. EPA also agrees that the willingness to pay for fuel
economy improvements is ``not zero.''
EPA agrees with the Global Automakers comment that while consumer
choice is an important consideration in determining the appropriate
level of the revised standards, the final rule analysis does not
support holding the standards constant. Although EPA proposed standards
at the level of 0 percent increase in stringency from MY 2021 and
later, after considering the comments received and based on the updated
analysis for this final rule, EPA is finalizing standards with a 1.5
percent per year improvement in stringency from MY 2021 to MY 2026. As
indicated in the comments on this topic, there is a range of views and
relevant information concerning the extent of consumers' interest in
fuel economy and on the role fuel savings plays in consumer purchase
decisions.\2488\ EPA's understanding is that some consumers value fuel
economy more than others, and EPA finds it unnecessary to identify the
precise role of fuel economy in consumer purchase decisions because the
Administrator believes that the standards should encourage a range of
vehicles meeting a range of consumer preferences. Further, as described
above, consumers are made aware of the relative fuel price impacts of
new vehicles, given the required information label on new vehicles,
thus indicating that, in all likelihood, consumers do take fuel
expenses into account when making new vehicle purchase decisions.
---------------------------------------------------------------------------
\2488\ Studies of the role of fuel economy in consumer purchase
decisions have found a wide range of values (Greene, D., A. Hossain,
J. Hofmann, G. Helfand, and R. Beach. ``Consumer Willingness to Pay
for Vehicle Attributes: What Do We Know?'' Transportation Research
Part A 118 (2018), p. 258-79). The National Academy of Sciences in
2015 judged that ``there is a good deal of evidence that the market
appears to undervalue fuel economy relative to its expected present
value, but recent work suggests that there could be many reasons
underlying this, and that it may not be true for all consumers.''
National Research Council of the National Academies (2015). Cost,
Effectiveness, and Deployment of Fuel Economy Technologies for
Light-Duty Vehicles. Washington, DC: National Academies Press, p. 9-
16.
---------------------------------------------------------------------------
EPA disagrees with Congressman DeSaulnier's assertion that EPA
seeks to set standards that do not affect what manufacturers produce--
instead, the agencies examine what consumers are purchasing in the
market to determine what standards are appropriate. The agency's
assumptions in 2012--that consumers would gravitate toward the purchase
of compact sedans and coupes in response to exceedingly high fuel
prices--have proved incorrect. Fuel prices have fallen and remained
relatively low, and are projected to remain relatively low throughout
the period covered by this rulemaking. EPA seeks to achieve
improvements in CO2 emissions, but it is not realistic to
expect the high demand for crossover vehicles to abate, or for those
vehicles to meet more-stringent standards set for compact sedans. That
said, EPA agrees with Congressman DeSaulnier that American consumers
are wary of the price of vehicles--popular reporting that consumers may
reference explain affordability concerns in crisis terms--even
indicating that the average price of a vehicle is now beyond that which
is affordable to the median household income of every city outside of
Washington, DC \2489\ This results in significant adverse economic
impacts--higher finance charges, taxes, registration fees, and
insurance costs, all of which result in challenges qualifying for
financing and longer finance terms, which increase the likelihood of
negative equity scenarios. EPA also agrees with Congressman DeSaulnier
that consumers want increased fuel efficiency and to reduce the impacts
of harmful air pollution. These are all true. But direct health impacts
of vehicles emissions stem more from criteria pollutant emissions than
from CO2 emissions. And CO2 emission technology
has a significant relationship to the price of vehicles for which
consumers are so wary. EPA, with this rulemaking, is attempting to
strike the correct balance between a number of factors, including
improving efficiency and affordability, which should yield additional
sales and an improved rate of fleet turnover to vehicles that have
better criteria pollutant emissions--particularly since the vehicles
sold subject to this rulemaking will be sold during the phase-in of
Tier 3 criteria pollutant emission standards.
---------------------------------------------------------------------------
\2489\ See., e.g., Car and Driver, ``For Middle-Class Shoppers,
New Cars Are Moving out of Reach'' November 30, 2019. Available at:
https://www.caranddriver.com/news/a30061910/middle-class-car-shoppers-priced-out/; New York Times, ``New Cars Are Too Expensive
for the Typical Family, Study Finds'' July 2, 2016. Available at:
https://www.nytimes.com/2016/07/02/your-money/new-cars-are-too-expensive-for-the-typical-family-study-finds.html.
---------------------------------------------------------------------------
In response to Berkeley, low-income consumers are even more
sensitive to upfront vehicle purchase prices than they are to the
smaller delta between weekly or monthly fuel costs
[[Page 25116]]
experienced over time between the previous standards and the standards
finalized today--they may well take note of the fact that one cannot
pay today's bills with tomorrow's savings. They may also want to take
note that the standards finalized today are projected to improve fleet
turnover into newer vehicles that emit reduced criteria pollutants.
EPA disagrees with the assertion by CBD and others that the agency
has not provided a rationale for its consideration of consumer choice
in determining the appropriate standards. EPA notes that despite a
variety of vehicles on the market today and over the past several
years, the fleet has failed to comply with standards based upon
performance beginning with the 2016 model year, and has fallen further
behind in the 2017 model year, when only three major automakers
complied with CO2 emission standards based upon performance
alone.
In response to IPI's comment that the deployment of more fuel-
efficient technologies, combined with manufacturer advertising, will
change consumer preference, this runs counter to historical trends.
Manufacturers have continuously deployed additional fuel efficiency
technology in each model year--which is why EPA continues to see
fleetwide improvements in CO2 emissions on new vehicles. And
manufacturers have consistently advertised the fuel economy performance
of their vehicles. Federal law requires the physical posting fuel
economy performance, as well as estimated and comparative fuel cost
information, on every new vehicle offered for sale. Notwithstanding
this activity, consumer demand, and willingness to pay for technology
that reduces CO2 emissions and improves fuel economy, has not matched
required standards--which is one of the reasons that EPA is revising
the standards today. As discussed in the proposal, EPA recognizes that
the diversity in the automotive customer base, combined with the facts
and analysis developed by the agency in this rulemaking, raises
concerns that the previous standards, if they are not adjusted, may not
continue to fulfill the agency's goal of providing sufficient
manufacturer flexibility to meet consumer needs and consumer choice
preferences in their vehicle purchasing decisions. In the 2012 Final
Rule and the Initial Determination, EPA expected that consumers would
readily accept fuel-saving technologies in their new vehicles, despite
the agency's uncertainty about the role of fuel savings in consumers'
purchase decisions. Given low fuel prices and the pronounced market
shift to crossovers and SUVs, notwithstanding required disclosers of
fuel costs and relative fuel economy performance, EPA now concludes
that it is appropriate to account for the shift in consumer preference
in concluding that the standards set in 2012 did not provide sufficient
lead time for manufacturers to achieve the standards set at that time.
EPA remains concerned that the projected level of hybridization and
other advanced technologies and the associated vehicle costs necessary
to achieve the previous standards are too high from a consumer-choice
perspective, and not sufficiently account for consumer acceptance of
such technology. While consumers have benefited from improvements over
several decades in traditional vehicle technologies, such as
advancements in transmissions and internal combustion engines,
electrification technologies are a departure from what consumers have
traditionally purchased. Strong hybrid and other advanced
electrification technologies have been available for many years (20
years for strong hybrids and eight years for plug-in and all electric
vehicles), and sales levels have been relatively low, in the 2-3
percent range.\2490\ As discussed above, the analysis projects that the
2012 EPA standards would be projected to require a significant increase
in hybridization (up to 8 percent for mild hybrids and 10 percent for
strong hybrids in MY 2030). This large increase in technology demand
over the next decade could lead to automotive companies needing to
change the choice of vehicle types they are able to offer to consumers,
compared to what the companies would otherwise have offered in the
absence of the previously issued standards. As discussed above,
manufacturers are, by and large, not meeting existing standards based
upon actual fleet performance in CO2 emissions and are
instead relying upon the use of earned or acquired credits. As the
previous standards were set to increase significantly through MY 2020
and thereafter, reducing the rate of increase is appropriate and
reasonable. Doing so will provide manufacturers with sufficient lead
time to meet the standards being set today.
---------------------------------------------------------------------------
\2490\ For instance, the 2019 calendar year saw only a 1.4%
penetration of battery electric vehicles in the light duty fleet,
following 1.2% for 2018, 0.6% for 2017, 0.5% for 2016, and 0.4% for
2015. Wards Auto Monthly Sales reports, available at https://wardsintelligence.informa.com/.
---------------------------------------------------------------------------
EPA recognizes that one possibility for automotive companies who
wish to retain their current vehicle offerings, but face compliance
challenges is to purchase GHG emissions credits. In EPA's annual
Automotive Trends Report, EPA has reported that credit trading has
occurred frequently in the past several years to achieve compliance
with the GHG standards.\2491\ Credit trading can lower a manufacturer's
costs of compliance, both for those selling and those purchasing
credits, and this program compliance flexibility is another tool
available to auto firms to allow them to continue offering the types of
vehicles that customers want. Between MY 2010 and MY 2017, these trades
have included 11 firms, with five firms selling CO2 credits
to seven firms.\2492\ The number of firms participating in the GHG
credits market represents about one-half of the automotive companies
selling vehicles in the U.S. market, but since several of these firms
are small players, they represent less than half of the vehicle
production volume. In total, approximately 48 million Megagrams of
CO2 credits have been traded between firms, which represents
19 percent of the MY 2017 industry-wide bank of credits. That said,
more manufacturers have relied upon previously earned credits to
achieve compliance. Between MY 2010 and MY 2017, 80% of firms applied
previously earned credits. However, long-term planning is an important
consideration for automakers, and an automaker who may need to purchase
credits as part of a future compliance strategy is not guaranteed to
find credits. The automotive industry is highly competitive, and firms
may be reluctant to base their future product strategy on an uncertain
future credit availability, but face struggles in achieving
CO2 emission reductions in a manner that meets consumer
expectations for cost, utility, and performance. Also, pools of
available credits continue to decline over time as the standards become
more stringent and previously banked credits are either used or expire;
indeed, this has happened in recent years.\2493\ EPA's views on the
availability of the credit market to aid in manufacturers' compliance
have changed since the Initial Determination. Based upon the
information available to the EPA in early January 2017, the auto
industry had outperformed its standards in the four previous compliance
years (MYs 2012-2015) and EPA had viewed that as
[[Page 25117]]
a positive trend.\2494\ Since then, however, overall manufacturer
performance failed to meet the standard fleetwide, and many
manufacturers relied on credits to meet their individual compliance
targets. Furthermore, recent experience suggests that availability of
the credit bank is becoming a more uncertain means to achieve
compliance.\2495\ Thus, while credit trading may be a useful
flexibility to reduce the overall costs of the program and to smooth
the pathway to compliance realizing necessary transitions from vehicle
redesign cycles, EPA believes it is important to set standards that
preserve consumer choice without relying on credit purchasing
availability as a compliance mechanism. As discussed in Section VII,
the agencies project that the EPA final standards (Alternative 3, 1.5
percent year over year stringency improvement), will require more
realistic penetration of advanced CO2 emission technologies
such as electrification--better ensuring that manufacturers will be
able to provide vehicles that meet consumer demand.
---------------------------------------------------------------------------
\2491\ 2018 EPA Automotive Trends Report at Figures 5.15 and
5.17.
\2492\ EPA Greenhouse Gas Emission Standards for Light-Duty
Vehicles: Manufacturer Performance Report for the 2016 Model Year.
EPA-420-R-18-002. January 2019.
\2493\ 2018 EPA Automotive Trends Report at Figure 5.17 and
Table 5.17.
\2494\ See Initial Determination at page 7-8.
\2495\ Id. at Figure ES-8.
---------------------------------------------------------------------------
(f) Consideration of Safety
As discussed above, EPA has long considered the safety implications
of its emission standards.\2496\ More recently, EPA has considered the
potential impacts of emission standards on safety in past rulemakings
on GHG standards, including the 2010 rule which established the 2012-
2016 light-duty vehicle GHG standards, and the 2012 rule which
previously established 2017-2025 light-duty vehicle GHG standards.
Indeed, section 202(a)(4)(A) specifically prohibits the use of an
emission control device, system or element of design that will cause or
contribute to an unreasonable risk to safety.\2497\ The relationship
between CO2 emissions and safety is more nuanced. Safety
impacts relate to changes in the use of vehicles in the fleet, relative
mass changes, and the turnover of fleet to newer and safer vehicles.
---------------------------------------------------------------------------
\2496\ See, e.g., 45 FR 14496, 14503 (1980) (``EPA would not
require a particulate control technology that was known to involve
serious safety problems.'').
\2497\ 42 U.S.C. 7521(a)(4)(A).
---------------------------------------------------------------------------
The analysis for the final rule projects that there will be a
change in vehicle miles traveled (VMT) under the final standards,
specifically 607 billion less miles traveled compared to the previous
standards case. Based on these projections about reduced VMT in the
light-duty fleet, the analysis estimates that fatalities will be
reduced by 2584 (out of a total impact of 3269) over the lifetime of MY
1977-2029 vehicles compared to the previous CO2
standards.\2498\ In other words, the reduction in fatalities under the
final standards compared to the previous standards is primarily driven
by the modeling's projected changes in VMT and associated changes in
mobility (i.e., people driving less). The details of the safety
assessment are discussed in Section VI of this preamble and in Section
VI of the FRIA. Under alternatives with stringency levels lower than
the final standards, the analysis projects greater reductions in VMT,
and thus projects somewhat greater reductions in fatalities based on
these VMT changes. Under alternatives with stringency levels higher
than the final standards, the analysis projects lower reductions in
VMT, and thus projects fewer fatalities reduced, See Table VI-271.
---------------------------------------------------------------------------
\2498\ The number of fatalities projected is a product of two
contributing factors: the number of miles driven (VMT) and the risk
of driving (i.e., fatalities per mile). Overall in this final rule
analysis, the change in fatalities projected is primarily caused by
the changes in VMT.
---------------------------------------------------------------------------
EPA notes that the magnitude of the changes in fatalities stemming
from changes in mobility projected in this final rule is less than what
was presented in the proposed rule. In response to comments, the
agencies took a conservative approach to modeling the effects of
standard stringency upon safety. The agencies held VMT constant across
alternatives. The reasons for the differences in fatality estimates in
the final rule compared to the proposed rule, including changes to the
modeling inputs and projections based on the agencies' assessment of
public comments.
The approach for reporting fatality impacts for this final rule is
different than the previous analyses for the Initial Determination and
the 2012 rulemaking. First, the analysis quantifies the number of
fatalities caused by changes in VMT between each Alternative and the
previous standards, whereas previous analyses did not. Second, the
safety analysis itself is different from previous analyses that assumed
that automakers would not reduce the weight of approximately the
lightest half of passenger cars--discounting the safety impacts of mass
reduction. Third, while the agencies qualitatively discussed the effect
of price increases attributable to increased stringency on vehicle
sales, fleet turnover, and the improved safety of newer vehicles, the
agencies never attempted to quantify these impacts.
With respect to public comments, the Alliance commented that ``EPA
has discretion to consider all the relevant factors in setting
appropriate emissions standards under Sec. 202(a)(1), including
vehicle safety. Moreover, given NHTSA's greater expertise in evaluating
motor vehicle safety, it is appropriate for EPA to respect the views of
its companion agency on those issues.'' The Alliance commented that
``[t]he new safety analysis likewise provides support for EPA's
conclusion that the MY 2021-2025 GHG standards are not appropriate and
should be reduced in stringency. Indeed, given that the `primary
purpose' of Sec. 202(a)(1) is `the protection of public health and
welfare,' EPA would be abdicating its statutory duty if it ignored
these concerns.''
Global Automakers commented that safety impacts due to the rebound
effect should not be attributed to the standards and should not serve
as a basis for keeping the standards flat. They further argued that the
dynamic scrappage model is flawed and should be removed from the
modeling for purposes of the final rule. They also argued, that
Congress expressed interest in improving efficiency, emissions, and
safety (without no recognition of cost as a factor), and that
therefore, improvement in all such areas should provide that
improvements in efficiency would not lead to negative safety impacts.
CBD and others commented that safety concerns should not be
considered because the record does not indicate that vehicles must be
unsafe to meet the previous standards. They further commented that EPA
cannot justify reduced stringency upon ``rebound'' fatalities, and they
argue that those fatalities cannot be considered by EPA, since they
``stem from voluntary choices by individuals to drive more--not the
`operation or function'of the technologies at issue'' (quoting CAA
Section 202(a)(4)(A)).
Environmental Defense Fund (EDF) similarly commented that the
estimates of fatalities are unsound, as is considering total fatalities
resulting from increased stringency, rather than fatality rates. They
added that the projected fatalities stem from consumer and manufacture
behaviors that are removed from the stringency requirements. They
further argue that considering fatalities that are attributable to the
standards--particularly rebound fatalities--are inappropriate. EDF,
UCS, and Consumers Union argue that fatalities attributable to
increased driving are not relevant to agency decisions.
In response to the Alliance comments, EPA has considered safety, as
described in this section, and agrees that the
[[Page 25118]]
potential impacts of emission standards on safety is an important
consideration in determining appropriate standards under CAA section
202(a). In response to comments from Global Automakers that the safety
analysis in the proposed rule did not support freezing the standards,
EPA agrees that safety considerations alone do not justify such an
approach, and notes that the safety analysis performed for this final
rule has changed from the analysis for the proposed rule based on
consideration of public comments. EPA is finalizing standards that are
more stringent (1.5 percent per year stringency improvement for MY
2021-2026) than the proposed rule's preferred alternative (0 percent
stringency improvement).
Several commenters argued that the proposal's claims of reduced
fatalities were based upon projected changes in driving, arguing that
that EPA should not decide the level of the standards based on these
assumed changes in travel. As discussed above, EPA acknowledged that
the reduction in fatalities under the final standards compared to the
previous standards are in large part driven by projected changes in
driving behavior (i.e., people driving less). While EPA is not seeking
to restrict mobility or driving, ignoring impacts associated with this
rule would be inappropriate. Moreover, the provisions of Section
202(a)(4) do not preclude EPA from considering such impacts. While EPA
has considered the safety assessment for this final rule, as discussed
in the following section below, safety was one of several factors
considered in deciding on the level of today's final standards.
g) Consideration of Energy Security Impacts
Among other factors EPA considered in selecting the previous
standards in the 2012 Final Rule was the effect of the standards on
U.S. petroleum imports and energy security.\2499\ As discussed in the
PRIA, Final RIA and in Section Energy Security, the energy security
position of the United States has changed dramatically since 2012. The
U.S. has become a net exporter of petroleum and additional payments by
United States consumers resulting from upward pressure on oil price due
to additional demand are a transfer that occurs within the United
States economy.\2500\ Additional petroleum use necessarily increases
demand and thus subjects the nation to additional risk of price shocks,
but this risk is significantly reduced as the United States has
dramatically increased domestic petroleum production and has additional
capacity to do so. Accordingly, energy security concerns are reduced
compared to the assessment in the 2012 rulemaking and do not alter
EPA's selection of final revised standards in this rule.
---------------------------------------------------------------------------
\2499\ See 77 FR 62938, et seq.
\2500\ The U.S. Energy Information Administration EIA estimates
that the United States exported more total crude oil and petroleum
products in September and October 2019, and expects the United
States to continue to be a net exporter. See Short Term Energy
Outlook November 2019, available at https://www.eia.gov/outlooks/steo/archives/nov19.pdf.
---------------------------------------------------------------------------
(h) Balancing of Factors and EPA's Revised Standards for MY 2021 and
Later
As discussed in this section, the Administrator is required to
consider a number of factors when establishing emission standards under
section 202(a)(2) of the Clean Air Act: The standards ``shall take
effect after such period as the Administrator finds necessary to permit
the development and application of the requisite technology, giving
appropriate consideration to the cost of compliance within such
period.'' \2501\ For this Final Rule, the Administrator has considered
a wide range of potential emission standards (Baseline/No Action
Alternative and Alternatives 1 through 7), ranging from the previous
EPA standards (Baseline/No Action Alternative), through a number of
less stringent alternatives, including the proposed preferred
alternative (Alternative 1, 0 percent per year stringency improvement)
and what has been chosen as the final standards (Alternative 3, 1.5
percent per year stringency improvement). The Administrator has
determined that the revised final standards, which would increase the
stringency of the MY 2020 standards by 1.5 percent per year for both
passenger cars and light-trucks from MY 2021 through 2026, are
appropriate under section 202(a) of the CAA. In addition to
technological feasibility, lead-time, and the costs of compliance, the
Administrator has also considered the impact of the standards on GHG
and non-GHG emissions reductions, the costs to consumers, and vehicle
safety.
---------------------------------------------------------------------------
\2501\ 42 U.S.C. 7521(a)(2).
---------------------------------------------------------------------------
In addition to comments on each of the factors the Administrator
considered discussed above, comments also were received on how the
Administrator should balance these factors in determining the
appropriate final standards.
The Alliance commented that the CAA provides EPA with significant
latitude to exercise its expert judgment in determining the level at
which emissions standards should be set. The Alliance commented further
that unlike other CAA provisions, Sec. 202(a)(1) does not require EPA
to set standards that will result in the greatest degree of emission
reduction achievable. Instead, the statute leaves EPA flexibility to
decide what factors are relevant, and how to weigh those factors, in
its decision-making process. The Alliance also commented ``EPA also has
'significant latitude' regarding the 'coordination of its regulations
with those of other agencies,' '' ``EPA has discretion to defer to the
judgment of other agencies regarding issues within their areas of
expertise,'' and the CAA ``gives the agency authority to engage in
reasoned decision-making, balancing all of the relevant factors in
light of the available facts. EPA has done that here and has provided a
reasoned explanation of its determination that the environmental
benefits of the existing MY 2021-2025 GHG standards are outweighed by
their negative effects on costs and safety.''
The American Iron and Steel Institute commented that it favors the
general direction taken in the SAFE proposal, including the preferred
option for CO2 standards, and that it believes a final SAFE
rule that ``balances the priorities of costs to consumers, safety
design considerations, employment impacts and total GHG emissions will
result in the best outcome.''
CBD and others claimed that the justifications EPA offered in the
notice are untethered from the statute, and that EPA used a flawed
analysis. Further, they claim that EPA did not exercise its own
judgment and delegated its responsibilities impermissibly to NHTSA,
failing to consider ``relevant EPA information.''
EPA's analysis is described in detail in this preamble. EPA decided
to use the CAFE model for a number of reasons, described in more detail
in Section IV, including that using two models results in an
inefficient use of resources, the CAFE model can analyze both EPA's and
NHTSA's statutory programs, the CAFE model is capable of modeling
incremental improvements of discrete technologies, and EPA believes
that the CAFE model provides reasonable results. Merely because EPA has
a set of its own analytical tools that model similar effects does not
mean that it must use those tools to perform the analysis, and doing so
would create unnecessary complication and lead to potential
inconsistencies. Since the agencies are establishing standards jointly
and seeking to avoid
[[Page 25119]]
inconsistencies in a manner consistent with Supreme Court direction,
using the same model for the analysis is reasonable. Nonetheless, EPA
has exercised its own judgment in this final rule.
The California Attorney General and others claim that EPA failed
adequately to acknowledge, explain, or justify its departure from the
prior determination. They claim that EPA failed to propose or make a
finding required by Section 202(a)(2) relating to adequate lead time,
inconsistent with EPA's prior explanation that it is provided with
limited flexibility in making such a determination.
The California Attorney General and others also claim that EPA's
analysis improperly weighs the factors it considers, and that it
insufficiently weighed certain factors required under the Clean Air
Act, including air pollution. In response, EPA notes that the Clean Air
Act does not specify how the Administrator should weigh the factors
considered, as discussed elsewhere in this section.
The California Attorney General and others further noted that the
purpose of the Clean Air Act is to is to ``protect and enhance the
quality of the Nation's air resources so as to promote the public
health and welfare and the productive capacity of its population.''
The Institute for Policy Integrity claimed that the agencies
balanced the factors in a way that conflicts with their controlling
statutes and weighed the statutory factors without regard for the
accuracy of the accompanying cost-benefit analysis.
The National Coalition for Advanced Transportation claimed that the
proposal appeared to be based on heightened concerns with cost,
consumer acceptance, and safety, and insufficiently on technology
availability and emissions reductions. As discussed in this section,
EPA is neither relying solely on cost or safety nor ignoring any
factors, but rather is balancing a number of factors.
Green Energy Institute at Lewis and Clark Law School et al.
commented that the Clean Air Act does not authorize the weakening or
freezing of existing standards due to industry costs or consumer
preferences. While EPA has broad discretion to revise standards based
upon a balancing of factors, the final rule will provide for increasing
stringency of 1.5 percent per year from MY 2021 through MY 2026.
Motor & Equipment Manufacturers Association (MEMA) commented that
the technology costs from their preferred alternative (Alternative 8 in
the notice) were not significant and did not justify holding MY 2020
standards flat in light of other elements, such as preserving
investments in fuel saving technology. EPA disagrees, and considers the
reductions in costs resulting from the revised final standards, $1,250
per vehicle by MY 2029, to be one important aspect of the justification
of these standards.
EPA believes the previously issued standards for MY 2021 and later,
considered as a whole, are too stringent. Factors in favor of reduced
stringency include manufacturer compliance costs, and the related per-
vehicle cost savings. As described above, the agencies project that the
final CO2 standards will reduce manufacturers' MY 2018-2029
compliance costs by $108 billion (when applying a 3% discount rate),and
will reduce average MY 2030 vehicle prices $977 (also applying a 3%
discount rate). Including other costs, such as financing and insurance,
consumers the standards finalized today will result in reduced costs of
$1,286 per-vehicle for a MY 2030 vehicle. EPA expects that the final
standards will not impede consumers from being able to purchase a new
vehicle of their choice or require significant changes in product lines
for any manufacturer. In fact, under the final standards, vehicle sales
are expected to increase by 2.2 million vehicles over MY 2017-2029
compared to projected sales under the augural standards, a significant
increase in vehicles sold over this timeframe see Table VI-155. EPA
views this projection of vehicle sales increases resulting from the
final standards as important in facilitating the turnover of the fleet
to newer, safer vehicles, all of which will be subject to increasingly
stringent criteria pollutant emission requirements as federal Tier 3
emission standards continue to phase in from MY 2017 through MY 2025.
Another factor weighing toward reduced stringency is safety. As
discussed previously, reduced stringency results in less pressure on
manufacturers to reduce mass in vehicles, which, for smaller passenger
cars has negative safety implications when involved in accidents with
heavier vehicles. Further, as vehicle prices decrease compared to the
previous standards, more consumers will be able to afford newer
vehicles, which are significantly safer. Lastly, as vehicles will not
be required to be as fuel efficient as under the previous standards,
``rebound'' driving will be reduced. The agencies project a reduction
in 605 billion miles traveled by light-duty vehicles produced through
MY 2029, and project that this reduced VMT will lead to 2,584 fewer
highway fatalities under the final standards compared to the previous
CO2 standards (i.e., people are projected to drive less
under the final standards with an associated reduction in driving-
related fatalities). While, notwithstanding EPA's involvement with
State and local Transportation Control Measures (TCMs), the
Administrator does not seek to change the way people drive--EPA's
intention is not to restrict mobility, or to discourage driving, based
on the level of the standards--EPA nonetheless believes it is
appropriate to consider this projection.\2502\ The agencies also
project that accelerated fleet turnover attributable to the change in
standards will lead to the avoidance of a further 447 fatalities, and
that the reduced need for reductions of vehicle mass will lead to the
avoidance of a further 238 fatalities. In other words, the agencies
project that the change in CO2 standards will lead to 3,269 fewer
fatalities over the useful lives of vehicles produced through MY 2029.
---------------------------------------------------------------------------
\2502\ Information regarding TCMs is available at https://www.epa.gov/statelocalenergy/transportation-control-measures.
---------------------------------------------------------------------------
Factors that weigh in favor of increased stringency options are
increased upstream criteria pollutant emissions attributable to
additional refining and other fuel-related activities, as well as
increased CO2 emissions and consumer fuel expenditures.
As described above, the agencies project that the revised final
standards will have a negative impact on air quality health outcomes,
including a projected increase of 444 to 1,000 premature deaths from
increased air pollution over the lifetime of the MY 1977-2029 vehicles
on the road after calendar year 2017 cumulative through CY 2068, under
EPA's CO2 program.\2503\ EPA recognizes that the final
standards are projected to increase CO2 emissions compared
to the previous EPA standards. However, EPA notes that, unlike other
provisions in Title II referenced above, section 202(a) does not
require EPA to set standards for light-duty vehicles which result in
the ``greatest degree of emission reduction achievable.'' EPA has not
chosen the standard that has the highest estimated net social benefits.
However, as discussed elsewhere in this preamble, from a cost benefit
perspective, the differences among the various alternatives are
relatively narrow. EPA believes consideration of costs and benefits is
certainly relevant to its
[[Page 25120]]
exercise of discretion in selecting appropriate standards, but also
recognizes that some costs and benefits are difficult to quantify, and
additional factors can prove material under the Clean Air Act as well
in those policy decisions. For example, EPA notes that the agency
decided against pursuing more stringent alternatives analyzed in both
the rulemaking establishing 2012-2016 standards and the rulemaking
establishing 2017-2025 standards.
---------------------------------------------------------------------------
\2503\ The agencies believe that these premature mortality
estimates may be over-estimated. Please see more detailed
discussions in Sections VI.D.3.d) and VIII.A.3.d) in this preamble,
and similar discussions in the final regulatory impact analysis.
---------------------------------------------------------------------------
EPA has also given weight to the policy goal of establishing
CO2 standards which are coordinated with NHTSA's CAFE
standards. While not a statutory requirement, EPA has considered the
importance of having coordinated and harmonized EPA CO2 and
CAFE programs, while recognizing the different statutory authorities
for those programs, since the establishment of the EPA CO2
program. The agencies discussed the importance of having one national
program in the SAFE Vehicles Part 1 joint action.\2504\ In today's
joint final rule, DOT is establishing CAFE standards for MY 2021-2026
which increase in stringency at a level of 1.5 percent per year. The
revised EPA standards will also increase in stringency at a rate of 1.5
percent per year. Coordinating revisions to the GHG and CAFE standards
in order to maintain one national program is a factor the Administrator
has consideration in determining the revised GHG standards.
---------------------------------------------------------------------------
\2504\ 84 FR 51,310 (Sept. 27, 2019).
---------------------------------------------------------------------------
In light of available statutory discretion and the range of factors
that the statute authorizes and permits the Administrator to consider,
and his consideration of the factors discussed above, the EPA concludes
that reducing the stringency of the MY 2021-2026 standards is an
appropriate approach under section 202(a). Therefore, based on the data
and analysis detailed in this final rule, the Administrator concludes
that the previous MY 2021 and later CO2 standards are too
stringent, and is establishing revised standards for MY 2021 through MY
2026 at a level of 1.5 percent per year improvement in stringency.
In response to comments concerned about EPA's proposal to freeze
the MY 2021-2026 standards at MY 2020 levels, EPA notes that it is
finalizing the 1.5 percent per year improvement in stringency level and
not the 0 percent improvement level proposed, after considering the
somewhat higher costs to industry and up-front vehicle costs to the
consumer and slightly lower GHG emissions and health-related impacts
compared to the proposed preferred alternative. The Administrator has
taken these tradeoffs into account in his balancing of factors under
section 202(a) of the CAA.
While the set of factors considered by EPA under section 202(a) of
the CAA in today's final rule and under the midterm evaluation
regulations \2505\ in the Initial Determination are similar and
overlapping, the Administrator recognizes that he is balancing these
factors differently in this final rule than in the Initial
Determination. In the Initial Determination, EPA's decision that the
previous MY 2022-2025 standards were appropriate was based on
conclusions that the standards were feasible within the lead time
provided at reasonable costs, the standards would result in significant
reductions in GHG emissions and oil consumption and associated fuel
savings for consumers, and the standards would yield significant
benefits to public health and welfare and positive net benefits
overall, without adverse impacts on industry, safety, or
consumers.\2506\
---------------------------------------------------------------------------
\2505\ 40 CFR 86.1818-12(h).
\2506\ Initial Determination, Section III, page 29-30.
---------------------------------------------------------------------------
Since the Initial Determination, EPA has completed its compliance
review of the first two model years covered by the 2012 final rule.
Notwithstanding widespread availability of vehicles that meet or exceed
their CO2 emission targets, consumers are not expressing
sufficient interest in fuel economy in their purchasing decisions to
enable manufacturers to meet the standards based upon fleet
performance. Although manufacturers earned significant credits in the
early years of the agency's CO2 regulation history, these
credits are being applied broadly across the industry and well in
advance of the more aggressive model year stringency increases. While
some manufacturers, including alternative fuel automakers are earning
significant tradable credits, they do not have to trade them. And
building a program around the potential for acquiring credits from
competing manufacturers is not the intention of this action. While EPA
is analyzing the differences between these standards and the previous
standards for this rulemaking, EPA cannot ignore that this rulemaking
was foreseen in the 2012 rulemaking. The prospect of revising the
standards was expressly envisioned in that rulemaking based upon the
uncertainty in the assumptions and future projections at that time.
When viewed from the perspective of the larger set of MY 2017 through
MY 2026 standards rulemakings, the standards finalized today fit the
pattern of gradual, tough, but feasible stringency increases that take
into account real world performance, shifts in fuel prices, and changes
in consumer behavior toward crossovers and SUVs and away from more
efficient sedans. This approach ensures that manufacturers are provided
with sufficient lead time to achieve standards, considering the cost of
compliance.
In this final rule, the EPA is placing greater weight on the costs
to industry and the up-front vehicle costs to consumers. EPA believes
that the costs to both industry and automotive consumers would have
been too high under the previous standards, and that the standards
should be revised to be less stringent to lower these costs. EPA
believes that by lowering the auto industry's costs to comply with the
program, with a commensurate reduction in per-vehicle costs to
consumers, the final rule is enhancing the ability of the fleet to turn
over to newer, cleaner and safer vehicles.
EPA believes that the characteristics and impacts of these and
other alternative standards generally reflect a continuum in terms of
technical feasibility, cost, lead time, consumer impacts, emissions
reductions, and oil savings, and other factors evaluated under section
202(a). In determining the appropriate standard to adopt in this
context, EPA judges that the final standards are appropriate and
preferable to more stringent alternatives based largely on
consideration of cost--both to manufacturers and to consumers--and the
potential for overly aggressive penetration rates for advanced
technologies relative to the penetration rates seen in the final
standards, especially in the face of an unknown degree of consumer
acceptance of both the increased costs and of the technologies
themselves--particularly given current projections of fuel prices
during that timeframe. At the same time, the final rule helps to
address these issues by maintaining incentives to promote broader
deployment of advanced technologies, and so provides a means of
encouraging their further penetration while leaving manufacturers
alternative technology choices. EPA thus judges that more stringent
alternatives, which would necessitate even more technology and more
cost, would not be appropriate. Instead, EPA is adopting a more gradual
increase in stringency to ensure that the benefits of reduced GHG
emissions are achieved without the potential for disruption to
automakers or consumers.
[[Page 25121]]
B. NHTSA's Statutory Obligations and Why the Selected Standards Are
Maximum Feasible as Determined by the Secretary
In this section, NHTSA discusses the factors, data and analysis
that the agency has considered in the selection of the CAFE standards
for MYs 2021 and later and the comments received on NHTSA's
consideration of these factors (see further discussion below on NHTSA's
summary and analysis of comments).
As discussed in more detail below, the primary purpose of EPCA, as
amended by EISA, and codified at 49 U.S.C. chapter 329, is energy
conservation, and fuel economy standards help to conserve energy by
requiring automakers to make new vehicles travel a certain distance on
a gallon of fuel.\2507\ The goal of the CAFE standards is to conserve
energy, while taking into account the statutory factors set forth at 49
U.S.C. 32902(f), as discussed below.
---------------------------------------------------------------------------
\2507\ While individual vehicles need not meet any particular
mpg level, as discussed extensively elsewhere in this preamble, it
is broadly true that fuel economy standards require vehicle
manufacturers' fleets to meet certain fuel economy levels as set
forth by NHTSA in regulation.
---------------------------------------------------------------------------
49 U.S.C. 32902(f) states when setting maximum feasible CAFE
standards for new vehicles, the Secretary of Transportation \2508\
``shall consider technological feasibility, economic practicability,
the effect of other motor vehicle standards of the Government on fuel
economy, and the need of the United States to conserve energy.'' In
previous rulemakings, including the 2012 final rule that established
CAFE standards for MY 2021 and set forth augural standards for MYs
2022-2025, NHTSA considered technological feasibility, including the
availability of various fuel-economy-improving technologies to be
applied to new vehicles in the timeframe of the standards depending on
the ultimate stringency levels, and also considered economic
practicability, including the differences between a range of regulatory
alternatives in terms of effects on per-vehicle costs, industry-wide
costs, the ability of both the industry and individual manufacturers to
comply with standards at various levels, as well as effects on vehicle
sales, industry employment, and consumer demand. NHTSA also considered
how compliance with other motor vehicle standards of the Government
might affect manufacturers' ability to meet CAFE standards represented
by a range of regulatory alternatives, and how the need of the U.S. to
conserve energy could be more or less met under a range of regulatory
alternatives, in terms of considerations like costs to consumers, the
national balance of payments, environmental implications like climate
and smog effects, and foreign policy effects like the likelihood that
U.S. military and other expenditures could change as a result of more
or less oil consumed by the U.S. vehicle fleet. These elements are
discussed in detail throughout this analysis. As will be discussed in
greater detail below, while NHTSA is considering all of the same
factors in setting today's CAFE standards that it considered in
previous rulemakings, and in many instances in a similar way as it
considered those factors in previous rulemakings, the facts on the
ground have changed and NHTSA is therefore choosing to set CAFE
standards at a different level from what the 2012 final rule set forth.
---------------------------------------------------------------------------
\2508\ By delegation, NHTSA.
---------------------------------------------------------------------------
NHTSA is not limited to consideration of the factors specified in
49 U.S.C. 32902(f) when establishing CAFE standards for passenger cars
and light trucks. In addition to the factors enumerated above, NHTSA
may (and historically has) considered such factors as safety and the
environment.
NHTSA also considers relevant case law. Critical to this series of
joint rulemakings with EPA, the Court in Massachusetts v. EPA,\2509\
recognized EPA's argument that ``it cannot regulate carbon dioxide
emissions from motor vehicles'' without ``tighten[ing] mileage
standards . . . .''--a task assigned to DOT. The Court found that
``[t]he two obligations may overlap, but there is no reason to think
the two agencies cannot both administer their obligations and yet avoid
inconsistency.'' \2510\ Accordingly, the agencies have worked closely
together in setting standards, and many of the factors that NHTSA
considers to set maximum feasible standards overlap with factors that
EPA considers under the Clean Air Act. Just as EPA considers energy use
and security, NHTSA considers these factors when evaluating the need of
the nation to conserve energy, as required by EPCA. Just as EPA
considers technological feasibility, the cost of compliance,
technological cost-effectiveness and cost and other impacts upon
consumers, NHTSA considers these factors when weighing the
technological feasibility and economic practicability of potential
standards. EPA and NHTSA both consider implications of the rulemaking
on CO2 emissions as well as criteria pollutant emissions.
And, NHTSA's role as a safety regulator inherently leads to the
consideration of safety implications when establishing standards. The
balancing of competing factors by both EPA and NHTSA are consistent
with each agency's statutory authority and recognize the overlapping
obligations the Supreme Court pointed to in directing collaboration.
NHTSA also considers the Ninth Circuit's decision in Center for
Biological Diversity v. NHTSA \2511\ which remanded NHTSA's 2006 final
rule establishing standards for MYs 2008-2011 light trucks and
underscored that ``the overarching purpose of EPCA is energy
conservation.''
---------------------------------------------------------------------------
\2509\ 549 U.S. 497, 531 (2007).
\2510\ Id. at 532.
\2511\ 538 F.3d 1172 (9th Cir. 2008).
---------------------------------------------------------------------------
The proposed rule presented an analysis of a wide range
alternatives as potential revisions of the existing standards for model
year 2021 and new standards for model years 2022-2026. These
alternatives ranged from a zero percent increase in stringency to a
stringency increase for passenger cars of 2 percent per year and for
light trucks of 3 percent per year, in addition to the baseline
alternative consisting of the augural standards.\2512\ The analysis
supported the range of alternative standards based on factors relevant
to NHTSA's exercise of its 49 U.S.C. 32902(f) authority, such as fuel
saved and emissions reduced, the technologies available to meet the
standards, the costs of compliance for automakers and their abilities
to comply by applying technologies, the impact on consumers with
respect to cost and vehicle choice, and effects on safety. The proposed
rule identified the alternative composed of a zero percent increase in
stringency as the preferred alternative.
---------------------------------------------------------------------------
\2512\ 83 FR 42990, Table I-4 (Aug. 24, 2018).
---------------------------------------------------------------------------
NHTSA received numerous public comments on the range of stringency
alternatives in the proposed rule and NHTSA's consideration of various
factors in determining maximum feasible CAFE standards under 49 U.S.C.
chapter 329. Below NHTSA responds to comments on these issues. NHTSA
notes that many comments concerned the technical foundation and
analysis upon which NHTSA was basing its regulatory decisions, such as
the modeling of fuel economy-improving technologies and costs, the
safety analysis, and consumer issues. Comments specific to these
analyses are discussed elsewhere in this preamble. The section below
addresses comments specifically addressing NHTSA's considerations in
finalizing maximum
[[Page 25122]]
feasible CAFE standards under 49 U.S.C. chapter 329.
NHTSA's conclusion, after consideration of the factors described
below, public comments, and other information in the administrative
record for this action is that 1.5 percent annual increases in
stringency from the MY 2020 standards through MY 2026 (Alternative 3 of
this final rule analysis) \2513\ are maximum feasible. Holding CAFE
standards for MY 2020 flat through MY 2026, as proposed, would unduly
weigh economic practicability concerns more heavily than the need of
the United States to conserve energy, while finalizing the MY 2021 and
augural standards first established and set forth in 2012 would place
undue weight on the need of the U.S. to conserve energy while being
beyond economically practicable, as described in more detail below.
---------------------------------------------------------------------------
\2513\ The numbered Alternatives presented in the SAFE proposed
rule (see Table I-4 at 83 FR 42990, August 24, 2018) were in some
cases defined differently than those presented in this final rule
(see Section V). Unless otherwise stated, the Alternatives described
in this section refer to those presented in this final rule.
---------------------------------------------------------------------------
The following sections discuss in more detail the statutory
requirements and considerations involved in NHTSA's determination of
maximum feasible CAFE standards, comments received on those issues, and
NHTSA's explanation of its balancing of factors for this final rule.
1. EPCA, as Amended by EISA
EPCA, as amended by EISA, contains a number of provisions regarding
how to set CAFE standards. DOT (by delegation, NHTSA) \2514\ must
establish separate CAFE standards for passenger cars and light trucks
\2515\ for each model year,\2516\ and each standard must be the maximum
feasible that the Secretary (again, by delegation, NHTSA) believes the
manufacturers can achieve in that model year.\2517\ In determining the
maximum feasible level achievable by the manufacturers, EPCA requires
that NHTSA consider four statutory factors of technological
feasibility, economic practicability, the effect of other motor vehicle
standards of the Government on fuel economy, and the need of the United
States to conserve energy.\2518\ In addition, NHTSA has the authority
to consider (and traditionally does) other relevant factors, such as
the effect of the CAFE standards on motor vehicle safety and consumer
preferences.\2519\ The ultimate determination of what standards can be
considered maximum feasible involves a weighing and balancing of
factors, and the balance may shift depending on the information before
NHTSA about the expected circumstances in the model years covered by
the rulemaking. The agency's decision must also support the overarching
purpose of EPCA, energy conservation, while balancing these
factors.\2520\
---------------------------------------------------------------------------
\2514\ EPCA and EISA direct the Secretary of Transportation to
develop, implement, and enforce fuel economy standards (see 49
U.S.C. 32901 et. seq.), which authority the Secretary has delegated
to NHTSA at 49 CFR 1.95(a).
\2515\ 49 U.S.C. 32902(b)(1) (2007).
\2516\ 49 U.S.C. 32902(a) (2007).
\2517\ Id.
\2518\ 49 U.S.C. 32902(f) (2007).
\2519\ Both of these additional considerations also can be
considered part of economic practicability, but NHTSA also has the
authority to consider them independently of that statutory factor.
\2520\ Center for Biological Diversity v. NHTSA, 538 F. 3d 1172,
1197 (9th Cir. 2008) (``Whatever method it uses, NHTSA cannot set
fuel economy standards that are contrary to Congress's purpose in
enacting the EPCA--energy conservation.'').
---------------------------------------------------------------------------
Besides the requirement that the standards be maximum feasible for
the fleet in question and the model year in question, EPCA/EISA also
contain several other requirements, as explained below.
(a) Lead Time
EPCA requires that NHTSA prescribe new CAFE standards at least 18
months before the beginning of each model year.\2521\ Thus, if the
first year for which NHTSA is proposing to set new standards in this
NPRM is MY 2022, NHTSA interprets this provision as requiring the
agency to issue a final rule covering MY 2022 standards no later than
April 1, 2020.
---------------------------------------------------------------------------
\2521\ 49 U.S.C. 32902(a) (2007).
---------------------------------------------------------------------------
For amendments to existing standards, EPCA requires that if the
amendments make an average fuel economy standard more stringent, at
least 18 months of lead time must be provided.\2522\ EPCA contains no
lead time requirement to amend standards if the amendments make an
average fuel economy standard less stringent. NHTSA therefore
interprets EPCA as allowing amendments to reduce a standard's
stringency up until the beginning of the model year in question. In the
NPRM, NHTSA proposed to amend the standards for model year 2021. NHTSA
explained that since the agency was proposing to reduce these
standards, the action was not subject to a lead time requirement.
---------------------------------------------------------------------------
\2522\ 49 U.S.C. 32902(g)(2) (2007).
---------------------------------------------------------------------------
The States and Cities commenters argued that NHTSA had counted 18
months incorrectly, and that ``18 months prior to September 1, 2021 is
in fact March 1, 2020.'' \2523\ NHTSA agrees that 18 months prior to
September 1 would be March 1 of the year prior; the statement in the
NPRM that ``NHTSA has consistently interpreted the ``beginning of the
model year'' as September 1 of the CY prior'' was a typographical
error. As prior Federal Register notices indicate, NHTSA has in fact
long interpreted the beginning of the model year for CAFE compliance
purposes as October 1 of the CY prior.\2524\ Thus, counting backwards,
18 months prior to October 1 is properly identified as April 1, meaning
that new standards for MY 2022 must be established by April 1, 2020.
---------------------------------------------------------------------------
\2523\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 78, fn. 211.
\2524\ See, e.g., 75 FR 25546 (May 7, 2010).
---------------------------------------------------------------------------
With regard to the amendments to the MY 2021 standards, a coalition
of environmental groups commented that NHTSA's legal construction of
EPCA's lead time requirement as not applying to MY 2021 was ``not . . .
permissible,'' arguing that section 32902(g)(1) only permits amendments
to existing CAFE standards that ``meet[ ] the requirement of subsection
(a) or (d) as appropriate,'' and that section 32902(a) requires fuel
economy standards to be prescribed 18 months before the beginning of
the model year.\2525\ The environmental group coalition therefore
argued that the two identified provisions must be read together to
compel all amendments to standards to be prescribed at least 18 months
before a model year, and concluded that because it was impossible to
finish a final rule 18 months before the start of MY 2021, that MY 2021
standards could not be amended.\2526\ The States and Cities group
provided similar comments, arguing that NHTSA's interpretation of
(g)(2) rendered the reference in (g)(1) to (a) ``a nullity,'' and that
the ``as appropriate'' language in (g)(1) referred to the determination
of whether providing 18 months of lead time was appropriate, rather
than to whether (a) or (d) was the relevant provision governing the
standards in question.\2527\ NCAT commented that ``Congress in Sec.
32902 has indicated that at least 18 months of lead time are
appropriate when setting standards,'' and stated that ``Manufacturers'
need for adequate lead time when designing products and developing
compliance strategies is the same regardless of whether the agency
[[Page 25123]]
is making standards more stringent, less stringent, or simply changing
the structure or compliance options provided under the standards.''
\2528\ NADA, in contrast, argued that NHTSA does ``have the authority
and discretion to reopen the MY 2021 standards,'' and that the
``mandate for at least 18 months of lead time before new standards may
take effect does not apply to instances, such as for MY 2021, where
standards are being relaxed.'' \2529\ CEI also agreed with NHTSA's
interpretation of lead time set forth in the NPRM.\2530\
---------------------------------------------------------------------------
\2525\ Center for Biological Diversity, Conservation Law
Foundation, Earthjustice, Environmental Defense Fund, Environmental
Law and Policy Center, Natural Resources Defense Council, Public
Citizen, Sierra Club, Union of Concerned Scientists (hereafter,
``environmental group coalition''), Appendix A, NHTSA-2018-0067-
12000, at 66.
\2526\ Id.
\2527\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 78-79.
\2528\ NCAT, NHTSA-2018-0067-11969, at 46.
\2529\ NADA, NHTSA-2018-0067-12064, at 9.
\2530\ CEI, NHTSA-2018-0067-12015, at 3-4.
---------------------------------------------------------------------------
NHTSA agrees that section 32902(g)(1) states that amendments must
meet the requirements of subsection (a) or (d) as appropriate, and that
32902(a) states that standards must be prescribed 18 months in advance
of the model year. However, NHTSA cannot agree that the 18-month lead
time requirement applies to amendments to existing standards that
reduce stringency. Section 32902(g)(2) clearly states that ``[w]hen the
Secretary of Transportation prescribes an amendment under this section
that makes an average fuel economy standard more stringent (emphasis
added), the Secretary shall prescribe the amendment . . . at least 18
months before the beginning of the model year to which the amendment
applies.'' Commenters' construction of the statute would render
superfluous the words ``more stringent'' in 32902(g)(2), and there is a
presumption against superfluity.\2531\ Congress purposely included the
words ``more stringent'' in order to exclude the contrary situation--
``less stringent''--from the 18-month lead time requirement. A plain
reading of (g)(1) simply provides that the Secretary (by delegation,
NHTSA) should refer to the correct provision depending on whether the
standard being amended is generally applicable (pointing to section
(a)) or a standard applicable to low-volume manufacturer pursuant to an
exemption (pointing to section (d)). Reading (g)(1) and (g)(2) together
is the appropriate way to give effect to both provisions. This reading
provides that NHTSA may amend the MY 2021 standard by following the
requirements for generally-applicable standards; this reading also
provides that 18 months' lead time is only required for amendments that
increase stringency. NHTSA also does not agree that (g)(1) can be read
to imply that the agency must provide 18 months of lead time ``if
appropriate,'' as the States and Cities suggest, nor that there is any
statutory basis to extend the lead time requirement to changes to the
``structure or compliance options provided under the standards'' as
NCAT suggests. If new off-cycle technologies could not be recognized
toward compliance without providing 18 months' lead time, manufacturer
efforts to rely on that compliance flexibility to redress past
shortfalls would be frustrated.
---------------------------------------------------------------------------
\2531\ See, e.g., Duncan v. Walker, 533 U.S. 167 (2001) (citing
U.S. v. Menasche, 348 U.S. 528, 538-539 (1955)).
---------------------------------------------------------------------------
Moreover, automakers need more time to respond when NHTSA amends
standards to be more stringent--doing so would likely require
automakers to change their product and/or sales plans to ensure that
they will meet more-stringent standards than those standards for which
they may have already prepared. But such product or sales plans would
not necessarily need to be changed if standards were amended to be less
stringent--in fact an automaker would be rewarded by keeping existing
plans to comply in place with additional bankable and tradable
overcompliance credits. However, the environmental group coalition
argued that ``[c]hanging the MY 2021 standard at this late date would
penalize technologically advanced automakers and parts suppliers, who
have already made significant investments in updating their
technology.'' \2532\ The States and Cities group made similar
comments,\2533\ as did NCAT.\2534\ The environmental group coalition
further suggested that amending the MY 2021 standard would reduce the
need for (and thus the value) of overcompliance credits, ``which would
be disruptive to the manufacturers that have done the most to further
EPCA's conservation goals.'' \2535\ NCAT made similar comments, arguing
that ``The practical and financial impact of the change accordingly is
not materially different from increasing the stringency of a standard
this late in the product cycle.'' \2536\
---------------------------------------------------------------------------
\2532\ Environmental group coalition, NHTSA-2018-0067-12000,
Appendix A, at 66.
\2533\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 78, fn. 213.
\2534\ NCAT, NHTSA-2018-0067-11969, at 46-47.
\2535\ Environmental group coalition, NHTSA-2018-0067-12000,
Appendix A. at 66-67.
\2536\ NCAT, NHTSA-2018-0067-11969, at 47.
---------------------------------------------------------------------------
NHTSA believes that to the extent that some manufacturers have
already invested in future fuel economy improvements, those
manufacturers will continue to be well-positioned both to respond to
increasing standards in the future, and to take advantage of any market
demand for higher fuel economy/reduced tailpipe CO2
emissions from consumers who put a premium on those aspects. NHTSA is
also aware that several companies have self-imposed emissions-reduction
goals which may drive their decisions on technology application
regardless of regulatory obligations. NHTSA does not believe that
companies which have already invested in higher levels of technology
consider those investments to be bad ones. The agencies note that
manufacturer commenters, despite the concerns expressed by others, did
not comment about a lack of lead time associated with changing the MY
2021 standards; rather, many manufacturer commenters expressly cited
the need to revise MY 2021 standards, arguing that the previously-
established values are beyond maximum feasible. Regarding the value of
overcompliance credits under more or less stringent standards, NHTSA
agrees that the need for credits may be less under less stringent
standards, but this is true regardless of the lead time question.
Further, NHTSA does not believe that this suggests only standards that
compel reliance on overcompliance credits (especially those earned by
competitors) can be maximum feasible; this topic will be addressed in
further detail below, and regardless, NHTSA is prohibited from
considering credit availability in determining maximum feasible CAFE
standards.
(b) Separate Standards for Cars and Trucks, and Minimum Standards for
Domestic Passenger Cars
As discussed above, EPCA requires NHTSA to set separate CAFE
standards for passenger cars and light trucks for each model
year.\2537\ NHTSA interprets this requirement as preventing the agency
from setting a single combined CAFE standard for cars and trucks
together, based on the plain language of the statute. Congress
originally required separate CAFE standards for cars and trucks to
reflect the different fuel economy capabilities of those different
types of vehicles,\2538\ and over the history of the CAFE program, has
never revised this requirement. Even as many cars and trucks have come
to resemble each other more closely over time--many crossover and
sport-utility models, for example, come in versions today that may be
subject to either the car standards or the truck standards depending on
their characteristics--it is still accurate to say that vehicles with
[[Page 25124]]
truck-like characteristics such as 4 wheel drive, cargo-carrying
capability, etc., consume more fuel per mile than vehicles without
these characteristics. Thus, NHTSA believes that the different fuel
economy capabilities of cars and trucks would generally make separate
standards appropriate for these different types of vehicles, regardless
of the plain language of the statute which requires such treatment.
---------------------------------------------------------------------------
\2537\ 49 U.S.C. 32902(b)(1) (2007).
\2538\ Indeed, EPCA initially only required NHTSA to establish
CAFE standards for passenger cars; establishment of light truck
standards was permissible.
---------------------------------------------------------------------------
EPCA, as amended by EISA, also requires another separate standard
to be set for domestically-manufactured \2539\ passenger cars. Unlike
standards for passenger cars and light trucks described above, the
compliance burden of the minimum domestic passenger car standard is the
same for all manufacturers: The statute clearly states that any
manufacturer's domestically-manufactured passenger car fleet must meet
the greater of either 27.5 mpg on average, or
---------------------------------------------------------------------------
\2539\ In the CAFE program, ``domestically-manufactured'' is
defined by Congress in 49 U.S.C. 32904(b). The definition roughly
provides that a passenger car is ``domestically manufactured'' as
long as at least 75% of the cost to the manufacturer is attributable
to value added in the United States, Canada, or Mexico, unless the
assembly of the vehicle is completed in Canada or Mexico and the
vehicle is imported into the United States more than 30 days after
the end of the model year.
92 percent of the average fuel economy projected by the Secretary
for the combined domestic and non-domestic passenger automobile
fleets manufactured for sale in the United States by all
manufacturers in the model year, which projection shall be published
in the Federal Register when the standard for that model year is
promulgated in accordance with [49 U.S.C. 32902(b)].\2540\
---------------------------------------------------------------------------
\2540\ 49 U.S.C. 32902(b)(4) (2007).
Since that requirement was promulgated, the ``92 percent'' has
always been greater than 27.5 mpg. NHTSA published the 92-percent
minimum domestic passenger car standards for model years 2017-2025 at
49 CFR 531.5(d) as part of the 2012 final rule. For MYs 2022-2025,
531.5(e) states that these were to be applied if, when actually
proposing MY 2022 and subsequent standards, the previously identified
standards for those years are deemed maximum feasible, but if NHTSA
determines that the previously identified standards are not maximum
feasible, the 92-percent minimum domestic passenger car standards would
also change. This is consistent with the statutory language that the
92-percent standards must be determined at the time an overall
passenger car standard is promulgated and published in the Federal
Register. Thus, any time NHTSA establishes or changes a passenger car
standard for a model year, the minimum domestic passenger car standard
for that model year will also be evaluated or reevaluated and
established accordingly. NHTSA explained this in the rulemaking to
establish standards for MYs 2017 and beyond and received no
comments.\2541\
---------------------------------------------------------------------------
\2541\ 77 FR 62624, 63028 (Oct. 15, 2012).
---------------------------------------------------------------------------
The 2016 Alliance/Global petition for rulemaking asked NHTSA to
revise the 92-percent minimum domestic passenger car standards
retroactively for MYs 2012-2016 ``to reflect 92 percent of the required
average passenger car standard taking into account the fleet mix as it
actually occurred, rather than what was forecast.'' The petitioners
stated that doing so would be ``fully consistent with the statute.''
\2542\
---------------------------------------------------------------------------
\2542\ Automobile Alliance and Global Automakers Petition for
Direct Final Rule with Regard to Various Aspects of the Corporate
Average Fuel Economy Program and the Greenhouse Gas Program (June
20, 2016) at 5, 17-18, available at https://www.epa.gov/sites/production/files/201609/documents/petition_to_epa_from_auto_alliance_and_global_automakers.pdf
(hereinafter Alliance/Global Petition).
---------------------------------------------------------------------------
NHTSA explained in the NPRM that NHTSA understood that determining
the 92 percent value ahead of the model year to which it applies, based
on the information then available to the agency, would result in a
different mpg number than if NHTSA determined the 92 percent value
based on the information available at the end of the model year in
question. NHTSA further explained that it understood that determining
the 92 percent value ahead of time could make the minimum domestic
passenger car standard more stringent than it could be if it were
determined at the end of the model year, if manufacturers end up
producing more larger-footprint passenger cars than what NHTSA had
originally anticipated.
Accordingly, NHTSA sought comment on the request by Alliance/
Global. Additionally, recognizing the uncertainty inherent in
projecting specific values far into the future, NHTSA also sought
comment on whether it is possible to define the 92 percent valueas a
range, if NHTSA defined the values associated with a CAFE standard
(i.e., the footprint curve) as a range rather than as a single number.
NHTSA referred to the sensitivity analysis included in the proposal and
in the accompanying PRIA as a basis for such an mpg range ``defining''
the passenger car standard in any given model year. If NHTSA took that
approach, 92 percent of that ``standard'' would also, necessarily, be a
range. NHTSA broadly sought comment on that approach or other similar
approaches.
The Alliance and FCA commented that they ``supported the NHTSA
proposal'' to calculate 92 percent as a range rather than as a single
value, with the ultimate minimum domestic passenger car standard to be
determined at the end of the MY to which it applies.\2543\ Both
organizations cited compliance difficulties when the 92 percent
calculated at the time of the rulemaking turns out to be more stringent
than 92 percent of the final MY compliance obligations for passenger
cars, and argued that minimum domestic passenger car standards should
be recalculated as part of this rulemaking for all model years, rather
than only MYs 2021-2026, in order to ameliorate that compliance
difficulty retroactively. The Alliance argued that the 18 month lead
time requirement should not be interpreted to apply to the minimum
domestic passenger car standards, because if the 92 percent value is a
range like the overall passenger car curve, then that value cannot be
determined until after the model year is completed.\2544\ Because
manufacturers' individual compliance obligations are not subject to the
18 month lead time requirement, the Alliance requested that the 92
percent should similarly not be.\2545\ Separately, Kreucher commented
that NHTSA should expand the credit transfer provision to allow
transferred credits to be used to meet the minimum domestic passenger
car standard.\2546\
---------------------------------------------------------------------------
\2543\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 41;
FCA, NHTSA-2018-0067-11943, at 64.
\2544\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 42-
43.
\2545\ Id.
\2546\ Kreucher, NHTSA-2018-0067-0444, at 11.
---------------------------------------------------------------------------
In contrast, the States and Cities and ACEEE opposed changes to the
minimum domestic passenger car standard, with the States and Cities
commenting that NHTSA ``is proposing to retroactively revise the 92
percent based on actual fleet mix'' \2547\ and ACEEE simply noting that
the Alliance/Global had requested that NHTSA do this.\2548\ ACEEE
stated that NHTSA did not have discretion to alter the statutory
requirement, and argued that calculating 92 percent at the end of the
model year was ``entirely counter to the intent of the law--the so-
called backstop is designed explicitly to protect against the market
shifts for which the [industry is] asking the standard to be
adjusted.'' \2549\
[[Page 25125]]
The States and Cities similarly argued that ``the 92 percent
requirement is expressly intended to be a projection, not a
retrospective recalculation,'' and ``the statute does not contemplate a
`range,' but rather a `minimum' with a set value--92 percent. If
Congress had intended the value to be a range, it would have included
that language in the statute, and would not have determined the value
with such specificity.'' \2550\
---------------------------------------------------------------------------
\2547\ States and Cities, NHTSA-2018-0067-11735, at 79.
\2548\ ACEEE, NHTSA-2018-0067-12122, Attachment (joint NGO
comment to manufacturer petition for flexibilities), at 15.
\2549\ Id. ACEEE cited a NHTSA statement in the 2010 final rule
establishing standards for MYs 2012-2016 in support of this
argument, noting that NHTSA had said ``this minimum standard was
intended to act as a `backstop,' ensurng that domestically-
manufactured passenger cars reached a given mpg level even if the
market shifted in ways likely to reduce overall fleet mpg.'' Id.
(emphasis added).
\2550\ States and Cities, NHTSA-2018-0067-11735, at 79.
---------------------------------------------------------------------------
NHTSA considered comments about setting the MDPCS as a range. NHTSA
recognizes that the approach discussed in the NPRM may not be within
our statutory authority and therefore is setting the standards as
specific values.
NHTSA agrees that setting the MDPCS after the model year is
completed and the total passenger car fleet standard is known would
provide standards that adapt with changes in consumer demand. However,
such an approach would not establish the final numerical value until
significantly after the model year completed, only after final
compliance data has been submitted by all manufacturers and EPA and
NHTSA have completed compliance work for the total passenger car fleet.
In addition, the standard would be based on the production of all
manufacturers of passenger cars, providing no means for an individual
manufacturer to have certainty over its final standard. Individual
manufacturers likewise would have no control over the value by
controlling their production mix. For these reasons, NHTSA is denying
the Alliance/Global petition that the 92 percent value for the MDPCS be
determined based on the information available at the end of the model
year in question.
That said, NHTSA agrees that the actual total passenger car fleet
standards have differed significantly the 2012 projection, and examined
the projections from past rulemakings in greater detail. NHTSA reviewed
the total passenger car fleet (all domestic and import passenger cars)
standard that was projected at the time of rulemakings for MYs 2011 to
2018 and compared those projections to the actual total fleet passenger
car standard for each of those model years from compliance data, based
on the actual footprints and production volume of the models produced
in those model years. Table VIII-1 shows the projected standards and
the actual standards on a fuel economy basis, and Table VIII-2 shows
the fuel economy values converted to fuel consumption values which was
used as the basis for and analyzing the differences between the
projected standards and actual standards.\2551\ Table VIII-2 also shows
the percentage difference between the total passenger car fleet
standard at the time of the rulemaking and the actual fleet standard
based on compliance data.
---------------------------------------------------------------------------
\2551\ Consistent with EPCA/EISA and corresponding regulations,
CAFE compliance calculations have been conducted on a mile per
gallon basis. However, engineering computations have almost
exclusively been conducted on a fuel consumption basis (i.e., in
gallons per mile), because the underlying engineering relationships
are more meaningfully defined on a fuel consumption basis.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 25126]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.729
[GRAPHIC] [TIFF OMITTED] TR30AP20.730
[[Page 25127]]
The data show that the standards projected in 2012 were
consistently more stringent than the actual standards, by an average of
1.9 percent. This difference indicates that in rulemakings conducted in
2009 through 2012, the agencies' projections of passenger car vehicle
footprints and production volumes consistently underestimated the
consumer demand for larger passenger cars over the MYs 2011 to 2018
period.
BILLING CODE 4910-59-C
To establish minimum standards for domestic passenger cars in these
past rulemakings, NHTSA computed the average of manufacturers'
requirements given the attribute-based standards being issued, and
given the projected distribution of passenger car footprints as
indicated in the analysis fleet (aka market forecast) used to analyze
impacts of the standards. The joint NHTSA-EPA rulemaking establishing
standards for MYs 2012-2016 presented analysis that, in turn, used a
``2008-based'' market forecast that combined detailed information
regarding the MY 2008 fleet with a commercial market forecast (by brand
and segment) and a range of agency assumptions. Importantly, the
commercial market forecast showed Chrysler's production falling
dramatically, and never recovering; as well as Chrysler passenger cars
being distributed more than most OEMs (other than Jaguar and Mercedes)
toward larger footprints, and this forecast impacted the NHTSA's
projection of overall average requirements for passenger cars under the
footprint-based standards. For example, the 2008-based forecast showed
production of Chrysler brands (Chrysler, Dodge, Jeep, and Ram) for the
U.S. market totaling 0.8 million units by MY 2017, and today's analysis
fleet uses a MY 2017 fleet showing 1.9 million Chrysler-branded units.
Also, among the agencies' assumptions, was that some manufacturers
(Chrysler, Ford, Subaru, Mazda, and Mitsubishi) would rapidly increase
production of small footprint vehicles not observed in the MY 2008
fleet.
The joint rulemaking establishing standards for MYs 2017-2025 also
used this 2008-based fleet for the NPRM, showing more than 1.3 million
units smaller than 41 square feet in MY 2017, far more than the 0.3m
units shown in the model inputs for today's analysis. For the 2012
final rule, the agencies conducted side-by-side analysis, one using the
2008-based fleet, and one using a 2010-based fleet. The 2010-based
fleet used a newer commercial forecast that was considerably more
sanguine regarding, for example, FCA's prospects. Minimum standards for
domestic passenger cars were based on an average of results for the
2008-based and 2010-based total passenger car fleets.
The analysis fleet underlying today's reference case analysis is
discussed above in Section VI.A.2 and available in full detail with the
model inputs and outputs accompanying today's notice.\2552\ For the
current rulemaking, NHTSA also considered that, unlike the passenger
car standards and light truck standards which are vehicle attribute-
based and automatically adjust with changes in consumer demand, that
MDPCSs are not attribute-based, and therefore do not adjust with
changes in consumer demand. They are fixed standards that are
established at the time of the rulemaking. The MYs 2011-2018 MDPCS were
more stringent and placed more burden on manufacturers of domestic
passenger cars than was projected and expected at the time of the
rulemakings. NHTSA agrees with the Alliance's concerns over the impact
of changes in consumer demand on manufacturers' ability to comply with
the MDPCS and in particular, manufacturers that produce larger
passenger cars domestically.
---------------------------------------------------------------------------
\2552\ https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
---------------------------------------------------------------------------
Additionally, as discussed in more detail in Section VIII.B.4
below, consumer demand may shift even more in the direction of larger
passenger cars if fuel prices continue to remain low. The fuel prices
used in the analysis for this final rule rely on EIA's future forecasts
of fuel prices, which were made prior to the recent collapse of oil
prices. If the former OPEC+ members continue to pursue market share,
fuel prices will likely continue to drop. If, instead of pursuing
market share, they try to control prices restricting supply, U.S. shale
production could begin to ramp back up and exert downward pressure on
price. If fuel prices end up even lower than our analysis assumes,
benefits from saving additional fuel will be worth even less to
consumers. Our analysis captures none of these effects. Sustained low
oil prices can be expected to have real effects on consumer demand for
additional fuel economy, and consumers may foreseeably be even less
interested in smaller passenger cars than they are at present.
To help avoid similar outcomes in the rulemaking timeframe to what
has happened with the MDPCS over the last several model years, NHTSA
determined it is reasonable and appropriate to consider the recent
projection errors as part of estimating the projected total passenger
car fleet fuel economy for MYs 2021-2026. As stated above the average
difference over MYs 2011-2018 was 1.9 percent. As explained above,
those differences are largely attributable to aspects of the forecasts
that turned out to be far different from reality. NHTSA is projecting
the total passenger car fleet fuel economy using the central analysis
value in each model year and applying an offset based on the historical
1.9 percent difference identified for MYs 2011-2018. Table VIII-3 hows
the calculation values used to determine the total passenger car fleet
fuel economy value for each model year.
NHTSA will continue its practice of determining the MDPCS as
specific values at the same time that it sets passenger car standards,
at 92 percent of the projected passenger cars standard in each model
year. Table VIII-3 also shows the computations for the MDPCS for each
model year. The new MDPCS are prescribed in the regulatory text below.
[[Page 25128]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.731
Table VIII-4 lists the minimum domestic passenger car standards
reflecting the updated analysis discussed above, and comparing these to
standards that would correspond to each of the other regulatory
alternatives considered. NHTSA has updated these to reflect its overall
analysis and resultant projection for the CAFE standards finalized
today, highlighted below as ``Preferred (Alternative 3),'' and has
calculated what those standards would be under the no action
alternative (as issued in 2012, as updated for the NPRM, and as further
updated by today's analysis) and under the other alternatives described
and discussed further in Section V, above. As explained in a separate
memorandum to the document, while the CAFE Model analysis underlying
the FEIS, FRIA, and final rule does not reflect this change, separate
analysis that does reflect the change demonstrates that doing so does
not change estimated impacts of any of the regulatory alternatives
under consideration.
[GRAPHIC] [TIFF OMITTED] TR30AP20.732
[[Page 25129]]
Attribute-Based and Defined by Mathematical Function
EISA requires NHTSA to set CAFE standards that are ``based on 1 or
more attributes related to fuel economy and express[ed] . . . in the
form of a mathematical function.'' \2553\ Historically, NHTSA has based
standards on vehicle footprint and proposes to continue to do so for
all the reasons described in previous rulemakings. As in previous
rulemakings, NHTSA proposed to define the standards in the form of a
constrained linear function that generally sets higher (more stringent)
targets for smaller-footprint vehicles and lower (less stringent)
targets for larger-footprint vehicles. These footprint curves are
discussed in much greater detail in Section V above. NHTSA sought
comment both on the choice of footprint as the relevant attribute and
on the rationale for the constrained linear functions chosen to
represent the standards; those comments and NHTSA's responses are
discussed above in Section V.
---------------------------------------------------------------------------
\2553\ 49 U.S.C. 32902(b)(3)(A).
---------------------------------------------------------------------------
d) Number of Model Years for Which Standards May Be Set at a Time
EISA also states that NHTSA shall ``issue regulations under this
title prescribing average fuel economy standards for at least 1, but
not more than 5, model years.'' \2554\ In the 2012 final rule, NHTSA
interpreted this provision as preventing the agency from setting final
standards for all of MYs 2017-2025 in a single rulemaking action, so
the MYs 2022-2025 standards were termed ``augural,'' meaning ``that
they represent[ed] the agency's current judgment, based on the
information available to the agency [then], of what levels of
stringency would be maximum feasible in those model years.'' \2555\
That said, NHTSA also repeatedly clarified that the augural standards
were in no way final standards and that a future de novo rulemaking
would be necessary in order both to propose and to promulgate final
standards for MYs 2022-2025.
---------------------------------------------------------------------------
\2554\ 49 U.S.C. 32902(b)(3)(B).
\2555\ 77 FR 62623, 62630 (Oct. 15, 2012).
---------------------------------------------------------------------------
In the NPRM, NHTSA proposed to establish new standards for MYs
2022-2026 and to revise the previously-established final standards for
MY 2021. NHTSA explained that legislative history suggests that
Congress included the five year maximum limitation so NHTSA would issue
standards for a period of time where it would have reasonably realistic
estimates of market conditions, technologies, and economic
practicability (i.e., not set standards too far into the future).\2556\
However, NHTSA suggested that the concerns Congress sought to address
by imposing those limitations are not present for nearer model years
where NHTSA already has existing standards, and noted that revisiting
existing standards is contemplated by both 49 U.S.C. 32902(c) and
32902(g). NHTSA stated that the agency therefore believed that it is
reasonable to interpret section 32902(b)(3)(B) as applying only to the
establishment of new standards rather than to the combined action of
establishing new standards and amending existing standards.
---------------------------------------------------------------------------
\2556\ See 153 Cong. Rec. 2665 (Dec. 28, 2007).
---------------------------------------------------------------------------
Moreover, NHTSA argued, it would be an absurd result if the five
year maximum limitation were interpreted to prevent NHTSA from revising
a previously-established standard that the agency had determined to be
beyond maximum feasible, while concurrently setting five years of
standards not so distant from today. The concerns Congress sought to
address are much starker when NHTSA is trying to determine what
standards would be maximum feasible 10 years from now as compared to
three years from now.
NADA commented that NHTSA has discretion and authority to set
standards for MY 2026 and that the ``statutory five-year rule is not a
barrier to doing so,'' \2557\ while the environmental group coalition
argued that NHTSA ``is limited to prescribing fuel economy standards
for only five model years at a time,'' but ``[h]ere, NHTSA is setting
standards for six model years, 2021 through 2026. This exceeds NHTSA's
statutory authority.'' \2558\ Consumers Union argued that ``[i]f
Congress had intended the statute to only apply to the establishment of
new standards, as the agencies contend, it certainly could have stated
as such. But Congress did not include any language even hinting at this
interpretation.'' \2559\
---------------------------------------------------------------------------
\2557\ NADA, NHTSA-2018-0067-12064, at 9.
\2558\ Environmental group coalition, NHTSA-2018-0067-12000, at
66.
\2559\ Consumers Union, NHTSA-2018-0067-12068, Attachment A, at
24.
---------------------------------------------------------------------------
NHTSA continues to believe, consistent with the legislative
history, that the five year limitation was intended to prevent NHTSA
from setting standards too far into the future, recognizing that
predicting the future is difficult. Consumers Union is correct that
nothing in the statute compels the interpretation that the five year
limitation applies only to the setting of new standards rather than to
the combined action of establishing new standards and amending existing
standards, but NHTSA does not believe that the statute precludes this
interpretation, either. The statute allows NHTSA to revisit existing
standards; the statute separately allows NHTSA to prescribe new
standards for at least 1, but not more than 5, model years when it
``issues regulations.'' It is not clear whether the statute precludes
multiple concurrent or quickly-sequential rulemakings ``issuing
regulations'' for different periods of time. If this approach were
used, for example, to try to set ten years' worth of CAFE standards
essentially at once, this would appear directly contrary to the
statute. If this approach were used to revisit an existing standard and
then (in a separate rulemaking) set five years' worth of standards for
the immediately ensuing model years, this would seem consistent with
Congressional intent, but an unnecessary use of tax dollars that could
be saved by consolidating agency (and commenter) work into a single
rulemaking action. NHTSA does not believe that Congress intended to
force the agency to waste resources, and continues to believe that the
current interpretation is reasonable and appropriate.
(e) Maximum Feasible Standards
As discussed above, EPCA requires NHTSA to consider four factors in
determining what levels of CAFE standards would be maximum feasible,
and NHTSA presents in the sections below its understanding of the
meaning of those four factors. All factors should be considered, in the
manner appropriate, and then the maximum feasible standards should be
determined.
(1) Technological Feasibility
``Technological feasibility'' refers to whether a particular method
of improving fuel economy is available for deployment in commercial
application in the model year for which a standard is being
established. Thus, NHTSA is not limited in determining the level of new
standards to technology that is already being commercially applied at
the time of the rulemaking. For the proposal, NHTSA explained that it
had considered a wide range of technologies that improve fuel economy,
subject to the constraints of EPCA regarding how to treat alternative
fueled vehicles, such as battery-electric vehicles, in determining
maximum feasible standards, and considering the need to account for
which technologies have already been applied to which vehicle model/
configuration, and the need to realistically estimate the cost and fuel
economy impacts of each technology.
[[Page 25130]]
NHTSA explained that it had not attempted to account for every
technology that might conceivably be applied to improve fuel economy
and considered it unnecessary to do so given that many technologies
address fuel economy in similar ways.\2560\ NHTSA noted that
technological feasibility and economic practicability are often
conflated, trying to explain that the question of whether a fuel-
economy-improving technology does or will exist (technological
feasibility) is a different question from what economic consequences
could ensue if NHTSA effectively requires that technology to become
widespread in the fleet and the economic consequences of the absence of
consumer demand for technology that are projected to be required
(economic practicability). NHTSA explained that it is therefore
possible for standards to be technologically feasible but still beyond
the level that NHTSA determines to be maximum feasible due to
consideration of the other relevant factors.
---------------------------------------------------------------------------
\2560\ For example, NHTSA has not considered high-speed
flywheels as potential energy storage devices for hybrid vehicles;
while such flywheels have been demonstrated in the laboratory and
even tested in concept vehicles, commercially available hybrid
vehicles currently known to NHTSA use chemical batteries as energy
storage devices, and the agency has considered a range of hybrid
vehicle technologies that do so.
---------------------------------------------------------------------------
The States and Cities commenters argued that NHTSA's interpretation
of the technological feasibility factor was unreasonable, stating that
``. . . fuel economy standards under EPCA are 'intended to be
technology forcing' because Congress recognized 'that 'market forces .
. . may not be strong enough to bring about the necessary fuel
conservation which a national energy policy demands.' '' \2561\ The
States and Cities commenters thus argued that all alternatives less
stringent than the baseline/augural standards alternative were
unacceptable because they would not force technologies to be developed
and applied, and NHTSA had ``conce[ded] that the technology already
exists that could meet the more stringent augural standards.'' \2562\
These commenters stated that ``NHTSA is therefore impermissibly and
unreasonably (and even implicitly) re-interpreting this factor in a
manner contrary to the plain meaning of 'feasibility' and ignoring
EPCA's technology-forcing purpose. See Chevron, 467 U.S. at 843; Fox
Television, 556 U.S. at 515 (`An agency may not . . . depart from a
prior policy sub silentio.').'' CARB \2563\ and CBD et al.\2564\ also
argued that EPCA was intended to be technology forcing.
---------------------------------------------------------------------------
\2561\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 66, citing CAS, 793 F.2d at 1339 (citing S. Rep. No.
179, 94th Cong., 1st Sess. 2 (1975) at 9).
\2562\ Id. at 66.
\2563\ CARB, NHTSA-2018-0067-11873, Detailed Comments, at 84
(``Since market inefficiencies may preclude sufficient improvement
without regulatory incentives, EPCA requires standards that advance
technology. (Citing CAS v. NHTSA, 793 F.2d 1322, 1339, citing S.
Rep. No. 179, 94th Cong., 1st Sess. 2 (1975), U.S.C.C.A.N. 1975 at
9)'').
\2564\ CBD et al., NHTSA-2018-0067-12057, at 2.
---------------------------------------------------------------------------
The States and Cities commenters also argued that NHTSA had
previously stated in rulemakings that it considered ``all types of
technologies that improve real-world fuel economy,'' but in the NPRM
NHTSA stated instead that it had ``not attempted to account for every
technology that might conceivably be applied to improve fuel economy
and consider[ed] it unnecessary to do so given that many technologies
address fuel economy in similar ways.'' \2565\ The States and Cities
commenters stated that ``[t]his is an unexplained departure from the
agency's past practice and prior interpretation of `technological
feasibility,A' citing Fox Television, and argued that NHTSA had not
explained ``1) what `similar ways' means, or 2) why the fact that a
technology that might improve fuel economy `in similar ways' to another
technology obviates NHTSA's obligation to consider its availability,
particularly given the differences in costs between different
technologies.'' \2566\ The States and Cities commenters pointed to the
examples of HCR1 and HCR2 as technologies ``already widely available in
the market'' that should have been considered, and claimed that NHTSA
had ``failed to even consult with EPA regarding which technologies the
agency considered,'' ``result[ing] in fundamentally flawed predictions
of what technology can be applied in model years 2021-2026.'' \2567\
---------------------------------------------------------------------------
\2565\ Id. at 67, referring to 83 FR at 43208.
\2566\ Id.
\2567\ Id.
---------------------------------------------------------------------------
Mazda, in contrast, stated that it agreed that ``mere development
and introduction of advanced fuel efficient technologies is not
sufficient for manufacturers to comply with established GHG and fuel
efficiency standards. The technologies must be widely adopted by
consumers for them to provide the expected environmental benefit.''
\2568\ Mr. Kreucher stated that manufacturers have been applying
``unprecedented levels of technology'' but are still falling short of
their compliance obligations, pointing in particular to light truck
compliance in MY 2016. Kreucher argued that ``[t]his indicates a
serious overestimation of technological feasibility in the prior [2012]
analysis that must be corrected.'' \2569\
---------------------------------------------------------------------------
\2568\ Mazda, NHTSA-2018-0067-11727, at 2.
\2569\ Kreucher, NHTSA-2018-0067-0444, at 7.
---------------------------------------------------------------------------
UCS stated that the NPRM analysis ``undermined'' an assessment of
``technical feasibility,'' by ``paint[ing] fuel-saving technologies as
less effective and more costly than real-world data indicate,'' through
several mechanisms.\2570\ First, UCS argued that the analysis had
underestimated ICE efficiency possibilities, ``frequently ignoring
technology that is already commercialized or is widely anticipated to
be readily available within the timeframe of the standards.'' \2571\
Second, UCS suggested that the NPRM analysis had ``overstate[d] the
degree to which manufacturers have deployed some of the most cost-
effective technologies, while errors in full vehicle simulation and
rampant disregard for the current state of technology underestimates
the potential for future improvement.'' \2572\ UCS claimed that
``[f]requently the agencies have departed from past precedence in
specific ways in order to increase technology costs associated with
technology deployment, sometimes failing to provide even a glimmer of
reasonable justification for such decisions.'' \2573\ (emphasis added)
Third, UCS argued that the model had been deliberately constructed to
avoid choosing the most cost-effective technology pathways, showing
higher costs and more future overcompliance than UCS analysis
showed.\2574\ Finally, UCS argued that better modeling of credit
trading and use would further reduce technology costs. UCS concluded
that ``The mischaracterization of technology and unrealistic model
construction lead to an inaccurate assessment of technological
feasibility, effectively undermining this factor's weight in
considering maximum feasible standards.'' \2575\
---------------------------------------------------------------------------
\2570\ UCS, NHTSA-2018-0067-12039, at 4.
\2571\ Id.
\2572\ Id.
\2573\ Id.
\2574\ Id.
\2575\ Id.
---------------------------------------------------------------------------
Contrary to the assertion by several commenters that NHTSA has
historically claimed that it must set technology-forcing standards,
NHTSA has previously described the technological feasibility factor as
allowing the agency to set standards that force the development and
application of new fuel-efficient technologies.\2576\ In the same
preamble section in which that description was set forth, NHTSA stated
[[Page 25131]]
that ``[i]t is important to remember that technological feasibility
must also be balanced with the other of the four statutory factors.
Thus, while 'technological feasibility' can drive standards higher by
assuming the use of technologies that are not yet commercial, 'maximum
feasible' is also defined in terms of economic practicability, for
example, which might caution the agency against basing standards (even
fairly distant standards) entirely on such technologies.'' \2577\ NHTSA
further stated that ``. . . as the `maximum feasible' balancing may
vary depending on the circumstances at hand for the model year in which
the standards are set, the extent to which technological feasibility is
simply met or plays a more dynamic role may also shift.'' \2578\
---------------------------------------------------------------------------
\2576\ See, e.g., 77 FR at 63015 (Oct. 15, 2012).
\2577\ Id.
\2578\ Id.
---------------------------------------------------------------------------
NHTSA continues to believe that, for purposes of this rulemaking
covering standards for MYs 2021-2026, the crucial question is not
whether technologies exist to meet the standards--they do. The question
is rather, given that the technology exists, how much of it should be
required to be added to new cars and trucks in order to conserve more
energy, and how to appropriately balance additional energy conserved
and additional cost for new vehicles. Regardless of whether
technological feasibility allows the agency to set technology-forcing
standards, technological feasibility does not require, by itself, NHTSA
to set technology-forcing standards if other statutory factors would
point the agency in a different direction. NHTSA has expressed this
interpretation of technological feasibility over the course of multiple
rulemakings.\2579\ The States and Cities commenters appear, at the
root, to be contesting the agency's determination of maximum feasible
standards, by way of arguing that NHTSA must interpret the
technological feasibility factor as necessarily driving greater energy
conservation. The balancing of factors to determine maximum feasible
standards is a separate issue, for which EPCA/EISA gives NHTSA
considerable discretion.
---------------------------------------------------------------------------
\2579\ Id., see also 75 FR at 25605 (May 7, 2010).
---------------------------------------------------------------------------
The States and Cities commenters focus on previous rulemaking
language when they suggest that the agency was arbitrary and capricious
for not explaining more fully why it need not expressly evaluate every
single technology that does or could exist in MYs 2021-2026. While
NHTSA stated in 2012 that it had ``considered all types of technologies
that improve real-world fuel economy, including air-conditioner
efficiency and other off-cycle technology, PHEVs, EVs, and highly-
advanced internal combustion engines not yet in production,'' \2580\
that statement was only one in a larger discussion. The 2012 final rule
also stated expressly that ``[t]here are a number of other potential
technologies available to manufacturers in meeting the 2017-2025
standards that the agencies have evaluated but have not considered in
our final analyses. These include HCCI, 'multi-air', and camless valve
actuation, and other advanced engines currently under development.''
\2581\ (emphasis added) Thus, even under the prior analysis that some
commenters appear to prefer, it is not entirely correct to say that
NHTSA had considered all technologies in existence or that could exist,
because some technologies were clearly and purposely left out of the
prior rule's analysis. In response to commenters' apparent confusion
regarding NHTSA's statement that it did not consider technologies that
improved fuel economy in ``similar ways'' as other technologies
discussed in the NPRM, the meaning behind that statement was discussed
at greater length in the section of the NPRM that substantively covered
those technologies. For example, in discussing the ``HCR2'' technology,
the agencies explained that while the agencies were not modeling HCR2
expressly due to concerns that it remained ``entirely speculative,''
``[t]he CAFE model allows for incremental improvement over existing
HCR1 technologies with the addition of improved accessory devices
(IACC), a technology that is available to be applied on many baseline
MY 2016 vehicles with HCR1 engines and may be applied as part of a
pathway of compliance to further improve the effectiveness of existing
HCR1 engines.'' \2582\ In this and in other instances, technologies
included in the analysis improved fuel economy in similar ways to other
technologies not included. Here, HCR1, when combined with IACC, results
in ``a step past'' HCR1, which is similar to the unproven HCR2. As in
the 2012 rule, the agencies explained in the NPRM why certain
technologies were not considered, and sought comment. In response to
comments received, some technologies have been added to the analysis
for the final rule. See Section VI for more information.
---------------------------------------------------------------------------
\2580\ 77 FR at 63037 (Oct. 15, 2012).
\2581\ 77 FR at 62706 (Oct. 15, 2012).
\2582\ 83 FR at 43038 (Aug. 24, 2018).
---------------------------------------------------------------------------
While the agencies respond to many of UCS's analytical concerns in
Sections IV and VI (which include extensive discussion of changes made
in response to comments), NHTSA recognizes that some commenters believe
that more technologies are ``available for deployment'' more widely,
and sooner, than the final rule's analysis reflects. This question has
long been a topic of debate in CAFE and CO2 rulemakings--the
agencies consider which technologies can be applied to which vehicles
in which model years in order to assess the costs and benefits of
pushing the industry to reach different levels of standards, which in
turn helps to inform stringency determinations. In response to
comments, the agencies have expanded the number of technologies and the
vehicles to which they may be applied for this final rule, but continue
to disagree that certain technologies can be applied widely in the
rulemaking timeframe. NHTSA does not believe, for example, that HCCI
will be unavailable for widespread application in the rulemaking
timeframe because it wishes to believe this prediction--NHTSA believes
it based on the fact that HCCI has been in the research phase for
several decades, and the only production applications to date use a
highly-limited version that restricts HCCI combustion to a very narrow
range of engine operating conditions. Section VI contains further
discussion of these issues.
(2) Economic Practicability
``Economic practicability'' has traditionally referred to whether a
standard is one ``within the financial capability of the industry, but
not so stringent as to'' lead to ``adverse economic consequences, such
as a significant loss of jobs or unreasonable elimination of consumer
choice.'' \2583\ In evaluating economic practicability, NHTSA considers
the uncertainty surrounding future market conditions and consumer
demand for fuel economy alongside consumer demand for other vehicle
attributes. NHTSA has explained in the past that this factor can be
especially important during rulemakings in which the auto industry is
facing significantly adverse economic conditions (with corresponding
risks to jobs). Consumer acceptability is also a major component of
economic practicability,\2584\ which can involve
[[Page 25132]]
consideration of anticipated consumer responses not just to increased
vehicle cost, but also to the way manufacturers may change vehicle
models and vehicle sales mix in response to CAFE standards. In
attempting to determine the economic practicability of attribute-based
standards, NHTSA considers a wide variety of elements, including the
annual rate at which manufacturers can increase the percentage of their
fleet that employs a particular type of fuel-saving technology,\2585\
and manufacturer fleet mixes. NHTSA also considers the effects on
consumer affordability resulting from costs to comply with the
standards, and consumers' valuation of fuel economy, among other
things.
---------------------------------------------------------------------------
\2583\ 67 FR 77015, 77021 (Dec. 16, 2002).
\2584\ See, e.g., Center for Auto Safety v. NHTSA (CAS), 793
F.2d 1322 (DC Cir. 1986) (Administrator's consideration of market
demand as component of economic practicability found to be
reasonable); see also Public Citizen v. NHTSA, 848 F.2d 256
(Congress established broad guidelines in the fuel economy statute;
agency's decision to set lower standards was a reasonable
accommodation of conflicting policies).
\2585\ For example, if standards effectively require
manufacturers to make technologies widely available that consumers
do not want, or to make technologies widely available before they
are ready to be widespread, NHTSA believes that these standards
could potentially be beyond economically practicable.
---------------------------------------------------------------------------
Prior to the MYs 2005-2007 rulemaking under the non-attribute-based
(fixed value) CAFE standards, NHTSA generally sought to ensure the
economic practicability of standards in part by setting them at or near
the capability of the ``least capable manufacturer'' with a significant
share of the market, i.e., typically the manufacturer whose fleet mix
was, on average, the largest and heaviest, generally having the highest
capacity and capability so as not to limit the availability of those
types of vehicles to consumers. In the first several rulemakings
establishing attribute-based standards, NHTSA applied marginal cost-
benefit analysis, considering both overall societal impacts and overall
consumer impacts. Whether the standards maximize net benefits has thus
been a significant, but not dispositive, factor in the past for NHTSA's
consideration of economic practicability. Executive Order 12866, as
amended by Executive Order 13563, states that agencies should ``select,
in choosing among alternative regulatory approaches, those approaches
that maximize net benefits . . .'' In practice, however, agencies,
including NHTSA, must consider that the modeling of net benefits does
not capture all considerations relevant to economic practicability.
Therefore, as in past rulemakings, NHTSA explained in the NPRM that it
was considering net societal impacts, net consumer impacts, and other
related elements in the consideration of economic practicability.
NHTSA's consideration of economic practicability depends on a
number of elements. Expected availability of capital to make
investments in new technologies matters; manufacturers' expected
ability to sell vehicles with certain technologies matters; likely
consumer choices matter; and so forth. NHTSA explained in the NPRM that
NHTSA's analysis of the impacts of the proposal incorporated
assumptions to capture aspects of consumer preferences, vehicle
attributes, safety, and other elements relevant to an impacts estimate;
but stated that it is difficult to capture every such constraint.
Therefore, NHTSA explained, it is well within the agency's discretion
to deviate from the level at which modeled net benefits are maximized
if the agency concludes that that level would not represent the maximum
feasible level for future CAFE standards. Economic practicability is
complex, and like the other factors must also be considered in the
context of the overall balancing and EPCA's overarching purpose of
energy conservation. Depending on the conditions of the industry and
the assumptions used in the agency's analysis of alternative standards,
NHTSA stated that it could well find that standards that maximize net
benefits, or that are higher or lower, could be at the limits of
economic practicability, and thus potentially the maximum feasible
level, depending on how the other factors are balanced.
NHTSA also stated in the NPRM that while the agency would discuss
safety as a separate consideration, NHTSA also considered safety as
closely related to, and in some circumstances a subcomponent of,
economic practicability. On a broad level, manufacturers have finite
resources to invest in research and development. Investment into the
development and implementation of fuel saving technology necessarily
comes at the expense of investing in other areas such as safety
technology. On a more direct level, when making decisions on how to
equip vehicles, manufacturers must balance cost considerations to avoid
pricing further consumers out of the market. As manufacturers add
technology to increase fuel efficiency, they may decide against
installing additional safety equipment to reduce cost increases. And as
the price of vehicles increase beyond the reach of more consumers, such
consumers continue to drive or purchase older, less safe vehicles. In
assessing practicability, NHTSA also considers the harm to the Nation's
economy caused by highway fatalities and injuries.
CARB, the States and Cities commenters, and UCS all commented that
the NPRM analysis, as the States and Cities put it, had ``inexplicably
inflat[ed] technology costs and rel[ied] on flawed models to predict
impacts on vehicle sales.'' \2586\ Both CBD et al. and UCS suggested
that it was incorrect to assume that manufacturers would pass on 100
percent of cost increases as price increases to consumers.\2587\ UCS
further stated that ``The agencies have then strategically excluded
well-established academic literature to limit the assumptions used to
define a consumer's willingness to pay in ways that further increase
costs to consumers and/or decrease the consumer benefits of fuel
economy and greenhouse gas emissions.'' \2588\ UCS argued that assuming
full pass-through of cost increases as price increases and assuming
that consumers may not fully value improvements in fuel economy
``arbitrar[ily] . . . depress the sales of highly fuel-efficient
vehicles in the model by systematically negating consumer benefits of
these vehicles.'' \2589\ The States and Cities further argued that
NHTSA had not ``substantiated its concern that an increase in new
vehicle prices would place a particular burden on `low-income
purchasers,' '' and stated that NHTSA had ``assume[d], without
explanation, that'' less-stringent fuel economy standards resulted in
greater net savings for consumers, which NHTSA ``acknowledge[d],
without justification, `is a significantly different analytical result
from the 2012 final rule.' '' \2590\ The States and Cities commenters
implied that this different result and NHTSA's ``failure to acknowledge
it'' was impermissible under the standard set forth in Fox
Television.\2591\
---------------------------------------------------------------------------
\2586\ CARB, NHTSA-2018-0067-11873, at 79-80; States and Cities,
NHTSA-2018-0067-11735, at 69-70; UCS, NHTSA-2018-0067-12039, at 4.
\2587\ CBD et al., NHTSA-2018-0067-12057, at 4; UCS, NHTSA-2018-
0067-12039, at 4.
\2588\ UCS, NHTSA-2018-0067-12039, at 5.
\2589\ Id.
\2590\ States and Cities, NHTSA-2018-0067-11735, at 70.
\2591\ Id.
---------------------------------------------------------------------------
A number of commenters stated that the NPRM's estimates of job
losses associated with the proposal conflicted with NHTSA's concerns
about job losses if more stringent standards were promulgated. CBD et
al. argued that NHTSA could not reasonably conclude that job losses
make less-stringent standards more economically practicable than more-
stringent
[[Page 25133]]
standards.\2592\ The States and Cities commenters stated that ``[b]y
declining to address its own findings of significant job losses in the
auto sector, NHTSA has ignored an important aspect of the problem and
failed to propose a `rational connection between the facts found and
the choice made.' '' \2593\ The States and Cities commenters also
argued that ``the agency failed to acknowledge or explain its break
with its own interpretation and practice of considering whether
standards would cause a `significant loss of jobs.' '' \2594\ Some
commenters argued that more-stringent standards would create more jobs
(and conversely, that less-stringent standards would result in job
losses), primarily for supplier companies,\2595\ and some noted that
other studies had concluded that more-stringent standards would
increase employment, citing, for example, the report by Synapse Energy
Economics, Inc. on ``Cleaner Cars and Job Creation.'' \2596\ Some
commenters further argued that less-stringent standards would hurt U.S.
GDP,\2597\ and some argued that they would hurt U.S. industry's
international competitiveness because other countries/regions have more
stringent standards, and investment may shift to those countries if
U.S. standards do not continue to compel it.\2598\ The States and
Cities commenters stated that failing to address fully ``the negative
employment and GDP impacts of the Proposed Rollback is an abdication of
NHTSA's clear statutory duty to consider the economic practicability of
its proposed standards, and an impermissible interpretation of the
statutory text.'' \2599\
---------------------------------------------------------------------------
\2592\ CBD et al., NHTSA-2018-0067-12057, at 4.
\2593\ States and Cities, NHTSA-2018-0067-11735, at 68 (citing
State Farm, 463 U.S. at 42).
\2594\ Id. (citing 83 FR at 43208; Fox Television, 556 U.S. at
515).
\2595\ CBD et al., NHTSA-2018-0067-12057; Alliance for Vehicle
Efficiency, NHTSA-2018-0067-11696, at 3-4; NESCAUM, NHTSA-2018-0067-
11691, at 5.
\2596\ States and Cities, NHTSA-2018-0067-11735, at 68; UCS,
NHTSA-2018-0067-12039, at 4.
\2597\ States and Cities, NHTSA-2018-0067-11735, at 68; UCS,
NHTSA-2018-0067-12039, at 4.
\2598\ NESCAUM, NHTSA-2018-0067-11691, at 5; Alliance for
Vehicle Efficiency, NHTSA-2018-0067-11696, at 4.
\2599\ States and Cities, NHTSA-2018-0067-11735, at 68 (citing
49 U.S.C. 32902(f); Chevron, 467 U.S. at 843).
---------------------------------------------------------------------------
Commenters disagreed on whether and how NHTSA should consider
consumer demand. Mr. Kreucher, the Texas Congressional
Delegation,\2600\ and Senator Inhofe,\2601\ among others, all argued
that considering consumer demand for fuel economy was important, while
other commenters argued that while it may be permissible for NHTSA to
consider consumer demand, NHTSA could not elevate that consideration
above others. CARB and the States and Cities commenters both cited
language from CAS v. NHTSA for the premise that ``Congress intended
energy conservation to be a long-term effort that would continue
through temporary improvements in energy availability. Thus, it would
clearly be impermissible for NHTSA to rely on consumer demand to such
an extent that it ignored the overarching goal of fuel conservation.''
\2602\ The Minnesota agencies stated that ``making sweeping assumptions
about consumer preferences should not trump the clear public benefit to
reducing GHG emissions through these standards.'' \2603\ Mr. Kreucher
commented, in contrast, that consumer preferences are driven entirely
by ``[l]ong term fuel price expectations and fuel price alone,'' and
disagreed with the historical ``implicit assumption that if you build
it customers will come.'' \2604\
---------------------------------------------------------------------------
\2600\ Texas Congressional Delegation, NHTSA-2018-0067-1421, at
1.
\2601\ Senator Inhofe, NHTSA-2018-0067-1422, at 1.
\2602\ CAS v. NHTSA, 793 F.2d 1322, 1340 (D.C. Cir. 1986), cited
by CARB, NHTSA-2018-0067-11873, at 79, and by States and Cities,
NHTSA-2018-0067-11735, at 69.
\2603\ Minnesota agencies, NHTSA-2018-0067-11706, at 4.
\2604\ Kreucher, NHTSA-2018-0067-0444, at 11-12.
---------------------------------------------------------------------------
The Minnesota agencies argued that focusing on consumer preferences
represented an ``unreasonable and unprecedented shift in
interpretation.'' \2605\ The States and Cities commenters stated
similarly that NHTSA had ``redefined `economically practicable' to
categorically exclude standards that, based on some unspecified metric,
`widely apply technologies that consumers do not want,' '' and argued
that ``NHTSA has offered no explanation for how it would define `wide
application,' much less how it would supposedly determine what
consumers do or do not want.'' \2606\ The States and Cities commenters
argued that it was internally inconsistent (and therefore arbitrary and
capricious) for NHTSA to rely in its justification on concerns about
consumer acceptance of technologies, while concurrently ``acknowledging
the `extensive debate over how much consumers do (and/or should) value
fuel savings and fuel economy as an attribute in new vehicles.' ''
\2607\ The States and Cities commenters stated that the NPRM's modeling
``assume[ed] that consumers assign no value to fuel savings
whatsoever,'' and that ``This assumption is not only implausible but
also flies in the face of the Agency's own statements that consumers
likely value between half of and all future fuel savings.'' \2608\
---------------------------------------------------------------------------
\2605\ Minnesota agencies, NHTSA-2018-0067-11706, at 4.
\2606\ States and Cities, NHTSA-2018-0067-11735, at 69 (citing
State Farm, 463 U.S. at 42-43).
\2607\ Id. (citing NPRM at 43216; Fox Television, 556 U.S. at
515, and United States Sugar Corp., 830 F.3d at 650).
\2608\ Id. at 70 (citing NPRM at 43073).
---------------------------------------------------------------------------
With regard to whether consumers do want more fuel economy, NESCAUM
stated that ``the most recent surveys indicate that consumers continue
to place a high value on fuel efficient vehicles of all types,'' \2609\
while Alliance for Vehicle Efficiency stated that ``Consumers have
adopted incremental changes to new vehicles that increase fuel economy
that don't compromise on power, size or safety.'' \2610\ The States and
Cities commenters argued that ``consumer choice is, in fact, enhanced
by providing consumers with the option of purchasing higher-efficiency
vehicles.'' \2611\ CBD et al. and the States and Cities commenters
stated that NHTSA had simply made assertions about consumer demands
without supporting evidence,\2612\ with the States and Cities
commenters also arguing that the fuel price assumptions in the NPRM
were ``unsupported'' and ``contradicted by recent evidence,'' despite
NHTSA's arguments that low fuel prices made ``fuel efficiency less
attractive to consumers.'' \2613\ Somewhat in contrast, NESCAUM stated
that ``[g]iven recent consumer preferences for larger vehicles,
maximizing fuel efficiency and GHG emission reductions in larger
footprint vehicles is even more important,'' noting that footprint
based standards ``are intentionally flexible to accommodate industry
and consumer preferences.'' \2614\ NESCAUM also stated that many HEV/
PHEV/EV models are now available and that their sales ``reflect[ ]
growing consumer acceptance of the technology, . . . despite the low
availability of electric vehicle models in the Northeast Section 177
States and the auto industry's continuing failure to actively market
[them].'' \2615\
---------------------------------------------------------------------------
\2609\ NESCAUM, NHTSA-2018-0067-11691, at 2.
\2610\ Alliance for Vehicle Efficiency, NHTSA-2018-0067-11696,
at 2.
\2611\ States and Cities, NHTSA-2018-0067-11735, at 70.
\2612\ CBD et al., NHTSA-2018-0067-12057, at 4; States and
Cities, NHTSA-2018-0067-11735, at 70.
\2613\ Id.
\2614\ NESCAUM, NHTSA-2018-0067-11691, at 2.
\2615\ Id. at 3.
---------------------------------------------------------------------------
[[Page 25134]]
Regarding the NPRM's statement that safety could be a subcomponent
of economic practicability, the States and Cities commenters stated
that this was ``an unreasonable interpretation of this factor, given
that safety concerns are not discussed in EPCA and have no direct
correlation to whether a standard is economically practicable.'' \2616\
The States and Cities commenters further stated that ``NHTSA has never
before analyzed safety considerations as falling under this factor, and
fails to explain its reason for doing so now,'' \2617\ and said that it
was ``unmoored from reality'' for NHTSA to state without support that
``[i]nvestment into the development and implementation of fuel saving
technology necessarily comes at the expense of investing in other areas
such as safety technology.'' \2618\ The States and Cities commenters
argued that investment in fuel economy rather than safety ``does not
explain why safety should be folded into a consideration of whether
standards are economically practicable.'' \2619\ IPI argued that ``[i]t
is arbitrary for NHTSA to count alleged safety costs as support for its
propose [sic] rollback both under the economic practicability factor
and as its own separate `bolster[ing] factor,' and yet never fully
monetize climate- and pollution-related deaths and other welfare
impacts under either the need to conserve energy factor nor under the
economic practicability factor.'' \2620\
---------------------------------------------------------------------------
\2616\ States and Cities, NHTSA-2018-0067-11735, at 70
(``arbitrary and capricious for agency to rely on factors `which
Congress has not intended it to consider' '') (citing Chevron, 467
U.S. at 843; State Farm, 463 U.S. at 43).
\2617\ Id. (citing Fox Television, 556 U.S. at 515).
\2618\ Id.
\2619\ Id.
\2620\ NYU IPI, NHTSA-2018-0067-12213, Appendix, at 6-7.
---------------------------------------------------------------------------
In response to these comments, NHTSA continues to believe that it
is reasonable to interpret ``economic practicability'' as the agency
has long interpreted it: As a question of whether a standard is one
``within the financial capability of the industry, but not so stringent
as to'' lead to ``adverse economic consequences, such as a significant
loss of jobs or the unreasonable elimination of consumer choice.''
\2621\ NHTSA disagrees that this interpretation is new or divergent
from past interpretations of economic practicability--this is, to the
word, the same interpretation set forth in the 2010 and 2012 final
rules, and in multiple earlier rules. Commenters disagreeing with the
NPRM's assessment of economic practicability seem, fundamentally, to be
disagreeing with how NHTSA applied this interpreted definition of
economic practicability to the information then before the agency, and
also with the agency's conclusion of how economic practicability
weighed against the other statutory factors.
---------------------------------------------------------------------------
\2621\ 67 FR 77015, 77021 (Dec. 16, 2002).
---------------------------------------------------------------------------
The following text explains why NHTSA continues to believe that the
pieces of the analysis it categorizes as relevant to economic
practicability fit within the long-standing definition of that factor.
Section VIII.B.4 below will explain how the agency has considered those
pieces of the analysis in balancing economic practicability with the
other statutory factors.
NHTSA has consistently described the manner in which it applies the
``economic practicability'' factor, and has given considerable weight
to the phrasing of this description. Parsing the words of this
description can be useful:
The core of the description is the phrase ``within the financial
capability of the industry,'' but not so stringent as to lead to
``adverse economic consequences.'' The following clause ``such as a
significant loss of jobs or the unreasonable elimination of consumer
choice'' is set off by a comma from ``consequences,'' and use of the
phrase ``such as'' indicates that it is a nonrestrictive clause.\2622\
A nonrestrictive clause means that ``significant loss of jobs'' and
``unreasonable elimination of consumer choice'' are examples of
``adverse economic consequences,'' but are not an exclusive list of the
possible adverse economic consequences that NHTSA may consider. Further
evidence that this clause was intended simply to offer examples comes
from the 1977 final rule establishing passenger car standards for MYs
1981-1984, in which NHTSA examined the potential meaning of ``economic
practicability'' at length and concluded that it should be interpreted
as ``requiring the standards to be within the financial capability of
the industry, but not so stringent as to threaten substantial economic
hardship for the industry,'' i.e., lacking the final clause.\2623\
---------------------------------------------------------------------------
\2622\ See Strunk, William and E.B. White, The Elements of
Style, Fourth Edition (2000), Rule 3, at 2-7.
\2623\ 42 FR 33534, 33537 (Jun. 30, 1977). It is worth noting
that the agency considered and rejected an interpretation of
economic practicability at that time based solely on cost-benefit
analysis, stating ``A cost-benefit analysis would be useful in
considering these factors [of economic practicability], but sole
reliance on such an analysis would be contrary to the mandate of the
act.'' Id.
---------------------------------------------------------------------------
A number of commenters took issue with NHTSA's consideration of
consumer demand, citing the 1986 D.C. Circuit decision CAS v. NHTSA for
the proposition that consumer demand cannot drive the balancing of
factors in determining maximum feasible standards. In that case, the
D.C. Circuit stated that ``[i]t is axiomatic that Congress intended
energy conservation to be a long term effort that would continue
through temporary improvements in energy availability. Thus, it would
clearly be impermissible for NHTSA to rely on consumer demand to such
an extent that it ignored the overarching goal of fuel conservation.''
\2624\ NHTSA agrees that the CAS decision makes this point, and that
the 9th Circuit decision in CBD v. NHTSA also underscored that the
overarching purpose of EPCA is energy conservation. That said, the CAS
decision also contains a number of other points that are relevant both
to the facts at hand in this rulemaking and NHTSA's current use of
consumer demand as an aspect of economic practicability and as a
consideration in determining maximum feasible standards. NHTSA will
discuss CAS more extensively below in Section VIII.B.4, but this
section will cover it briefly, specifically with respect to NHTSA's
interpretation of economic practicability.
---------------------------------------------------------------------------
\2624\ CAS, 793 F.2d 1322, 1340 (D.C Cir. 1986).
---------------------------------------------------------------------------
As noted in the NPRM and in the 2012 final rule, the CAS decision
found NHTSA's consideration of market demand as a component of economic
practicability reasonable.\2625\ In CAS, petitioners the Center for
Auto Safety, Public Citizen, Union of Concerned Scientists, and
Environmental Policy Institute sued NHTSA over CAFE standards for MY
1986, arguing that NHTSA could not determine stringency on the basis of
low expected consumer demand for fuel economy, and ``that technology
permitted greater fuel savings and that the statutorily required
`maximum feasible' level of fuel economy is higher than the standard''
determined by NHTSA.\2626\ The court followed Chevron in evaluating
whether NHTSA could consider consumer demand, and found that Congress
had not directly spoken to the consideration of consumer demand. The
court then assessed whether NHTSA's interpretation of the statute
``represents a reasonable accommodation of conflicting policies that
were committed to the agency's care by statute,'' stating that ``The
agency's interpretation of the statutory requirements is due
considerable deference and must be
[[Page 25135]]
found adequate if it falls within the range of permissible
constructions.'' \2627\
---------------------------------------------------------------------------
\2625\ 83 FR at 43208, fn. 402; 77 FR at 62668, fn. 111 (both
citing CAS, 793 F.2d 1322, 1338 (D.C. Cir. 1986)).
\2626\ CAS, at 1328.
\2627\ CAS, at 1338.
---------------------------------------------------------------------------
In assessing NHTSA's interpretation, the court stated that
``Consumer demand is not specifically designated as a factor, but
neither is it excluded from consideration; the factors of
`technological feasibility' and `economic practicability' are each
broad enough to encompass the concept. Thus, the unadorned language of
the statute does not indicate a congressional intent concerning the
precise objections raised by the petitioners.'' The court then examined
EPCA's legislative history and concluded that ``this language neither
precludes nor requires lower standards when consumer demand for heavy
vehicles is strong. The agency is directed to weigh the `difficulties
of individual automobile manufacturers;' there is no reason to conclude
that difficulties due to consumer demand for a certain mix of vehicles
should be excluded.'' \2628\ The court even noted that ``the
petitioners [did] not challenge the consideration of consumer demand
per se, but rather the weight the agency has given the factor in
downgrading standards . . . .'' \2629\
---------------------------------------------------------------------------
\2628\ CAS, at 1338-1339.
\2629\ CAS, at 1340.
---------------------------------------------------------------------------
NHTSA continues to believe that it is reasonable to consider
consumer demand as an element of economic practicability, as the CAS
court recognized. Comments objecting to the consideration of consumer
demand appear to focus more, like the petitioners in CAS, on the
agency's focus on consumer demand in the overall balancing of factors
to determine what CAFE standards would be maximum feasible, insofar as
they are expressing concern about consumer demand undermining energy
conservation. Again, this question will be addressed further in Section
VIII.B.4 below. To the extent that commenters dispute any consideration
of consumer demand, the D.C. Circuit put that question to rest decades
ago.
Related to the agency's consideration of consumer demand, a number
of commenters took issue with the agencies' estimates of the cost of
meeting higher fuel economy standards, arguing essentially that the
analysis was deliberately constructed to inflate costs and minimize
consumer willingness to pay for fuel economy improvements in order to
arrive at a policy conclusion that higher fuel economy standards would
not be economically practicable. NHTSA does not believe that commenters
mean to argue with the agency's legal interpretation (i.e., the
consideration of cost as an aspect of economic practicability), but
rather with the agencies' analytical findings which inform that
consideration. Comments on those analytical findings, and the agencies'
responses and changes to the analysis in response to those comments,
are discussed in Sections VI and VII above. Consumer willingness to pay
for additional fuel economy in their new vehicles, in particular, is
represented throughout the final rule analysis as 2.5 years--that is,
that consumers value, and manufacturers will voluntarily add, fuel
economy-improving technology that pays for itself in fuel savings
within 2.5 years.
More generally, NHTSA believes that the cost of meeting CAFE
standards is inherently relevant to assessing whether those standards
are ``within the financial capability of the industry but not so
stringent as to lead to adverse economic consequences,'' for two
primary reasons. First, vehicle manufacturers tend to have relatively
fixed budgets for R&D and production, which are tied to overall
revenues. If more of those budgets are spent on improving fuel economy,
less of those budgets are available to spend on other vehicle
characteristics (such as advanced safety features, or better
performance or utility) that might improve sales. Offering less of
those other vehicle characteristics in a market where many consumers
are not particularly focused on fuel economy could lead to adverse
economic consequences for those manufacturers. Manufacturers cannot
simply increase budgets or turn limited resources toward supplying more
of vehicle characteristics that do not motivate most sales. To the
extent that more stringent standards drive manufacturing costs higher
and those costs are passed forward to consumers in the form of price
increases, those price increases can affect vehicle sales to some
extent. NHTSA understands that some commenters disagree that higher
manufacturing costs are necessarily passed forward to consumers in the
way that the agencies have modeled them being passed forward, but the
agencies do not have adequate information on which to base a different
approach. Commenters disagreeing with this approach generally object on
two fronts: First, because they believe that automakers cross-subsidize
cost increases by raising the prices of certain models rather than all
models, and second, because they believe that automakers could absorb
regulatory costs and reduce profits. The agencies do not have enough
information to model either of those issues in a meaningful way. Some
amount of cross-subsidization no doubt occurs, but automakers closely
hold pricing strategy information. The agencies do not attempt to model
automakers voluntarily reducing profits in response to standards, again
in part because the agencies do not have sufficient information, but
also because these companies are publicly-traded and taking losses is
not a long-term solution for companies whose success is measured by
profitability. NHTSA believes that the analytical approach used today
is reasonable given the information available to the agencies. While
today's analysis does not show large sales effects due to price
increases, and even accounting for fuel economy differences in this
final rule still does not show large sales effects, it seems reasonable
to call negative sales effects ``adverse economic consequences.''
Also related to consumer demand, NHTSA has previously considered
manufacturer ``shortfalls'' as an aspect of economic
practicability.\2630\ The CAFE standards are corporate average
standards, by definition, giving manufacturers the flexibility to
decide how to distribute fuel economy-improving technologies throughout
their fleet. In other words, no given vehicle need, itself, meet a
standard or even its ``target'' on the target curve, as long as the
fleet as a whole meets the standard. However, CAFE compliance is
measured on a sales-weighted basis, so if a manufacturer ultimately
sells more vehicles that perform poorly relative to their targets than
it sells vehicles that beat their targets, the manufacturer may fall
short of its compliance obligation despite having applied fuel economy-
improving technologies in amounts that the manufacturer originally
anticipated would result in compliance. Recent compliance trends have
illustrated this phenomenon, as discussed in Section IV above. When
fuel is relatively inexpensive, Americans tend to be less interested in
saving money on fuel, and thus less interested in fuel economy as
compared to other vehicle attributes. Compliance shortfalls represent
this consumer decision-making playing out in the market, and can thus
be evidence of economic impracticability if sufficiently
widespread.\2631\
---------------------------------------------------------------------------
\2630\ See 77 FR at 63040-43 (Oct. 15, 2012).
\2631\ See, e.g., Alliance comments (Full Comment Set) at 25-29,
describing automaker shortfalls in terms of fleet fuel economy
increases required by augural and prior standards.
---------------------------------------------------------------------------
As with the above-discussed aspects of economic practicability,
commenters who objected to NHTSA's consideration
[[Page 25136]]
of employment impacts disagreed less with the principle of considering
employment impacts, and more with how NHTSA discussed employment
impacts in the proposal's justification given the NPRM's findings on
employment. Namely, the NPRM included a simplistic analysis that
converted reduced technology costs under the preferred alternative
relative to the augural standards into ``job years'' metric and
estimated U.S. auto sector labor would be slightly reduced under the
proposal as compared to under the augural standards (reflecting those
reduced technology costs). Although new vehicle sales increased
slightly under the NPRM's preferred alternative, this was offset
because ``manufacturing, integrating, and selling less technology means
using less labor to do so.'' \2632\ However, NHTSA expressed concern in
the proposal justification section that ``there could be potential for
. . . loss of U.S. jobs . . . under nearly all if not all of the
regulatory alternatives considered . . . .'' \2633\ A number of
commenters argued that if more stringent standards led to higher
employment, as the NPRM (and also outside analyses) appeared to show,
there was no way that less stringent standards could be more
economically practicable.
---------------------------------------------------------------------------
\2632\ 83 FR at 43436 (Aug. 24, 2018).
\2633\ Id. at 43216.
---------------------------------------------------------------------------
As in the NPRM, NHTSA recognizes that the employment analysis for
this final rule does not capture certain potential effects that may be
important. NHTSA explained in the NPRM that the NPRM's employment
analysis did not account for the risks that vehicle sales may be facing
a bubble situation, or that manufacturers facing higher production
costs might choose to move production overseas.\2634\ This topic is
discussed at greater length in Section VIII.B.4 below.
---------------------------------------------------------------------------
\2634\ Id. at 43224-25.
---------------------------------------------------------------------------
Commenters addressing NHTSA's consideration of safety as an aspect
of economic practicability argued generally that EPCA did not call for
discussion of safety concerns, and that it was unreasonable to assume
that requiring higher levels of fuel economy might preclude investment
in further vehicle safety improvements. NHTSA has already explained
above that the long-standing definition of ``economic practicability''
lists example ``adverse economic consequences'' in a nonrestrictive
clause format, meaning that other things besides employment and
consumer choice impacts could cause economic consequences and be
relevant to economic practicability. NHTSA believes that it is
reasonable and appropriate to consider some aspects of safety as part
of its consideration of economic practicability, because NHTSA
continues to believe that vehicle manufacturers have finite budgets for
R&D and production that may be spent on fuel economy improvements when
they may otherwise be spent on safety improvements, among other things
that consumer value. Some commenters said that that was not a
reasonable assumption, but it is supported by statements from vehicle
manufacturers,\2635\ and NHTSA does not have a reason to disbelieve
that companies have limited budgets. Moreover, case law does not object
to consideration of safety as an aspect of economic
practicability.\2636\ With regard to IPI's comment about monetization
of climate and pollution-related deaths and other welfare impacts, the
social cost of carbon and criteria pollutant damages estimates are
intended to account for these impacts, and are considered both as part
of the cost-benefit analysis and under the environmental implications
aspect of the need of the U.S. to conserve energy. Given that the
decision about what standards are ``maximum feasible'' is made by
considering all of the factors, it is therefore less relevant under
which factor a given issue is considered, so long as it is
appropriately considered. To the extent that IPI disagrees with those
estimated valuations, Section VI discusses comments on those topics and
the agencies' responses.
---------------------------------------------------------------------------
\2635\ See, e.g., Toyota comments at 6, NHTSA-2018-0067-12098
(``There are now more realistic limits placed on the number of
engines and transmissions in a powertrain portfolio which better
recognizes manufacturers must manage limited engineering resources
and control supplier, production, and service costs.'').
\2636\ Competitive Enterprise Institute v. NHTSA, 901 F.2d 107,
120, n. 11 (``Petitioners have never clearly identified the precise
statutory basis on which safety concerns should be factored into the
CAFE scheme, although they alluded to occupant safety as part of the
`economic practicability' criterion in their MY 1989 petition to
NHTSA and at oral argument. We do not find this failure fatal,
however, because NHTSA has always examined the safety consequences
of the CAFE standards in its overall consideration of relevant
factors since its earliest rulemaking under the CAFE program,
(citations omitted). Moreover, NHTSA itself believes Congress was
cognizant of safety issues when it enacted the CAFE program. As
evidence, NHTSA discusses a congressional report that dealt with the
safety consequences of a downsized fleet of cars which had been
considered by Congress during its enactment of the CAFE program.'').
---------------------------------------------------------------------------
Based on the above, NHTSA continues to believe that its
interpretation of economic practicability is reasonable. Section
VIII.B.4 will discuss how NHTSA has considered and balanced economic
practicability for this final rule, and also respond to comments that
addressed the NPRM's application of economic practicability to the
information before the agency at that time.
(3) The Effect of Other Motor Vehicle Standards of the Government on
Fuel Economy
``The effect of other motor vehicle standards of the Government on
fuel economy'' involves analysis of the effects of compliance with
emission, safety, noise, or damageability standards on fuel economy
capability and thus on average fuel economy. In many past CAFE
rulemakings, NHTSA has said that it considers the adverse effects of
other motor vehicle standards on fuel economy. It said so because, from
the CAFE program's earliest years \2637\ until recently, the effects of
such compliance on fuel economy capability over the history of the CAFE
program have been negative ones. For example, safety standards that
have the effect of increasing vehicle weight thereby lower fuel economy
capability, thus decreasing the level of average fuel economy that
NHTSA can determine to be feasible. In the analyses for both the NPRM
and this final rule, NHTSA has considered the additional weight that it
estimates would be added in response to new safety standards during the
rulemaking timeframe.\2638\ NHTSA has also accounted for EPA's ``Tier
3'' standards for criteria pollutants in its estimates of technology
effectiveness in both the NPRM and final rule analyses.\2639\
---------------------------------------------------------------------------
\2637\ 42 FR 63184, 63188 (Dec. 15, 1977). See also 42 FR 33534,
33537 (Jun. 30, 1977).
\2638\ PRIA, Chapter 5; FRIA, Section 5.
\2639\ PRIA, Chapter 6; FRIA, Section 6.
---------------------------------------------------------------------------
NHTSA discussed in the NPRM whether to consider EPA's
CO2 standards as an ``other motor vehicle standard of the
Government'' among the other regulations typically considered, and if
so, how. NHTSA explained that in the 2012 final rule establishing CAFE
standards for MYs 2017-2021, NHTSA recognized that ``To the extent the
GHG standards result in increases in fuel economy, they would do so
almost exclusively as a result of inducing manufacturers to install the
same types of technologies used by manufacturers in complying with the
CAFE standards.'' \2640\ NHTSA concluded in 2012 that ``no further
action was needed'' because ``the agency had already considered EPA's
[action] and the harmonization benefits of the National Program in
developing its own [action].'' \2641\
---------------------------------------------------------------------------
\2640\ 77 FR 62624, 62669 (Oct. 15, 2012).
\2641\ Id.
---------------------------------------------------------------------------
[[Page 25137]]
In the NPRM, NHTSA considered the issue afresh, and determined that
it was clear based on a purely textual analysis of the statutory
language that EPA's CO2 standards applicable to light-duty
vehicles are literally ``other motor vehicle standards of the
Government,'' in that they are standards set by a Federal agency that
apply to motor vehicles. Basic chemistry makes fuel economy and
tailpipe CO2 emissions two sides of the same coin, as
discussed at length above, and when two agencies functionally regulate
both (because when regulating fuel economy, CO2 emissions
are necessarily also regulated, and vice versa), it would be absurd not
to link the standards.\2642\ The global warming potential of
N2O, CH4, and HFC emissions are not closely
linked with fuel economy, but neither do they affect fuel economy
capabilities. Simply concluding that EPA's CO2 standards
were ``other motor vehicle standards of the Government,'' however, did
not answer how should NHTSA should consider them.
---------------------------------------------------------------------------
\2642\ In fact, EPA includes tailpipe CH4, CO, and
CO2 in the measurement of tailpipe CO2 for
CO2 compliance using a carbon balance equation so that
the measurement of tailpipe CO2 exactly aligns with the
measurement of fuel economy for the CAFE compliance.
---------------------------------------------------------------------------
NHTSA acknowledged in the NPRM that some stakeholders had
previously suggested that NHTSA should implement this statutory factor
by letting EPA decide what CO2 standards are appropriate and
reasonable under the CAA and then simply setting CAFE standards with
reference to CO2 stringency. NHTSA disagreed that such an
approach would be a reasonable interpretation of EPCA, explaining that
while EPA and NHTSA consider some similar factors under the CAA and
EPCA/EISA, respectively, they are not identical, and standards that are
appropriate under the CAA may not be ``maximum feasible'' under EPCA/
EISA, and vice versa. Moreover, NHTSA explained, considering EPCA's
language in the context in which it was written, it seemed unreasonable
to conclude that Congress intended EPA to dictate CAFE stringency. In
fact, Congress clearly separated NHTSA's and EPA's responsibilities for
CAFE under EPCA by giving NHTSA authority to set standards and EPA
authority to measure and calculate fuel economy. If Congress had wanted
EPA to set CAFE standards, it could have given that authority to EPA in
EPCA or at any point since Congress amended EPCA.\2643\
---------------------------------------------------------------------------
\2643\ The NPRM noted, for instance, that EISA was passed after
the Massachusetts v. EPA decision by the Supreme Court. If Congress
had wanted to amend EPCA in light of that decision, it would have
done so at that time, but did not.
---------------------------------------------------------------------------
NHTSA explained that NHTSA and EPA are obligated by Congress to
exercise their own independent judgment in fulfilling their statutory
missions, even though both agencies' regulations affect both fuel
economy and CO2 emissions. Because of this relationship, it
is incumbent on both agencies to coordinate and look to one another's
actions to avoid unreasonably burdening industry through inconsistent
regulations,\2644\ but both agencies' programs must stand on their own
merits. As with other recent CAFE and CO2 rulemakings, NHTSA
explained that the agencies were continuing do all of these things in
the proposal.
---------------------------------------------------------------------------
\2644\ Massachusetts v. EPA, 549 U.S. 497, 532 (2007) (``[T]here
is no reason to think the two agencies cannot both administer their
obligations and yet avoid inconsistency.'').
---------------------------------------------------------------------------
With regard to standards issued by the State of California, the
NPRM explained that State tailpipe standards (whether for
CO2 or for other pollutants) do not qualify as ``other motor
vehicle standards of the Government'' under 49 U.S.C. 32902(f), and
that therefore, NHTSA would not consider them as such in proposing
maximum feasible average fuel economy standards. NHTSA explained that
States may not adopt or enforce standards related to fuel economy
standards, which are preempted under EPCA, regardless of whether EPA
granted any waivers under the Clean Air Act (CAA).
NHTSA and EPA agreed in the NPRM that State tailpipe CO2
emissions standards do not become Federal standards and qualify as
``other motor vehicle standards of the Government,'' when subject to a
CAA preemption waiver. NHTSA stated that EPCA's legislative history
supports that position, as follows:
EPCA, as initially passed in 1975, mandated average fuel economy
standards for passenger cars beginning with model year 1978. The law
required the Secretary of Transportation to establish, through
regulation, maximum feasible fuel economy standards \2645\ for model
years 1981 through 1984 with the intent to provide steady increases to
achieve the standard established for 1985 and thereafter authorized the
Secretary to adjust that standard.
---------------------------------------------------------------------------
\2645\ As is the case today, EPCA required the Secretary to
determine ``maximum feasible average fuel economy'' after
considering technological feasibility, economic practicability, the
effect of other Federal motor vehicle standards on fuel economy, and
the need of the Nation to conserve energy. 15 U.S.C. 2002(e)
(recodified July 5, 1994).
---------------------------------------------------------------------------
For the statutorily-established standards for model years 1978-
1980, EPCA provided each manufacturer with the right to petition for
changes in the standards applicable to that manufacturer. A petitioning
manufacturer had the burden of demonstrating a ``Federal fuel economy
standards reduction'' was likely to exist for that manufacturer in one
or more of those model years and that it had made reasonable technology
choices. ``Federal standards,'' for that limited purpose, included not
only safety standards, noise emission standards, property loss
reduction standards, and emission standards issued under various
Federal statutes, but also ``emissions standards applicable by reason
of section 209(b) of [the CAA].'' \2646\ (Emphasis added). Critically,
all definitions, processes, and required findings regarding a Federal
fuel economy standards reduction were located within a single self-
contained subsection of 15 U.S.C. 2002 that applied only to model years
1978-1980.\2647\
---------------------------------------------------------------------------
\2646\ Section 202 of the CAA (42 U.S.C. 7521) requires EPA to
prescribe air pollutant emission standards for new vehicles; Section
209 of the CAA (42 U.S.C. 7543) preempts state emissions standards
but allows California to apply for a waiver of such preemption.
\2647\ As originally enacted as part of Public Law 94-163, that
subsection was designated as section 502(d) of the Motor Vehicle
Information and Cost Savings Act.
---------------------------------------------------------------------------
In 1994, Congress recodified EPCA. As part of this recodification,
the CAFE provisions were moved to Title 49 of the United States Code.
In doing so, unnecessary provisions were deleted. Specifically, the
recodification eliminated subsection (d). The House report on the
recodification declared that the subdivision was ``executed,'' and
described its purpose as ``[p]rovid[ing] for modification of average
fuel economy standards for model years 1978, 1979, and 1980.'' \2648\
It is generally presumed, when Congress includes text in one section
and not in another, that Congress knew what it was doing and made the
decision deliberately.
---------------------------------------------------------------------------
\2648\ H.R. Rep. No. 103-180, at 583-584, tbl. 2A.
---------------------------------------------------------------------------
NHTSA stated in the NPRM that it had previously considered the
impact of California's Low Emission Vehicle standards in establishing
fuel economy standards and occasionally has done so under the ``other
standards'' sections.\2649\ During the 2012 rulemaking, NHTSA sought
comment on the appropriateness of considering California's tailpipe
CO2 emission standards in this section and concluded that
doing so was unnecessary.\2650\ In light of the legislative history
discussed above, however, NHTSA stated in the NPRM that such
consideration would be inappropriate, and confirms that consideration
of California's LEV
[[Page 25138]]
standards as among the ``other standards of the Government'' was
inappropriate.
---------------------------------------------------------------------------
\2649\ See, e.g., 68 FR 16896, 71 FR 17643.
\2650\ See 77 FR 62669.
---------------------------------------------------------------------------
Commenters addressing criteria pollutant standards generally
supported NHTSA's approach in the NPRM. AFPM commented that NHTSA
``must consider the effect on fuel economy of EPA's Title II standards,
including the use of catalytic converters, PM traps and other
technologies that address emissions and have a fuel economy impact.''
\2651\ Ford also stated that previous analyses ``did not assess the
impact of the criteria pollutant emission standards that were adopted
subsequent to the [2012 final rule],'' which Ford said ``increased the
challenge of meeting the fuel economy and GHG targets and should be
taken into consideration.'' \2652\ Ford stated that the NPRM
appropriately included ``updat[ed] core engine maps using correct,
regular-grade octane test fuel,'' and that it accounts for ``ultra-low
2025 MY Tier 3 and LEVIII emissions standards [which] will require
aggressive cold start strategies [that] consume additional fuel at
start-up in order to rapidly heat the catalyst to an effective
operating temperature, which degrades CO2 and fuel economy
performance on the FTP test [and] was not considered previously. . .
.'' \2653\
---------------------------------------------------------------------------
\2651\ AFPM, NHTSA-2018-0067-12078, at 52.
\2652\ Ford, NHTSA-2018-0067-11928, at 7.
\2653\ Id.
---------------------------------------------------------------------------
Regarding how NHTSA should consider EPA's CO2 standards
as ``other motor vehicle standards of the Government,'' ACEEE suggested
amongst its comments that, in considering EPA's CO2
standards, ``NHTSA should not weaken its program . . . to compensate
for . . . inevitable, modest differences'' between EPA's and NHTSA's
programs.\2654\ ``Indeed, to the extent that differences in the
requirements of the two programs remain, it is clear that the more
stringent requirement in any given respect should govern the
obligations of the manufacturer.'' \2655\ AFPM commented similarly that
``Although NHTSA must consider the effect of other governmental
regulations, Congress intended that NHTSA would have exclusive
authority over a single set of national fuel economy standards.''
\2656\ Mr. Dotson expressed his belief that ``Congress was cognizant of
the relationship between EPCA and the Clean Air Act when crafting
EISA'' and cited and discussed various types of legislative history for
the proposition that EISA had not limited EPA's CAA authority, and that
various legislative efforts to do so had been put forth in some fashion
and had failed.\2657\
---------------------------------------------------------------------------
\2654\ ACEEE, NHTSA-2018-0067-12122, joint NGO comment to
Alliance/Global petition for flexibilities, at 3.
\2655\ Id.
\2656\ AFPM, NHTSA-2018-0067-12078, at 52.
\2657\ Dotson, EPA-HQ-OAR-2018-0283-4132, Appendix A, at A2-A23.
NHTSA disagrees with the persuasiveness of the legislative history
cited by Mr. Dotson, which includes floor debates, colloquies, and
other similar information that does not reflect the agreement of the
Congress as a whole. NHTSA looks to the language Congress actually
passed and the President signed into law.
---------------------------------------------------------------------------
NHTSA agrees that while it is appropriate for NHTSA to coordinate
with and look to EPA's actions to avoid unreasonably burdening industry
through inconsistent regulations, it would not be appropriate for NHTSA
to reduce stringency below levels it believes to be maximum feasible
solely for purposes of accommodating differences between programmatic
flexibilities. The 2012 final rule clearly stated that while the
agencies had made efforts to align their standards, programmatic
differences existed, and how manufacturers chose to rely on compliance
flexibilities could affect the relative stringency of NHTSA's and EPA's
standards:
We note, however, that the alignment is based on the assumption
that manufacturers implement the same level of direct A/C system
improvements as EPA currently forecasts for those model years, and
on the assumption of PHEV, EV, and FCV penetration at specific
levels. If a manufacturer implements a higher level of direct A/C
improvement technology (although EPA predicts 100% of manufacturers
will use substitute refrigerants by MY 2021, and the GHG standards
assume this rate of substitution) and/or a higher penetration of
PHEVs, EVs and FCVs, then NHTSA's standards would effectively be
more stringent than EPA's. Conversely, if a manufacturer implements
a lower level of direct A/C improvement technology and/or a lower
penetration of PHEVs, EVs and FCVs, then EPA's standards would
effectively be more stringent than NHTSA's. Several manufacturers
commented on this point and suggested that this meant the standards
were not aligned, because NHTSA's standards might be more stringent
in some years than EPA's. This reflects a misunderstanding of the
agencies' purpose. The agencies have sought to craft harmonized
standards such that manufacturers may build a single fleet of
vehicles to meet both agencies' requirements. That is the case for
these final standards. Manufacturers will have to plan their
compliance strategies considering both the NHTSA standards and the
EPA standards and assure that they are in compliance with both, but
they can still build a single fleet of vehicles to accomplish that
goal.\2658\
---------------------------------------------------------------------------
\2658\ 77 FR at 63054-55 (Oct. 15, 2012) (emphasis added).
Thus, NHTSA has been consistent in its position that CO2
stringency does not and should not, by itself, dictate CAFE stringency.
That said, consideration of EPA's standards was inherent in development
of this final rule, given that the same technologies improve fuel
economy and reduce CO2 emissions, and given that
CO2 emissions represent the majority of GHGs produced by
light-duty vehicles, and given that the agencies have conducted the
analysis for this rulemaking jointly. NHTSA believes that EPA's
standards have been fully and appropriately considered as part of its
decision on these final standards. To be clear, NHTSA did not assert in
the NPRM that EISA constrained EPA's authorities under the CAA and do
---------------------------------------------------------------------------
not disagree with that aspect of Mr. Dotson's comment.
Chemours argued that, contrary to the NPRM's statements about
having considered EPA's GHG standards in developing the proposal, NHTSA
had not adequately considered EPA's GHG standards because only the no-
action alternative reflected EPA regulation of the non-CO2
GHGs, and the analysis did not otherwise account for the non-
CO2 GHG standards.\2659\ Chemours stated that those
standards were ``required, pursuant to CAA section 202(a), to address
`air pollution' from mobile sources,'' and that ``No assessment was
done as to whether such standards could be made less stringent in order
to avoid the various issues identified (e.g., changes in technology
since the 2012 final rule, costs to consumers, the effect of
`diminishing returns,' a changed petroleum market and other factors.''
\2660\
---------------------------------------------------------------------------
\2659\ Chemours, NHTSA-2018-0067-12018, at 25.
\2660\ Id. at 25-26.
---------------------------------------------------------------------------
NHTSA disagrees that it was necessary for NHTSA to consider EPA's
standards for non-CO2 GHG emissions any further than as
discussed above. Regulation of CH4, N2O, and HFCs
affects fuel economy only indirectly, if at all. As explained above and
in the 2012 final rule, while NHTSA recognizes that some manufacturers
may choose paths to compliance with EPA's GHG standards that make their
compliance with CAFE standards more challenging, the agencies previewed
this possibility and stated their expectation that manufacturers could
make these decisions for themselves. To the extent that Chemours is
asking NHTSA to examine regulatory alternatives reflecting less
stringent CAFE standards in light of changed conditions since the 2012
final rule, that is exactly what the NPRM and final rule analyses have
done.
A number of commenters disagreed with NHTSA's explanation of how
State standards need not be considered under this factor. The States
and Cities
[[Page 25139]]
commenters stated that NHTSA was required to consider State tailpipe
standards because 49 U.S.C. 32902(f) does not specify that
``Government'' refers only to ``Federal'' government; because NHTSA had
not offered compelling evidence or arguments that Congress did not
intend NHTSA to consider State tailpipe standards; and because ``case
law . . . states unequivocally that California's standards must be
considered by NHTSA under this factor [citing Green Mountain Chrysler's
``federalizing'' language].'' \2661\ The States and Cities commenters
further argued that NHTSA was trying to argue simultaneously that it
could not consider State standards under the ``other standards'' factor
but could consider State standards ``under other EPCA factors, if and
when it sees fit'' (citing NPRM language that technological feasibility
and economic practicability are broad factors allowing NHTSA to
consider elements not specifically designated by Congress).\2662\ The
States and Cities commenters further argued, citing Fox Television,
that NHTSA was deviating from past practice without a reasoned
explanation by not specifically requesting comment in the NPRM on the
fact that it was not considering California's standards as ``other
motor vehicle standards of the Government.'' \2663\
---------------------------------------------------------------------------
\2661\ States and Cities, NHTSA-2018-0067-12018, at 71.
\2662\ Id. at 71-72.
\2663\ Id. at 72. Fox Television did not involve a rulemaking,
and does not require agencies to specifically seek public comment
when they deviate from past practice. In any event, by articulating
in the NPRM that NHTSA was not considering California's standards as
``other motor vehicle standards of the Government'' the public had
ample opportunity to provide comment on this issue, and commenters
in fact did so as discussed above.
---------------------------------------------------------------------------
With regard to NHTSA's analysis of EPCA's original language for MYs
1978-80 and the 1994 positive law recodification, the States and Cities
commenters stated that ``NHTSA's statutory and legislative history
arguments related to standards for model years 1978-1980 lack merit, as
NHTSA has provided no reasonable argument that Congress meant NHTSA to
consider a wider range of standards for those years than for others,''
and stated that the section in question ``was removed from the statute
because it expired, not because Congress took issue with NHTSA's
consideration of California's waiver standards.'' \2664\ Mr. Dotson
commented similarly that NHTSA could not rely on the 1994 positive law
codification as basis to conclude that State tailpipe standards
(whether for GHGs or other emissions) do not qualify as ``other motor
vehicle standards of the Government,'' because it said ``without
substantive change. . . .'' \2665\
---------------------------------------------------------------------------
\2664\ Id. at 71.
\2665\ Dotson, EPA-HQ-OAR-2018-0283-4132, Appendix A, at A23-
A24.
---------------------------------------------------------------------------
Additionally, the States and Cities commenters stated that NHTSA
could not argue that California's emissions standards are not ``other
motor vehicle standards of the Government'' because they are preempted,
because NHTSA ``has no authority to decide whether or not California's
standards are preempted,'' and ``one of the reasons California's
Advanced Clean Cars program is not preempted by EPCA is because those
standards are `other motor vehicle standards of the Government' within
the meaning of EPCA.'' \2666\ Besides this comment, a number of
comments were submitted regarding NHTSA's statements in the NPRM about
EPCA's preemption provision and how it applied to California's
standards. Those comments have been addressed \2667\ as part of the
separate final rule published on September 27, 2019,\2668\ and will not
be discussed further as part of this action.
---------------------------------------------------------------------------
\2666\ States and Cities, NHTSA-2018-0067-12018, at 71.
\2667\ To the extent that any individual comment was not
specifically addressed, NHTSA believes that the substance and themes
of all substantive comments on EPCA preemption were addressed as
part of that final rule.
\2668\ 84 FR 51310.
---------------------------------------------------------------------------
NHTSA affirms that its interpretation set forth in the NPRM that
``other motor vehicle standards of the Government'' does not apply to
State emissions standards that relate to fuel economy. NHTSA does not
understand how 49 U.S.C. 32919 could be given effect if the purpose of
the ``other motor vehicle standards of the Government'' provision is to
compel their inclusion in NHTSA's decision-making. NHTSA continues to
disagree with the two district court cases suggesting that the ``other
motor vehicle standards of the Government'' provision obviates 49
U.S.C. 32919, as explained at some length in the ``One National
Program'' final rule preceding this regulatory action.\2669\ NHTSA
refers readers to that document for more detail on this topic.
---------------------------------------------------------------------------
\2669\ See, e.g., 84 FR at 51323 (Sep. 27, 2019).
---------------------------------------------------------------------------
With regard to State tailpipe standards that do not directly relate
to fuel economy, NHTSA continues to believe that Congress's original
direction to consider ``emissions standards applicable by reason of
section 209(b) of [the CAA]'' applied only to CAFE standards for MYs
1978-1980, as discussed in the NPRM. NHTSA agrees that the 1994
positive law recodification was not intended to make substantive
changes to EPCA; the NPRM explained that, in dropping Section 502(d),
Congress made clear that that provision was executed, and that
provision expressly directed NHTSA to consider State standards that had
been granted preemption waivers under CAA 209(b). In order for States
even to have their own emissions standards for motor vehicles,
California must be granted a waiver of preemption under CAA section
209(b). If Congress had intended for NHTSA to continue to consider
State tailpipe standards post-MY 1980, the direction to consider
emissions standards that had been granted Section 209 waivers could
have been placed elsewhere in the statute. Congress did not do
so.\2670\ While NHTSA may have considered State tailpipe standards in
the past, it is not bound to do so, and NHTSA does not believe that it
is unreasonable to consider those standards under technological
feasibility or economic practicability if they are to be considered.
---------------------------------------------------------------------------
\2670\ The negative inference canon is logically and reasonably
employed here, particularly given that, as a factual matter and as
discussed further below, considering EPA's Tier 3 standards (which
are clearly ``other motor vehicle standards of the Government'')
effectively accounts for the technological implications of
California's LEVIII standards.
---------------------------------------------------------------------------
State tailpipe standards primarily affect fuel economy by requiring
gasoline ICE vehicles to burn additional fuel when the engine first
starts. For most gasoline engines on the road today, the majority of
tailpipe NOX, NMOG, and CO emissions occur during ``cold
start,'' before the three-way catalyst has reached the very high
temperature (e.g., 900-1000 [deg]F), at which point it is able to
convert (through oxidation and reduction reactions) those emissions
into less harmful derivatives. By strictly limiting the amount of those
emissions, tailpipe smog standards require the catalyst to be brought
to temperature extremely quickly, so modern vehicles employ cold start
strategies that intentionally release fuel energy into the engine
exhaust to heat the catalyst to the relevant temperature as quickly as
possible. The additional fuel that must be used to heat the catalyst is
typically referred to as a ``cold-start penalty,'' meaning that
vehicle's fuel economy (over a test cycle) is reduced because the fuel
consumed to heat the catalyst did not go toward the goal of moving the
vehicle forward.\2671\ The Autonomie
[[Page 25140]]
work employed to develop technology effectiveness estimates for this
final rule does, in fact, account for cold-start penalties.\2672\ The
Autonomie model documentation discusses the fact that cold-start
penalties were derived from an EPA database of MY 2016 vehicles, which
would have met both EPA and California smog standards. Moreover, EPA
regulations allow manufacturers to employ LEVIII data for Tier 3
compliance. Based on all of these factors, NHTSA believes that the
negative fuel economy effects of California's tailpipe standards for
smog-related emissions are reasonably represented in the analysis for
the final rule, regardless of whether NHTSA was obligated by law to
consider them expressly.
---------------------------------------------------------------------------
\2671\ For more information on this, see, e.g., Pihl, Josh A.,
et al., ``Development of a Cold Start Fuel Penalty Metric for
Evaluating the Impact of Fuel Composition Changes on SI Engine
Emissions Control,'' Oak Ridge National Laboratory, 2018. Available
at https://www.osti.gov/biblio/1462896-development-cold-start-fuel-penalty-metric-evaluating-impact-fuel-composition-changes-si-engine-emissions-control.
\2672\ See ANL Model Documentation, Section 6.1.5, available in
Docket No. NHTSA-2018-0067.
---------------------------------------------------------------------------
Ultimately, it would be illogical for NHTSA to consider legally
unenforceable standards to be ``other motor vehicle standards of the
Government.'' That is the case for State standards preempted by EPCA.
While NHTSA understands that certain commenters disagree with a
separate final rule that NHTSA issued concerning EPCA preemption, and
the particular State standards that NHTSA considers preempted by EPCA,
those issues are outside the scope of this final rule.
(4) The Need of the United States To Conserve Energy
NHTSA has historically interpreted ``the need of the United States
to conserve energy'' to mean ``the consumer cost, national balance of
payments, environmental, and foreign policy implications of our need
for large quantities of petroleum, especially imported petroleum.''
\2673\
---------------------------------------------------------------------------
\2673\ 42 FR 63184, 63188 (Dec. 15, 1977).
---------------------------------------------------------------------------
(a) Consumer Costs and Fuel Prices:
NHTSA explained in the NPRM that fuel for vehicles costs money for
vehicle owners and operators. All else equal--a critical caveat--
consumers benefit from vehicles that need less fuel to perform the same
amount of work. Future fuel prices are a critical input into the
economic analysis of potential CAFE standards because they determine
the value of fuel savings both to new vehicle buyers and to society,
the amount of fuel economy that the new vehicle market is likely to
demand in the absence of new standards, and they inform NHTSA about the
``consumer cost . . . of our need for large quantities of petroleum.''
In the proposal, NHTSA's analysis relied on fuel price projections from
the U.S. Energy Information Administration's (EIA) Annual Energy
Outlook (AEO) for 2017; in the final rule, on fuel price projections
derived from the version of NEMS used to produce AEO 2019. Federal
government agencies generally use EIA's price projections in their
assessment of future energy-related policies.
Several commenters stated that consumer costs for fuel were an
important consideration. ACEEE stated that ``The average U.S. household
still spent nearly $2,000 on gasoline and motor oil (directly) in 2017,
making oil savings very relevant for consumers,'' and argued that ``Oil
price volatility remains a threat to U.S. consumers and businesses--the
price of crude oil has more than doubled since 2016, belying the
theoretical suggestion in the notice that conditions for oil price
shocks no longer exist,'' suggesting that further fuel efficiency
improvements were necessary to protect consumers.\2674\ NESCAUM
commented that prior analyses had suggested that consumers would save
$6,000 on net, after paying more for their vehicles upfront, and that
the proposal would cost consumers more in fuel.\2675\ Both NESCAUM and
the States and Cities commenters stated that higher fuel costs would
disproportionately affect low-income consumers, who spend a higher
share of their income on fuel costs.\2676\ The Congressional Tri-Caucus
commented that ``As we see oil prices rising again, it makes no sense
for DOT to roll back these standards.'' \2677\ The States and Cities
commenters argued that increased gas expenditures would result ``in
negative economy-wide effects'' for many years ``given that cars sold
in the model years for which NHTSA proposes to freeze standards will,
according to the Agencies, be on the road for decades,'' and stated
that ``NHTSA's analysis is arbitrary and capricious because it entirely
fails to consider how the Proposed Rollback would impact consumers and
the economy as a whole due to increased gasoline expenditures.'' \2678\
The States and Cities commenters further argued that NHTSA was
incorrect in the NPRM when it interpreted ``the relevant question for
the need of the U.S. to conserve energy is not whether there will be
any movement in prices but whether that movement will be sudden and
large,'' \2679\ and cited State Farm to say that NHTSA had ``failed to
consider an important aspect of the problem'' by ``failing to analyze
the likely impact of even moderate future increases and volatility in
fuel prices.'' \2680\
---------------------------------------------------------------------------
\2674\ ACEEE, NHTSA-2018-0067-12122, at 2.
\2675\ NESCAUM, NHTSA-2018-0067-11691, at 4.
\2676\ NESCAUM, NHTSA-2018-0067-11691, at 5; States and Cities,
NHTSA-2018-0067-11735, at 75, citing Synapse Report.
\2677\ Congressional Tri-Caucus, NHTSA-2018-0067-1424, at 2.
\2678\ States and Cities, NHTSA-2018-0067-11735, at 75.
\2679\ 83 FR at 43214, n. 444.
\2680\ States and Cities, NHTSA-2018-0067-11735, at 75.
---------------------------------------------------------------------------
A number of commenters addressed consumer willingness to pay more
money upfront in order to save money on fuel costs. Many of these
comments are addressed in Section VI.C as part of the discussion of how
sales are modeled. More specifically in the context of how NHTSA
interprets the need of the U.S. to conserve energy, IPI commented that
NHTSA was incorrect that ``consumers' need to save money is now `less
urgent' and no longer supports a strong overall need to conserve
energy. The agencies assert that past rulemakings were overly and
paternalistically focused on `myopia.' This statement ignores all the
other pathways through which the 2012 standards benefit consumers' need
to save money, including by correcting informational asymmetries,
attention costs, and other informational failures; positional
externalities; and various other supply-side and demand-side
explanations for consumers' inability to achieve in an unregulated
market the level of fuel economy that they desire. These components of
the national need to conserve energy are discussed at length throughout
these comments, and were specifically considered by the agencies in the
2012 rule.'' \2681\
---------------------------------------------------------------------------
\2681\ IPI, NHTSA-2018-0067-12213, Appendix, at 5-6.
---------------------------------------------------------------------------
Several commenters disagreed with NHTSA's suggestion in the NPRM
that increasing U.S. production and exports reduced volatility in the
oil market. Securing America's Energy Future stated that ``. . . recent
events are an important validation of public policies that support
long-term goals like efficiency and fuel diversity. Indeed, in the
absence of fuel-efficiency standards, global oil price volatility would
likely render the country even more exposed to oil price shocks than it
is currently.'' \2682\ Mr. Bordoff, IPI, the States and Cities
commenters, and UCS all commented that the oil market is global, so
increasing U.S. production does not prevent price shocks that occur
[[Page 25141]]
due to non-U.S. events or circumstances. Mr. Bordoff stated that ``In a
globalized oil market, the consequence of a supply disruption anywhere
is a price increase everywhere--regardless of how much oil the U.S.
imports.'' \2683\ UCS made similar comments.\2684\ Mr. Bordoff further
commented that U.S. gasoline prices still follow the fluctuations in
global crude oil prices regardless of the U.S. oil import/export
balance,\2685\ and stated that ``Gasoline prices at the pump are
especially sensitive to changes in the global crude oil price due to
the relatively low level of fuel taxation [in the U.S.] compared to
other OECD countries.'' \2686\ Mr. Bordoff stated that gas price spikes
are still possible due to ongoing geopolitical challenges in major oil
producing areas, and concluded that ``Continuing with planned fuel
economy increases through CAFE standards is one effective way to reduce
the oil intensity of the economy and mitigate the adverse impact of
future oil price increases on American drivers.'' \2687\ The States and
Cities commenters cited to and echoed Mr. Bordoff's comments on this
point.\2688\ CARB commented that the proposal had relied on AEO 2017,
which reflected fuel prices that still assumed the augural standards
remained in place, but that AEO 2018 assumes ``no new fuel efficiency
standard'' and held fuel economy flat after 2021, and showed fuel
prices would be higher.\2689\
---------------------------------------------------------------------------
\2682\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 7.
\2683\ Bordoff, EPA-HQ-OAR-2018-0283-3906, at 6.
\2684\ UCS, NHTSA-2018-0067-12039, at 7.
\2685\ IPI cited and echoed these comments. IPI, NHTSA_2018-
0067-12213, Appendix, at 3.
\2686\ Bordoff, EPA-HQ-OAR-2018-0283-3906, at 7.
\2687\ Id. at 10-12.
\2688\ States and Cities, NHTSA-2018-0067-11735, at 74-75.
\2689\ CARB, NHTSA-2018-0067-11783, at 318.
---------------------------------------------------------------------------
Mr. Bordoff also commented that the future of shale oil in the U.S.
was uncertain, and therefore increased U.S. oil production was not a
basis on which to assume future global price stability.\2690\ Mr.
Bordoff argued that ``Although shale oil is more responsive to price
changes than conventional supply, it cannot serve as a swing supplier
to stabilize oil markets in the way true spare capacity (held by Saudi
Arabia) can. It takes at least 6-12 months for U.S. shale to respond to
price changes.'' \2691\ Bordoff continued, stating that ``For example,
although shale oil is more responsive to oil prices, oil prices still
plunged below $30 per barrel at the start of 2016 and soared to $80 per
barrel earlier this year. Shale oil could not swing quickly enough to
stabilize markets. This role fell to OPEC instead in both cases, first
to put a floor under prices by cutting supply and, more recently, to
provide relief by ramping up production.'' \2692\ Bordoff further
commented that political or popular pressures due to environmental
concerns may significantly increase the cost and/or difficulty of
expanding shale infrastructure,\2693\ and that even disregarding
uncertainty in supply, ongoing uncertainty in demand (both U.S. and
abroad) also contributed to global price uncertainty.\2694\
---------------------------------------------------------------------------
\2690\ Bordoff, EPA-HQ-OAR-2018-0283-3906, at 3.
\2691\ Id., at 7.
\2692\ Id., at 7-8.
\2693\ Id., at 9-10.
\2694\ Id., at 3.
---------------------------------------------------------------------------
NHTSA agrees with commenters that consumer costs for fuel are
relevant to the need of the U.S. to conserve energy. NHTSA also agrees
that future fuel prices are uncertain, and that shale oil development
in the U.S. is (1) still proceeding and subject to uncertainty, (2)
very different from traditional sources like Saudi Arabia, and (3) not
enough, by itself, to preclude any possibility of major swings in
future global oil prices. That said, NHTSA continues to believe that
U.S. shale development may reduce the negative price effects of global
price swings due to events and situations outside of our borders. Shale
represents a large, new, relatively-geopolitically-stable oil supply
source, and traditional oil producers appear to understand that
stabilizing prices below the price at which shale production starts to
ramp up faster helps those traditional producers take market advantage
of their lower cost of production.\2695\ The net effect of this, for
American drivers, should be greater fuel price stability, at least at
the upper end of fuel prices. NHTSA also continues to believe that, for
purposes of considering consumer cost of fuel as part of the need of
the U.S. to conserve energy, the fact that Americans' gasoline costs
might be minutely lower under more stringent CAFE standards and
minutely higher under comparatively less stringent CAFE standards is
not dispositive by itself. There is some tolerance in the market for
some amount of fluctuation in fuel prices, as evidenced by the
discussion in Section VI. Slow increases in fuel prices are relatively
easy for households to absorb; sharp increases are more difficult.
---------------------------------------------------------------------------
\2695\ Since 1995, EIA data indicates that OPEC production
roughly stabilized in late 2016 and has either remained steady or
fallen since then. See https://www.eia.gov/opendata/qb.php?category=1039874&sdid=STEO.PAPR_OPEC.M. See also Ilya
Arkhipov, Will Kennedy, Olga Tanas, and Grant Smith, ``Putin Dumps
MBS to Start a War on America's Shale Oil Industry,'' March 7, 2020,
Bloomberg News, available at https://www.bloomberg.com/news/articles/2020-03-07/putin-dumps-mbs-to-start-a-war-on-america-s-shale-oil-industry (describing the collapse of the OPEC+ coalition);
EIA, ``This Week in Petroleum--OPEC shift to maintain market share
will result in global inventory increases and lower prices,'' March
11, 2020, https://www.eia.gov/petroleum/weekly/; DOE, ``DOE Responds
to Recent Oil Market Activity,'' March 9, 2020, https://www.energy.gov/articles/doe-responds-recent-oil-market-activity.
---------------------------------------------------------------------------
Increases in CAFE stringency reduce the effects of all types of
increases in fuel prices, at least to the extent that people can buy
new cars and trucks, but as discussed below in Section VIII.B.4, fuel
costs and per-vehicle costs balance against one another for many
buyers. With respect to relatively low U.S. gasoline taxes creating
more pass-through effects of global oil price fluctuations, that would
be true regardless of stringency. Broadly speaking, while consumer fuel
costs are an important consideration of the need of the U.S. to
conserve energy, at this time NHTSA believes, as discussed in Section
VI, that American consumers generally understand fuel costs and their
tolerance for fluctuations, and tend to purchase vehicles accordingly.
Requiring consumers to save more fuel over the longer term by spending
more money upfront on new vehicle purchases may involve more tradeoffs
than suggested in prior rulemakings, and this rulemaking seeks to keep
these possible tradeoffs in mind.
(b) National Balance of Payments:
As the NPRM explained, the need of the United States to conserve
energy has historically included consideration of the ``national
balance of payments'' because of concerns that importing large amounts
of oil created a significant wealth transfer to oil-exporting countries
and left the U.S. economically vulnerable.\2696\ As recently as 2009,
nearly half the U.S. trade deficit was driven by petroleum,\2697\ yet
this concern has largely laid fallow in more recent CAFE actions,
arguably in part because other factors besides petroleum consumption
have since played a bigger role in the U.S. trade deficit. Given
[[Page 25142]]
recent significant increases in U.S. oil production and corresponding
decreases in oil imports, this concern seems likely to remain fallow
for the foreseeable future.\2698\ Increasingly, changes in the price of
fuel have come to represent transfers between domestic consumers of
fuel and domestic producers of petroleum rather than gains or losses to
foreign entities. NHTSA explained in the NPRM that some commenters have
lately raised concerns about potential economic consequences for
automaker and supplier operations in the U.S. due to disparities
between CAFE standards at home and their counterpart fuel economy/
efficiency and CO2 standards abroad. NHTSA finds these
concerns more relevant to technological feasibility and economic
practicability than to the national balance of payments. Moreover, to
the extent that an automaker decides to globalize a vehicle platform to
meet more stringent standards in other countries, that automaker would
comply with United States' standards and additionally generate
overcompliance credits that it can save for future years if facing
compliance concerns, or sell to other automakers. While CAFE standards
are set at maximum feasible rates, efforts of manufacturers to exceed
those standards are rewarded not only with additional credits but a
market advantage in that those consumers who place a large weight on
fuel savings will find such vehicles that much more attractive.
---------------------------------------------------------------------------
\2696\ See 42 FR 63184, 63192 (Dec. 15, 1977) (``A major reason
for this need [to reduce petroleum consumption] is that the
importation of large quantities of petroleum creates serious balance
of payments and foreign policy problems. The United States currently
spends approximately $45 billion annually for imported petroleum.
But for this large expenditure, the current large U.S. trade deficit
would be a surplus.'').
\2697\ See Today in Energy: Recent improvements in petroleum
trade balance mitigate U.S. trade deficit, U.S. Energy Information
Administration (July 21, 2014), https://www.eia.gov/todayinenergy/detail.php?id=17191.
\2698\ For an illustration of recent increases in U.S.
production, see, e.g., U.S. crude oil and liquid fuels production,
Short-Term Energy Outlook, U.S. Energy Information Administration
(June 2018), https://www.eia.gov/outlooks/steo/images/fig13.png.
While it could be argued that reducing oil consumption frees up more
domestically-produced oil for exports, and thereby raises U.S. GDP,
that is neither the focus of the CAFE program nor consistent with
Congress' original intent in EPCA. EIA's Annual Energy Outlook (AEO)
series provides midterm forecasts of production, exports, and
imports of petroleum products, and is available at https://www.eia.gov/outlooks/aeo/.
---------------------------------------------------------------------------
Several commenters addressed how much oil the U.S. imports, and the
assumptions about imports in the NPRM analysis. Securing America's
Energy Future commented that ``Because there are no readily available
substitutes to oil in the U.S. transportation sector, volatile crude
oil and petroleum product prices represent an enduring threat to the
U.S. economy.'' \2699\ ACEEE commented that overall U.S. oil imports
are higher now than they were in 1975, and nearly as high as they were
in 2012, and also stated that compared to a small overall trade surplus
in 1975, ``the U.S. now runs a large overall trade deficit.'' \2700\
The States and Cities commenters made a similar point, arguing that the
U.S. still imports large amounts of petroleum; that imports made up
about 25 percent of total U.S. oil consumption in 2017; and that EIA
indicates that ``imports as a share of oil consumption in the United
States are only about 10% lower today as compared to 1975, and we are
producing the same amount of crude oil domestically today as we were in
1970.'' \2701\ IPI stated that EIA analysis shows that the ``U.S. will
continue to import crude oil through 2050 and `remains a net importer
of petroleum and other liquids on an energy basis.' '' \2702\ CARB
disagreed that the U.S. was projected to become a net petroleum
exporter, and stated that even if it were, the rollback would have
negative effects on the U.S., because (1) it ignores short-run damages
caused by increased oil consumption and imports; (2) relies on
projections of net imports of oil which also do not take account of the
effects of the proposed rule; and (3) is not supported by the
evidence.\2703\
---------------------------------------------------------------------------
\2699\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 6.
\2700\ ACEEE, NHTSA-2018-0067-12122, at 2.
\2701\ States and Cities, NHTSA-2018-0067-11735, at 76.
\2702\ IPI, NHTSA-2018-0067-12213, Appendix, at 3.
\2703\ CARB, NHTSA-2018-0067-11873, at 317.
---------------------------------------------------------------------------
Regarding assumptions about oil imports in the NPRM analysis, the
States and Cities commented that in 2016 the agencies had assumed that
``90% of fuel savings from existing standards would lead directly to a
reduction in imported oil,'' and argued that the NPRM analysis had
ignored that previous assumption and ``la[id] great emphasis on the
fact that `oil imports have declined while exports have increased'
since 2005.'' \2704\ IPI argued that the NPRM analysis was internally
inconsistent, assuming in NHTSA's need of the nation discussion that
``additional gasoline consumption will be entirely domestic,'' while
``upstream emissions calculations assume that 95% of increased
consumption will either be from foreign refining or from foreign crude
imports,'' and suggested that this inconsistency was purposeful to make
the NPRM analysis look more favorable to the proposal.\2705\ ACEEE
commented that ``The EIA AEO side cases suggest that reduced oil demand
will primarily reduce oil imports, thus improving the overall balance
of trade regardless of the narrow balance of trade in petroleum.''
\2706\
---------------------------------------------------------------------------
\2704\ States and Cities, NHTSA-2018-0067-11735, at 75.
\2705\ IPI, NHTSA-2018-0067-12213, Appendix, at 3-4.
\2706\ ACEEE, NHTSA-2018-0067-12122, at 2.
---------------------------------------------------------------------------
Regarding the effects on the U.S. economy of increasing U.S. oil
production, Mr. Morris agreed with the NPRM's suggestion that U.S.
self-sufficiency in petroleum supply meant that higher consumer
payments for fuel under less-stringent CAFE standards would be
transfers within the U.S. economy, and stated that ``[a]t that point,
the initial purpose of EPCA is entirely obviated.'' \2707\ The States
and Cities commenters, in contrast, argued that focusing on this effect
meant that NHTSA essentially claims that increasing revenues of oil
companies--which report annual profits in the billions--is an even
trade-off for adding cost pressures and oil-price shock exposure to
American households.'' \2708\ The States and Cities commenters stated
that ``. . .this assertion ignores the negative economic impacts that
would result from increasing the cost burden on oil consumers,'' and
was ``. . .so implausible that it could not be ascribed to a difference
of view or the product of agency expertise,' citing State Farm, 463
U.S. at 43.\2709\
---------------------------------------------------------------------------
\2707\ Morris (GWU RSC), EPA-HQ-OAR-2018-0283-4028, at 15.
\2708\ States and Cities, NHTSA-2018-0067-11735, at 76.
\2709\ Id.
---------------------------------------------------------------------------
As discussed above, NHTSA agrees that oil is a global commodity.
Living in a globalized economy necessarily means that supply
disruptions (and thus, price effects) can come from a great variety of
sources--this was why the CAFE program was created, in recognition of
this risk. Increasing U.S. energy independence reduces this risk. There
are two ways to increase petroleum independence: To use less petroleum,
and to produce more of our own petroleum and use less petroleum
purchased from abroad. Both approaches work, and both are being
followed today.
NHTSA also agrees that the Draft TAR text describes the analytical
assumption that for every gallon of fuel not consumed as a result of
more stringent standards, imported crude would be reduced by 0.9
gallons. The Draft TAR stated that this assumption was based on
``changes in U.S. crude oil imports and net petroleum products in the
AEO 2015 Reference Case in comparison [sic] the Low (i.e., Economic
Growth) Demand Case,'' and also on a 2013 paper by Paul Leiby which
``suggests that `Given a particular reduction in oil demand stemming
from a policy or significant technology change, the fraction of oil use
savings that shows up as reduced U.S. imports, rather than reduced
U.S., supply, is actually quite
[[Page 25143]]
close to 90 percent, and probably close to 95 percent.' '' \2710\
---------------------------------------------------------------------------
\2710\ Draft TAR, 2016, Chapter 10, Endnote 39, p. 10-59.
---------------------------------------------------------------------------
EIA data clearly states that while the U.S. still relies on oil
imports, it is producing an increasingly large share of the petroleum
it consumes.\2711\ In 2018, domestic petroleum production made up 86
percent of domestic consumption, while imports made up 11 percent. EIA
data also clearly states that U.S. reliance on petroleum imports peaked
in 2005 and has declined since then, and that the import-percentage-of-
consumption in 2018 was the lowest it has been since 1957--this despite
the fact that overall U.S. petroleum consumption has increased
significantly over that time period as the on-road fleet has grown and
VMT (both individual and collective) has increased. Of the 11 percent
of oil consumed that was imported, 43 percent came from Canada, and 16
percent came from Persian Gulf countries. AEO 2019 states that under
its Reference case assumptions, which it describes as a ``best
assessment'' and ``a reasonable baseline case,'' \2712\ the U.S.
remains projected to become a net exporter of petroleum liquids by
2020.\2713\ During several weeks in 2019, the U.S. also exported more
oil than it imported.\2714\
---------------------------------------------------------------------------
\2711\ EIA, ``Oil: Crude and Petroleum Products Explained, Oil
Imports and Exports,'' updated May 29, 2019, available at https://www.eia.gov/energyexplained/oil-and-petroleum-products/imports-and-exports.php.
\2712\ AEO 2019, at 5.
\2713\ AEO 2019, at 14.
\2714\ See https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=wttntus2&f=4.
---------------------------------------------------------------------------
U.S. Census data indicate that the U.S. balance of trade has
generally grown over time, although it has fluctuated since peaking in
2006.\2715\ U.S. Census data further indicate that the U.S. petroleum
balance of trade, in particular, has fluctuated over time, peaking in
2008 at roughly -$386 million and decreasing to -$50 million in 2018.
2019 trends demonstrate further decreases. In percentage terms,
petroleum trade as a percentage of total trade went from roughly 52
percent in 1992 (the earliest year for which Census appears to have
data online), to 47 percent in 2008, to less than 6 percent in 2018. In
terms of national balance of payments, this is fairly clear evidence
that petroleum has decreased rapidly as part of the problem. Part of
this is due to improvements in fleet fuel economy over time, and part
is due to increases in U.S. production, particularly in the last
several years.
---------------------------------------------------------------------------
\2715\ ``U.S. Trade in Goods and Services--Balance of Payments
(BOP) Basis,'' June 6, 2019, available at https://www.census.gov/foreign-trade/statistics/historical/gands.pdf.
---------------------------------------------------------------------------
NHTSA notes also that the Draft TAR previewed the possibility of
this outcome, discussing the ``Shale Oil Revolution'' and the fact that
``[t]he recent economics literature on whether oil shocks are the
threat to economic stability that they once were is mixed.'' \2716\ The
Draft TAR stated that because of increased U.S. shale oil production,
``The resulting decrease in foreign imports . . . effectively permits
U.S. supply to act as a buffer against artificial or other supply
restrictions (the latter due to conflict or a natural disaster, for
example).'' \2717\
---------------------------------------------------------------------------
\2716\ See Draft TAR at 10-30--10-33.
\2717\ Draft TAR at 10-31.
---------------------------------------------------------------------------
Since the Draft TAR was issued, U.S. shale production has developed
even further, and U.S. petroleum imports have continued to fall. If
more oil is being produced in the U.S., and more of domestic
consumption comes from domestic production, then even though oil is a
global commodity and thus subject to price changes resulting from non-
U.S. events, the U.S. economy is inherently better off. When money
moves around within the U.S. instead of having to leave the U.S., and
everyone's needs are being met, U.S. citizens are better off when
things outside the U.S. go wrong--this is what NHTSA means when it
refers to within-U.S. transfers not being a bad thing as compared to
greater reliance on imports for consumption needs. To the extent that
some commenters find within-U.S. transfers problematic because they
increase U.S. oil company revenues without reducing fuel cost burdens
on consumers, NHTSA notes that, as discussed above, consumers seem
willing and able to tolerate some amount of fuel price increases and
fluctuation risk, as evidenced by their purchasing decisions. Prices
may still fluctuate, but shortages may foreseeably be reduced.
The Draft TAR stated that ``despite continuing uncertainty about
oil market behavior and outcomes and the sensitivity of the U.S.
economy to oil shocks, it is generally agreed that it is beneficial to
reduce petroleum fuel consumption from an energy security standpoint.
It is not just imports alone, but both imports and consumption of
petroleum from all sources and their role in economic activity, that
may expose the U.S. to risk from price shocks in the world oil price.
Reducing fuel consumption reduces the amount of domestic economic
activity associated with a commodity whose price depends on volatile
international markets.'' NHTSA continues to agree with these
statements, but cannot ignore the fact that increased U.S. petroleum
production represents the other side of the coin. Again, both national
balance of payments and energy security can be improved on both the
supply side and the demand side. While today's final rule continues to
improve on the demand side by setting standards that continue to push
CAFE levels upward, it also recognizes that supply side improvements
are playing a role.
(c) Environmental Implications
The NPRM explained that higher fleet fuel economy can reduce U.S.
emissions of CO2 as well as various other pollutants by
reducing the amount of oil that is produced and refined for the U.S.
vehicle fleet, but can also increase emissions by reducing the cost of
driving, which can result in increased vehicle miles traveled (i.e.,
the rebound effect). Thus, the net effect of more stringent CAFE
standards on emissions of each pollutant depends on the relative
magnitudes of its reduced emissions in fuel refining and distribution
and increases in its emissions from vehicle use. Fuel savings from CAFE
standards also necessarily result in lower emissions of CO2,
the main gas emitted as a result of refining, distribution, and use of
transportation fuels. Reducing fuel consumption directly reduces
CO2 emissions because the primary source of transportation-
related CO2 emissions is fuel combustion in internal
combustion engines.
NHTSA has considered environmental issues, both within the context
of EPCA and the context of the National Environmental Policy Act
(NEPA), in making decisions about the setting of standards since the
earliest days of the CAFE program. As courts of appeal have noted in
three decisions stretching over the last 20 years,\2718\ NHTSA defined
``the need of the United States to conserve energy'' in the late 1970s
as including, among other things, environmental implications. In 1988,
NHTSA included climate change concepts in its CAFE notices and prepared
its first environmental assessment addressing that subject.\2719\ It
cited concerns about climate change as one of its reasons for limiting
the extent of its reduction of the CAFE standard for MY 1989 passenger
[[Page 25144]]
cars.\2720\ Since then, NHTSA has considered the effects of reducing
tailpipe emissions of CO2 in its fuel economy rulemakings
pursuant to the need of the United States to conserve energy by
reducing petroleum consumption.
---------------------------------------------------------------------------
\2718\ CAS, 793 F.2d 1322, 1325 n. 12 (D.C. Cir. 1986); Public
Citizen, 848 F.2d 256, 262-63 n. 27 (D.C. Cir. 1988) (noting that
``NHTSA itself has interpreted the factors it must consider in
setting CAFE standards as including environmental effects''); CBD,
538 F.3d 1172 (9th Cir. 2007).
\2719\ 53 FR 33080, 33096 (Aug. 29, 1988).
\2720\ 53 FR 39275, 39302 (Oct. 6, 1988).
---------------------------------------------------------------------------
Many commenters addressed the environmental implications of CAFE
standards and the proposal. ACEEE stated that ``The environmental need
to save energy is much greater than we realized in 1975,'' and that
``The notice argues that since improved standards will not by
themselves solve global warming, they are not necessary. That logic
would equally suggest that since no one soldier would win a war, we
should never deploy any troops. No one measure will solve global
warming. . . . vehicle standards have been the most important.'' \2721\
The Harvard environmental law clinic commenters similarly stated that
``It is illogical to argue against taking a single step on the basis
that a single step is insufficient to reach one's goal,'' and commented
that it was unreasonable for the DEIS to state that ``[t]he emission
reductions necessary to keep global emissions within this carbon budget
could not be achieved solely with drastic reductions in emissions from
the U.S. passenger car and light truck fleet.'' \2722\ UCS also argued
that with respect to the environmental implications of the standards,
NHTSA's ``argument that the augural standards would only limit global
warming by 0.02 degrees C in 2100 actually supports the need to
maintain the standards. That a single U.S. policy could make that much
difference in limiting global warming is, in fact, quite significant.''
\2723\
---------------------------------------------------------------------------
\2721\ ACEEE, NHTSA-2018-0067-12122, main comments, at 2.
\2722\ Harvard environmental law clinic, EPA-HQ-OAR-2018-0283-
5486, at 13.
\2723\ UCS, NHTSA-2018-0067-12039, at 7.
---------------------------------------------------------------------------
The States and Cities commenters objected to NHTSA's consideration
in the NPRM of ``whether rapid ongoing increases in CAFE stringency . .
. can sufficiently address climate change to merit their costs,''
arguing that NHTSA had ``completely disregard[ed] environmental costs''
contrary to NHTSA's own long-standing approach to CAFE standards.\2724\
The States and Cities commenters then framed the CO2 impacts
of the proposal in tons (specifically, 7,400 million metric tons
additional CO2 emitted by 2100 as compared to the augural
standards) and argued that ``the agency effectively ignores its own
findings, in a sharp and unexplained break with the agency's past
practice of considering climate impacts,'' citing Fox Television, 556
U.S. at 515 and the 2010 and 2012 final CAFE rules which discussed
reduced economic damages from lower climate impacts for those standards
compared to their baselines.\2725\ IPI also argued that if NHTSA had
focused on economic damages rather than fractions of degrees Celsius,
``Once climate damages are fully monetized (as the agencies are
required to do), it will become apparent that the proposed rollback
will cause billions of dollars in climate damages. Billions of dollars
lost to avoidable climate damages is not a small effect, and it very
clearly is a `destructive and wasteful' effect.'' \2726\ CARB also
argued that the NPRM had ``wholly fail[ed] to analyze the economic
effects of the climate change and public health implications of the
rollback,'' stating that [t]he Agencies assert these are insignificant,
but that is only because the Agencies' projections of climate change
are so extreme. An appropriate analysis of a proposal that speeds
progress toward such a calamitous condition must acknowledge and
analyze the expected effects.'' \2727\
---------------------------------------------------------------------------
\2724\ States and Cities, NHTSA-2018-0067-11735, at 73.
\2725\ Id.
\2726\ IPI, NHTSA-2018-0067-12213, Appendix, at 4-5.
\2727\ CARB, NHTSA-2018-0067-11873, Detailed Comments, at 84.
---------------------------------------------------------------------------
The States and Cities commenters also argued that NHTSA had not
explained what the NPRM's definition of ``conservation'' as meaning
``avoid[ing] wasteful or destructive use'' ``actually means and how it
changes the agency's past practice of considering environmental
impacts,'' citing State Farm, 463 U.S. at 43, and Fox Television, 556
U.S. at 515.\2728\
---------------------------------------------------------------------------
\2728\ States and Cities, NHTSA-2018-0067-11735, at 73.
---------------------------------------------------------------------------
Regarding non-climate impacts, IPI commented that the NPRM ``only
briefly mention[ed] the possible effects on other emissions without
detailing any of the myriad non-climate public health and welfare
consequences from pollution associated with petroleum production and
combustion for motor vehicles.'' \2729\ The States and Cities
commenters similarly stated that ``NHTSA's evaluation of this factor
fails to include any analysis of environmental costs related to air
quality,'' and that the NPRM/DEIS analysis substantially understates
the actual impacts of the Proposed Rollback on criteria air pollutants
(such as NOX and PM) and air toxics (such as benzene),
making it inappropriate to rely upon.'' \2730\
---------------------------------------------------------------------------
\2729\ IPI, NHTSA-2018-0067-12213, Appendix, at 5.
\2730\ States and Cities, NHTSA-2018-0067-11735, at 73-74.
---------------------------------------------------------------------------
NHTSA agrees that the NPRM considered environmental implications of
the standards somewhat differently from past rulemaking discussions.
The 2012 final rule, for example, stated that ``[t]he need of the
nation to conserve energy has long operated to push the balancing
toward more stringent standards,'' and asked ``[i]n this final rule,
then, the question raised by this factor, combined with technological
feasibility, becomes `how stringent can NHTSA set standards before
economic practicability considerations intercede?' '' \2731\ The NPRM
discussed the dictionary definition of ``to conserve,'' tentatively
concluded that thousandths of a degree centigrade in 2100 did not rise
to the level of being ``wasteful,'' and suggested that ultimately ``we
no longer view the need of the U.S. to conserve energy as nearly
infinite.'' \2732\ This is an evolution in interpretation that was
expressly acknowledged in the NPRM--the words ``we no longer view''
clearly indicate acknowledgement of a change in view, i.e.,
interpretation. The NPRM's climate findings were not ignored, they were
directly examined and discussed at 83 FR 43215-16 in the context of
NHTSA's interpretation of their significance. The NPRM also discussed
overall costs and benefits and net benefits in the context of the
proposed maximum feasible determination, and the cost of carbon
emissions was included in those values. This final rule similarly
directly examines and discusses the analytical findings below.
---------------------------------------------------------------------------
\2731\ 77 FR at 63038-39.
\2732\ 83 FR at 43215-16.
---------------------------------------------------------------------------
Moreover, contrary to commenters' statements that NHTSA did not
acknowledge that its interpretation of the effect of the ``need of the
U.S. to conserve energy'' factor was changing, or that the balancing of
factors was different, the NPRM directly stated that:
NHTSA well recognizes that the decision it proposes to make in
today's NPRM is different from the one made in the 2012 final rule
that established standards for MY 2021 and identified `augural'
standard levels for MYs 2022-2025. Not only do we believe that the
facts before us have changed, but we believe that those facts have
changed sufficiently that the balancing of the EPCA factors and the
other considerations must also change.
The standards that we are proposing today reflect that
balancing.\2733\
---------------------------------------------------------------------------
\2733\ 83 FR at 43213. See also 83 FR at 43226 (``In the 2012
final rule . . . , NHTSA stated that `maximum feasible standards
would be represented by the mpg levels that we could require of the
industry before we reach a tipping point that presents risk of
seriously adverse economic consequences.' [citation omitted]
However, the context of that rulemaking was meaningfully different
from the current context. At that time, NHTSA understood the need of
the U.S. to conserve energy as necessarily pushing the agency toward
setting stricter and stricter standards. Combining a then-paramount
need of the U.S. to conserve energy with the perception that
technological feasibility should no longer be seen as a limiting
factor, NHTSA then concluded that only significant economic harm
would be the basis for controlling the pace at which CAFE stringency
increased over time. Today, the relative importance of the need of
the U.S. to conserve energy has changed . . . a great deal even
since the 2012 rulemaking. [T]he need of the U.S. to conserve energy
may no longer disproportionately outweigh other statutorily-mandated
considerations such as economic practicability--even when
considering fuel savings from potentially more-stringent
standards.'').
[[Page 25145]]
---------------------------------------------------------------------------
NHTSA believes that this is clear acknowledgement of the differences in
interpretation and the effect of those differences on policy decisions.
[[Page 25146]]
That said, NHTSA agrees (indeed, has always agreed) with commenters
that environmental implications exist as a result of changes in CAFE
stringency. While CO2 emissions will be higher under this
final rule than if NHTSA had determined that the augural standards were
maximum feasible, they will be lower than they would have been under
the proposal--for the ``standard setting'' runs, which are what NHTSA
looks at for assistance in determining maximum feasible standards,
NHTSA estimates that, accounting for both tailpile and upstream
emissions, CO2 emissions in 2050 under the final standards
will total 1,134 mmt, as compared to 1,149 mmt under the proposed
standards, or 1,020 mmt under the augural standards. According to the
Final EIS, which uses a ``real-world'' analysis that incorporates
models and modeling approaches that permit the agency to take a hard
look at the potential environmental impacts of the rule,\2734\ NHTSA
estimates that these amounts of CO2 emissions would lead to
the following global temperature, sea level, and ocean acidification
effects: \2735\
---------------------------------------------------------------------------
\2734\ See Kleppe v. Sierra Club, 427 U.S. 390, 410, n. 21
(1976).
\2735\ As discussed in Section 5.3.1 of the FEIS, NHTSA used the
Global Change Assessment Model (GCAM) Reference scenario to
represent the No Action Alterantive (Alternative 0) in the modeling
runs used to create Table I-1. The GCAM Reference Scenario is based
on a set of assumptions about drivers such as population,
technology, and socioeconomic changes, in the absence of global
action to mitigate climate change. It can be described as a
``business-as-usual'' scenario. NHTSA also conducted an analysis in
Chapter 8 of the FEIS using the GCAM6.0 scenario, which assumes a
moderate level of global GHG reductions and corresponds to
stabilization, by 2100, of total radiative forcing and associated
CO2 concentrations at roughly 678 ppm. Several commenters
argued that NHTSA presented climate results in the NPRM/DEIS in the
context of a ``doomsday scenario,'' in which no actions at all are
taken to mitigate carbon emissions, but NHTSA emphasizes that this
is simply the GCAM Reference Scenario, which is a reasonable
scenario to run given that GCAM is a widely accepted climate model.
Running the analysis using the GCAM Reference Scenario and GCAM6.0
Scenario results in different absolute values for the climate
variables presented in this table and Table 8.6.4-1 of the FEIS, but
again, this is because of the underlying scenarios, which reflect
very different levels of global action. When the differences in
levels of global action are accounted for, the relative impact of
each action alternative as compared to the No Action Alternative is
very similar. Thus, regardless of what GCAM scenario the agencies
consider regarding global action to mitigate climate change, it is
still meaningful to draw conclusions about the relative impacts of
the alternatives, because the alternatives are what is within the
agencies' authority to affect.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 25147]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.733
[[Page 25148]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.734
BILLING CODE 4910-59-C
NHTSA understands that some commenters view climate change as an
imminent existential threat. NHTSA does not agree, however, that
Congress
[[Page 25149]]
intended for NHTSA to set aside other statutory factors in determining
what CAFE standards would be maximum feasible. Even the maximum
feasible discussion for the 2012 final rule stated that
We recognize that higher standards would help the need of the
nation to conserve more energy . . ., but based on our analysis and
the evidence presented by the industry, we conclude that higher
standards would not represent the proper balancing for MYs 2017-2025
cars and trucks. [footnote omitted] We conclude that the correct
balancing recognizes economic practicability concerns as discussed
above, and sets standards at the levels that the agency is
promulgating in this final rule for MYs 2017-2021 and presenting for
MYs 2022-2025.\2736\
---------------------------------------------------------------------------
\2736\ 77 FR at 63055.
The footnote following the last sentence quoted above further stated
that ``We underscore that the agency's decision regarding what
standards would be maximum feasible for MYs 2017-2025 is made with
reference to the rulemaking time frame and the circumstances of this
final rule. Each CAFE rulemaking (indeed, each stage of any given CAFE
rulemaking) presents the agency with new information that may affect
how the agencies we balance the relevant factors.'' \2737\ NHTSA has
been consistent over time, despite commenters' suggestions to the
contrary, that maximum feasible is a balancing of factors; that all
factors must be considered; and that information before the agency may
change how the agency both understands and balances the statutory
factors.
---------------------------------------------------------------------------
\2737\ Id at fn. 1275.
---------------------------------------------------------------------------
With regard to criteria and toxic air pollutant emissions, NHTSA
agrees with commenters that the NPRM discussion of environmental
implications did not specifically identify these emissions, but notes
that air quality issues were discussed in a variety of places in the
NPRM, DEIS, and PRIA, and that the monetized effects of air quality
impacts were included in the overall cost-benefit analysis which
informed NHTSA's balancing of factors, as discussed above. To the
extent that commenters disagreed with the values or the agency's air
quality analyses, those topics will be addressed in Section VII and
VIII and in the FEIS. NHTSA has considered all of these findings along
with other factors, as discussed below.
(d) Foreign Policy Implications
In the NPRM, NHTSA explained that U.S. consumption and imports of
petroleum products impose costs on the domestic economy that are not
reflected in the market price for crude petroleum or in the prices paid
by consumers for petroleum products such as gasoline. These costs
include (1) higher prices for petroleum products resulting from the
effect of U.S. oil demand on world oil prices, (2) the risk of
disruptions to the U.S. economy caused by sudden increases in the
global price of oil and its resulting impact of fuel prices faced by
U.S. consumers, and (3) expenses for maintaining the strategic
petroleum reserve (SPR) to provide a response option should a
disruption in commercial oil supplies threaten the U.S. economy, to
allow the U.S. to meet part of its International Energy Agency
obligation to maintain emergency oil stocks, and to provide a national
defense fuel reserve.\2738\ Higher U.S. consumption of crude oil or
refined petroleum products increases the magnitude of these external
economic costs, thus increasing the true economic cost of supplying
transportation fuels above the resource costs of producing them.
Conversely, reducing U.S. consumption of crude oil or refined petroleum
products (by reducing motor fuel use) can reduce these external costs.
---------------------------------------------------------------------------
\2738\ While the U.S. maintains a military presence in certain
parts of the world to help secure global access to petroleum
supplies, that is neither the primary nor the sole mission of U.S.
forces overseas. Moreover, the scale of oil consumption reductions
associated with CAFE standards would be insufficient to alter any
existing military missions focused on ensuring the safe and
expedient production and transportation of oil around the globe.
Chapter 7 of the PRIA discussed this topic in more detail.
---------------------------------------------------------------------------
The NPRM stated that while these costs are considerations, the
United States has significantly increased oil production capabilities
in recent years to the extent that the U.S. is currently producing
enough oil to satisfy nearly all of its energy needs and is projected
to continue to do so or become a net energy exporter. This has added
new stable supply to the global oil market and reduced the urgency of
the U.S. to conserve energy. The NPRM referred readers to the balancing
discussion for more detail on this issue.
Securing America's Energy Future commented that continuing to raise
stringency would be good for energy security, spur innovation, and
``advance the administration's energy dominance agenda.'' \2739\ CARB
argued that the proposal would ``significantly diminish U.S. energy
security,'' ``. . . contrary to the President's recent executive order
to promote national security, and contrary to the intent of Congress in
EPCA.'' \2740\
---------------------------------------------------------------------------
\2739\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 6.
\2740\ CARB, NHTSA-2018-0067-11783, at 316.
---------------------------------------------------------------------------
Several commenters disagreed with the NPRM's suggestion that
increases in U.S. oil production reduced the foreign policy
implications relevant to the need of the U.S. to conserve energy. ACEEE
commented that because the market for oil is global, ``. . . regardless
of actual imports, the nation is still affected by what happens to oil
worldwide, and oil remains a foreign policy concern . . . .'' \2741\
Securing America's Energy Future commented that increased U.S.
production ``. . . has reduced some of the negative consequences of oil
dependence, energy security is primarily a function of consumption, not
production.'' \2742\ IPI argued that ``. . . the agencies falsely and
inconsistently argue that the need to conserve energy has diminished
because U.S. reliance on foreign oil has decreased,'' disagreeing with
the NPRM's assumption that monopsony and military security costs
resulting from the proposal would be zero.\2743\ The States and Cities
commenters raised similar points, stating that ``U.S. military and
foreign policy institutes'' place emphasis on ``global oil market
stability and the stability of major oil-exporting nations,'' which the
States and Cities argued had not changed as
[[Page 25150]]
U.S. exports have risen.\2744\ The States and Cities commenters further
argued that if a quarter of U.S. oil consumed is still imported, then
increases in consumption would necessarily raise imports, and thus also
monopsony and military security costs associated with those
imports.\2745\
---------------------------------------------------------------------------
\2741\ ACEEE, NHTSA-2018-0067-12122, main comments, at 2.
\2742\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 6.
\2743\ IPI, NHTSA-2018-0067-12213, Appendix, at 2-3.
\2744\ States and Cities, NHTSA-2018-0067-11735, at 76-77.
\2745\ Id.
---------------------------------------------------------------------------
CARB questioned whether it was accurate to assume that the U.S.
would ever reach net exporter status, and commented that even if
becoming a net exporter helped to insulate the Nation from the effects
of reducing CAFE stringency, it would not lead to greater energy
security until at least 2029, the first year for which AEO 2018
forecasts that the U.S. will stop being a net importer.\2746\ CARB
further argued that increased domestic oil production did not insulate
the U.S. from risk, and that in fact ``. . . current conditions are
more prone to risk due to lower available spare oil production capacity
in major oil producing countries, meaning that a supply disruption is
more likely to have a more pronounced effect on oil prices and U.S.
energy security.'' \2747\
---------------------------------------------------------------------------
\2746\ CARB, NHTSA-2018-0067-11783, at 317.
\2747\ Id., at 319.
---------------------------------------------------------------------------
Mr. Bordoff commented that geopolitical risk can still affect
global oil prices, citing U.S. withdrawal from the Iran nuclear
agreement and the reimposition of sanctions on Iranian oil sales; the
collapse of Libyan oil production following conflict there; ongoing
problems in Venezuela; a variety of short-term production outages in
other producing areas; and even situations where geopolitics can result
in lower prices rather than higher prices.\2748\
---------------------------------------------------------------------------
\2748\ Bordoff, EPA-HQ-OAR-2018-0283-3906, at 3-4.
---------------------------------------------------------------------------
IPI stated that ``. . . the protective value that the SPR offers
given its size does automatically change as total U.S. petroleum
consumption changes,'' and argued that it was not sufficient to
consider only ``the budgetary costs for maintaining [the size of] the
SPR.'' IPI thus argued that ``The agencies have failed to assess how
much the relative protective value of the SPR will change as total U.S.
consumption rises following the proposed rollback, and therefore have
failed entirely to consider one important element of the national need
to conserve energy.'' \2749\
---------------------------------------------------------------------------
\2749\ IPI, NHTSA-2018-0067-12213, Appendix, at 4.
---------------------------------------------------------------------------
Total energy independence for any country is only possible if it
does not participate in the global energy markets, either because it
consumes no energy (which is unrealistic) or because it produces enough
energy to meet all of its energy needs and uses only energy that is
produced domestically. As discussed above, NHTSA agrees with commenters
that the oil market is global, and that events and situations abroad
can affect oil prices even as U.S. oil production increases. The fact
that the U.S. became a net oil exporter, at least on a weekly basis, in
November 2019, and the evidence indicates that it will become a net oil
exporter on a longer-term basis in MY 2020 does not change geopolitics
in many parts of the world. Striving for energy independence in a
global market necessarily means reducing risks, because even if the
U.S. consumed only domestically-produced petroleum and continued to
export, the U.S. economy would still be subject to oil price
fluctuations due to external events and situations. The NPRM was clear
on all of these points.\2750\ The NPRM and PRIA repeatedly emphasized
that changes in the oil market meant that the risk of damage to the
U.S. economy and of additional pain for U.S. drivers is lower than it
was at the beginning of the CAFE program, not that it was eliminated
entirely. NHTSA agrees with commenters that risk still exists, and that
both production and consumption of oil are relevant to how big that
risk might be. NHTSA simply believes, as explained in the NPRM and as
explained again below, that the risk is lower than it would have been
in the absence of the rapid growth in U.S. oil production, and that the
lower risk means that the need of the U.S. to conserve energy, from
this perspective, is less dire than it was at earlier points in the
program.
---------------------------------------------------------------------------
\2750\ See 83 FR at 43213-15.
---------------------------------------------------------------------------
The analyses for both the NPRM and the final rule account for the
ongoing economic risk of participating in the global oil market by
placing a value on energy security. The energy security value is made
of several components. While commenters are correct that neither the
NPRM nor the final rule analyses attributed a positive cost to the
monopsony or military security components, the agencies do employ a
cost for macroeconomic shock risk as part of energy security. Section
VI discusses these estimates in more detail; for purposes of this
discussion, NHTSA only notes that these issues are accounted for in the
agencies' cost-benefit analysis, and to the extent that zero values are
used for some elements, the reason for that is explained at length in
those sections and public comments received on these issues did not
present new information to change the agencies' minds on those values.
With regard to the comment that NHTSA should be accounting for the
``protective value'' of the SPR along with the literal cost of
maintaining it, NHTSA is not in a position at this time to attempt to
estimate such a value, and notes that the commenter provided no
suggestions as to how to do so. The Department of Energy's website
states that the maximum number of days of import protection provided by
the SPR is 143 days, and that it takes 13 days from Presidential
decision for SPR fuel to enter the market.\2751\ The 1973 OPEC oil
embargo lasted from October 1973 to March 1974, roughly 150 days. As
explained, NHTSA continues to believe that the effect of increased U.S.
oil production is to stabilize, broadly, global oil markets. The longer
a sustained spike in prices due to geopolitical events continues, the
greater incentive U.S. shale production has to respond. NHTSA believes
that it is foreseeable that the SPR could be utilized to help mitigate
a price shock in the interim, for the majority of foreseeable shock
situations.
---------------------------------------------------------------------------
\2751\ See https://www.energy.gov/fe/services/petroleum-reserves/strategic-petroleum-reserve/spr-quick-facts-and-faqs.
---------------------------------------------------------------------------
(5) Factors That NHTSA Is Prohibited From Considering
The NPRM explained that EPCA also provides that in determining the
level at which it should set CAFE standards for a particular model
year, NHTSA may not consider the ability of manufacturers to take
advantage of several EPCA provisions that facilitate compliance with
CAFE standards and thereby reduce the costs of compliance.\2752\ As
discussed further in Section IX below, NHTSA cannot consider compliance
credits that manufacturers earn by exceeding the CAFE standards and
then use to achieve compliance in years in which their measured average
fuel economy falls below the standards. NHTSA also cannot consider the
use of alternative fuels by dual fuel vehicles nor the availability of
dedicated alternative fuel vehicles--including battery-electric
vehicles--in any model year. EPCA encourages the production of
alternative fuel vehicles by specifying that their fuel economy is to
be determined using a special calculation procedure that results in
those vehicles being assigned a higher equivalent fuel economy level
than they actually achieve.
---------------------------------------------------------------------------
\2752\ 49 U.S.C. 32902(h).
---------------------------------------------------------------------------
The NPRM further explained that the effect of the prohibitions
against
[[Page 25151]]
considering these statutory flexibilities in setting the CAFE standards
is that the flexibilities remain voluntarily-employed measures. If
NHTSA were instead to assume manufacturer use of those flexibilities in
setting new standards, higher standards would appear less costly and
therefore more feasible, which would thus effectively require
manufacturers to use those flexibilities in order to meet higher
standards. By keeping NHTSA from including them in our stringency
determination, the provision ensures that these statutory credits
remain true compliance flexibilities.
Additionally, for the non-statutory fuel economy improvement value
program that NHTSA developed by regulation, the NPRM stated that NHTSA
does not consider these subject to the EPCA prohibition on considering
flexibilities. EPCA is very clear as to which flexibilities are not to
be considered. When the agency has introduced additional flexibilities
such as A/C efficiency and ``off-cycle'' technology fuel economy
improvement values, NHTSA has considered those technologies as
available in the analysis. Thus, today's analysis includes assumptions
about manufacturers' use of those technologies, as detailed in Section
VI.
Michalek and Whitefoot commented that ``[w]e find [the statutory
prohibition on considering certain flexibilities in determining maximum
feasible CAFE standards] problematic because the automakers use these
flexibilities as a common means of complying with the regulation, and
ignoring them will bias the cost-benefit analysis to overestimate
costs.'' \2753\ IPI commented that ``it is not clear that the statutory
prohibition on considering credit availability was intended to apply to
banked credits,'' because 49 U.S.C. 32902(h)(3) was
---------------------------------------------------------------------------
\2753\ Michalek and Whitefoot, NHTSA-2018-0067-11903, at 10-11.
added . . . as a `conforming amendment' to EISA, which was the
statute that gave NHTSA authority to allow credit trading and
transferring; meanwhile, banking and borrowing have been part of
NHTSA's authority since EPCA in 1975. In 1989, e.g., NHTSA
explicitly relied on the availability of `credit banks' to justify
maintaining the MY 1990 standard at 27.5 mpg instead of lowering its
stringency. NHTSA has not explained why it now believes it may not
more fully consider banking.\2754\
---------------------------------------------------------------------------
\2754\ IPI, NHTSA-2018-0067-12213, Appendix, at 19.
NHTSA agrees, as explained in the NPRM, that if the agency was able
to consider the compliance flexibilities in determining maximum
feasible standards, more-stringent standards would appear less costly
and therefore more feasible. NHTSA is nevertheless bound by the
statutory prohibition on considering the above-mentioned flexibilities.
As for IPI's disagreement that 32902(h)(3) should apply to banked
credits because it was labeled a ``conforming amendment,'' NHTSA looks
to the specific statutory language provided, which prohibits
``[consideration], when prescribing a fuel economy standard, [of] the
trading, transferring or availability of credits . . . .'' (Emphasis
added.) IPI's suggested interpretation would render ``availability'' as
surplusage. If Congress had meant the prohibition to apply only to
traded and transferred credits, it would have said so. Instead,
Congress also prohibited consideration of the ``availability of
credits,'' which must be read reasonably to refer to ``what credits are
available,'' i.e., banked credits. The fact that NHTSA considered the
availability of banked credits in 1989, prior to establishment of this
statutory prohibition, has no bearing in a post-EISA world.
Nonetheless, NHTSA notes that it is informed by the ``real-world''
analysis presented in the FRIA, which accounts for credit availability
and usage, and manufacturers' ability to employ alternative fueled
vehicles--for purpose of conformance with E.O. 12866. Under the real-
world analysis, compliance does, in fact, appear less costly. For
example, today's ``real world'' analysis shows manufacturers' costs
averaging about $1,420 in MY 2029 under the final standards, as
compared to the $1,640 shown by the ``standard setting'' analysis.
However, for purposes of determining maximum feasible CAFE levels,
NHTSA considers only the ``standard-setting'' analysis shown in the
NPRM, consistent with Congress's direction.
(f) EPCA/EISA Requirements That No Longer Apply Post-2020
The NPRM explained that Congress amended EPCA through EISA to add
two requirements not yet discussed in this section relevant to
determination of CAFE standards during the years between MY 2011 and MY
2020 but not beyond. First, Congress stated that, regardless of NHTSA's
determination of what levels of standards would be maximum feasible,
standards must be set at levels high enough to ensure that the combined
U.S. passenger car and light truck fleet achieves an average fuel
economy level of not less than 35 mpg no later than MY 2020.\2755\ And
second, between MYs 2011 and 2020, the standards must ``increase
ratably'' in each model year.\2756\ Neither of these requirements apply
after MY 2020, so given that this rulemaking concerns the standards for
MY 2021 and after, the NPRM stated that they are not relevant to this
rulemaking.
---------------------------------------------------------------------------
\2755\ 49 U.S.C. 32902(b)(2)(A).
\2756\ 49 U.S.C. 32902(b)(2)(C).
---------------------------------------------------------------------------
CARB commented that because the proposal did not ``provide for
improved efficiency of motor vehicles'' over the long term,
``Stagnating the standards violates Congressional direction to ratably
increase fuel economy when the technology for doing so has been
demonstrated to exist (which it does . . .) or could be developed in
the necessary time.'' \2757\
---------------------------------------------------------------------------
\2757\ CARB, NHTSA-2018-0067-11873, Detailed Comments, at 84.
---------------------------------------------------------------------------
NHTSA notes, again, that the statutory language is clear that
Congress only directed ratable increases in stringency through MY 2020.
After MY 2020, the statutory language is clear that standards simply
need be ``maximum feasible, as determined by the Secretary.'' Some
commenters may have disagreed that the proposal represented maximum
feasible levels, but there is no statutory basis for arguing that the
``ratable increase'' requirement extends beyond MY 2020.
(g) Other Considerations in Determining Maximum Feasible Standards
The NPRM explained that NHTSA has historically considered the
potential for adverse safety consequences in setting CAFE standards.
This practice has been consistently approved in case law. As courts
have recognized, ``NHTSA has always examined the safety consequences of
the CAFE standards in its overall consideration of relevant factors
since its earliest rulemaking under the CAFE program.'' Competitive
Enterprise Institute v. NHTSA, 901 F.2d 107, 120 n. 11 (D.C. Cir. 1990)
(``CEI-I'') (citing 42 FR 33534, 33551 (June 30, 1977)). The courts
have consistently upheld NHTSA's implementation of EPCA in this manner.
See, e.g., Competitive Enterprise Institute v. NHTSA, 956 F.2d 321, 322
(D.C. Cir. 1992) (``CEI-II'') (in determining the maximum feasible fuel
economy standard, ``NHTSA has always taken passenger safety into
account'') (citing CEI-I, 901 F.2d at 120 n. 11); Competitive
Enterprise Institute v. NHTSA, 45 F.3d 481, 482-83 (D.C. Cir. 1995)
(``CEI-III'') (same); Center for Biological Diversity v. NHTSA, 538
F.3d 1172, 1203-04 (9th Cir. 2008) (upholding NHTSA's analysis of
vehicle safety issues associated with weight in connection with the MYs
2008-2011
[[Page 25152]]
light truck CAFE rulemaking). Thus, NHTSA explained that in evaluating
what levels of stringency would result in maximum feasible standards,
NHTSA assesses the potential safety impacts and considers them in
balancing the statutory considerations and to determine the maximum
feasible level of the standards.
The attribute-based standards that Congress requires NHTSA to set
help to mitigate the negative safety effects of the historical single
number standards originally required in EPCA, and in past rulemakings,
NHTSA constrained its modeling so as not to consider possible mass
reduction in lower weight vehicles in its analysis, which affected the
resulting assessment of potential adverse safety impacts. That
analytical approach did not reflect, however, the likelihood that
automakers may pursue the most cost-effective means of improving fuel
efficiency to comply with CAFE requirements. For the NPRM, as for the
final rule, the modeling did not limit the amount of mass reduction
that is applied to any segment, but rather considered that automakers
may apply mass reduction based upon cost-effectiveness, similar to most
other technologies. NHTSA does not, of course, mandate the use of any
particular technology by manufacturers in meeting the standards. The
NPRM and today's final rule, like the Draft TAR, also considered the
safety effect associated with the additional vehicle miles traveled due
to the rebound effect.
NHTSA explained that the NPRM considered the safety effects of
vehicle scrappage rates on the fleet as a whole. The NPRM also
explained NHTSA's consideration of the effect of additional expenses in
fuel savings technology on the affordability of vehicles--the
likelihood that increased standards will result in consumers being
priced out of the new vehicle market and choosing to keep their
existing vehicle or purchase a used vehicle. Since new vehicles are
significantly safer than used vehicles, slowing fleet turnover to newer
vehicles results in older and less safe vehicles remaining on the roads
longer. NHTSA stated that this significantly affects the safety of the
United States light duty fleet, as described more fully in in the
safety section of the NPRM and in Chapter 11 of the PRIA. Furthermore,
as fuel economy standards become more stringent, and more fuel
efficient vehicles are introduced into the fleet, fueling costs are
reduced. This results in consumers driving more miles, which results in
more crashes and increased highway fatalities.
A number of commenters disagreed with a variety of aspects of the
NPRM's analysis of safety, and several also disagreed with how NHTSA
considered safety along with the other factors in the proposal. The
States and Cities commenters, for example, agreed that ``NHTSA has
historically considered safety impacts when setting maximum feasible
standards,'' but argued that:
in the Proposed Rollback, NHTSA departs from its past practice by
relying on completely novel and unsupported theories regarding the
linkages between fuel economy and safety that do not reflect
reality. In the past, NHTSA has considered the safety of the
technologies that improve fuel economy. [citations omitted] In the
Proposed Rollback, however, NHTSA has linked safety concerns with
rebound and scrappage effects of more stringent fuel economy
standards. [citations omitted] As discussed [elsewhere], these
theories are unsupported, implausible, and contradicted by numerous
experts--rendering them arbitrary and capricious. The agency has
also failed to acknowledge or adequately justify its break with past
analyses of safety. See Fox Television, 556 U.S. at 515.'' \2758\
---------------------------------------------------------------------------
\2758\ States and Cities, NHTSA-2018-0067-11735, at 77.
EDF commented that NHTSA cannot ``. . . lawfully rely upon the
repercussions of increased driving as a justification. . . . The fact
that the standards do not `compel' this driving prevents such reliance,
and . . . [EPCA/EISA] nowhere indicate that [NHTSA] can refuse to
comply with [its] statutory obligations by pointing to a projection
that individuals might drive more and in doing so, some of them will
get into traffic accidents.\2759\ EDF further argued that:
---------------------------------------------------------------------------
\2759\ EDF, NHTSA-2018-0067-12137, Supplemental Safety Comments,
at 3.
It is especially unlikely that Congress intended for NHTSA to
consider potential increases in driving (or . . . `VMT'). Under
basic economic theory and under the Agency's traditional analysis
(including their analysis of this proposal), an improvement in fuel
economy--which makes driving cheaper--would be expected to lead to
some increase in driving for households that are sensitive to and
conscious of that effect on their budgets. Thus, consideration of
VMT impacts could be used to undermine any fuel economy standard.
Because VMT is `a factor [that] is both so indirectly related to
[fuel economy] and so full of potential for canceling the
conclusions drawn from [a fuel economy analysis] . . . it would
surely have been expressly mentioned in [the statute] had Congress
meant it to be considered.' Whitman v. Am. Trucking Associations,
531 U.S. 457, 469 (2001).'' \2760\
---------------------------------------------------------------------------
\2760\ Id.
Other comments on safety as part of the legal justification varied.
NESCAUM claimed that NHTSA's safety justification ``is disputed by
EPA's technical staff based on their identification of flaws in NHTSA's
analysis,'' suggesting that it was therefore invalid and not a basis
for decision-making.\2761\ Global commented that there was no policy
reason for freezing the level of standards due to mass reduction
concerns (i.e., safety), given footprint standards.\2762\ IPI argued
that it was inappropriate to account for vehicle safety-related deaths
and injuries ``without an adequate discussion of the health and safety
impacts of the Proposed Rule's increased emissions or without an
accurate estimate of the actual safety impact of the rollback versus
the 2012 standards.'' \2763\
---------------------------------------------------------------------------
\2761\ NESCAUM, NHTSA-2018-0067-11691, at 3.
\2762\ Global, NHTSA-2018-0067-12032, Attachment A, at A-32.
\2763\ IPI, NHTSA-2018-0067-12213, Appendix, at 11.
---------------------------------------------------------------------------
NHTSA agrees with commenters that the safety analysis conducted to
inform this rulemaking (both NPRM and final rule) is different from--
broader than--past safety analyses conducted to inform CAFE and
CO2 rulemakings. NHTSA disagrees, however, that the agency
failed to acknowledge or explain this fact. The NPRM directly
acknowledges and explains the evolution of the safety analysis over
time and why, specifically, the NPRM included the safety effects of
rebound and scrappage phenomena.\2764\ The NPRM also expressly sought
comment on these elements of the safety analysis and the safety
analysis generally, before explaining how they worked and describing
their tentative findings in considerable detail. It is inaccurate for
commenters to claim that the agency did not acknowledge or explain
these changes. Commenters' disagreement with the substance of the
safety analysis does not create a valid process complaint here. Section
VI discusses in detail the comments received on the substance of the
safety analysis, including a number of comments citing deliberative
feedback provided by some members of EPA staff during NPRM development,
and contains the agencies' responses. With regard to the comment from
EDF, as explained above, the premise that vehicles may be driven more
or less in response to more or less stringent CAFE (or CO2)
standards is called the rebound effect, and it is discussed at length
in Section VI above. The rebound effect has been factored into
rulemaking cost-benefit analyses and reduced CAFE and CO2
standard benefits in such analyses for well over a decade,\2765\ and
EPA and NHTSA have
[[Page 25153]]
written repeatedly about and considered the magnitude of this effect.
NHTSA is aware that some commenters disagree that a rebound effect even
exists for fuel economy, and understands how such commenters would
correspondingly disagree that VMT-related safety effects could arise
from differences in CAFE standards. But NHTSA does not agree that the
rebound effect is zero, and correspondingly believes that safety
effects from additional driving (due to exposure to crashes) exist and
are capable of quantification for analytical purposes.
---------------------------------------------------------------------------
\2764\ See 83 FR at 43106-07.
\2765\ See, e.g., 68 FR 16868, 16878 (Apr. 7, 2003).
---------------------------------------------------------------------------
Moreover, if EDF were correct that agencies may consider only the
behavior that regulations directly ``compel,'' then CAFE analysis would
be challenged to consider even fuel savings--the purpose of CAFE
standards--because the standards do not compel Americans to drive, or
to buy new vehicles, or to buy any vehicles at all. Reasonable
assumptions about how much Americans drive (depending on how much it
costs to drive, among other things), and what vehicles Americans buy
and how often they buy them (depending on how much those vehicles cost,
among other things), are useful and important for including in analyses
that help decision-makers distinguish between different levels of
potential CAFE standards. Circular A-4 additionally directs agencies to
consider ancillary effects of rulemakings.\2766\ NHTSA believes that it
is reasonable to consider these effects as part of the safety analysis,
and to consider safety effects as part of its determination of maximum
feasible standards.
---------------------------------------------------------------------------
\2766\ See OIRA, ``Regulatory Impact Analysis: A Primer,'' at 7,
https://www.reginfo.gov/public/jsp/Utilities/circular-a-4_regulatory-impact-analysis-a-primer.pdf (``In addition to the
direct benefits and costs of each alternative, the list should
include any important ancillary benefits and countervailing risks.
An ancillary benefit is a favorable impact of the alternative under
consideration that is typically unrelated or secondary to the
purpose of the action (e.g., reduced refinery emissions due to more
stringent fuel economy standards for light trucks). A countervailing
risk is an adverse economic, health, safety, or environmental
consequence that results from a regulatory action and is not already
accounted for in the direct cost of the action (e.g., adverse safety
impacts from more stringent fuel-economy standards for light
trucks). As with other benefits and costs, an effort should be made
to quantify and monetize both ancillary benefits and countervailing
risks.'')
---------------------------------------------------------------------------
(2) Administrative Procedure Act
To be upheld under the ``arbitrary and capricious'' standard of
judicial review in the APA, an agency rule must be rational, based on
consideration of the relevant factors, and within the scope of the
authority delegated to the agency by the statute. The agency must
examine the relevant data and articulate a satisfactory explanation for
its action including a ``rational connection between the facts found
and the choice made.'' \2767\
---------------------------------------------------------------------------
\2767\ Burlington Truck Lines, Inc., v. United States, 371 U.S.
156, 168 (1962).
---------------------------------------------------------------------------
Statutory interpretations included in an agency's rule are subject
to the two-step analysis of Chevron, U.S.A. v. Natural Resources
Defense Council.\2768\ Under step one, where a statute ``has directly
spoken to the precise question at issue,'' id. at 842, the court and
the agency ``must give effect to the unambiguously expressed intent of
Congress.'' \2769\ If the statute is silent or ambiguous regarding the
specific question, the court proceeds to step two and asks ``whether
the agency's answer is based on a permissible construction of the
statute.'' \2770\
---------------------------------------------------------------------------
\2768\ 467 U.S. 837 (1984).
\2769\ Id. at 843.
\2770\ Id.
---------------------------------------------------------------------------
If an agency's interpretation differs from the one that it has
previously adopted, the agency need not demonstrate that the prior
position was wrong or even less desirable. Rather, the agency would
need only to demonstrate that its new position is consistent with the
statute and supported by the record and acknowledge that this is a
departure from past positions. The Supreme Court emphasized this in FCC
v. Fox Television.\2771\ When an agency changes course from earlier
regulations, ``the requirement that an agency provide a reasoned
explanation for its action would ordinarily demand that it display
awareness that it is changing position,'' but ``need not demonstrate to
a court's satisfaction that the reasons for the new policy are better
than the reasons for the old one; it suffices that the new policy is
permissible under the statute, that there are good reasons for it, and
that the agency believes it to be better, which the conscious change of
course adequately indicates.'' \2772\ The APA also requires that
agencies provide notice and comment to the public when proposing
regulations,\2773\ as the agencies did when publishing the NPRM for
this rulemaking.
---------------------------------------------------------------------------
\2771\ 556 U.S. 502 (2009).
\2772\ Id., at 1181.
\2773\ 5 U.S.C. 553.
---------------------------------------------------------------------------
a) Requests To Extend the Comment Period
On August 2, 2018, the agencies published the NPRM on the agencies'
respective websites, soliciting public comments.\2774\ On August 24,
2018, the Federal Register published the NPRM, which began a 60-day
public comment period.\2775\ The public comment period would have ended
on October 23, 2018, but the agencies extended the comment period until
October 26, 2018.\2776\ In the Federal Register notice extending the
comment period, the agencies explained that they were denying requests
for an extension of the comment period by at least 60 days, explaining
that ``[a]utomakers will need maximum lead time to respond to the final
rule[.]'' \2777\ Although the comment period ultimately closed on
October 26, 2018, the agencies' dockets remained open, and the agencies
continued to accept and consider comments, to the extent possible, for
more than one year after the comment period began.\2778\
---------------------------------------------------------------------------
\2774\ https://www.nhtsa.gov/corporate-average-fuel-economy/safe; https://www.epa.gov/newsreleases/us-epa-and-dot-propose-fuel-economy-standards-my-2021-2026-vehicles.
\2775\ 83 FR 42986 (Aug. 24, 2018).
\2776\ See 83 FR 48578 (Sept. 26, 2018) (extending comment
period).
\2777\ Id.
\2778\ The agencies notified the public of this possibility in
the NPRM, stating that: ``To the extent practicable, we will also
consider comments received after'' the close of the comment period.
83 FR 42986, 43471 (Aug. 24, 2018).
---------------------------------------------------------------------------
After publishing the NPRM, the agencies received a number of
requests to extend the comment period, generally for an additional 60
days.\2779\ For example, seventeen States and the District of Columbia
jointly requested a 60-day extension of the comment period.\2780\ That
request cited the voluminous record, the complexity of the material,
and the profound potential impact on human health and the environment,
among other things.\2781\ The City of Los Angeles and New York State
Department of Environmental Conservation also requested a 60-day
extension, for similar reasons.\2782\ In addition, 32 United States
Senators jointly requested a 60-day extension of the comment
period.\2783\ The Senators argued that an extension was appropriate to
ensure adequate public participation with such an important rule.\2784\
Several non-government organizations similarly requested a 60-day
extension of the comment period due to the complexity of the issues and
[[Page 25154]]
the importance of the proposed rule.\2785\ Other organizations also
requested a 60-day extension, stressing the complexity of the issues
and the significance of the proposed rule's impact on the
environment.\2786\ The American Lung Association also requested a 60-
day extension of the comment period, asserting that it needed more time
to analyze the impact of the proposed rule on human health.\2787\ The
California Air Resources Board (CARB) likewise requested a 60-day
extension, in part, based on information that it asserted should have
been included in the NPRM.\2788\ New York University School of Law's
Institute for Policy Integrity similarly requested a 60-day extension
based on information that it contended should have been included in the
NPRM's ``sensitivity analysis table for the `Cumulative Changes in
Fleet Size, Travel (VMT), Fatalities, Fuel Consumption and C02
Emissions through MY2029.' '' \2789\
---------------------------------------------------------------------------
\2779\ See 83 FR 48578 (Sept. 26, 2018).
\2780\ See comments from the State of California et al., Request
for an extension, Docket No. NHTSA-2018-0067-3458.
\2781\ See id.
\2782\ Also for similar reasons, the Minnesota Pollution Control
Agency and the Minnesota Department of Transportation submitted a
joint request for a 120-day extension of the comment period. See
comments from the Minnesota Pollution Control Agency and Minnesota
Department of Transportation, Docket No. NHTSA-2018-0067-3580.
\2783\ See comments from 32 U.S. Senators (Kamala D. Harris et
al.), Docket No. NHTSA-2018-0067-5643.
\2784\ See id.
\2785\ See, e.g., comments from the Alliance of Automobile
Manufacturers, Docket No. NHTSA-2018-0067-3619; Communities for a
Better Environment, Docket No. EPA-HQ-OAR-2018-0283-1095; Consumer
Federation of America, NHTSA-2018-0067-3400; Edison Electric
Institute, received by mail; and South Coast Air Quality Management
District, Docket No. EPA-HQ-OAR-2018-0283-0885.
\2786\ See, e.g., comments from the Environmental Law and Policy
Center, NHTSA-2018-0067-2728; Georgetown Climate Center, Docket No.
NHTSA-2018-0067-3610; Center for Biological Diversity, Conservation
Law Foundation, Earthjustice, Environmental Defense Fund, Natural
Resources Defense Council, Public Citizen,
Sierra Club, and Union of Concerned Scientists, Docket No.
NHTSA-2018-0067-3278; and National Governors Association, Docket No.
EPA-HQ-OAR-2018-0283-0871.
\2787\ See comments from American Lung Association, Docket No.
NHTSA-2018-0067-3615.
\2788\ See comments from California Air Resources Board, Docket
No. NHTSA-2018-0067-4166.
\2789\ See comments from New York University School of Law's
Institute for Policy Integrity, NHTSA-2018-0067-5641.
---------------------------------------------------------------------------
The agencies do not believe a further extension of the comment
period was warranted under the circumstances.\2790\ The APA does not
specify a minimum number of days for a comment period.\2791\ Two
Executive Orders also provide direction to Federal agencies with
respect to the length of a comment period for a proposed rule.\2792\
Executive Order 12,866 states that ``[e]ach agency shall (consistent
with its own rules, regulations, or procedures) provide the public with
meaningful participation in the regulatory process . . . . In addition,
each agency should afford the public a meaningful opportunity to
comment on any proposed regulation, which in most cases should include
a comment period of not less than 60 days.'' \2793\ Additionally,
Executive Order 13,563 reaffirmed Executive Order 12,866's directive
that comment periods should generally not be less than 60 days,
stating: ``To the extent feasible and permitted by law, each agency
shall afford the public a meaningful opportunity to comment through the
internet on any proposed regulation, with a comment period that should
generally be at least 60 days.'' \2794\ More recently, in December of
2018, the Department of Transportation implemented DOT Order 2100.6,
which provides its operating administrations, including NHTSA, with
direction on appropriate rulemaking processes and procedures.\2795\
While not yet effective at the time the proposal was published, the
Order provides that ``the comment period for significant DOT rules
should be at least 45 days.'' \2796\ The 63 day comment period for the
proposal far exceeded this amount.
---------------------------------------------------------------------------
\2790\ See 83 FR 48578 (Sept. 26, 2018) (extending comment
period until October 26, 2018 and denying requests for longer
extensions).
\2791\ See 5 U.S.C. 553(c).
\2792\ The Executive Orders do not create any enforceable right
or benefit by a party against any federal agency. E.O. 12,866 Sec.
10; E.O. 13,563 Sec. 7(d).
\2793\ Executive Order 12,866 Sec. 6(a)(1).
\2794\ Executive Order 13,563 Sec. 2(b).
\2795\ DOT Order 2100.6, ``Policies and Procedures for
Rulemakings,'' available at: https://www.transportation.gov/sites/dot.gov/files/docs/regulations/328561/dot-order-21006-rulemaking-process-signed-122018.pdf.
\2796\ Id., at (11)(i)(3).
---------------------------------------------------------------------------
Consistent with these principles, courts give broad discretion to
agencies in determining the reasonableness of a comment period. Courts
have frequently upheld comment periods that were significantly less
than the 63-day comment period here. See Connecticut Light & Power Co.
v. Nuclear Regulatory Comm'n, 673 F.2d 525, 534 (D.C. Cir. 1982)
(upholding a 30-day comment period and stating that ``neither statute
nor regulation mandates that the agency do more''); see also North
American Van Lines v. ICC, 666 F.2d 1087, 1092 (7th Cir. 1981)
(upholding a 45-day comment period).\2797\ In addition to the length of
a comment period, courts consider the number of comments received and
whether comments had an effect on an agency's final rule, in assessing
whether the public had a meaningful opportunity to comment.\2798\
---------------------------------------------------------------------------
\2797\ In certain circumstances, particularly urgent ones,
courts have even upheld comment periods of less than 30 days. See
Omnipoint Corp. v. FCC, 78 F.3d 620, 629-30 (D.C. Cir. 1996)
(holding that a 14-day comment period was sufficient given the
``urgent necessity for rapid administrative action under the
circumstances''); see also Fla. Power & Light Co. v. United States,
846 F.2d 765, 772 (D.C. Cir. 1988) (upholding a 15-day comment
period given a deadline that Congress imposed on the Nuclear
Regulatory Commission to finalize its rule).
\2798\ See Florida Power & Light, Co. v. United States, 846 F.2d
765, 772 (D.C. Cir. 1988); see also Conference of State Bank Sup'rs
v. Office of Thrift Supervision, 792 F. Supp. 837, 844 (D.D.C.
1992).
---------------------------------------------------------------------------
These principles are easily satisfied here. Here, the agencies
initially provided a 60-day comment period and then further extended it
to ensure compliance with the Clean Air Act. The Clean Air Act requires
that the record of proceedings allowing oral presentation of data,
views, and arguments on a proposed rule be kept open for 30 days after
completion of a proceeding to provide an opportunity for submission of
rebuttal and supplementary information.\2799\ Because the final
``proceeding allowing oral presentation of data, views, and arguments''
was expected to be on September 26, 2018, the comment period for the
proposed rule was extended by three days to meet that
requirement.\2800\
---------------------------------------------------------------------------
\2799\ 42 U.S.C. 7607(d)(5).
\2800\ See 83 FR 48578, 48581 (Sept. 26, 2018).
---------------------------------------------------------------------------
The 63-day comment period was consistent with what the law
requires.\2801\ While the agencies understand and agree with commenters
about the importance and complexity of the issues here, the public
docket demonstrates that the public had a meaningful opportunity to
comment on the proposed rule.\2802\ The agencies received a total of
more than 750,000 public comments, many of which commented on detailed,
technical portions of the proposed rule. For instance, the California
Air Resources Board provided 415 pages of detailed comments involving
very specific aspects of the proposal,\2803\ and the Auto Alliance
filed 202 pages of detailed comments, and commissioned a separate
econometric study analyzing the effects of multiple alternatives.\2804\
This is clear evidence that the public had not only the opportunity to
review and comment on the proposal, but to do so with an extraordinary
level of detail.
---------------------------------------------------------------------------
\2801\ In any event, the two Executive Orders explicitly state
that they do not create any enforceable right or benefit by a party
against any federal agency. See Executive Order 12,866 Sec. 10; see
also Executive Order 13,563 Sec. 7(d).
\2802\ See Rural Cellular Ass'n v. FCC, 588 F.3d 1095, 1101
(D.C. Cir. 2009).
\2803\ NHTSA-2018-0067-11873.
\2804\ NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Finally, notwithstanding the sufficiency of the agencies' 63-day
comment period, the agencies published their NPRM on their websites on
August 2, 2018, more than three weeks before the comment period
formally opened on August 24, and this effectively provided the public
with 22 additional days in
[[Page 25155]]
which to review the proposal and draft comments.\2805\
---------------------------------------------------------------------------
\2805\ The agencies' public dockets also remained open for more
than one year after the start of the comment period, and the
agencies considered some late comments received, to the extent
practicable, although many late comments were simply too untimely to
be considered.
---------------------------------------------------------------------------
b) Other Comments on Public Participation
Several commenters objected to NHTSA's 15-page limit on primary
comments, asserting that it impacted the public's ability to
meaningfully participate in the rulemaking process.\2806\ However, as
certain of the commenters acknowledged, the NPRM also explicitly stated
that commenters could also submit attachments--without any page
limit.\2807\ Thus, the page limit on primary comments did not prevent
commenters from presenting any information they deemed relevant to the
agencies. Both primary comments and their attachments are available in
the agencies' public dockets, and were considered by the agencies in
this rulemaking as demonstrated by the responses to comments discussed
throughout this final rule.
---------------------------------------------------------------------------
\2806\ See States of California et al., Attachment1_States and
Cities Detailed Comments, Docket No. NHTSA-2018-0067-11735, at 46;
Center for Biological Diversity, et al., NHTSA-2018-0067-12088;
CARB, NHTSA-2018-0067-1187; Environmental Defense Fund, NHTSA-2018-
0067-12108; BlueGreen Alliance, NHTSA-2018-0067-12440; Connecticut
Department of Energy and Environmental Protection (DEEP), EPA-HQ-
OAR-2018-0283-4202.
\2807\ 83 FR 43470 (Aug. 24, 2018) (citing 49 CFR 553.21).
---------------------------------------------------------------------------
NHTSA's 15-page limit simply prescribed the form that comments
should take: A concise summary comment of up to 15 pages, with optional
attachments with no page limit. Many commenters submitted extensive
attachments to their comments, including commenters that objected to
the 15-page limit for primary comments. For example, several States and
cities that jointly commented submitted a 13-page primary comment,
accompanied by 145 pages of ``detailed comments'' and three appendices
totaling 101 additional pages.\2808\ The 15-page limit had the effect
of creating executive summaries of otherwise voluminous comments, which
increased efficiency during the rulemaking process. This was NHTSA's
stated purpose for the 15-page limit. As explained in the NPRM: ``NHTSA
established this limit to encourage you to write your primary comments
in a concise fashion.'' \2809\ In any event, no commenter was prevented
from submitting information to the agencies based on NHTSA's page
limitation for primary comments. The agencies strongly disagree that
public participation was impeded by NHTSA's specification that primary
comments were limited to 15 pages.
---------------------------------------------------------------------------
\2808\ States of California et al., NHTSA-2018-0067-11735.
\2809\ 83 FR 43470 (Aug. 24, 2018).
---------------------------------------------------------------------------
On August 2, 2018, the agencies published a joint Notice of
Proposed Rulemaking (NPRM) on the agencies' respective websites, which
solicited public comments on ``The Safer Affordable Fuel-Efficient
(SAFE) Vehicles Rule for Model Years 2021-2026 Passenger Cars and Light
Trucks.'' \2810\ The NPRM indicated that the public may submit written
comments by any of the following methods: Online through the Federal
eRulemaking Portal at www.regulations.gov, by fax, by mail, or by hand
delivery. The NPRM also notified the public that the agencies planned
to hold three joint public hearings, and would accept oral and written
comments at the hearings. The NPRM indicated that the agencies planned
to hold the hearings in Washington, DC; the Detroit, Michigan area; and
the Los Angeles, California area, but indicated that the specific
addresses and dates for the hearings would be announced in a
supplemental Federal Register notice.\2811\ On August 24, 2018, the
agencies published a notice in the Federal Register, which provided new
locations for two of the three hearings and added dates for each
hearing.\2812\ That notice informed the public that the agencies
planned to hold three joint public hearings during the comment period:
(1) On September 24, 2018 in Fresno, California; (2) on September 25,
2018 in Dearborn, Michigan; and (3) on September 26, 2018 in
Pittsburgh, Pennsylvania.\2813\
---------------------------------------------------------------------------
\2810\ https://www.nhtsa.gov/corporate-average-fuel-economy/safe; https://www.epa.gov/newsreleases/us-epa-and-dot-propose-fuel-economy-standards-my-2021-2026-vehicles. The Agencies subsequently
published the NPRM in the Federal Register on August 24, 2018. 83 FR
42986 (August 24, 2018).
\2811\ 83 FR 42986 (August 24, 2018).
\2812\ 83 FR 42817 (August 24, 2018).
\2813\ Id.
---------------------------------------------------------------------------
The agencies also received several comments with respect to the
sufficiency of the agencies' public hearings during the comment period.
For example, the South Coast Air Quality Management District asserted
that EPA failed to meet its obligation to hold public hearings under
the Clean Air Act, claiming that an EPA ``political appointee'' did not
have the legal authority to change hearing locations.\2814\ The comment
also claimed that holding certain of the hearings in smaller
metropolitan areas than originally announced resulted in 15 million
fewer potential participants in the hearings.\2815\ Additionally, the
comment noted that the NPRM and the notice that set the new locations
of two of the public hearings were both published in the Federal
Register on the same day, yet those documents contained conflicting
hearing locations (the NPRM listed the originally planned hearing
locations).\2816\
---------------------------------------------------------------------------
\2814\ See comments from the South Coast Air Quality Management
District, Attachment 1--SCAQMD Combined NHTSA Waiver Comment (Oct.
25, 2018), Docket No. NHTSA-2018-0067-11813, at 37-38.
\2815\ See id. at 37.
\2816\ See id.
---------------------------------------------------------------------------
Similarly, seventeen States and the District of Columbia submitted
a joint comment requesting that the agencies reinstate the hearing
locations that were initially listed in the NPRM, with the stated goal
of maximizing the number of public participants.\2817\ Similarly, a
group of environmental organizations jointly submitted a comment
stating that the new hearing locations failed to maximize the potential
participants for the agencies' public hearings.\2818\ That group also
asserted that the agencies failed to provide a reason for the agencies'
denial of requests to hold more than three public hearings.\2819\
---------------------------------------------------------------------------
\2817\ See comments from the State of California et al., Request
for an extension, Docket No. NHTSA-2018-0067-3458.
\2818\ See comments from the Center for Biological Diversity,
Conservation Law Foundation, Environmental Defense Fund,
Earthjustice, Environmental Law and Policy Center, Natural Resources
Defense Council, Public Citizen, Inc., Sierra Club, and Union of
Concerned Scientists, Appendix A--Coalition Comment Letter (10-26-
2018), Docket No. NHTSA-2018-0067-12000, at 213. A number of other
commenters also requested that the Agencies hold additional public
hearings. See, e.g., comments from the Georgetown Climate Center,
20180906--GCC Comments to NHTSA and EPA, Docket No. NHTSA-2018-0067-
3610; The City of Los Angeles, Docket No. NHTSA-2018-0067-4159, at
2-3; California Air Resources Board, 2018-09-11 SAFE Rule DEIS--CARB
Req Add Info, Docket No. NHTSA-2018-0067-4166, at 1; Northeast
States for Coordinated Air Use Management, NESCAUM SAFE rule request
for comment extension and hearing_20180824, Docket No. NHTSA-2018-
0067-2158, at 1-2.
\2819\ Id.
---------------------------------------------------------------------------
The agencies more than satisfied their legal obligation with
respect to holding public hearings, and the three hearings provided
substantial additional opportunity for public participation. While the
agencies understand that some commenters were disappointed with some
aspects of the process, those commenters did not demonstrate that the
agencies' process was legally deficient, nor that any party suffered
prejudice from the changes the agencies made to their public hearing
arrangement.
[[Page 25156]]
The APA does not require agencies to hold public hearings during
the rulemaking process, unless the opportunity for a public hearing is
required by a governing statute.\2820\ NHTSA's governing fuel economy
statute does not require a public hearing during the rulemaking
process.\2821\ The Clean Air Act requires EPA to ``give interested
persons an opportunity for the oral presentation of data, views, or
arguments, in addition to an opportunity to make written submissions .
. . .'' 42 U.S.C. 7607(d)(5)(ii). The agencies' three joint public
hearings satisfied this statutory requirement.
---------------------------------------------------------------------------
\2820\ See 5 U.S.C. 553(c). Absent a statutory requirement, the
APA gives agencies the discretion whether or not to hold a public
hearing, stating that ``the agency shall give interested persons an
opportunity to participate in the rule making through submission of
written data, views, or arguments with or without opportunity for
oral presentation.'' Id.
\2821\ See 49 U.S.C. 32902.
---------------------------------------------------------------------------
The agencies note that it was clear from the NPRM that the hearings
were not yet finalized. No addresses or dates were announced for the
hearings, and the NPRM indicated that information on the hearings would
be forthcoming in a supplemental Federal Register notice. The NPRM
(signed by the EPA Administrator) indicated that three hearings would
be held, and the fact that specific details about those hearings were
announced in a later notice signed by a different political appointee
does not itself make the hearings themselves invalid. The Clean Air Act
does not mandate hearings in any particular location and the public was
aware from the NPRM that additional information on the hearings would
be forthcoming. To the extent that any individual person or group was
inconvenienced by the change in location announced in the supplemental
notice, they still had ample time to submit public comments through any
of the multiple other available methods indicated in the NPRM.\2822\
---------------------------------------------------------------------------
\2822\ Executive Order 13,563 offers guidance to agencies with
respect to how to maximize public participation. The Executive Order
states that agencies should ``afford the public a meaningful
opportunity to comment through the internet on any proposed
regulation . . . .'' The vast majority of the comments the agencies
received in this rulemaking were submitted through the internet.
---------------------------------------------------------------------------
The agencies regret any confusion that resulted from publication of
the NPRM in the Federal Register on the same date as publication of the
notice that updated the hearing locations and provided additional
information, including hearing dates. However, because the NPRM did not
include dates for the hearings, and the NPRM informed interested
parties to look for an additional notice that would announce specific
dates and addresses for the hearings, no one could have relied on the
NPRM to the exclusion of the supplemental notice.\2823\
---------------------------------------------------------------------------
\2823\ Additionally, as a matter of fairness, the agencies gave
interested parties notice about the change in public hearing
locations one month prior to the first public hearing. See 83 FR
42817 (August 24, 2018).
---------------------------------------------------------------------------
The agencies ultimately held three public hearings, as was
originally announced. There is no Clean Air Act requirement for a
particular number of hearings, and by holding the hearings in locations
throughout the United States (including in California), the agencies
offered a meaningful opportunity for participation. Moreover, the
public docket remained open for two months subsequent to the
announcement of the final hearing locations, providing any interested
party who was unable to attend a public hearing ample opportunity to
submit comments in writing. As evidence of this meaningful opportunity
to comment on the proposed rule, the agencies received a total of more
than 750,000 public comments.
Several commenters also asserted that the agencies delayed posting
the hearing transcripts to the public docket until October 25, which
was one day before the close of the public comment period.\2824\ The
Environmental Defense Fund claimed that this was inconsistent with the
Clean Air Act's requirements that ```[t]he transcript of public
hearings, if any, on the proposed rule shall also be included in the
docket promptly upon receipt from the person who transcribed such
hearings.' 42 U.S.C. 7607(d)(4)(B).'' \2825\ As one commenter
acknowledged, the transcripts were certified by the reporters on
September 26, 2018 (Pittsburgh hearing), September 27, 2018 (Dearborn
hearing), and October 1, 2018 (Fresno hearing).\2826\ The agencies made
the transcripts publicly available within a reasonable period.
Moreover, it was reasonable for the agencies to have an opportunity to
review the transcripts for errors prior to making them publicly
available. While the concern expressed by these commenters was an
inadequate ability to offer responsive comments to the transcripts, the
rulemaking process would be never-ending if every commenter had an
opportunity to respond to every other commenter. There is no such
requirement in the APA, the Clean Air Act, or otherwise. The public had
sufficient opportunity to comment on the agencies' proposals, as
described above.
---------------------------------------------------------------------------
\2824\ Environmental Defense Fund, NHTSA-2018-0067-12108, NHTSA-
2018-0067-12327, NHTSA-2018-0067-12371; State of California et al.,
NHTSA-2018-0067-11735.
\2825\ Environmental Defense Fund, NHTSA-2018-0067-12371.
\2826\ State of California et al., NHTSA-2018-0067-11735.
---------------------------------------------------------------------------
A few commenters requested that the agencies host a workshop or
webinar to help commenters better understand the agencies' modeling and
analyses.\2827\ The commenters pointed to similar activities undertaken
by EPA for other complex rulemakings. While the agencies did not
conduct a live workshop or webinar regarding the proposal, they did
make extensive information publicly available beyond the contents of
the NPRM. To assist the public, NHTSA hosted a dedicated web page with
information on the modeling.\2828\ The web page included a video
introduction to the CAFE model.\2829\ The web page enabled members of
the public to download the model software, its system documentation,
source code, and input files.\2830\ Many commenters commented in detail
on the modeling and analyses. However, the agencies recognize that
public stakeholders vary in their experience and understanding of the
modeling and analyses and will continue to consider ways to facilitate
public participation in future rulemakings, which could include the use
of workshops or webinars.
---------------------------------------------------------------------------
\2827\ See Minnesota Pollution Control Agency (MPCA), NHTSA-
2017-0069-0528; Minnesota Pollution Control Agency (MPCA) et al.,
NHTSA-2018-0067-11706.
\2828\ https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
\2829\ Id.
\2830\ Id.
---------------------------------------------------------------------------
Some comments criticized the agencies for the agencies'
untimeliness in adding materials to the rulemaking dockets, for
example, identifying material ``that was not added to the rulemaking
docket until the end of the original comment period or, in some cases,
added either after that period already had closed or not at all.''
\2831\
---------------------------------------------------------------------------
\2831\ CBD et. al, Supplemental Comments, Docket No. NHTSA-2018-
0067-12371, at 8.
---------------------------------------------------------------------------
The critical question is ``whether the final rule changes
critically from the proposed rule rather than on whether the agency
relies on supporting material not published for comment.'' \2832\ In
other words, ``[t]he question is typically whether the agency's final
rule so departs from its proposed rule as to constitute more surprise
than notice.'' \2833\ To that end, agencies are allowed--as the
agencies here did--to
[[Page 25157]]
rely on supplemental data that clarified, expanded on, or confirmed
information in the proposed rule, even if that supplemental data was
not disclosed in the proposed rule.\2834\ In any event, the commenters
have failed to show how they were prejudiced by any information posted
later than they would have preferred.\2835\
---------------------------------------------------------------------------
\2832\ Air Transp. Ass'n of Am. v. F.A.A., 169 F.3d 1, 7 (D.C.
Cir. 1999).
\2833\ Id. (citing Air Transp. Ass'n of Am., 732 F.2d 219, 225
n.12 (D.C. Cir. 1984)).
\2834\ See Air Transp. Ass'n of Am. v. F.A.A., 169 F.3d 1, 7
(D.C. Cir. 1999) (citing Solite Corp. v. EPA, 952 F.2d 473, 485
(D.C. Cir. 1991); Air Transp. Ass'n of Am. v. CAB, 732 F.2d 219, 224
(D.C. Cir. 1984)).
\2835\ See Solite Corp. v. U.S. E.P.A., 952 F.2d 473, 484 (D.C.
Cir. 1991) (citing Cmty. Nutrition Inst. v. Block, 749 F.2d 50, 57-
58 (D.C. Cir. 1984)). Parties also could have submitted comments
after the end of the comment period on any of these materials. See
49 CFR 553.23 (NHTSA regulation providing that ``[l]ate filed
comments will be considered to the extent practicable.'').
---------------------------------------------------------------------------
Some commenters noted that certain aspects of the CAFE model used
for the proposal were not previously subject to peer review.\2836\
Certain commenters asserted that the proposal was legally flawed
because the full CAFE model was not peer reviewed prior to the
proposal.\2837\ In support of this argument, commenters cited the
Information Quality Act and related OMB guidance that states that
``each agency shall have a peer review conducted on all influential
scientific information that the agency intends to disseminate.'' \2838\
Commenters also cited EPA's Peer Review Handbook, which states: ``For
highly influential scientific assessments, external peer review is the
expected procedure.'' \2839\
---------------------------------------------------------------------------
\2836\ See, e.g., Center for Biological Diversity et al., NHTSA-
2018-0067-12000; Environmental Defense Fund, NHTSA-2018-0067-12327;
Environmental Defense Fund et al., NHTSA-2018-0067-12371;
Environmental Defense Fund et al., NHTSA-2018-0067-12406; Center for
Biological Diversity, Environment America, Environmental Defense
Fund, Environmental Law Policy Center, Public Citizen, Inc., Sierra
Club, and Union of Concerned Scientists, NHTSA-2018-0067-12439;
States of California et al., NHTSA-2018-0067-11735.
\2837\ See, e.g., Center for Biological Diversity et al., NHTSA-
2018-0067-12000.
\2838\ See Center for Biological Diversity et al., NHTSA-2018-
0067-12000.
\2839\ See Center for Biological Diversity et al., NHTSA-2018-
0067-12000.
---------------------------------------------------------------------------
The agencies agree that peer review is appropriate for the CAFE
model, and the CAFE model has been peer reviewed. As discussed in the
NPRM, and as certain commenters acknowledged, the CAFE model was peer
reviewed in 2017.\2840\ NHTSA included peer review materials in the
public docket as well as on its web page regarding the model.\2841\ As
described in those materials: ``In 2017, the Volpe Center arranged for
a formal peer review of the version of the CAFE model released and
documented in 2016 . . . . All of the peer reviewers supported much
about the model's general approach, and supported many of the model's
specific characteristics. Peer reviewers also provided a variety of
general and specific recommendations regarding potential changes to the
model, inputs, outputs, and documentation. NHTSA and Volpe Center staff
agree with many of these recommendations and have either completed or
begun work to implement many of them; implementing others would require
further research, testing, and development not possible at this time,
but we are considering them for future model versions.''\2842\
---------------------------------------------------------------------------
\2840\ 83 FR 43000 (Aug. 24, 2018) (``A report available in the
docket for this rulemaking presents peer reviewers' detailed
comments and recommendations, and provides DOT's detailed
responses.''); see Center for Biological Diversity et al., NHTSA-
2018-0067-12000.
\2841\ NHTSA-2018-0067-0055; https://www.nhtsa.gov/corporate-average-fuel-economy/compliance-and-effects-modeling-system.
\2842\ NHTSA-2018-0067-0055.
---------------------------------------------------------------------------
However, certain new elements of the CAFE model were not completed
at the time of the 2017 peer review.\2843\ NHTSA subsequently obtained
a peer review of significant new elements added to the model after the
2017 peer review.\2844\ As described in the new peer review charge,
included in a July 2019 report included in the rulemaking docket, NHTSA
explained:
---------------------------------------------------------------------------
\2843\ NHTSA-2018-0067-0055 (explaining, in responses to 2017
peer review, that ``[t]he model has been updated to including
procedures to estimate impacts on new vehicle sales, and on older
vehicle scrappage'').
\2844\ NHTSA-2018-0067-0055.
To inform the proposed rule announced in August 2018, DOT staff
introduced significant new elements to the model, including methods
to estimate changes in vehicle sales volumes, vehicle scrappage, and
automotive sector labor usage. Each of these regulatory actions
involved consideration of and response to significant public comment
on model results, as well as comments on the model itself. In
addition to DOT staff's own observations, these comments led DOT
staff to make a wide range of improvements to the model. Insofar as
a formal peer review could identify additional potential
opportunities to improve the model, DOT sponsored a review of the
entire model in 2017. At this time, DOT seeks review of some of the
---------------------------------------------------------------------------
significant new elements added to the model after that review.
This subsequent peer review of the new elements was not complete at
the time the proposal was published, and therefore materials concerning
the peer reviewers' comments and NHTSA's responses were not available
until later.\2845\ Although the comment period on the proposal had
closed at that time, the agencies continued to receive comments on the
new peer review materials, which they have considered in issuing this
final rule.\2846\ Of course, the new elements of the modeling were also
described in detail in the NPRM and commenters also directly commented
on them in great detail. Thus, the public was fully apprised of all
aspects of the modeling and had a robust opportunity to provide
comment.
---------------------------------------------------------------------------
\2845\ NHTSA-2018-0067-0055 (July 2019 report).
\2846\ See, e.g., Center for Biological Diversity et al., NHTSA-
2018-0067-12439; Environment America et al., NHTSA-2018-0067-12441.
---------------------------------------------------------------------------
To the extent commenters are suggesting the Information Quality Act
required a full peer review of all aspects of the CAFE model prior to
the proposal, the agencies disagree.\2847\ Peer review of the new
elements of the CAFE model helped ensure that the model is
scientifically sound, and the peer reviewers provided feedback that
helped improve the model and may help develop additional improvements
to the model in the future. In this sense, the peer review of the new
elements of the model functioned similarly to public comments from
commenters with specific scientific expertise. Much of the feedback
from the peer reviewers were in fact similar in nature to comments
received from public commenters on the model. By engaging in both peer
review and notice-and-comment procedures, the agencies ensured that
they had information from a wide variety of sources, including those
with specific expertise, to validate and improve the model.\2848\ The
technical aspects of the model, including improvements made to the
model following the proposal, are described in detail in this final
rule. Moreover, as the Center for Biological Diversity noted, the
Information Quality Act does not create third-party rights.\2849\
---------------------------------------------------------------------------
\2847\ See, e.g., Center for Biological Diversity et al., NHTSA-
2018-0067-12000; Environment America et al., NHTSA2018-0067-12441.
\2848\ The timing of the peer review of new elements of the
model also did not require a second cycle of notice and comment.
See, e.g., Alto Dairy v. Veneman, 336 F.3d 560, 569-70 (7th Cir.
2003) (``The law does not require that every alteration in a
proposed rule be reissued for notice and comment. If that were the
case, an agency could `learn from the comments on its proposals only
at the peril of subjecting itself to rulemaking without end.''').
\2849\ Center for Biological Diversity et al., NHTSA-2018-0067-
12000.
---------------------------------------------------------------------------
The agencies also disagree that EPA needed to obtain a separate
peer review of the CAFE model.\2850\ The peer review addressed aspects
of the model relevant to the analysis by both agencies under their
respective statutory schemes. The agencies have expertise in their
[[Page 25158]]
statutory requirements and discussed in detail both in the proposal and
this final rule how the CAFE model was used to inform the decision-
making under both EPCA and the CAA.
---------------------------------------------------------------------------
\2850\ Center for Biological Diversity et al., NHTSA-2018-0067-
12000.
---------------------------------------------------------------------------
(c) Other APA Comments
Many commenters suggested that the record of evidence developed for
the 2016 Draft TAR and EPA's Original Determination was a better basis
for NHTSA to determine maximum feasible standards than the record of
evidence for the current rulemaking. These commenters also argued that,
in the NPRM, NHTSA ignored the findings and analysis in the TAR and the
Technical Support Document and contradicted the pre-existing record
without explanation. Lastly, these commenters argued that the NPRM did
not have a reasoned basis under the APA, particularly in light of the
agency's change in position and the reliance interests at stake.
Agencies always have authority under the Administrative Procedure
Act to revisit previous decisions in light of new facts, as long as
they provide notice and an opportunity for comment--as the agencies did
here. Indeed, it is the best practice to do so when changed
circumstances so warrant.\2851\
---------------------------------------------------------------------------
\2851\ See FCC v. Fox Television, 556 U.S. 502 (2009).
---------------------------------------------------------------------------
``Changing policy does not, on its own, trigger an especially
`demanding burden of justification.' '' \2852\ ``Agencies are free to
change their existing policies as long as they provide a reasoned
explanation for the change.'' \2853\ Providing this explanation ``would
ordinarily demand that [the agency] display awareness that it is
changing position.'' \2854\ Beyond that, however, ``[w]hen an agency
changes its existing position, it `need not always provide a more
detailed justification than what would suffice for a new policy created
on a blank slate.' '' \2855\ The agency ``need not demonstrate to a
court's satisfaction that the reasons for the new policy are better
than the reasons for the old one.'' \2856\ For instance, ``evolving
notions'' about the appropriate balance of varying policy
considerations constitute sufficiently good reasons for a change in
position.\2857\ A change in policy is ``well within an agency's
discretion:'' Agencies are permitted to conduct a ``reevaluation of
which policy would be better in light of the facts,'' without
``rely[ing] on new facts.'' \2858\
---------------------------------------------------------------------------
\2852\ Mingo Logan Coal Co. v. Envtl. Prot. Agency, 829 F.3d
710, 718 (DC Cir. 2016) (quoting Ark Initiative v. Tidwell, 816 F.3d
119, 127 (DC Cir. 2016)).
\2853\ Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125
(2016) (citations omitted).
\2854\ FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009) (emphasis in original) (``An agency may not, for example,
depart from a prior policy sub silentio or simply disregard rules
that are still on the books.'').
\2855\ Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125-
26 (2016) (quoting FCC v. Fox Television Stations, Inc., 556 U.S.
502, 515 (2009)).
\2856\ FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009) (emphasis in original).
\2857\ N. Am.'s Bldg. Trades Unions v. Occupational Safety &
Health Admin., 878 F.3d 271, 303 (D.C. Cir. 2017) (quoting the
agency's rule).
\2858\ Nat'l Ass'n of Home Builders v. E.P.A., 682 F.3d 1032,
1037-38 (D.C. Cir. 2012).
---------------------------------------------------------------------------
To be sure, providing ``a more detailed justification'' is
appropriate in some cases.\2859\ But when ``a more detailed
justification'' is needed, all that is required is for the agency to
explain how ``new information arising after'' the previous
determination ``informed its conclusion'' that a change was
appropriate: ``Explanations relying on new data are sufficient to
satisfy the more detailed explanatory obligation.'' \2860\ As one of
the critical comments itself noted, ``[a]gencies must use `the best
information available' in reaching their conclusions, and cannot
lawfully rely on outdated information as circumstances change.'' \2861\
Accordingly, when new information became available, the agencies relied
on it expressly, resulting in a fully-explained change in their
analysis and ultimately their conclusions.
---------------------------------------------------------------------------
\2859\ FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009) (``Sometimes [the agency] must [provide a more detailed
justification than what would suffice for a new policy created on a
blank slate]--when, for example, its new policy rests upon factual
findings that contradict those which underlay its prior policy; or
when its prior policy has engendered serious reliance interests that
must be taken into account.'').
\2860\ Mingo Logan Coal Co. v. Envtl. Prot. Agency, 829 F.3d
710, 727 (D.C. Cir. 2016).
\2861\ CBD et. al, Appendix A, Docket No. NHTSA-2018-0067-12000,
at 11 (quoting Flyers Rights Education Fund v. FAA, 864 F. 3d 738,
745 (D.C. Cir. 2017)).
---------------------------------------------------------------------------
While ``[i]t would be arbitrary or capricious to ignore such
matters,''\2862\ the agencies have not ignored them. NHTSA has
satisfied these standards. The NPRM expressly and repeatedly
acknowledged that it represented a change from the 2012 final rule, the
Draft TAR, and EPA's Original Determination, appropriately justifying
the change by citing shifts in policy priorities or new facts and
changed circumstances that became apparent since the Original
Determination.\2863\ The agencies are fully cognizant of the facts and
circumstances that have changed since the Original Determination,
expressly acknowledged them in the Revised Determination and SAFE Rule
NPRM, and adapted to accept them now in the final rule.
---------------------------------------------------------------------------
\2862\ FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515
(2009).
\2863\ See, e.g., 83 FR at 43213 (Aug. 24, 2018).
---------------------------------------------------------------------------
Several commenters invoked requests to the agencies under the
Freedom of Information Act (``FOIA'') regarding material sought in
connection with the rulemaking.\2864\ These comments ranged from simple
references to existing FOIA requests to the agencies, to the actual
submission of the FOIA requests as a comment posted to the rulemaking
docket.\2865\ These commenters sought a variety of information, which
included calendars and internal correspondence of specific agency
personnel, communications with non-governmental stakeholders, and
technical materials and clarifications relating to aspects of the
agencies' analysis.\2866\
---------------------------------------------------------------------------
\2864\ See, e.g., Environmental Defense Fund, NHTSA-2018-0067-
12371.
\2865\ Compare, e.g., Joint Submission from the States of
California et al. and the Cities of Oakland et al., NHTSA NHTSA-
2018-0067-11735, with, e.g., Office of the Attorney General of the
State of New York, NHTSA-2018-0067-3613.
\2866\ See, e.g., Environmental Defense Fund, NHTSA-2018-0067-
12397; Office of the Attorney General of the State of New York,
NHTSA-2018-0067-3613; California Air Resources Board, NHTSA-2018-
0067-4166.
---------------------------------------------------------------------------
To the extent these requests sought substantive material, those
matters are addressed in other sections herein that pertain to the
respective underlying issues implicated. Although the submission of
FOIA requests through an online rulemaking docket is a very unusual
form of submitting a FOIA request to an agency, the agencies
nevertheless processed the comments that requested materials by
invoking FOIA as formal FOIA requests. As such, once identified, those
comments were forwarded to the agencies' respective FOIA offices, which
commenced the intake process of the letters as FOIA requests. In turn,
the agencies' FOIA offices transmitted receipt acknowledgement letters
to the requestors and conducted searches for the applicable material.
The agencies responded to the requestors by producing the responsive
non-exempt records identified, applying the appropriate FOIA standards
applicable to the records and requests. Like all other typical FOIA
requests, the requestors were provided with an opportunity to
administratively appeal the FOIA decision and, if desired, subsequently
seek judicial review of the agencies' decisions. Several commenters
availed themselves of this procedure.\2867\
---------------------------------------------------------------------------
\2867\ See generally, e.g., New York v. U.S. Envtl. Prot. Agency
and Nat'l Highway Traffic Safety Admin., Case No. 1:19-cv-00712
(S.D.N.Y.) (FOIA litigation concerning a FOIA request submitted as a
comment from the Office of the Attorney General of the State of New
York, NHTSA-2018-0067-3613).
---------------------------------------------------------------------------
[[Page 25159]]
Thus, the agencies fully satisfied their obligations under the
governing FOIA provisions. In fact, other commenters noted the
agencies' responses to these FOIA requests and incorporated information
disclosed in the responses into their comments.\2868\ Moreover, several
of the FOIA requests submitted as comments requested information that
had already been published on the agencies' websites for the rulemaking
or in the rulemaking dockets.
---------------------------------------------------------------------------
\2868\ See James H. Stock, Kenneth Gillingham & Wade Davis, EPA-
HQ-OAR-2018-0283-6220, at p. 6.
---------------------------------------------------------------------------
Although the agencies fulfilled their obligations under all
applicable FOIA law, the agencies also stress that FOIA compliance is
wholly irrelevant to conformity to governing APA standards in the
rulemaking process. FOIA arises from an independent statutory
framework, which contains unique provisions for judicial review.\2869\
These provisions for judicial review provide ``an adequate form of
relief'' such that the APA is not typically even an appropriate
mechanism to seek the disclosure of further information requested under
FOIA.\2870\ Likewise, the APA's principles governing rulemaking
procedures, including disclosures of information for such rulemakings,
exist as autonomous statutory and jurisprudential concepts totally
untethered from the principles of disclosure under FOIA.
---------------------------------------------------------------------------
\2869\ 5 U.S.C. 552(a)(4)(B).
\2870\ See, e.g., Feinman v. FBI, 713 F. Supp. 2d 70, 76 (D.D.C.
2010) (``This court and others have uniformly declined jurisdiction
over APA claims that sought remedies made available by FOIA.'').
---------------------------------------------------------------------------
Similarly, as an independent statutory framework from the APA, the
susceptibility of materials and records for production under FOIA has
no bearing on whether such materials should have been made public under
the APA as part of a rulemaking. The scope of materials for production
under FOIA arises from the Agency's reasonable interpretation of the
language of the FOIA request, as well as the exemptions potentially
applicable to the records under the applicable FOIA statutes and
implementing regulations.\2871\ In contrast, in an APA review of
rulemaking procedures, separate standards exist to govern the scope of
materials an agency must make available during the rulemaking
process.\2872\ Thus, records may be responsive to a FOIA request, but
not appropriate for publication under the APA--even if the FOIA request
concerns the proposed rule in question. The FOIA requests at issue here
are illustrative of this distinction. For example, one of the specific
FOIA requests identified by commenters describes the requests as
pertaining to the NPRM, but seeks Outlook calendars of DOT and NHTSA
personnel.\2873\ While such materials may be responsive to the
underlying FOIA requests, which expressly mention the calendars, an
employee's entire list of calendar appointments--including appointments
unrelated to the rulemaking--is clearly not contemplated by the APA as
material necessary for publication along with a proposed rule. Thus,
while the agencies sought to comply with their independent statutory
obligations under FOIA, to the extent commenters invoke purported FOIA
noncompliance, the agencies consider such arguments irrelevant to the
rulemaking analysis. Likewise, any production of records in connection
with any FOIA request that invokes the proposed rule is not a
recognition by the agencies that the material should have also been
made available during the rulemaking under the APA.
---------------------------------------------------------------------------
\2871\ See 5 U.S.C. 552. See also, e.g., Weisberg v. U.S. Dep't
of Justice, 745 F.2d 1476, 1485 (DC Cir. 1984) (discussing standards
applicable to the scope of an Agency's search for records under
FOIA).
\2872\ See Air Transp. Ass'n of Am. v. F.A.A., 169 F.3d 1, 7 (DC
Cir. 1999) (discussing the scope of materials for an agency to make
available during a notice and comment period).
\2873\ See Environmental Defense Fund, NHTSA-2018-0067-12397.
---------------------------------------------------------------------------
Several commenters also criticized the agencies, and specifically
the EPA, for not publishing an updated version of the Optimization
Model for Reducing Emissions of Greenhouse Gases from Automobiles
(``OMEGA'') along with the proposed rule.\2874\ As described in further
detail in Section IV herein, OMEGA is a fleet compliance model
developed by the EPA and used in previous rulemakings. While many
commenters raised technical arguments comparing the OMEGA model to the
CAFE Model utilized in this rulemaking, such technical analysis and
comments are addressed elsewhere in this final rule analysis. See
Section IV. Likewise, while several comments refer to FOIA requests for
OMEGA model materials, the Agencies' discussion of FOIA comments are
addressed above.
---------------------------------------------------------------------------
\2874\ See, e.g., International Council on Clean Transportation,
NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
Most other commenters who raised more procedural arguments
concerning the unavailability of an updated version of the OMEGA model
argued that an updated version of the model should have been released
because the EPA utilized the model during an interagency review of the
proposed rule.\2875\ In considering these comments, the agencies
emphasize that neither NHTSA, the EPA, nor any other interagency
reviewer relied upon the OMEGA model for the preparation of either the
proposed or the final versions of the SAFE Vehicles Rule. Instead, as
clearly expressed in rulemaking descriptions and documents accompanying
both this final rule and the proposed rule, the agencies relied on a
separate model to perform the analysis that helped to inform the
agencies regarding potential effects of various fuel economy standards.
This independent model, the CAFE Model, was developed by the Department
of Transportation's Volpe National Transportation Systems Center.
---------------------------------------------------------------------------
\2875\ See, e.g., Sallie E. Davis, NHTSA-2018-0067-12430.
---------------------------------------------------------------------------
In fact, most commenters discussing the OMEGA model understood and
expressly acknowledged that the agencies relied upon the CAFE Model
rather than the OMEGA model for this rulemaking.\2876\ Several
commenters even paradoxically argued both that the agencies
unreasonably failed to utilize the OMEGA model and that the agencies
denied meaningful opportunity for comment by utilizing but failing to
publish an updated OMEGA model.\2877\ Nevertheless, the analysis and
universe of documents published for the proposed rule made abundantly
clear that the CAFE Model--not the OMEGA model--performed the
applicable analysis for this rulemaking. Likewise, the agencies'
proposed rule published voluminous analyses and supporting documents to
describe the CAFE Model and explain the underlying methodologies
incorporated into the model's operation for this rulemaking. The
agencies also released the full version of the CAFE Model employed in
this rulemaking, as well as its respective inputs and outputs, in order
to provide commenters with ample opportunities to understand the
model's function and operation.
---------------------------------------------------------------------------
\2876\ See, e.g., Union of Concerned Scientists, NHTSA-2018-
0067-12303-016; Center for Biological Diversity, NHTSA-2018-0067-
12000.
\2877\ See, e.g., Environmental Defense Fund, NHTSA-2018-0067-
12108.
---------------------------------------------------------------------------
The extensive comments on the modeling conducted for this
rulemaking confirm that the agencies provided the public with
sufficient information to comment on the modeling process for the
rulemaking. Comments regarding the OMEGA and CAFE models were
expansive, spanning hundreds of pages of technical analysis and
submissions from a variety of commenters. Many of these comments even
consisted of detailed and technical comparisons of
[[Page 25160]]
the CAFE model used in this rulemaking with past versions of OMEGA
models used for prior rulemakings.\2878\ Even if certain of these
commenters disagreed with the Agencies' ultimate approach to the
modeling, they evidently understood the applicable methodologies and
performance of the CAFE Model for this rulemaking sufficiently to
substantively engage with the Agencies on these topics through their
comments. Therefore, the agencies consider the detailed comments on the
OMEGA and CAFE models as clear indicia that the extensive information,
materials, and explanations provided by the agencies in the proposed
rule enabled significant opportunity for the public to comment on the
modeling for the rule.
---------------------------------------------------------------------------
\2878\ See, e.g., California Air Resources Board, NHTSA-2018-
0067-11873; Union of Concerned Scientists, NHTSA-2018-0067-12039;
Alliance of Automobile Manufacturers, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
To the extent that commenters allege an insufficient opportunity to
comment by claiming that the EPA actually utilized the OMEGA model in
the rulemaking process, the agencies consider such comments
unfounded.\2879\ The agencies did not rely on the OMEGA model during
the rulemaking process, including during the analysis for the proposed
and final rules. In past rulemakings, the EPA developed a complete
final version of the OMEGA model to perform the rulemaking analysis.
Here, the EPA did not even finalize a completed updated version of the
OMEGA model, much less rely on such a model in the course of the
rulemaking. Therefore, no completed version of an updated OMEGA model
even existed for the agencies to publish as part of the notice of
proposed rulemaking.
---------------------------------------------------------------------------
\2879\ See, e.g., Center for Biological Diversity, NHTSA-2018-
0067-12000.
---------------------------------------------------------------------------
To the extent commenters argue that the EPA should have updated the
model for this rulemaking, the APA's facilitation of a meaningful
opportunity to comment neither requires nor contemplates a mandate that
the agencies develop computational modeling alternatives for the
public, which were not even incorporated into the agencies' own
rulemaking analysis.\2880\ In fact, doing so would actually detract
from the notice and comment process because it would convolute the
rulemaking docket and inhibit the public's ability to identify the
modeling materials actually used in the rulemaking process. Thus, such
extraneous materials would only dilute the rulemaking docket with
voluminous and complex materials, such as modeling files, input files,
and statistical figures, that had no influence on the rulemaking in
question. Indeed, several commenters already claimed that the
voluminous and complex supporting materials in the rulemaking docket
required significant time for review, so the introduction of extensive
totally extraneous material would have been only counterproductive to
the process.\2881\
---------------------------------------------------------------------------
\2880\ See, e.g., Center for Biological Diversity et al., NHTSA-
2018-0067-12000.
\2881\ See, e.g., Institute for Policy Integrity, NHTSA-2018-
0067-5641; Northeast States for Coordinated Air Use Management,
NHTSA-2018-0067-2158.
---------------------------------------------------------------------------
Moreover, requiring the EPA to perform the work necessary to fully
update the OMEGA model solely for a public release--when it did not
otherwise intend to consider the model in the rulemaking--would divert
valuable and finite agency resources away from actual rulemaking
analyses in favor of efforts that further no progress in the
rulemaking.\2882\ Such an approach would detract from the agencies'
opportunities to devote time to other considerations that actually
influenced the rulemaking, such as the substantive analysis
incorporated into the proposed rule and the drafting of extensive
language to explain to the public the methodologies applied by the
agencies for the proposal. Such an inefficient allocation of resources
undermines both the rulemaking process envisioned by the APA and the
very notice and comment procedures utilized by these commenters.
---------------------------------------------------------------------------
\2882\ See, e.g., Environmental Defense Fund, NHTSA-2018-0067-
12108.
---------------------------------------------------------------------------
Several commenters also argued that even if the agencies did not
rely on the model for this rulemaking, the OMEGA model still informed
the EPA's analysis and interagency review by providing general
background experience in regulating greenhouse gas emissions--either
through the agency's work with prior versions of the model or ongoing
efforts to update the OMEGA model for purposes unrelated to this
rulemaking. However, even assuming the model provided background
experience to the EPA in regulating in this arena, federal
jurisprudence makes clear that ``[t]he Administrative Procedure Act
does not require that every bit of background information used by an
administrative agency be published for public comment.'' See B. F.
Goodrich Co. v. Dep't of Transp., 541 F.2d 1178, 1184 (6th Cir. 1976).
This is particularly the case when, as here, ``[t]he basic data upon
which the agency relied in formulating the regulation was available . .
. for comment.'' Id.; see also Am. Min. Cong. v. Marshall, 671 F.2d
1251, 1261 (10th Cir. 1982) (``These documents consist of background
information and data as well as several internal memoranda. There is
nothing to indicate that the Secretary actually relied on any of these
documents in promulgating the rule or that the data they contain was
critical to the formulation of the rule.''). In fact, publishing such
background information not only exceeds the requirements of the APA,
but would actually affirmatively undermine the APA's notice and comment
procedure. If every piece of information ever referenced by the
agencies or upon which the Agencies drew regulatory experience were
required to be published, rulemaking dockets would expand to an absurd
scope of nearly infinite materials, spanning arguably back to even the
school textbooks the rulemaking personnel used to learn the underlying
disciplines employed in the rulemaking analysis. Clearly such a scope
would frustrate rather than further the provision of proper notice to
the public about a proposed rule.\2883\
---------------------------------------------------------------------------
\2883\ To the extent commenters seek to understand the manner in
which the OMEGA model informed prior rulemaking efforts, the EPA has
released the full versions of prior OMEGA models and applicable
materials along with the prior rulemakings. In fact, several
commenters referenced such materials in submitting detailed comments
comparing the CAFE Model with the OMEGA model. Manufacturers of
Emission Controls Association, NHTSA-2018-0067-11994. Thus, any
commenters that were interested in such extraneous background
information had ample opportunity to access the material.
---------------------------------------------------------------------------
Moreover, even assuming the premise of several commenters'
challenges--that the EPA consulted updates to the OMEGA model during
the interagency review--such a predicate still would not require the
publication of the model during the rulemaking process.\2884\ As the
agencies have made clear, the OMEGA model did not affect any part of
the rule, including the methodologies and analysis underlying the
formulation of the rule. Therefore, even if consulted, the OMEGA model
would exist as, at most, supplementary material which had no influence
on the rulemaking methodologies, all of which were fully disclosed.
See, e.g., Chamber of Commerce of U.S. v. SEC., 443 F.3d 890, 900 (DC
Cir. 2006) (``When the agency relies on supplementary evidence without
a showing of prejudice by an interested party, the procedural
requirements of the APA are satisfied without further opportunity for
comment, provided that the agency's response constitutes a logical
outgrowth
[[Page 25161]]
of the rule initially proposed'') (internal citations omitted).
---------------------------------------------------------------------------
\2884\ See, e.g., Environmental Defense Fund, NHTSA-2018-0067-
12406.
---------------------------------------------------------------------------
3. National Environmental Policy Act
As discussed above, EPCA requires NHTSA to determine the level at
which to set CAFE standards for each model year by considering the four
factors of technological feasibility, economic practicability, the
effect of other motor vehicle standards of the Government on fuel
economy, and the need of the United States to conserve energy. The
National Environmental Policy Act (NEPA) directs that environmental
considerations be integrated into that process.\2885\ To explore the
potential environmental consequences of this rulemaking action, NHTSA
prepared a Draft Environmental Impact Statement (``DEIS'') for the NPRM
and a Final Environmental Impact Statement (``FEIS'') for the final
rule. The purpose of an EIS is to ``provide full and fair discussion of
significant environmental impacts and [to] inform decisionmakers and
the public of the reasonable alternatives which would avoid or minimize
adverse impacts or enhance the quality of the human environment.''
\2886\
---------------------------------------------------------------------------
\2885\ NEPA is codified at 42 U.S.C. 4321-47. The Council on
Environmental Quality (CEQ) NEPA implementing regulations are
codified at 40 CFR parts 1500-08.
\2886\ 40 CFR 1502.1.
---------------------------------------------------------------------------
As explained in the NPRM, NEPA is ``a procedural statute that
mandates a process rather than a particular result.'' \2887\ The
agency's overall EIS-related obligation is to ``take a `hard look' at
the environmental consequences before taking a major action.'' \2888\
Significantly, ``[i]f the adverse environmental effects of the proposed
action are adequately identified and evaluated, the agency is not
constrained by NEPA from deciding that other values outweigh the
environmental costs.'' \2889\ The agency must identify the
``environmentally preferable'' alternative but need not adopt it.\2890\
``Congress in enacting NEPA . . . did not require agencies to elevate
environmental concerns over other appropriate considerations.'' \2891\
Instead, NEPA requires an agency to develop and consider alternatives
to the proposed action in preparing an EIS.\2892\ The statute and
implementing regulations do not command the agency to favor an
environmentally preferable course of action, only that it make its
decision to proceed with the action after taking a hard look at the
potential environmental consequences and consider the relevant factors
in making a decision among alternatives.\2893\
---------------------------------------------------------------------------
\2887\ Stewart Park & Reserve Coal., Inc. v. Slater, 352 F.3d
545, 557 (2d Cir. 2003).
\2888\ Baltimore Gas & Elec. Co. v. Natural Resources Defense
Council, Inc., 462 U.S. 87, 97 (1983).
\2889\ Robertson v. Methow Valley Citizens Council, 490 U.S.
332, 350 (1989).
\2890\ 40 CFR 1505.2(b).
\2891\ Baltimore Gas, 462 U.S. at 97.
\2892\ 42 U.S.C. 4332(2)(C)(iii).
\2893\ 40 CFR 1505.2(b).
---------------------------------------------------------------------------
NHTSA received many comments on the DEIS. Among the comments
received, many commenters stated that the baseline/no-action standards
were the environmentally preferable alternative and argued that the
environmental benefits of the proposal were (1) insufficient and/or (2)
incorrectly assessed in a variety of ways. Comments regarding the
environmental analyses presented in this preamble are addressed in
Section VI above, while those regarding the DEIS are addressed in
Chapter 10 of the FEIS.
When preparing an EIS, NEPA requires an agency to compare the
potential environmental impacts of its proposed action and a reasonable
range of alternatives. In the DEIS, NHTSA analyzed a No Action
Alternative and eight action alternatives. In the FEIS, NHTSA analyzed
the same No Action Alternative and seven action alternatives, including
a new alternative (the Preferred Alternative) within the range of the
alternatives considered in the DEIS and FEIS.\2894\ The alternatives
represent a range of potential actions the agency could take, and they
are described more fully in Section V above, below in this section, and
Chapter 2 of the FEIS. The environmental impacts of these alternatives,
in turn, represent a range of potential environmental impacts that
could result from NHTSA's setting maximum feasible fuel economy
standards for passenger cars and light trucks.
---------------------------------------------------------------------------
\2894\ In its scoping notice, NHTSA indicated that the action
alternatives analyzed would bracket a range of reasonable annual
fuel economy standards, allowing the agency to select an action
alternative in its final rule from any stringency level within that
range. 82 FR 34740, 34743 (July 26, 2017).
---------------------------------------------------------------------------
To derive the direct and indirect impacts of the action
alternatives, NHTSA compared each action alternative to the No Action
Alternative, which reflects baseline trends that would be expected in
the absence of any further regulatory action other than finalizing the
augural standards. More specifically, the No Action Alternative in the
DEIS and FEIS assumed that NHTSA would not amend the CAFE standards for
MY 2021 passenger cars and light trucks. In addition, the No Action
Alternative assumed that NHTSA would finalize the MY 2022-2025 augural
CAFE standards that were described in the 2012 final rule. Finally, for
purposes of its analysis, NHTSA assumed that the MY 2025 augural
standards would continue indefinitely. The augural standards also serve
as a proxy for EPA's CO2 standards for MYs 2022-2025, which
were also finalized in the 2012 final rule. The No Action Alternative
provides an analytical baseline against which to compare the
environmental impacts of other alternatives presented in the EIS.\2895\
---------------------------------------------------------------------------
\2895\ See 40 CFR 1502.2(e), 1502.14(d). CEQ has explained that
``[T]he regulations require the analysis of the no action
alternative even if the agency is under a court order or legislative
command to act. This analysis provides a benchmark, enabling
decision makers to compare the magnitude of environmental effects of
the action alternatives [See 40 CFR 1502.14(c).] . . . Inclusion of
such an analysis in the EIS is necessary to inform Congress, the
public, and the President as intended by NEPA. [See 40 CFR
1500.1(a).]'' Forty Most Asked Questions Concerning CEQ's National
Environmental Policy Act Regulations, 46 FR 18026 (Mar. 23, 1981).
---------------------------------------------------------------------------
For the DEIS, NHTSA analyzed eight action alternatives,
Alternatives 1 through 8, which ranged from amending the MY 2021
standards to match the MY 2020 standards and holding those standards
flat for passenger cars and light trucks through MY 2026 (Alternative
1) to maintaining the existing MY 2021 standards and subsequently
requiring average annual increases in fuel economy by 2.0 percent
(passenger cars) and 3.0 percent (light trucks) (Alternative 8). The
action alternatives analyzed in the DEIS also reflected different
options regarding air conditioning efficiency and off-cycle technology
adjustment procedures, with some alternatives phasing out these
adjustments in MYs 2022-2026. For the FEIS, NHTSA analyzed seven action
alternatives, Alternatives 1 through 7, which range from amending the
MY 2021 standards to match the MY 2020 standards and holding those
standards flat for passenger cars and light trucks through MY 2026
(Alternative 1) to maintaining the existing MY 2021 standards and
subsequently requiring average annual increases in fuel economy by 2.0
percent (passenger cars) and 3.0 percent (light trucks) (Alternative 7)
from year to year. The primary differences between the action
alternatives for the DEIS and FEIS is that the FEIS did not analyze
alternatives that phased out the air conditioning efficiency and off-
cycle technology adjustments (see Section V above for further
discussion), and the FEIS added an alternative under which fuel economy
increased at 1.5 percent per year for both cars and light trucks
(Alternative 3). Both of the ranges of action alternatives, as well as
the No
[[Page 25162]]
Action Alternative, in the DEIS and FEIS encompassed a spectrum of
possible standards NHTSA could determine was maximum feasible based on
the different ways the agency could weigh EPCA's four statutory
factors. Throughout the FEIS, estimated impacts were shown for all of
these action alternatives, as well as for the No Action Alternative.
For a more detailed discussion of the environmental impacts associated
with the alternatives, see Chapters 3-8 of the FEIS, as well as Section
VII above.
NHTSA's FEIS describes potential environmental impacts to a variety
of resources, including fuel and energy use, air quality, climate, land
use and development, hazardous materials and regulated wastes,
historical and cultural resources, noise, and environmental justice.
The FEIS also describes how climate change resulting from global carbon
emissions (including CO2 emissions attributable to the U.S.
light duty transportation sector under the alternatives considered)
could affect certain key natural and human resources. Resource areas
are assessed qualitatively and quantitatively, as appropriate, in the
FEIS, and the findings of that analysis are summarized here.\2896\
---------------------------------------------------------------------------
\2896\ The impacts described in this section come from NHTSA's
FEIS, which is being publicly issued simultaneously with this final
rule. As described in Section VII.A.4.c.1 above, the FEIS is based
on ``unconstrained'' modeling rather than ``standard setting''
modeling; NHTSA conducts modeling both ways in order to reflect the
various statutory requirements of EPCA and NEPA. The preamble
employs the ``standard setting'' modeling in order to ensure that
the decision-maker does not consider things that EPCA/EISA prohibit,
but as a result, the impacts reported here may differ from those
reported elsewhere in this preamble. However, NHTSA considers the
impacts reported in the FEIS, in addition to the other information
presented in this preamble, as part of its decision-making process.
---------------------------------------------------------------------------
As the stringency of the alternatives increases, total U.S.
passenger car and light truck fuel consumption for the period of 2020
to 2050 decreases. Total light-duty vehicle fuel consumption from 2020
to 2050 under the No Action Alternative is projected to be 3,371
billion gasoline gallon equivalents (GGE). Light-duty vehicle fuel
consumption from 2020 to 2050 under the action alternatives is
projected to range from 3,598 billion GGE under Alternative 1 to 3,456
billion gallons GGE under Alternative 7. Under the Alternative 3,
light-duty vehicle fuel consumption from 2020 to 2050 is projected to
be 3,571 GGE. All of the action alternatives would increase fuel
consumption compared to the No Action Alternative, with fuel
consumption increases that range from 226 billion GGE under Alternative
1 to 85 billion GGE under Alternative 7.
The relationship between stringency and air pollutant emissions is
less straightforward, reflecting the complex interactions among the
tailpipe emissions rates of the various vehicle types, the technologies
assumed to be incorporated by manufacturers in response to the CAFE
standards, upstream emissions rates, the relative proportions of
gasoline and diesel in total fuel consumption, and changes in VMT from
the rebound effect. In general, emissions of criteria and toxic air
pollutants increase across all action alternatives, with some
exceptions. Further, the action alternatives would result in increased
incidence of PM2.5-related adverse health impacts (including
increased incidences of premature mortality, acute bronchitis,
respiratory emergency room visits, and work-loss days) due to the
emissions increases.\2897\
---------------------------------------------------------------------------
\2897\ As discussed in Section X.E.1, NHTSA also performed a
national-scale photochemical air quality modeling and health benefit
assessment for the FEIS, which is included as Appendix E. This
analysis affirms the estimates that appeared in the DEIS and
explains conclusions that may be drawn from the FEIS air quality
discussion.
---------------------------------------------------------------------------
For CO (in 2025), NOX (in 2025), and SO2,
emissions generally decrease under the action alternatives compared to
the No Action Alternative. For CO in 2025, the largest decrease occurs
under Alternative 1 and the emissions decreases get smaller from
Alternative 1 through Alternative 7. For NOX in 2025, the
largest decrease occurs under Alternative 6. For SO2 in
2025, the largest decrease occurs under Alternative 6; however,
SO2 emissions under Alternative 7 are greater than under the
No Action Alternative. For SO2 in 2035, the largest decrease
occurs under Alternative 2. For SO2 in 2050, the largest
decrease occurs under Alternative 1 and the emissions decreases get
smaller from Alternative 1 through Alternative 7. Across all criteria
pollutants, action alternatives, and analysis years, the smallest
decrease in emissions is less than 0.1 percent and occurs for
NOX under Alternative 7 in 2025; the largest decrease is 12
percent and occurs for SO2 under Alternative 2 in 2050.
For CO (in 2035 and 2050), NOX (in 2035 and 2050),
PM2.5, and VOCs, emissions show increases across action
alternatives compared to the No Action Alternative, with the largest
increases occurring under Alternative 1 (except CO in 2035, for which
the largest increase occurs under Alternative 4). The emissions
increases get smaller from Alternative 1 through Alternative 7.
Exceptions to this trend are for PM2.5 and VOCs in 2025,
which show the smallest emissions increase under Alternative 6. Across
all criteria pollutants, action alternatives, and analysis years, the
smallest increase in emissions is 0.1 percent and occurs for
SO2 under Alternative 7 in 2025; the largest increase is 12
percent and occurs for VOCs under Alternative 1 in 2050.
Under each action alternative in 2025 compared to the No Action
Alternative, decreases in emissions would occur for all toxic air
pollutants except for DPM, for which emissions would increase by as
much as 2 percent. For 2025, the largest relative decreases in
emissions would occur for 1,3,-butadiene, for which emissions would
decrease by as much as 0.5 percent. Percentage reductions in emissions
of acetaldehyde, acrolein, benzene, and formaldehyde would be less.
Under each action alternative in 2035 and 2050 compared to the No
Action Alternative, increases in emissions would occur for all toxic
air pollutants. The largest relative increases in emissions would occur
for DPM, for which emissions would increase by as much as 9 percent.
Percentage increases in emissions of acetaldehyde, acrolein, benzene,
1,3,-butadiene, and formaldehyde would be less.
In addition, the action alternatives would result in increased
incidence of PM2.5-related adverse health impacts due to the
emissions increases. Increases in adverse health outcomes include
increased incidences of premature mortality, acute bronchitis,
respiratory emergency room visits, and work-loss days. In 2025 and
2035, all action alternatives except for Alternative 6 would result in
increased adverse health impacts nationwide compared to the No Action
Alternative as a result of increases in emissions of NOX,
PM2.5, and DPM. The increases in adverse health impacts are
largest for the least stringent alternative (Alternative 1). The
increases get smaller from Alternative 1 to Alternative 4, get larger
from Alternative 4 to Alternative 5, then smaller from Alternative 5 to
Alternative 6, and larger again from Alternative 6 to Alternative 7. In
2050, all action alternatives would result in decreased adverse health
impacts nationwide compared to the No Action Alternative as a result of
decreases in emissions of SOX. The decreases in adverse
health impacts get smaller from Alternative 1 to Alternative 7.
The action alternatives would increase U.S. passenger car and light
truck fuel consumption and CO2 emissions compared with the
No Action
[[Page 25163]]
Alternative, resulting in minor increases to the anticipated increases
in global CO2 concentrations, temperature, precipitation,
and sea level, and minor decreases in ocean pH that would otherwise
occur, as described below. They could also, to a small degree, increase
the impacts and risks of climate change. Uncertainty exists regarding
the magnitude of impact on these climate variables, as well as to the
impacts and risks of climate change. Still, the impacts of the action
alternatives on global mean surface temperature, precipitation, sea
level, and ocean pH would be extremely small in relation to global
emissions trajectories. This is because of the global and multi-
sectoral nature of climate change. These effects would be small, would
occur on a global scale, and would not disproportionately affect the
United States.
According to the FEIS, passenger cars and light trucks are
projected to emit 85,900 million metric tons of carbon dioxide
(MMTCO2) from 2021 through 2100 under the No Action
Alternative. Alternative 1 would increase these emissions by 10 percent
through 2100 (approximately 8,800 MMTCO2). Alternative 7
would increase these emissions by 4 percent through 2100 (approximately
3,100 MMTCO2). Emissions increases would be highest under
Alternative 1 and would decrease across the action alternatives, with
emissions being the lowest under the No Action Alternative.
In the FEIS, NHTSA presented two different analyses based on these
emissions changes to illustrate potential impacts to certain climate
variables. In the first analysis, to represent the direct and indirect
impacts of this action, NHTSA used the Global Change Assessment Model
(GCAM) Reference scenario (i.e., future global emissions assuming no
additional climate policy [``business-as-usual'']) to represent the
reference case emissions scenario. Under that analysis, total global
CO2 emissions from all sources are projected to be 4,950,865
MMTCO2 under the No Action Alternative from 2021 through
2100, which means that the action alternatives are expected to increase
global CO2 emissions between 0.06 (Alternative 7) and 0.17
(Alternative 1) percent by 2100. The estimated CO2
concentrations in the atmosphere for 2100 would range from 789.89 parts
per million (ppm) under Alternative 1 to approximately 789.11 ppm under
the No Action Alternative, indicating a maximum atmospheric
CO2 increase of approximately 0.78 ppm compared to the No
Action Alternative.
Changes in CO2 emissions translate to changes in global
mean surface temperature, sea levels, global mean precipitation, and
ocean pH, among other things. Under the first analysis, global mean
surface temperature is projected to increase by approximately
3.48[deg]C (6.27 [deg]F) under the No Action Alternative by 2100.
Implementing the lowest-emissions action alternative (Alternative 7)
would increase this projected temperature rise by 0.001[deg]C (0.002
[deg]F), while implementing the highest-emissions alternative
(Alternative 1) would increase projected temperature rise by
0.003[deg]C (0.005 [deg]F). Projected sea-level rise in 2100 ranges
from a low of 76.28 centimeters (30.03 inches) under the No Action
Alternative to a high of 76.35 centimeters (30.06 inches) under
Alternative 1. Alternative 1 would result in an increase in sea level
equal to 0.07 centimeter (0.03 inch) by 2100 compared with the level
projected under the No Action Alternative, compared to an increase
under Alternative 7 of 0.02 centimeter (0.001 inch) compared with the
No Action Alternative. Global mean precipitation is anticipated to
increase by 5.85 percent by 2100 under the No Action Alternative. Under
the action alternatives, this increase in precipitation would be
increased further by 0.01 percent. Finally, ocean pH in 2100 is
anticipated to be 8.2715 under Alternative 7, about 0.0001 less than
the No Action Alternative. Under Alternative 1, ocean pH in 2100 would
be 8.2712, or 0.0004 less than the No Action Alternative.
In the second analysis, NHTSA used the GCAM6.0 scenario instead of
the default scenario to represent the reference case emissions
scenario. The GCAM6.0 scenario assumes a moderate level of global GHG
reductions and corresponds to stabilization, by 2100, of total
radiative forcing and associated CO2 concentrations at
roughly 678 ppm. By assuming a moderate level of global GHG reduction,
NHTSA attempts to capture the cumulative impacts of this action (i.e.,
the impact on the environment which results from the incremental impact
of the action when added to other past, present, and reasonably
foreseeable future actions). In the FEIS, NHTSA documented a number of
domestic and global actions that indicate that a moderate reduction in
the growth rate of global GHG emissions is reasonably foreseeable in
the future.
Under the second analysis, compared with projected total global
CO2 emissions of 4,044,005 MMTCO2 from all
sources from 2021 to 2100, the incremental impact of this rulemaking is
expected to increase global CO2 emissions between 0.08
(Alternative 7) and 0.22 (Alternative 1) percent by 2100. Estimated
atmospheric CO2 concentrations in 2100 range from a low of
687.3 ppm under the No Action Alternative to a high of 688.04 ppm under
Alternative 1. Alternative 7, the lowest CO2 emissions
alternative, would result in CO2 concentrations of 687.55
ppm, an increase of 0.26 ppm compared with the No Action Alternative.
Global mean surface temperature increases for the action alternatives
compared with the No Action Alternative in 2100 range from a low of
0.001[deg]C (0.002 [deg]F) under Alternative 7 to a high of 0.004[deg]C
(0.007 [deg]F) under Alternative 1. Global mean precipitation is
anticipated to increase by 4.77 percent by 2100 under the No Action
Alternative. Under the action alternatives, this increase in
precipitation would be increased further by 0.01 percent. Projected
sea-level rise in 2100 ranges from a low of 70.22 centimeters (27.65
inches) under the No Action Alternative to a high of 70.30 centimeters
(27.68 inches) under Alternative 1, indicating a maximum increase of
sea-level rise of 0.07 centimeter (0.03 inch) by 2100. Sea-level rise
under Alternative 7 would be 70.25 centimeters (27.66 inches), a 0.03
centimeter (0.01-inch) increase compared to the No Action Alternative.
Ocean pH in 2100 is anticipated to be 8.2721 under Alternative 7, about
0.0001 less than the No Action Alternative. Under Alternative 1, ocean
pH in 2100 would be 8.2719, or 0.0004 less than the No Action
Alternative.
For several other resources, NHTSA is unable to provide a
quantitative measurement of potential impacts. Instead, the FEIS
presents a qualitative discussion on potential impacts. In most cases,
NHTSA presents the findings of a literature review of scientific
studies, such as in Chapter 6, where NHTSA provides a literature
synthesis focusing on existing credible scientific information to
evaluate the most significant lifecycle environmental impacts from some
of the fuels, materials, and technologies that may be used to comply
with the alternatives. In Chapter 7, NHTSA discusses land use and
development, hazardous materials and regulated waste, historical and
cultural resources, noise, and environmental justice. Finally, in
Chapter 8, NHTSA discusses cumulative impacts related to energy, air
quality, and climate change, and provides a literature synthesis of the
impacts on key natural and human resources of changes in climate change
variables. In these chapters, NHTSA concludes that impacts would be
proportional to changes in emissions that would result
[[Page 25164]]
under the alternatives. As a result, among the action alternatives,
Alternative 1 would have the highest impact on these resources while
Alternative 7 would have the lowest.
Based on the foregoing, NHTSA concludes from the FEIS that the No
Action Alternative is the overall environmentally preferable
alternative because, assuming full compliance were achieved regardless
of the agency's assessment of the costs to industry and society, it
would result in the largest reductions in fuel use and CO2
emissions among the alternatives considered. In addition, the No Action
Alternative would result in the lowest overall emissions levels of
criteria air pollutants (with the exception of sulfur dioxide) and of
the toxic air pollutants studied by NHTSA. Impacts on other resources
(especially those described qualitatively in the FEIS) would be
proportional to the impacts on fuel use and emissions, as further
described in the FEIS, with the No Action Alternative expected to have
the fewest negative impacts.\2898\ Although the CEQ regulations require
NHTSA to identify the environmentally preferable alternative,\2899\ the
agency need not adopt it, as described above. The following section
(Section VIII.B.4) explains how NHTSA balanced the relevant factors to
determine which alternative represented the maximum feasible standards,
including why NHTSA does not believe that the environmentally
preferable alternative is maximum feasible.
---------------------------------------------------------------------------
\2898\ Among the action alternatives considered, Alternative 7
would be the environmentally preferable alternative, as it is
closest in stringency to the No Action Alternative.
\2899\ 40 CFR 1505.2(b).
---------------------------------------------------------------------------
4. Evaluating the EPCA Factors and Other Considerations To Arrive at
the Proposed Standards
As discussed in this section, NHTSA is required to consider four
enumerated factors when establishing maximum feasible CAFE standards
under 49 U.S.C. chapter 329: ``technological feasibility, economic
practicability, the effect of other motor vehicle standards of the
Government on fuel economy, and the need of the United States to
conserve energy.'' \2900\ For this final rule, NHTSA has considered a
wide range of potential CAFE standards (Baseline/No Action Alternative
and Alternatives 1 through 7), ranging from the augural standards set
forth in 2012 (Baseline/No Action Alternative), through a number of
less stringent alternatives, including the proposed preferred
alternative (Alternative 1, 0 percent per year stringency improvement)
and what has been chosen as the final standards (Alternative 3, 1.5
percent per year stringency improvement). NHTSA has determined that
Alternative 3, which would increase the stringency of the MY 2020
standards by 1.5 percent per year for both passenger cars and light
trucks from MY 2021 through 2026, represents the maximum feasible CAFE
standards under 49 U.S.C. 39202. In addition to technological
feasibility, economic practicability, the effects of other motor
vehicle standards of the Government on fuel economy, and the need of
the United States to conserve energy, NHTSA has also considered the
impact of the standards on safety and the environment.
---------------------------------------------------------------------------
\2900\ 49 U.S.C. 32902(f).
---------------------------------------------------------------------------
How did the Agency balance the factors for the NPRM?
In the NPRM, NHTSA began its discussion of the tentative balancing
of factors by explaining that ``NHTSA well recognizes that the decision
it proposes to make in today's NPRM is different from the one made in
the 2012 final rule that established standards for MY 2021 and
identified ``augural'' standard levels for MYs 2022-2025. Not only do
we believe that the facts before us have changed, but we believe that
those facts have changed sufficiently that the balancing of the EPCA
factors and other considerations must also change. The standards we are
proposing today reflect that balancing.'' \2901\ NHTSA highlights this
discussion at the outset in response to the number of commenters who
claimed that NHTSA had not acknowledged or explained in the NPRM how or
why the proposal was different from past work or policy decisions.
---------------------------------------------------------------------------
\2901\ 83 FR at 43213.
---------------------------------------------------------------------------
The NPRM balancing discussion went on to explore the definition of
``to conserve'' in the context of what ``energy conservation'' and
``the need of the U.S. to conserve energy'' should be interpreted to
mean, in recognition of the major structural changes in global oil
markets since EPCA was originally passed, and even since the 2012 final
rule that set forth the augural standards. NHTSA examined these changes
from both a demand perspective and a supply perspective. On the demand
side, U.S. demand and global demand have both changed over time. The
NPRM discussed the fact that the U.S. consumes a much smaller share of
global oil output than it did at the CAFE program's outset, both
because U.S. fleet fuel economy has improved, and because other
countries that were not major petroleum consumers in the 1970s have
rapidly increased their share of consumption, and continue to do so. A
more globalized market means that risk of price spikes is spread
around--making the U.S. in particular less likely to bear a
disproportionate burden of price spikes. The NPRM also discussed the
decreasing energy intensity of the U.S. economy over time and the
improving balance of payments in petroleum, including the likelihood
that the U.S. is poised to become a net petroleum exporter in the near
future. Related to the decreasing energy intensity of the U.S. economy,
on the demand side, the NPRM discussed the proliferation of fuel-
efficient vehicle options in the market in response to CAFE increases
over time, and the fact that consumers who wish to purchase more fuel
efficient vehicles have largely done so, and may continue to do so over
time if they wish.
On the supply side, the NPRM explained, vast increases in U.S.
petroleum production, largely from shale formations, have introduced a
major new stable supply into the global market. Shale oil production
costs may be higher than the cost (for example, to OPEC members) to
produce traditional oil, but that itself acts as a lever on global
prices. Prices of goods like oil are affected by demand and supply--
given that global demand trends increase relatively steadily, if OPEC
States want to increase revenues by selling more of the total oil
consumed globally, they have to try to control global supply volume by
controlling production volumes (to avoid shale production increasing in
response to higher prices). In short, the higher global prices trend,
the more U.S. shale production increases in response, and as supply
increases, prices fall. The NPRM discussed the responsiveness of U.S.
shale production and suggested it could be higher than traditional
producers in some instances. Traditional oil producers seeking to
maintain market share have a new incentive to keep prices below a
certain threshold, and U.S supply helps to buffer the impact of
geopolitical events. The NPRM looked at then-current EIA oil price
forecasts, under which U.S. gasoline prices were not forecast to exceed
$4/gallon through 2050, and acknowledged that while price shocks could
still occur, NHTSA tentatively concluded that from the supply side, it
is possible that the oil market conditions that created the price
shocks in the 1970s may no longer exist.
In light of these changes in global oil markets, the NPRM
tentatively concluded that many aspects of the need of the U.S. to
conserve energy had
[[Page 25165]]
improved enough over time to merit further consideration of what the
need of the United States is to conserve oil today and going forward.
With regard to environmental considerations, the NPRM returned to the
definition of ``to conserve'' and suggested that differences of
thousandths of a degree Celsius in 2100 resulting from higher levels of
carbon dioxide emissions under the proposal as compared to the augural
standards might not rise to the level of ``wasteful,'' given the other
considerations discussed. With regard to consumer costs, the NPRM
discussed the interplay of oil market conditions with prior arguments
about consumer ``myopia'' with regard to the benefits of fuel savings,
and tentatively concluded that U.S. consumers may be valuing fuel
savings appropriately and purchasing the vehicles they want to
purchase--i.e., that using CAFE standards as a tool to compel consumers
to save money may not be necessary.
Given the discussion above, NHTSA tentatively concluded that the
need of the U.S. to conserve energy may no longer function as assumed
in previous considerations of what CAFE standards would be maximum
feasible. In that discussion, NHTSA stated that the overall risks
associated with the need of the U.S. to conserve oil have entered a new
paradigm with the risks substantially lower today and projected into
the future than when CAFE standards were first issued and in the recent
past. NHTSA explained that the effectiveness of CAFE standards in
reducing the demand for fuel combined with the increase in domestic oil
production have contributed significantly to the current situation and
outlook for the near- and mid-term future. NHTSA tentatively concluded
that the world has changed, and the need of the U.S. to conserve
energy, at least in the context of the CAFE program, has also changed.
Of two other factors under 32902(g), the NPRM explained that the
changes were perhaps less significant. NHTSA suggested that all of the
alternatives appear as though they could narrowly be considered
technologically feasible, in that they could be achieved based on the
existence or the projected future existence of technologies that could
be incorporated on future vehicles. With regard to the effect of other
motor vehicle standards of the Government on fuel economy, the NPRM
explained that it was similarly not heavily limiting during this
rulemaking time frame. The NPRM analysis projected that neither safety
standards nor Tier 3 compliance obligations appeared likely to make it
significantly harder for industry to comply with more stringent CAFE
standards, and that EPA's CO2 standards should have no
greater effect on difficulty in meeting CAFE standards than already
existed.
For economic practicability, the NPRM considered the traditional
definition used by the agency, and expressed concern that all of the
alternatives considered in the proposal could raise economic
practicability concerns. NHTSA stated that it believed there could be
potential for unreasonable elimination of consumer choice, loss of U.S.
jobs, and a number of adverse economic consequences under nearly all if
not all of the regulatory alternatives considered in the NPRM. NHTSA
explored consumer choice issues given a foreseeable future of
relatively low fuel prices and the likelihood that more stringent CAFE
standards could cause automakers to add technology to new vehicles that
consumers do not want, or prevent the addition of technology to new
vehicles that consumers do want, and suggested that there could be risk
that such elimination of consumer choice could be unreasonable. NHTSA
explained its assumption, based on repeated manufacturer input, that
fuel-saving technologies that paid for themselves within 2.5 years
would be added regardless of CAFE stringency, meaning that the power of
CAFE standards (by themselves) to compel fuel savings was reduced.
NHTSA suggested that requiring more technology to be added than
consumers were willing to pay for could have dampening effects on
vehicle sales, particularly given forecasted relatively low gas prices,
increasing the likelihood of automaker non-compliance with more
stringent standards due to difficulty in selling higher-fuel-economy
models. NHTSA examined the levels of electrification necessary to meet
the various regulatory alternatives evaluated in the NPRM and compared
them with information about consumers' willingness to purchase vehicles
with these technologies and even to spend money on fuel economy
improvements generally. NHTSA suggested that if the market for higher
fuel-economy vehicles exists and is already possibly saturated,
increasing fuel economy requirements could create economic
practicability concerns by affecting sales and consumer choice.
NHTSA recognized that automakers cross-subsidize regulation-driven
cost increases and expressed concern about their ability to do that
under sustained, ongoing increases over many years, and the
corresponding concern that continued cross-subsidizing could create
affordability problems for lower-income consumers if manufacturers pass
costs forward to consumers more broadly rather than concentrating them
in high-volume, higher-profit vehicles. NHTSA suggested that higher
vehicle prices and monthly vehicle payments could outweigh, for at
least some new vehicle purchasers, the benefit of fuel savings, because
vehicle payments are fixed costs and fuel costs may be less fixed.
NHTSA expressed concern that as vehicles get more expensive in response
to higher CAFE standards, it will become more and more difficult for
finance companies and dealers to continue creating loan terms that keep
monthly payments low and do not result in consumers' still owing
significant amounts of money on the vehicle by the time they can be
expected to be ready for a new vehicle. This situation may imply a
bubble in new vehicle sales, the effects of which could fall
disproportionately on new and low-income buyers. NHTSA suggested that
these effects could impact both fleet-wide safety (by slowing fleet
turnover) and consumer choice. The NPRM also expressed concern that the
sales and employment analyses were unable to capture (1) the risk that
manufacturers and dealers may not be able to continue keeping monthly
new vehicle payments low, or (2) the risk that manufacturing could
shift overseas as manufacturing costs rise.
NHTSA also examined the net benefits of the various regulatory
alternatives, and noted that the analysis showed that consumers recoup
only a portion of the costs associated with increasing stringency under
all of the alternatives, because the fuel savings resulting from each
of the alternatives was substantially less than the costs associated
with the alternative, meaning that net savings for consumers improved
as stringency decreased. NHTSA explained that it recognized that this
was a significantly different analytical result from the 2012 rule,
which showed the opposite trend, and explained that the result was
different because the facts and analysis underlying the result were
also different, and enumerated the noteworthy differences, such as
payback assumptions; fleet composition; what levels of technologies had
already been applied; the costs and effectiveness values for some of
those technologies; fuel price forecasts; the value of the rebound
effect; the value of the social cost of carbon; accounting for price
impacts on fleet turnover; not limiting mass reduction to only the
largest vehicles; and the value of a statistical life having increased.
NHTSA explained that all of these changes, together, meant
[[Page 25166]]
that the standards under any of the regulatory alternatives (compared
to the preferred alternative) were more expensive and had lower
benefits than if they had been calculated using the inputs and
assumptions of the 2012 analysis. This assessment, in turn, contributed
to the agency's decision to reevaluate what standards might be maximum
feasible in the model years covered by the rulemaking. NHTSA explained
that it had thus both relied on new facts and circumstances in
developing the proposal and reasonably rejected prior analyses relied
on in the 2012 final rule.\2902\
---------------------------------------------------------------------------
\2902\ See FCC v. Fox Television Stations, 556 U.S. at 514-515;
see also NAHB v. EPA, 682 F.3d 1032 (D.C. Cir. 2012).
---------------------------------------------------------------------------
NHTSA then considered that ``maximum feasible'' may change over
time as the agency assessed the relative importance of each factor that
Congress requires it to consider, and tentatively concluded that
proposing CAFE standards that hold the MY 2020 curves for passenger
cars and light trucks constant through MY 2026 would be the maximum
feasible standards for those fleets and would fulfill EPCA's
overarching purpose of energy conservation in light of the facts before
the agency and as the agency expected them to be in the rulemaking time
frame. NHTSA recognized that this was a different interpretation from
the 2012 final rule and explained that the context of that rulemaking
was meaningfully different from the current context, because the facts
had changed the importance of the need of the U.S. to conserve energy,
and NHTSA recognized that under that circumstance, while more stringent
standards may be possible, insofar as production-ready technology
exists that the industry could physically employ to reach higher
standards, it was not clear that higher standards would be economically
practicable in light of current U.S. consumer needs to conserve energy.
Therefore, NHTSA stated, it viewed the determination of maximum
feasible standards as a question of the appropriateness of standards
given that their need--either from the societal-benefits perspective in
terms of risk associated with fuel price shocks or other related
catastrophes, or from the private-benefits perspective in terms of
consumer willingness to purchase new vehicles with expensive
technologies that may allow them to save money on future fuel
purchases--seems likely to remain low for the foreseeable future. NHTSA
also considered the effects of the standards on highway safety and
expressed concern that because more stringent standards could depress
sales and slow fleet turnover, and because higher fuel economy leads to
more driving and more exposure to crash risk, all regulatory
alternatives would improve safety as compared to the augural standards.
(b) What comments did NHTSA receive regarding how it balanced the
factors in the NPRM?
In addition to comments on each of the factors NHTSA considered
discussed above, comments also were received on how NHTSA should
balance these factors in determining the maximum feasible final
standards. Hundreds of thousands of comments addressed stringency and,
thus, the agency's evaluation of what standards were maximum feasible.
Most of those focused on the augural standards: Many individual
commenters supported reducing the stringency of the standards from
augural levels--some citing estimates of cost, and some citing concerns
about consumer choice. Many comments by other individual commenters
supported retaining stringency at augural levels or increasing
stringency beyond that level--generally citing concerns about climate
change and increased fuel costs under less stringent standards. A few
commenters, like CEI, expressly supported the proposal, and even
suggested that stringency should be decreased further. Many other
commenters, including environmental and consumer groups, health
advocacy organizations, and a number of State organizations, argued
that the proposal was flawed and/or that the augural standards should
be finalized because more stringent standards help to reduce climate
change and address other air quality issues.\2903\ The Congressional
Tri-Caucus commenters supported maintaining the augural standards,
stating that they contribute to employment and protect low income
communities and communities of color.\2904\
---------------------------------------------------------------------------
\2903\ See, e.g., Harvard Environmental Law Clinic, EPA-HQ-OAR-
2018-0283-5486, at 1; University of San Francisco graduate students,
EPA-HQ-OAR-2018-0283-2676, at 1-2; Vanderbilt student organizations,
EPA-HQ-OAR-2018-0283-4189, at 1-2; Blue Planet Foundation, EPA-HQ-
OAR-2018-0283-4207, at 1; Green Energy Institute (Lewis and Clark
Law School), et al., EPA-HQ-OAR-2018-0283-4193, at 1-3; CBD et al.,
NHTSA-2018-0067-12057, at 2; NESCAUM, NHTSA-2018-0067-11691, at 3-4.
\2904\ Congressional Tri-Caucus, NHTSA-2018-0067-1424, at 1.
---------------------------------------------------------------------------
The Alliance and Global Automakers both supported final standards
that increased in stringency year over year. The Alliance stated that
it could support stringency increases between 0 percent per year and 2-
3 percent per year ``along with the inclusion of appropriate
flexibilities.'' \2905\ Global stated that increases should be
``meaningful'' \2906\ and suggested that ``[i]n order for the U.S. auto
industry to remain competitive and continue to export vehicles to the
rest of the world, industry is best served by a reasonable, steady ramp
rate that accounts for investments made and the global nature of the
market. Steady increases allow for long-term planning and create an
environment of security that fosters ongoing investment in vehicle
technology and consumer confidence in purchasing new vehicles. It also
provides a level playing field upon which automakers can compete.''
\2907\ Toyota made similar points, and argued that while the standards
set in 2012 are beyond maximum feasible today, the ``statutes support
an adjustment to those standards that reflect the realities of the
market, consumer choice, and the pace of technological advancement
acceptable to consumers.'' \2908\ Mazda stated that it supported
``increasing requirements for fuel efficiency. . ., if they are
sensible and achievable under changing market conditions.'' \2909\
---------------------------------------------------------------------------
\2905\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 8.
\2906\ Global, NHTSA-2018-0067-12032, at 3.
\2907\ Global, NHTSA-2018-0067-12032, Attachment A, at A-11.
\2908\ Toyota, NHTSA-2018-0067-12150, at 31.
\2909\ Mazda, NHTSA-2018-0067-11727, at 2.
---------------------------------------------------------------------------
NADA commented that it was willing to support standards that
increased in stringency (i.e., more stringent than the proposal) if
they were economically practicable and technologically feasible, based
on the evidence before the agencies; if they ensured consumer choice
and ``the strongest possible rate of fleet turnover;'' and if passenger
car and light truck standards increased at the same rate.\2910\ The
Alliance for Vehicle Efficiency (AVE) argued that compliance shortfalls
are evidence that the current rate of stringency increase is beyond
maximum feasible, and that the assumptions that enabled those rates to
be chosen ``are no longer feasible based on consumer adoption.'' \2911\
AVE suggested that a rate of increase of 2.5 percent per year for both
cars and trucks, retroactively imposed beginning in MY 2018, would be
feasible given sufficient flexibilities.\2912\
---------------------------------------------------------------------------
\2910\ NADA, NHTSA-2018-0067-12064, at 12.
\2911\ AVE, NHTSA-2018-0067-11696, at 6-8.
\2912\ Id., at 10.
---------------------------------------------------------------------------
NADA also stressed the importance of flexibilities as a compliance
tool for meeting standards that increase faster
[[Page 25167]]
than the proposal.\2913\ The Minnesota agencies supported maintaining
standards at the augural levels, commenting that automakers has simply
``requested additional flexibility . . ., not a wholesale rollback of
the standards,'' and suggesting that additional flexibilities would
enable augural levels.\2914\ IPI disagreed with the suggestion in the
NPRM that heavy automaker reliance on credits for compliance might
indicate that standards were beyond maximum feasible, arguing that
automakers must be either using credits about to expire, or counting on
future standards being cheaper to meet due to rising consumer demand
for fuel economy, technology costs decreasing over time, and the cost-
effectiveness of EPA's EV multiplier incentive.\2915\
---------------------------------------------------------------------------
\2913\ NADA, NHTSA-2018-0067-12064, at 12.
\2914\ 1 Minnesota agencies, NHTSA-2018-0067-11706, at 6-7.
\2915\ IPI, NHTSA-2018-0067-12213, Appendix, at 25-26.
---------------------------------------------------------------------------
With regard to analysis of costs and benefits, IPI argued that the
final rule needed, like the 2012 rule, to cite costs and benefit
expressly in discussing balancing of statutory factors, but with a
``proper'' accounting of costs and benefits. IPI claimed that in the
NPRM the factors were balanced ``in a way that conflicts with the . .
.controlling statute[ ] and weighed . . .without regard for the
accuracy of the accompanying cost-benefit analysis.'' \2916\ IPI stated
that ``. . . the agencies' analysis produced biased and irrational
results at each of the steps in that causal chain, leading to a
Proposed Rule that vastly overstates the benefits of the rollback and
understates the benefits society foregoes with the rollback,'' and that
``[a] full and balanced analysis of all the costs and benefits that the
agencies are charged with considering would reveal--as the midterm
review recently confirmed--that the baseline standards will deliver
massive net social benefits, and the proposed rollback is
unjustified.'' \2917\
---------------------------------------------------------------------------
\2916\ Id.
\2917\ IPI, NHTSA-2018-0067-12213, Appendix, at 1-2.
---------------------------------------------------------------------------
With regard to net benefits, the States and Cities commenters
stated that prior analyses had concluded that the net benefits of the
augural standards were extremely high,\2918\ while the Alliance stated
that ``[t]he NERA-Trinity Assessment confirms the Agencies' findings
that Alternatives 1, 5, and 8 result in increased net benefits relative
to the no-action alternative augural CAFE standards.''\2919\ Michalek
and Whitefoot commented that ``maximizing net benefits is among the
most important factors to consider in policy selection because it is an
effort to weigh a variety of policy implications on a common basis and
seek decisions that are beneficial to society overall,'' but also
cautioned that estimates are inherently uncertain and should be
transparent and clearly justified; that sensitivity analysis is
necessary; that a net benefits analysis will not be able to capture
distributional effects or changes in behavior caused by the policy; and
that ``it is not clear that there is necessarily any relationship
between MNB and setting the `maximum feasible' criteria while
considering `economic practicability.' ''\2920\ IPI disagreed with the
NPRM's suggestion that feasibility concerns could lead NHTSA not to
maximize net benefits, stating that ``if a standard were truly not
feasible, then its costs would be prohibitively high, and a full and
fair cost-benefit analysis would reflect that.'' \2921\
---------------------------------------------------------------------------
\2918\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 6.
\2919\ Alliance, NHTSA-2018-0067-12073, Full Comment Set, at 13.
\2920\ Michalek and Whitefoot, NHTSA-2018-0067-11903, at 14-15.
\2921\ IPI, NHTSA-2018-0067-12213, Appendix, at 11.
---------------------------------------------------------------------------
CARB argued that ``[a]lthough EPCA provides NHTSA with some
discretion with respect to balancing the four factors, that discretion
is nevertheless constrained by EPCA's overriding mandate of conserving
energy.'' \2922\ CARB further stated that EPCA ``envision[s] the
promulgation of increasingly stringent requirements to ensure the
continued reductions of both emissions and fuel consumption from motor
vehicles.'' \2923\ Michalek and Whitefoot similarly commented that the
requirement that standards be maximum feasible necessarily means that
stringency must increase over time, because technology capabilities and
cost are constantly improving; international regulations are constantly
increasing in stringency; and if standards are held constant,
automakers will always exceed them.\2924\ The States and Cities
commenters cited the CAS language from the D.C. Circuit that ``[i]t is
axiomatic that Congress intended energy conservation to be a long term
effort that would continue through temporary improvements in energy
availability,'' and argued that ``[w]hile NHTSA purports to acknowledge
this purpose and the importance of improving fuel economy over time,
NHTSA proposes to do the opposite: roll back fuel economy standards for
a period of at least six years.'' \2925\ The States and Cities
commenters further argued that NHTSA had ``departed sharply from its
past interpretations and practice without an adequate explanation,
often without even an acknowledgement,'' citing Fox Television, insofar
as the 2012 final rule justification had noted that less stringent
regulatory alternatives would have conserved less energy than the then-
finalized standards, as compared to ``[w]ith the Proposed Rollback,
NHTSA has radically changed positions--assuming energy conservation
provides little, if any, benefits, for example--without explaining or
even acknowledging this complete reversal of course.'' \2926\ The
States and Cities commenters concluded that it was ``impermissible''
for NHTSA to balance ``the factors in a manner that contravenes EPCA's
central purpose of energy conservation.'' \2927\
---------------------------------------------------------------------------
\2922\ CARB, NHTSA-2018-0067-11783, Detailed Comments, at 78.
\2923\ Id., at 80.
\2924\ Michalek and Whitefoot, NHTSA-2018-0067-11903, at 3-4.
\2925\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 64-65.
\2926\ Id., at 65.
\2927\ Id.
---------------------------------------------------------------------------
ACEEE commented that NHTSA did not have discretion to assess
whether the need of the U.S. to conserve energy was as great as when
EPCA was first passed, arguing that ``[t]he statute does not ask for a
determination on whether the nation needs to save energy. It assumes
the need and directs that the need be taken into account along with
other considerations.'' \2928\ Securing America's Energy Future
commented that the need of the U.S. to conserve energy continued, and
that ``[a]lthough the nation is undoubtedly more energy secure than it
was before the start of the U.S. shale oil revolution ten years ago,''
\2929\ ``[u]ntil the U.S. transportation sector is no longer beholden
to oil, the country will be vulnerable to oil price volatility.
Improving the fuel efficiency of the U.S. vehicle fleet is a valuable
insurance policy against this volatility.'' \2930\ IPI also commented
that fuel efficiency standards act as insurance, but against
unpredictable future fuel prices.\2931\ IPI stated that anticipating
relatively low future fuel prices was not an appropriate basis for
finalizing the proposal, both because fuel costs may rise in the
future, and also because
[[Page 25168]]
EPA's Final Determination ``found that that even with the lowest prices
projected in AEO 2016 of close to $2, the `lifetime fuel savings
significantly outweigh the increased lifetime costs' of the GHG
standards.'' \2932\ IPI further argued that ``[i]n ignoring the [FD]
analysis, the Proposed Rule has failed to provide a `reasoned
explanation' for dismissing the `facts and circumstances that underlay'
the original rule, rendering its analysis arbitrary and capricious.''
\2933\ IPI also argued that NHTSA had not adequately explained its
``shift since 2012 in its interpretation and application of the need to
conserve energy factor,'' stating that ``[a]ctual fuel savings, and the
associated benefits to consumers, the environment, and society, were at
the heart of NHTSA's analysis of the need to conserve energy factor
back in 2012. Now the agency ignores those conclusions from 2012 and
relies on mistaken and inconsistent interpretations of petroleum import
projections and the urgency of climate change to justify ignoring this
statutory factor and giving primacy instead to economic practicability
and safety effects. The failure to explain this shift in approach is
arbitrary.'' \2934\
---------------------------------------------------------------------------
\2928\ ACEEE, NHTSA-2018-0067-12122, main comments, at 1.
\2929\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 17.
\2930\ Id., at 7, 8.
\2931\ IPI, NHTSA-2018-0067-12213, Appendix, at 31.
\2932\ Id., at 32.
\2933\ Id.
\2934\ Id., at 6.
---------------------------------------------------------------------------
UCS argued that the need of the United States to conserve energy is
``the most important of the four required factors'' according to CBD v.
NHTSA, and claimed that ``NHTSA has manipulated the evaluation of the
factors to produce a result that supports the preferred option in the
NPRM.'' \2935\ The States and Cities commenters argued that it was
``[c]ynical. . .'' for NHTSA to justify the proposal on the basis that
``the oil intensity of U.S. GDP has continued to decline'' in part as a
result of increasingly stringent CAFE standards, and on the basis that
``[m]anufacturers have responded to fuel economy standards and to
consumer demand over the last decade to offer a wide array of fuel-
efficient vehicles in different segments and with a wide array of
features.'' \2936\
---------------------------------------------------------------------------
\2935\ UCS, NHTSA-2018-0067-12039, at 3, 7.
\2936\ States and Cities, NHTSA-2018-0067-11735, Detailed
Comments, at 64-65.
---------------------------------------------------------------------------
CARB and CBD et al. argued that if NHTSA's analysis indicates that
automakers will voluntarily exceed the standards, then the standards
cannot be maximum feasible.\2937\ Robertson commented relatedly that
standards should not be set below augural levels because ``Much higher
fuel economy and reduced emissions have been achieved by several lower
priced makes and models using hybrid technology.'' \2938\ Blue Planet
Foundation stated that the augural standards are feasible because
automakers have already invested in technologies, and electrification
is projected to continue to grow cheaper over time, so that ``even the
up-front cost of an EV will begin to reach parity with gas-powered cars
by 2024.'' \2939\ ACEEE also cited the voluntary overcompliance in the
NPRM analysis as evidence that there could not be diminishing returns
from higher fuel efficiency standards, because ``the list of [cost-
effective] technology [must] continually regenerate itself'' if
manufacturers would continue applying it in the absence of future
standards. Moreover, ACEEE argued, past analyses had always found
plenty of available cost-effective technologies, and automakers would
find a way to apply them.\2940\
---------------------------------------------------------------------------
\2937\ CARB, NHTSA-2018-0067-11873, Detailed Comments, at 84;
CBD et al., NHTSA-2018-0067-12057, at 2.
\2938\ Robertson, EPA-HQ-OAR-2018-0283-0787, at 3.
\2939\ Blue Planet Foundation, EPA-HQ-OAR-2018-0283-4207, at 1-
2.
\2940\ ACEEE, NHTSA-2018-0067-12122, main comments, at 9.
---------------------------------------------------------------------------
c) How is NHTSA Balancing the Factors to Determine the Maximum Feasible
Final CAFE Standards?
EPCA/EISA grants the Secretary (by delegation, NHTSA) discretion in
how to balance the relevant statutory factors, while bearing in mind
EPCA's overarching purpose of energy conservation. Many commenters
cited the Ninth Circuit's language in CBD v. NHTSA that ``the
overarching purpose of EPCA is energy conservation,'' \2941\ and the
D.C. Circuit's language in CAS v. NHTSA that ``[i]t is axiomatic that
Congress intended energy conservation to be a long term effort that
would continue through temporary improvements in energy availability.''
\2942\ NHTSA has considered those comments and those court decisions
carefully as it made the decision set forth in the final rule. Based on
the information before the agencies and considering carefully the
comments received, NHTSA has determined that the preferred alternative
identified in the proposal--amending the MY 2021 standards to match MY
2020, and holding those standards flat through MY 2026--does not
represent the maximum feasible standards, and that the maximum feasible
standards for MYs 2021-2026 passenger cars and light trucks increase in
stringency by 1.5 percent per year from the MY 2020 standards. The
following discussion walks through NHTSA's evaluation and balancing of
the relevant factors in light of the information before it.
---------------------------------------------------------------------------
\2941\ CBD, 508 F.3d 508, 537 (9th Cir. 2007), opinion vacated
and superseded on denial of reh'g, 538 F.3d 1172 (9th Cir. 2008).
\2942\ CAS, 793 F.2d 1322, 1340 (D.C. Cir. 1986).
---------------------------------------------------------------------------
(1) Need of the U.S. to Conserve Energy
NHTSA agrees with commenters that energy conservation remains
important, and that changed conditions, even significantly changed
conditions, do not obviate NHTSA's obligation to set maximum feasible
CAFE standards as directed by Congress. Many commenters disagreed
strongly with NHTSA's suggestion in the NPRM that increased U.S.
petroleum production, and the U.S.'s likely imminent status as a net
petroleum exporter, decreased the need of the U.S. to conserve energy.
NHTSA agrees that there is still a need to conserve energy, and oil in
particular. Like an insurance policy or a savings account, continuing
to move the needle forward on CAFE helps position Americans better to
weather certain types of possible future uncertainty. NHTSA believes
that it is reasonable to be somewhat conservative about this risk, and
thus to set CAFE standards that increase in stringency year over year
through MY 2026.
That said, NHTSA believes that there are limits to how much
uncertainty the CAFE program can mitigate--continuing to make progress
is important, but it is also important to be transparent and realistic
about what is being accomplished, even if NHTSA were able to set
standards beyond levels that NHTSA considers maximum feasible. NHTSA
also continues to believe that structural changes in global oil markets
over the last 10 years, driven in part by changes in demand both in the
U.S. and abroad, and in part by the significant growth in U.S.
petroleum production, have led to a fundamental shift in the dynamics
of global oil prices, which has in turn improved U.S. (and possibly,
global) energy security. NHTSA believes that this shift is important to
consider as NHTSA weighs the need of the Nation to conserve energy.
NHTSA acknowledges that price shocks can still happen. The large
scale attack on Saudi Arabia's Abqaiq processing facility--the world's
largest crude oil processing and stabilization plant--on September 14,
2019 caused ``the largest single-day [crude oil] price increase in the
past decade,'' of between $7 and $8, according to EIA.\2943\ The Abqaiq
facility has a capacity to process
[[Page 25169]]
7 million barrels per day, or about 7 percent of global crude oil
production capacity. By September 17, however, also according to EIA,
---------------------------------------------------------------------------
\2943\ https://www.eia.gov/todayinenergy/detail.php?id=41413.
Saudi Aramco reported that Abqaiq was producing 2 million
barrels per day, and they expected its entire output capacity to be
fully restored by the end of September. In addition, Saudi Aramco
stated that crude oil exports to customers will continue by drawing
on existing inventories and offering additional crude oil production
from other fields. Tanker loading estimates from third-party data
sources indicate that loadings at two Saudi Arabian export
facilities were restored to the pre-attack levels. Likely driven by
news of the expected return of the lost production capacity, both
Brent and WTI crude oil prices fell on Tuesday, September 17.\2944\
---------------------------------------------------------------------------
\2944\ Id.
Thus, the largest single-day oil price increase in the past decade
was largely resolved within a week, and assuming very roughly that
average crude oil prices were $70/barrel in September 2019 (slightly
higher than actual), an increase of $7/barrel would represent a 10
percent increase as a result of the Abqaiq attack. Contrast this with
the 1973 Arab oil embargo, which lasted for months and raised prices
350 percent.\2945\ Saudi Arabia could have benefited, revenue-wise,
from higher prices following the Abqaiq attack, but instead moved
rapidly to restore production and tap reserves to control the risk of
resulting price increases, likely recognizing that long-term sustained
price increases would reduce their ability to control global supply
(and thus prices, and thus their own revenues) by relying on their
lower cost of production.\2946\ Even if the NPRM discussion was perhaps
overconfident about the ability of U.S. shale producers to act as
``swing'' supply, as some commenters suggested, it seems clear from
events that the existence of U.S. production has a stabilizing effect
on global oil prices. This has played out in important ways in the
first quarter of 2020, with the dissolution of the ``OPEC+'' coalition
as Russia and Saudi Arabia compete for market share in response to U.S.
shale production and also in the wake of global demand downturn.\2947\
---------------------------------------------------------------------------
\2945\ See Jeanne Whalen, ``Saudi Arabia's oil troubles don't
rattle the U.S. as they used to,'' Washington Post, September 19,
2019, available at https://www.washingtonpost.com/business/2019/09/19/saudi-arabias-oil-troubles-dont-rattle-us-like-they-used/.
\2946\ See, e.g., ``Dynamic Delivery: America's Evolving Oil and
Natural Gas Transportation Infrastructure,'' National Petroleum
Council (2019) at 18, available at: https://dynamicdelivery.npc.org/downloads.php. See also ``Oil prices plunge as Trump speech eases
Iran fears,'' CNN, available at https://www.cnn.com/2020/01/07/business/oil-prices-iran-attack-iraq/index.html.
\2947\ See, e.g., EIA, ``This Week in Petroleum--OPEC shift to
maintain market share will result in global inventory increases and
lower prices,'' March 11, 2020, https://www.eia.gov/petroleum/weekly/; DOE, ``DOE Responds to Recent Oil Market Activity,'' March
9, 2020, https://www.energy.gov/articles/doe-responds-recent-oil-market-activity; Reid Standish, Keith Johnson, ``No End in Sight to
the Oil Price War Between Russia and Saudi Arabia,'' March 14, 2020,
https://foreignpolicy.com/2020/03/14/oil-price-war-russia-saudi-arabia-no-end-production/; Alex Ward, ``The Saudi Arabia-Russia oil
war, explained,'' March 9, 2020, https://www.vox.com/2020/3/9/21171406/coronavirus-saudi-arabia-russia-oil-war-explained.
---------------------------------------------------------------------------
Even though the effect of significant supply disruptions appears
much lower than was the case several years ago, the analysis for this
final rule (like the NPRM analysis) does, in fact, explicitly account
for the possible occurrence of price shocks. The cost penalty used in
the analysis to represent the consequences of those shocks attempts to
quantify the negative impact on U.S. GDP created by abrupt, short-term
increases in the world oil price. The values used in the NPRM were
based on arguably outdated work, and commenters cited more recent
studies of relevance in their comments on the NPRM--one of which formed
the basis for the estimates in today's analysis. The final rule
estimate of this cost are based on a recent study which states that
``[i]n recent years, the United States has become much more self-
reliant in producing oil, and a newer economics literature suggests
that oil demand may be more elastic and U.S. GDP may be less sensitive
to world oil price shocks than was previously estimated. These
developments suggest somewhat lower security costs may be associated
with U.S. oil consumption.'' \2948\ These more recent studies concede
that the fact that ``the world has not seen a major oil supply
disruption since 2003,'' and that therefore ``we have no reliable
method to quantify the effects of these disruptions,'' \2949\ but even
the range of uncertainty suggests that the risk has decreased relative
to prior estimates. The price shock cost estimate employed in the NPRM
was at least twice as large as the upper bound of the range in Brown's
new estimates, and consistently close to the upper bound of the range
of his more conservative estimates. The approach taken today, which
relies on median estimates in Brown's study, implies that risk is more
properly estimated here than in the NPRM.
---------------------------------------------------------------------------
\2948\ Brown, Stephen, ``New estimates of the security costs of
U.S. oil consumption,'' Energy Policy 113 (2018) 171-192, at 171.
Cited in Securing America's Energy Future, NHTSA-2018-0067-12172, at
29.
\2949\ Brown, at 181.
---------------------------------------------------------------------------
Commenters (Bordoff, SAFE, CARB, IPI) argued that increased U.S.
petroleum production, which improves the stability of the global supply
and reduces the probability of supply interruptions, does not reduce
U.S. exposure to petroleum price shocks, which are still determined by
the dynamics of the global market. By reducing the probability of
supply disruptions in the global market, the U.S. does reduce its
vulnerability to price shocks. However, to the extent that the
vulnerability to price shocks is a function of exposure, commenters are
correct that looming petroleum independence does not entirely insulate
the U.S. economy from the consequences of global oil price shocks. Some
commenters further argued that the proposed standard would leave the
U.S. more exposed to oil price shocks, which would harm consumers.
Basic mathematics means that a less efficient on-road fleet necessarily
would spend more on fuel than a more efficient on-road fleet in the
event of a sudden, unexpected, and dramatic increase in oil price. The
suggestion in these comments, however, is that finalizing the augural
standards would sufficiently insulate U.S. consumers from harm during
such an event, while finalizing any other regulatory alternative would
not. NHTSA disagrees that finalizing the augural standards, as compared
to the standards we are finalizing, would make a meaningful difference
in this case.
A continuous, but slow, price increase over several years is
fundamentally different from the kinds of acute price shocks over which
commenters have expressed understandable concern. Long-term price
increases signal consumers to make investments in fuel economy, in both
the new and used vehicle markets, and to diversify the vehicles in
their household fleets. In a side analysis using outputs from the CAFE
Model, the agencies examined the consequences of a gasoline price spike
in 2030--increasing the price from $3.40/gallon to $6/gallon for eight
months, then reverting back to $3.40/gallon.\2950\ By choosing a year
so far in the future, the agencies consider a larger gap in fleet fuel
efficiency than is attributable to this action. If the agencies
increase stringency again after MY 2026, the efficiency gap between the
on-road fleet in the final standards and baseline would be smaller than
simulated here. This side analysis showed that even a nearly doubling
of the fuel price, sustained for more than half a year, would result in
less than 1 percent savings in fuel expenditures for that
[[Page 25170]]
year under the final standards (relative to the proposal), compared to
about 5 percent reduction in expenditures under the augural standards.
This demonstrates that even though finalizing the augural standards
would mitigate American drivers' increase in fuel expenditures by more
than the standards the agencies are finalizing today, it would only do
so by a few percent. This is important to understanding concerns about
differences in the amount of fuel saved under today's final standards
versus if the augural standards were finalized, as will be discussed
more below. And as also discussed below, NHTSA believes the augural
standards are beyond maximum feasible at this time.
---------------------------------------------------------------------------
\2950\ Docketed in NHTSA-2018-0067.
---------------------------------------------------------------------------
Some commenters raised the possibility that the U.S. might ban
fracking at some point in the future, and suggested that therefore the
need of the U.S. to conserve energy could not be assumed away. NHTSA
acknowledges that the future is uncertain. Without the supply of U.S.
oil in the global market, NHTSA agrees that it is foreseeable that
conditions could revert somewhat to how global oil market conditions
were before the ramp-up in U.S. supply--i.e., that the global market as
a whole could be somewhat less stable and thus fuel prices could be
somewhat more prone to change unexpectedly and for longer periods.
Pulling out of the market on the supply side means that the agencies
would lose the ability to influence the market on that side.
Presumably, part of the policy objective of banning fracking would be
to accelerate a transition to a post-oil transportation system. In that
scenario, presumably decision-makers would consider higher fuel prices
to be an acceptable tradeoff for less driving and lower emissions. That
said, the availability of shale oil resources does exist today, and is
not realistically in question. And, even if the future availability of
that capacity was realistically doubtful, any increase in fuel economy
above current levels, like the final rule will require, will help
somewhat to mitigate the economic pain to drivers of that event were it
to occur, as shown above.\2951\ To the extent that current events cause
pauses or consolidation in the shale industry's development, while that
may lead to transitory difficulty for the shale industry, the resources
will continue to exist, and U.S. shale will continue to be able to act
as a lever to keep global prices from rising very high for very long.
---------------------------------------------------------------------------
\2951\ See also Letter from Alliance for Automotive Innovation,
NADA, and MEMA to Congress, Mar. 23, 2020, available at https://www.autosinnovate.org/wp-content/uploads/2020/03/COVID-19-Letter-to-Congress-NADA-MEMA-AAI-March-23.pdf.
---------------------------------------------------------------------------
As noted above, Securing America's Energy Future commented that
``[a]lthough the nation is undoubtedly more energy secure than it was
before the start of the U.S. shale oil revolution ten years ago,''
\2952\ ``[u]ntil the U.S. transportation sector is no longer beholden
to oil, the country will be vulnerable to oil price volatility.
Improving the fuel efficiency of the U.S. vehicle fleet is a valuable
insurance policy against this volatility.'' \2953\ (Emphasis added.)
NHTSA agrees fully with this comment. Energy security concerns were the
driving force behind the creation of the CAFE program, as discussed in
the NPRM. U.S. energy security has improved, but the only way to
resolve petroleum-related energy security concerns entirely would be
for the U.S. vehicle fleet to stop using oil. And doing so would not
avoid energy-related concerns entirely, but rather shift them away from
petroleum (and the Middle East) and toward battery-related security
(and lithium-, nickel-, cobalt-, and other metals-producing
countries).\2954\
---------------------------------------------------------------------------
\2952\ Securing America's Energy Future, NHTSA-2018-0067-12172,
at 17.
\2953\ Id., at 7, 8.
\2954\ While progress is being made on developing and improving
domestic sources for many of the minerals necessary for battery
development, the U.S. is still heavily dependent on imports of both
raw materials and batteries. Regarding minerals production and
import dependence, see Schulz, K.J., DeYoung, J.H., Jr., Seal, R.R.,
II, and Bradley, DC, eds., Critical mineral resources of the United
States--Economic and environmental geology and prospects for future
supply: U.S. Geological Survey Professional Paper 1802 (see
particularly Chapter K, p. K1-K21 on lithium), available at https://www.commerce.gov/sites/default/files/2020-01/Critical_Minerals_Strategy_Final.pdf and https://pubs.usgs.gov/pp/1802/k/pp1802k.pdf. Regarding vehicle battery supply chains, see
Coffin, D., and J. Horowitz, ``The Supply Chain for Electric Vehicle
Batteries,'' Journal of International Commerce and Economics,
December 2018, available at https://www.usitc.gov/publications/332/journals/the_supply_chain_for_electric_vehicle_batteries.pdf.
---------------------------------------------------------------------------
Our relationship to the global energy market has changed
significantly since the CAFE program was created, with most of this
change occurring over the last decade. The United States has become
energy independent, and is currently a net exporter of petroleum
products. Rising world oil prices no longer only mean a financial
burden on U.S. drivers and a wealth transfer to foreign nations. While
rising prices continue to affect U.S. motorists, we have taken steps to
insulate our transportation system from exogenous price shocks. CAFE
standards (and, recently, CO2 standards) have increased the
efficiency of new vehicles for more than a decade, and these
increasingly efficient vehicles are still working their way into the
on-road fleet as older models are retired. Accompanying any increase in
the global oil price is an increase in revenue to the U.S. oil
industry. To the extent that motorists are spending more on oil
everywhere, the dollars spent on domestically produced petroleum
products stay within the U.S. and additional revenue from foreign
buyers flows into our domestic energy industry. To the extent that the
U.S. transportation system is able to further reduce its dependence on
petroleum in a cost-effective manner, it is sensible to do so. But in
the current environment, in which motorized transportation is
increasingly energy efficient and U.S. energy producers are not only
supplying our demand but exporting petroleum products to other nations,
the nationwide benefits of reducing petroleum consumption are
substantially diminished.
There is also the opposite concern to bear in mind--that energy
security is not just about oil becoming more expensive, but also about
other changes in oil prices. Major fluctuations in either direction, as
well as oil price collapse, can potentially have seriously
destabilizing geopolitical effects. Many major oil producing countries
(some of whom are allies) rely heavily on oil revenues for public
revenue, and sustained losses in public revenue in certain countries
and regions can foreseeably create new energy-related security risks,
not only for the U.S. As the world works toward transitioning away from
oil for transportation, keeping prices reasonably stable may best help
that transition remain peaceful and steady. In short, energy security
can cut both ways, and the current estimates of price shock that we
model inherently do not account for the longer-term stabilizing effect
of steady global oil consumption (of which the U.S. is a part) on
global security. Steady trends in consumption can facilitate steady
changes in production, which can facilitate a steady security
situation.
NHTSA does not interpret EPCA/EISA to mean that Congress expected
the CAFE program to take the U.S. auto fleet off of oil entirely--
indeed, EISA renders doing so impossible because it amended EPCA to
prohibit NHTSA from considering the fuel economy of dedicated
alternative fuel vehicles, including electric vehicles, when setting
maximum feasible standards. This means that standards cannot be set
that assume increased usage of full electrification for compliance.
Reading that prohibition together with the obligation to set maximum
feasible standards by considering (which is hard
[[Page 25171]]
to do without balancing) factors like economic practicability with the
need of the U.S. to conserve energy, NHTSA believes that Congress
intended CAFE to try to mitigate the risk of gas lines, but not to
shift the fleet entirely off of oil. Moreover, the EISA-added
requirement that standards ``increase ratably'' for MYs through 2020
ceases to apply beginning in MY 2021. While NHTSA unquestionably has
discretion to determine that standards should continue to increase
post-MY 2020, NHTSA does not interpret EPCA/EISA as requiring that they
do, as long as they are maximum feasible. Several commenters suggested
that standards that do not continue to increase, by definition, cannot
be maximum feasible, but NHTSA believes that this interpretation does
not account for the clear requirement that maximum feasible standards
be determined with reference to the four statutory factors. The statute
does not preclude an interpretation that non-increasing standards could
be maximum feasible, depending on the facts before the agency. Neither
does the statute preclude an interpretation that amending standards
downward can be maximum feasible, as has occurred in the past in
response to changes in consumer demand.\2955\
---------------------------------------------------------------------------
\2955\ See, Center for Auto Safety v. NHTSA (CAS), 793 F.2d 1322
(D.C. Cir. 1986).
---------------------------------------------------------------------------
Nevertheless, for purposes of this final rule, NHTSA does believe
that standards that increase in stringency are maximum feasible; the
question remains by how much those standards should increase. While
NHTSA agrees that CAFE standards must conserve energy, the improvement
in energy security discussed above is entirely relevant to how much
energy should be conserved. If the marginal improvement in energy
security of increasing CAFE stringency from one regulatory alternative
to another is very small, as it appears to be based on the above
discussion, then other aspects of the need of the U.S. to conserve
energy must be considered next to see what effect they have.
Consumer costs, as discussed above, is another aspect of the need
of the U.S. to conserve energy. The final rule analysis estimates that
all alternatives besides the baseline/augural standards would result in
higher fuel costs for consumers than the baseline/augural standards
would result in, as follows:
[GRAPHIC] [TIFF OMITTED] TR30AP20.735
A number of commenters stated that the 2012 rulemaking had relied
on fuel savings as part of its justification, and argued that the NPRM
had not adequately grappled with the fact that the proposal would have
cost consumers more in fuel expenditures than if NHTSA finalized the
augural standards. In fact, NHTSA explained in the NPRM that while fuel
costs would be higher, NHTSA believed that the higher upfront (and
ongoing, if financed) costs of new vehicles and associated taxes and
registration fees--as well as the opportunity cost associated with
those upfront costs--would outweigh, for many consumers, the additional
fuel costs that would be incurred if standards were less stringent than
augural. That continues to be the case under the final rule analysis,
as discussed below. In addition, Section VI.D. discusses how past
rulemaking analyses assumed that consumers were `myopic' and/or did not
have adequate information about the benefits of fuel savings, which led
them to choose to purchase less efficient vehicles than they otherwise
would if they better understood the costs or savings they would accrue.
As Section VI.D. explains, the agencies are less certain today that
consumers improperly value fuel savings. Vehicle buyers today have more
information about fuel costs than ever before, including right on the
window sticker when considering a new vehicle purchase, and it is
ultimately a private choice whether consumers prefer improvements in
other vehicle attributes over additional fuel economy. When fuel costs
are expected to rise manageably over time, it may be that consumers are
comfortable choosing to absorb an additional $1,375 over the vehicle's
lifetime, the estimated difference in lifetime expenditures between the
proposal and if NHTSA was choosing to finalize the augural standards,
and are even more comfortable choosing to absorb an additional $1,125,
the estimated difference in lifetime expenditures between the final
standards and what
[[Page 25172]]
the augural standards would have required. If fuel prices rise less
than anticipated, as they have done since the 2012 final rule, or even
decrease over time, buyers face an even smaller tradeoff between
foregone fuel savings and the value of improvements in other aspects of
new cars.
Consumer expenditures on fuel are important to understanding the
benefits (and net benefits) of CAFE and CO2 standards. Every
analysis of CAFE/CO2 standards relies on hundreds of
assumptions, and estimates of costs and benefits developed as part of
those analyses, by their very nature, depend on those assumptions.
Specifically, the net benefits associated with each alternative result
from the assumptions used and the relationships between vehicle
production, ownership, and usage in which the assumptions interact. Put
more simply, inputs affect outputs. As discussed in the section above
on economic practicability, net benefits may be a consideration in the
determination of maximum feasible standards, among the many other
things the agency considers. While some commenters have asserted that
the analysis for this rulemaking has ``put a thumb on the scale by
undervaluing the benefits and overvaluing the costs of more stringent
standards,'' \2956\ this final rule has identified a number of critical
assumptions in the 2012 final rule that were problematic in the other
direction (i.e., undervaluing the costs and overvaluing the benefits),
for a variety of reasons. For example, the projected fuel prices in the
2012 analysis inflated the value of fuel savings relative to what has
actually occurred. That assumption about how fuel prices were projected
to rise over time was solidly grounded at the time, but is no longer
so, and continuing to use it would not be reasonable, even if that
means that the benefits of all of the regulatory alternatives decrease
as compared to what the 2012 analysis showed. Lower oil prices mean
that fuel savings benefits for consumers are lower under any CAFE
standards, whether the augural standards or the standard being
finalized today--consumers may yet spend less on fuel under more
stringent standards, but how much less matters.
---------------------------------------------------------------------------
\2956\ See CBD v. NHTSA, 538 F.3d 1172, 1189 (9th Cir. 2008).
---------------------------------------------------------------------------
Additionally, the assumption in 2012 that no market exists for fuel
economy improvements at any fuel price or technology cost artificially
inflated the value of fuel savings attributable to the standards in
each regulatory alternative. The combination of assumptions and
relationships (the examples above, and others) in the 2012 final rule
produced estimates of net benefits that continued to increase with
stringency from 1 percent per year through 6 percent per year.\2957\
Under some alternatives, benefits actually would have appeared to be
infinite, growing faster than the discount rate, if the analysis had
been extended far enough into the future. No market works this way, and
there is no reasonable set of assumptions under which costs could never
exceed benefits no matter how much technology was deployed or how much
stringency was required. Rather than demonstrating meaningfully that
more stringent standards are always more beneficial to society, the
result from the 2012 analysis suggests that that analysis was
critically flawed. That said, while the 2012 analysis appeared to show
that more technology, at a faster pace, is always preferable from the
perspective of net benefits, the agencies ultimately relied on other
features of the analysis and considerations of impacts in choosing a
preferred alternative. While today's analysis produces an inflection
point at a 3 percent discount rate--a level of stringency where further
increases reduce net benefits as the tradeoff between regulatory costs
and resulting net benefits tips the other way \2958\--the agencies
similarly rely on considerations beyond net benefits in choosing the
preferred alternative.\2959\
---------------------------------------------------------------------------
\2957\ The 7 percent per year alternative happened to be
indistinguishable from the 6 percent alternative in that analysis.
\2958\ See Table VII-95.
\2959\ See CBD v. NHTSA, 538 F.3d 1172, 1188 (9th Cir. 2008).
---------------------------------------------------------------------------
NHTSA also agrees with many commenters that environmental (both
climate and air quality) concerns are relevant to the need of the U.S.
to conserve oil, as explained above. As the Supreme Court stated in
Massachusetts v. EPA, ``[a] reduction in domestic emissions would slow
the pace of global emissions increases,'' \2960\ and there is no
question that CAFE standards directly affect CO2 emissions.
Besides providing information on differences between the regulatory
alternatives in terms of million metric tons of CO2 emitted,
the NPRM also provided a chart illustrating the difference between the
estimated atmospheric CO2 concentration (789.76 ppm) in 2100
under the proposal as compared to the estimated level under the augural
standards (789.11 ppm) in a scenario where no CO2 emissions
reduction measures are implemented throughout the planet.\2961\ The
NPRM noted that this translated to 3/1000ths of a degree Celsius
increase in global average temperatures by 2100, relative to the
augural standards. Many commenters strongly objected to the framing of
these findings, as discussed above in the section on the environmental
implications of the need of the U.S. to conserve energy. Changing the
framing does not change the agency's findings.\2962\ For this final
rule, the Preferred Alternative would result in 922.5 million metric
tons of CO2 more than the estimated emissions if the augural
standards were to be finalized (for MY 2017-MY 2029 vehicles between
calendar years 2017 and 2070), which is 160.2 million fewer tons than
if the proposed Preferred Alternative were to be finalized. It is
reasonable to consider these raw million-metric-ton estimates in terms
of their effects, namely, on estimated temperature change and sea level
rise, which are the primary climate effects referred to and estimated.
The FEIS accompanying today's rule estimates that, by 2100, global mean
surface temperature will increase by 3.487 degrees (Celsius) under
either the proposed or final standards, versus 3.484 degrees under the
augural standards. The FEIS shows corresponding sea level rise in 2011
reaching 76.34 cm under the final standards, 76.35 cm under the
proposed standards, and 76.28 cm under the augural standards. This is
accounted for in economic terms (i.e., translated from fractions of a
degree temperature rise and from millimeters of sea level rise, among
other things, into dollar-based effects) in the measure of the social
cost of carbon, described in Section VI.D.1.b)(13).
---------------------------------------------------------------------------
\2960\ Mass. v. EPA, 549 U.S. at 526.
\2961\ 83 FR at 42996-97 (Aug. 24, 2018).
\2962\ In fact, NHTSA's analysis in Section 8.6.4.2 of the FEIS
illustrates that the differences between alternatives are similar in
reference to other GCAM scenarios. Regardless of whether there will
be widespread global efforts to mitigate climate change, the impacts
of this action are roughly the same.
---------------------------------------------------------------------------
NHTSA is mindful of the language in Massachusetts v. EPA that
``[a]gencies . . . do not generally resolve massive problems in one
fell regulatory swoop,'' \2963\ and acknowledges the concerns of many
commenters that standards less stringent than augural may result in
higher CO2 emissions. In response, it is important to
remember that even under the proposal, sales of new vehicles would,
over time, have continued to improve the fuel economy and reduce the
CO2 emissions of the on-road fleet through fleet turnover
effects, as discussed in Section IV. Under the final rule, those rates
of improvement will likely be faster than they would have been if NHTSA
were finalizing the
[[Page 25173]]
proposal. Emissions are still being reduced under the final rule, and
the on-road fleet will be less energy and carbon intensive than it is
today. NHTSA is taking the impacts of CO2 emissions into
account, while also considering the other statutory factors in its
balancing.
---------------------------------------------------------------------------
\2963\ Mass. v. EPA, 549 U.S. at 524.
---------------------------------------------------------------------------
It is also important to note that the science of climate change and
the models used to assess effects on climate variables (and other
effects discussed in Section VII.A.4.b, and in the DEIS/FEIS) are
subject to various types and degrees of uncertainty. In light of this,
NHTSA also conducted climate sensitivity analyses in the FEIS.\2964\ In
these analyses, NHTSA considered a range of climate sensitivities (1.5
[deg]C, 2.0 [deg]C, 2.5 [deg]C, 3.0 [deg]C, 4.5 [deg]C, and 6.0 [deg]C)
for a doubling of CO2 compared to preindustrial atmospheric
concentrations (278 ppm CO2). Even under the least stringent
alternative considered (the proposal) and assuming the highest level of
climate sensitivity (6.0 [deg]C), the global mean surface temperature
increase in 2100 was 0.006 [deg]C higher than under the augural
standards. Thus, accounting for some of this uncertainty, impacts on
global mean surface temperature resulting from this action remain very
small.
---------------------------------------------------------------------------
\2964\ See Sections 5.4.2.3 and 8.6.4.2 of the FEIS.
---------------------------------------------------------------------------
NHTSA received many comments about the costs of delaying
CO2 emissions reductions and the potential of crossing
climate tipping points and triggering abrupt climate change. Many of
these costs and risks are factored in to the social cost of carbon, and
are therefore considered as part of the agency's cost-benefit analysis.
And many of these costs and risks cannot be quantified at all: The
current state of science does not allow for quantifying how increased
emissions from a specific policy or action might affect the probability
and timing of abrupt climate change. However, NHTSA does recognize that
while these costs cannot be quantified, they do exist and must also be
taken into account. Ultimately, the costs of delaying CO2
emissions reductions (both the ones that can be accounted for
quantitatively and those that can only be considered qualitatively)
must also be balanced against the costs associated with more stringent
alternatives. Some of the costs associated with more stringent
alternatives are direct, such as the additional costs passed on to
consumers for technology that improves fuel economy. Other costs are
indirect, such as environmental costs associated with more stringent
fuel economy standards. For example, the increased electrification of
motor vehicles can result in localized impacts associated with the
production and recycling of lithium-ion batteries. Similarly, the
increased reliance on material substitution for vehicle mass reduction
could result in various environmental impacts associated with
manufacture and recycling. Certainly, the benefits of these
technologies in reducing carbon emissions outweighs the other life-
cycle environmental impacts, but that does not mean NHTSA can just
ignore those impacts, either.
Many commenters claimed that NHTSA ignored the effects of climate
change or determined they were inevitable, not urgent enough to act
upon, or not worth the effort to address at all. NHTSA makes none of
those determinations here. On the contrary, NHTSA has considered the
material on this subject in the administrative record and the plethora
of public comments we received on the topic. The agency recognizes what
is at stake, but we also recognize that NHTSA is not charged by
Congress to single-mindedly address carbon emissions at the expense of
all other considerations. The question before NHTSA is not whether to
conserve energy (and thereby reduce carbon emissions, which drive
climate change) but by how much each year. Taking climate change into
account elevates the importance of the ``need of the United States to
conserve energy'' criterion in NHTSA's balancing. However, in light of
the limits in what the agency can achieve, the potential offsetting
impacts to the environment, and the statutory requirement to consider
other factors, the impacts of carbon emissions alone cannot drive the
outcome of NHTSA's decision-making.
NHTSA also recognizes the potential impacts of this rulemaking on
air quality. To be clear, this final rule does not directly involve the
regulation of pollutants such as carbon monoxide, smog-forming
pollutants (nitrogen oxides and unburned hydrocarbons), or ``air
toxics'' (e.g., formaldehyde, acetaldehyde, benzene). Nevertheless,
NHTSA recognizes that this rule is expected to impact such emissions
indirectly (by reducing travel demand and accelerating fleet turnover
to newer and cleaner vehicles on one hand while, on the other,
increasing activity at refineries and in the fuel distribution system).
Based on a review of Section VII.A.4.c. above and the FEIS, NHTSA
believes these impacts are much smaller than impacts on fuel use and
CO2 emissions, and therefore factor in less to the need of
the U.S. to conserve energy.\2965\
---------------------------------------------------------------------------
\2965\ For an explanation of how NHTSA considers environmental
impacts and the differences between the preamble and FEIS analyses,
see Section VII.A.4.c.1 above.
---------------------------------------------------------------------------
For criteria pollutants, NHTSA estimates that emissions over the
lifetimes of vehicles through MY 2029 under the alternatives will not
change significantly. Tailpipe emissions of most pollutants will
generally decrease, while upstream emissions will generally increase.
Overall emissions under the action alternatives for most pollutants
will increase over time. Changes are not uniform year-to-year, however,
reflecting the complex interaction of the amount of highway travel, the
distribution of that travel among different vehicles, upstream
processes, etc. Generally, tailpipe air toxic emissions decrease while
upstream air toxic emissions increase. Over the long term, however, the
upstream emissions increase further while the decreases in tailpipe
emissions become less pronounced. Overall, NHTSA anticipates that air
toxic emission will increase over time under the action alternatives.
Most alternatives result in cumulative increases in adverse health
impacts associated with total upstream and tailpipe pollutant
emissions. Although some alternatives would have resulted in decreases,
the differences among alternatives across the lifetime of vehicles
through MY 2029 are not large.
NHTSA also considered the various impacts reported qualitatively in
the FEIS and described briefly above in Section VIII.B.3. Although the
agency cannot compare the impacts of the alternatives quantitatively
(except to the degree that they are otherwise covered by the agency's
monetary cost-benefit analysis, such as through the social cost of
carbon), NHTSA recognizes that such impacts would generally increase
under all the action alternatives compared to the augural standards. In
Chapter 8 of the FEIS, for example, NHTSA provides a qualitative
discussion of the long-term impacts of climate change on key natural
and human resources. While these impacts would be expected to increase
under the action alternatives, the change is expected to be very small.
In contrast, the FEIS also discusses some environmental impacts that
would decrease with the lower stringencies considered in this
rulemaking. For example, in Chapter 6 of the FEIS, NHTSA provides a
literature review of potential lifecycle impacts as a result of
manufacturer use of various materials and technologies to meet the
standards. NHTSA can account for the benefits to
[[Page 25174]]
tailpipe emissions of these technologies as part of its evaluation of
technology effectiveness. However, as discussed in the FEIS, accounting
for the upstream emissions associated with the processes used in the
manufacture of these technologies can be complicated. Because the
adoption of these materials and technologies would vary across
alternatives, and each has varying upstream impacts, the agency cannot
provide meaningful comparisons across alternatives. Still, any benefit
to tailpipe CO2, criteria pollutant, or air toxic emissions
of more stringent alternatives would be offset by the increased
upstream impacts reported in that section.\2966\
---------------------------------------------------------------------------
\2966\ In most cases, tailpipe emissions benefits offset
upstream environmental impacts associated with materials and
technologies NHTSA considered in its analysis. However, in some
cases, results may not align with conventional wisdom. For example,
while EVs can offer significant life-cycle GHG emissions savings
over conventional vehicles, this is highly dependent on the time and
location of charging. In some regions, life-cycle impacts are
similar for EVs and conventional vehicles.
---------------------------------------------------------------------------
In total, environmental impacts factor into the need of the U.S. to
conserve energy and potentially elevate that criterion, but those
impacts cannot be considered in isolation. While some impacts are more
significant than others, NHTSA must consider how much weight to place
on this factor as well as the relative weight of other factors.
Thus, even if the agency no longer interprets the need of the U.S.
to conserve energy as necessarily boundless as it once did, as it
explained in the NPRM and again in the discussion above, NHTSA
continues to believe that the factor functions in the overall balancing
to push toward increases in stringency, and notes that any increase in
stringency over the last binding standards--not in question at this
point, the standards for MY 2020--does conserve energy and reduce
negative environmental impacts. In fact, fleet turnover over time means
that less energy is being consumed by the fleet over time even if
standards did not increase year over year. Even if new vehicles are not
all as efficient as would have been required under more stringent
standards, they are still more efficient on average than the older
vehicles they are replacing, particularly after a decade of successive
increases in CAFE standard stringency, as Section IV above discusses.
The on-road fleet has well over 250 million vehicles, dwarfing the
roughly 16 million new vehicles sold each year. Comprehensive energy
savings come from turning over legacy vehicles in the fleet so that
overall fleet fuel economy increases. If the NPRM's preferred
alternative were finalized, the fuel consumption of the passenger car
and light truck fleet would have fallen from roughly 8.5 million
barrels per day (currently) to roughly 7 million barrels per day by
2050 as the fleet turned over. Finalizing the 1.5 percent alternative
reduces that number to 6.3 million barrels per day. That breaks the
trend of increasing oil consumption over time, and conserves energy.
(2)Technological Feasibility and the Effect of Other Motor Vehicle
Standards of the Government on Fuel Economy
As in the 2012 final rule, technological feasibility and the effect
of other motor vehicle standards of the Government on fuel economy do
weigh in NHTSA's balancing of the relevant factors, but they play a
less significant role because they vary less across regulatory
alternatives than the other factors vary. Technological feasibility, as
explained above and as similarly explained in 2012, relates to whether
technologies exist and can be commercially applied during the
rulemaking timeframe. None of the regulatory alternatives under
consideration today would require brand new technologies to be
invented--they can all be met with technology that exists currently.
However, as recognized in the 2012 final rule, ``some technologies that
currently have limited commercial use cannot be deployed on every
vehicle model in MY [2021], but require a realistic schedule for
widespread commercialization to be feasible. . . . Any of the
alternatives could thus be achieved on a technical basis alone if the
level of resources that might be required to implement the technologies
is not considered.'' As explained above in the discussion of economic
practicability, however, resources must be, and are, considered. The
2012 final rule further explained that ``If all alternatives are at
least theoretically technologically feasible in the [rulemaking]
timeframe, and the need of the nation is best served by pushing
standards as stringent as possible, then the agency might be inclined
to select the alternative that results in the very most stringent
standards considered.'' The 2012 final rule stated, however, that such
a selection would be inappropriate because ``the agency must also
consider what is required to practically implement technologies, which
is part of economic practicability, and to which the most stringent
alternatives give little weight.''
NHTSA considers technological feasibility similarly to how it has
long considered that factor--for the most part, the question of what
standards are maximum feasible is less about technological feasibility
than about economic practicability. All of the regulatory alternatives
considered in this final rule are likely technologically feasible, but
that does not mean that any of them could be maximum feasible, just as
we concluded in evaluating alternatives in 2012. NHTSA must now account
for how the need of the U.S. to conserve oil has changed, and this
consideration tips our balancing away from the most stringent
standards.
For the effect of other motor vehicle standards of the Government
on fuel economy, there is relatively little variation across regulatory
alternatives, as discussed in the FRIA. As in the 2012 final rule, in
developing this final rule NHTSA considered the effects of compliance
with known and possible NHTSA safety standards and known EPA emission
standards in developing this final rule, and has accounted for those
effects in the analysis. The effect of other motor vehicle standards of
the Government does not, therefore, have a noticeable effect on NHTSA's
balancing of factors to determine maximum feasible standards.
(3) Economic Practicability
Economic practicability remains a complex factor to consider and
balance, as discussed above, encompassing a variety of different issues
that are each captured to various degrees through the analysis. As
NHTSA stated in the 2012 final rule, ``The agency does not necessarily
believe that there is a bright-line test for whether a regulatory
alternative is economically practicable, but there are several metrics
. . . that we find useful for making the assessment.'' \2967\ In 2012,
as today, NHTSA looks to factors like:
---------------------------------------------------------------------------
\2967\ 77 FR at 63038 (Oct. 15, 2012).
---------------------------------------------------------------------------
Per-vehicle cost, in terms of ``even if the technology
exists and it appears that manufacturers can apply it consistent with
their product cadence, if meeting the standards will raise per-vehicle
cost more than we believe consumers are likely to accept, which could
negatively impact sales and employment in this sector, the standards
may not be economically practicable;'' \2968\
---------------------------------------------------------------------------
\2968\ Id.
---------------------------------------------------------------------------
Application rate of technologies, because ``even if
shortfalls are not extensive, whether it appears that a regulatory
alternative would impose undue burden on manufacturers in either or
both the near and long term in terms of how much and which technologies
might be required'' can be
[[Page 25175]]
relevant to manufacturers' difficulty with meeting standards; \2969\
---------------------------------------------------------------------------
\2969\ Id.
---------------------------------------------------------------------------
Consumer demand, which NHTSA described in 2012 as ``other
. . . considerations related to the application rate of technologies,
whether it appears that the burden on several or more manufacturers
might cause them to respond to the standards in ways that compromise .
. . other aspects of performance that are important to consumer
acceptance of new products'' \2970\
---------------------------------------------------------------------------
\2970\ Id.
---------------------------------------------------------------------------
Manufacturer compliance shortfalls, because ``If it
appears, in our modeling analysis, that a significant portion of the
industry cannot meet the standards defined by a regulatory alternative
in a model year, given that our modeling analysis accounts for
manufacturers' expected ability to design, produce, and sell vehicles
(through redesign cycle cadence, technology costs and benefits, etc.),
then that suggests that the standards may not be economically
practicable;'' \2971\
---------------------------------------------------------------------------
\2971\ Id.
---------------------------------------------------------------------------
Uncertainty and consumer acceptance of technologies, which
the 2012 final rule said was ``not accounted for expressly in our
modeling analysis, but [was] important to an assessment of economic
practicability given the time frame of this rulemaking.'' \2972\
---------------------------------------------------------------------------
\2972\ Id.
---------------------------------------------------------------------------
Thus, estimated impacts on per-vehicle cost are one issue;
estimated sales and employment impacts are issues; uncertainty
surrounding future market conditions and consumer demand for fuel
economy (versus consumer demand for other vehicle attributes) are other
issues. Consumers may respond to per-vehicle cost increases by choosing
to keep their current vehicle or buy used vehicles instead of new
vehicles, with consequent effects on new vehicle sales and the overall
fleet makeup; consumers may respond to new fuel-economy-improving
technologies on certain models by choosing to buy other models,
especially when fuel costs are not expected to increase significantly
in the ownership timeframe and consumers value other vehicle attributes
more than they value fuel economy. Either of these responses may cause
manufacturers both to lose money and to face further difficulties in
meeting the CAFE standards. While there are significant benefits for
both manufacturers and consumers under attribute-based standards,
manufacturers must still sell enough ``target-beaters'' to balance out
sales of less-fuel-efficient vehicles and meet their overall fleet-
average compliance obligations. If consumer demand shifts strongly away
from target-beaters, CAFE compliance will be a struggle, even if the
target-beaters are widely available. Section IV above discusses this
phenomenon in more detail. And if consumers buy fewer new vehicles in
response to per-vehicle cost increases, which the agencies are
beginning to see already, \2973\ the fleet as a whole will turn over
more slowly, and fuel conservation gains may also be slowed. NHTSA does
not believe that that is EPCA's goal. Manufacturers struggling to sell
new vehicles will have less capital to devote to further technological
improvements; may choose to move manufacturing jobs outside the U.S. to
places with lower labor costs; and so forth. A net benefits analysis
may be informative to attempting to quantify some of the issues
described above, but not all of these issues lend themselves to clear
quantification. The following discussion will evaluate what the
agencies believe has been reasonably accounted for.
---------------------------------------------------------------------------
\2973\ See, e.g., Jackie Charniga, ``Prime buyers flood used-
vehicle market in Q4,'' Automotive News, March 4, 2020, https://www.autonews.com/finance-insurance/prime-buyers-flood-used-vehicle-market-q4.
---------------------------------------------------------------------------
(a) Per-Vehicle Costs, Sales, and Employment as Part of Economic
Practicability
Per-vehicle cost estimates are relevant to NHTSA's consideration of
economic practicability because, when cost increases associated with
more stringent standards are passed through to consumers as price
increases, they affect consumers' willingness and ability to purchase
new vehicles, and thus influence vehicle sales and fleet turnover. A
similar effect occurs in reverse when stringency is decreased. Table
VIII-7 below shows the estimated effects on per-vehicle costs by
regulatory alternative in MY 2029:
[GRAPHIC] [TIFF OMITTED] TR30AP20.736
Generally speaking, per-vehicle costs increase as stringency
increases. The agencies estimate that, by MY 2029, costs for additional
fuel-saving technology (beyond that present on vehicles in MY 2017)
would average about $2,800 under the augural CAFE standards, as
compared to about $1,400 under the proposed CAFE standards,
[[Page 25176]]
and about $1,650 under the final CAFE standards for MYs 2021-2026. The
next most stringent alternative beyond the 1.5 percent alternative is
the ``2%/3%'' alternative. Under 2%/3%, the agencies estimate that
costs would increase by $2,000 per vehicle on average. NHTSA
understands that many readers may not find an extra $350 per vehicle to
be a compelling reason to reject the 2%/3% alternative, or even find an
additional $1,125 per vehicle a reason to reject the baseline/augural
standards. As the NPRM discussed, ``. . . the corresponding up-front
and monthly costs may pose a challenge to low-income or credit-
challenged purchasers. . . . such increased costs will price many
consumers out of the market--leaving them to continue driving an older,
less safe, less efficient, and more polluting vehicle, or purchasing
another used vehicle that would likewise be less safe, less efficient,
and more polluting than an equivalent new vehicle.'' \2974\ This
continues to be a concern: For example, the average MY 2025 prices
estimated here under the baseline, final, and 2%/3% CAFE standards are
about $38,100, $36,850, and $37,150, respectively. The buyer of a new
MY 2025 vehicle might thus avoid the following purchase and first-year
ownership costs under the final standards as compared to the baseline
standards or 2%/3% standards:
---------------------------------------------------------------------------
\2974\ 83 FR at 43222 (Aug. 24, 2018).
[GRAPHIC] [TIFF OMITTED] TR30AP20.737
While the buyer of the average vehicle would also purchase somewhat
more fuel under the final standards than the baseline standards, this
difference might average less than four gallons per month during the
first year of ownership. Some purchasers may consider it more important
to avoid these very certain (e.g., being reflected in signed contracts)
cost savings than the comparatively uncertain (because, e.g., some
owners drive considerably less than others, and may purchase fuel in
small increments as needed) fuel savings. For some low-income
purchasers or credit-challenged purchasers, the cost savings may make
the difference between being able or not to purchase the desired
vehicle. As vehicles get more expensive in response to higher CAFE
standards, it will get more and more difficult for manufacturers and
dealers to continue creating loan terms that both keep monthly payments
low and do not result in consumers still owing significant amounts of
money on the vehicle by the time they can be expected to be ready for a
new vehicle. These considerations were discussed in the NPRM and they
remain true for this final rule.
---------------------------------------------------------------------------
\2975\ Edmunds estimates that the average down payment for a new
vehicle in 2019 was 11.7% of the vehicle's price, see https://www.edmunds.com/car-buying/how-much-should-a-car-down-payment-be.html.
---------------------------------------------------------------------------
Per-vehicle cost and fuel economy both affect sales estimates in
the final rule analysis. Table VIII-9 below shows the estimated effects
on fleet-wide sales by regulatory alternative from 2017-2030, where the
augural standards represent absolute sales and all other alternatives
represent increases relative to the augural sales:
BILLING CODE 4910-59-P
[[Page 25177]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.738
BILLING CODE 4910-59-C
The final rule analysis indicates that industry sales decrease as
stringency increases, and increase as stringency decreases. While sales
under both the
[[Page 25178]]
proposal and the final rule are comparable, each represents about a 1.5
percent reduction in total sales over the period from 2017--2030. In
the context of 16-17 million new vehicle sales annually, NHTSA does not
believe that the sales volume effects here, while significant, are
necessarily determinative for economic practicability, even after
accounting for fuel economy effects in the sales analysis as some
commenters recommended. That said, NHTSA recognizes that the final rule
sales analysis does not account for a number of factors that could
cause differences between alternatives to result in changes in new
vehicle sales (perhaps greater). For example, as explained above, NHTSA
remains concerned that significant increases in fixed upfront prices
(which for many people translate to monthly financing costs) are harder
for certain segments of new vehicle buyers to manage than fuel costs,
which can be managed to some extent through vehicle switching or travel
decisions. The sales analysis for this final rule indicates that more
stringent standards tend to result in higher light truck sales and
lower passenger car sales. While NHTSA does not have specific
information (or a vehicle choice model) to inform the agency about
which consumers (by income) buy which vehicles, and while NHTSA
acknowledges that it does not account for price cross-subsidization by
manufacturers to keep ``entry-level'' new vehicle (often, passenger
car) prices low, NHTSA continues to be concerned about the possibility
of a bubble in the market for new vehicles. As the Wall Street Journal
reported in November 2019, ``Some 33% of people who traded in cars to
buy new ones in the first nine months of 2019 had negative equity,
compared with 28% five years ago and 19% a decade ago, according to
car-shopping site Edmunds . . . . Rising car prices have exacerbated an
affordability gap that is increasingly getting filled with auto debt.''
\2976\ The sales analysis for this final rule does not directly account
for these effects, but NHTSA is concerned that they may be
considerable. NHTSA notes that this analysis does not take into account
potential economic turmoil or recession, which may have a significant
impact on vehicle sales and industry viability.\2977\
---------------------------------------------------------------------------
\2976\ AnnaMaria Andriotis and Ben Eisen, ``A $45,000 Loan for a
$27,000 Ride: More Borrowers are Going Underwater on Car Loans,''
Wall Street Journal, November 9, 2019.
\2977\ Letter from Alliance for Automotive Innovation, NADA, and
MEMA to Congress, Mar. 23, 2020, available at https://www.autosinnovate.org/wp-content/uploads/2020/03/COVID-19-Letter-to-Congress-NADA-MEMA-AAI-March-23.pdf.
---------------------------------------------------------------------------
The final rule analysis also looked at employment effects under the
different regulatory alternatives. A number of commenters argued that
more stringent standards improved employment opportunities, as shown in
the NPRM analysis and in other analyses, due to the need for workers to
manufacture the additional technology needed to meet those more
stringent standards. Similar to the NPRM analysis, the agencies'
updated analysis shows labor utilization, on balance, increasing
slightly with stringency, as this effect outweighs the opposing effect
of changes in vehicle sales. Table VIII-11 below shows the estimated
effects on U.S. auto industry employment by regulatory alternative in
MY 2029:
[GRAPHIC] [TIFF OMITTED] TR30AP20.739
[GRAPHIC] [TIFF OMITTED] TR30AP20.740
It is important to note, however, that the reduction in person-
years described in this table merely reflects the fact that, when
compared to the standards set in 2012, fewer jobs will be specifically
created to meet infeasible regulatory requirements. It is also
important to note that the $15 billion in avoided required technology
costs (in MY 2029) can be invested by manufacturers into other areas,
or passed on to consumers. Moreover, consumers can either take those
cost savings in the form of a reduced vehicle price, or used toward the
purchase of specific automotive features that they desire (potentially
including a more-efficient vehicle or optional safety features that can
reduce risk of injury or death for all vehicle occupants on the road),
which would increase employment among suppliers and manufacturers.
Generally speaking, the agencies' analysis shows net labor
utilization increasing with stringency, because the additional labor
utilization involved with producing additional fuel-saving
[[Page 25179]]
technology outweighs the foregone labor utilization involved with the
foregone sales. As indicated above, for the scope of labor utilization
accounted for in today's analysis, the agencies show about 1.20 million
person-years under the augural CAFE standards and about 1.19 million
person-years under either the proposed or final standards. As for
sales, it is arguably instructive to consider these estimates in the
broader context of U.S. employment. BLS data indicates that roughly 129
million people in the U.S. are employed full-time at the time of
writing,\2978\ and that roughly 1.4 million people were employed in
motor vehicle and motor vehicle equipment manufacturing in 2018.\2979\
The agencies estimate that, compared to the augural standards, the
final standards will reduce automotive labor utilization associated
with production of the MY 2029 fleet by about 1.1%, a slightly smaller
reduction than the 1.4% estimated to occur under the proposed
standards. For comparison, the Synapse Report cited often by commenters
concluded that vehicle standards result in ``nationwide employment
increases of more than 100,000 in 2025 and more than 250,000 in 2035. .
. these increases represent less than 0.2 percent of current U.S.
employment levels.'' \2980\ Even at these levels, which NHTSA does not
necessarily agree are accurate, the employment effects of standards are
in the range of the average of more than 216,000 jobs added to the U.S.
economy during each month of 2018.\2981\ That said, as for sales, NHTSA
recognizes that the final rule labor utilization analysis does not
account for a number of factors that could cause differences between
alternatives to be different (perhaps greater), as discussed further
below.
---------------------------------------------------------------------------
\2978\ https://www.bls.gov/cps/cpsaat08.htm.
\2979\ https://www.bls.gov/cps/cpsaat18b.htm.
\2980\ https://www.synapse-energy.com/sites/default/files/Cleaner-Cars-and%20Job-Creation-17-072.pdf, at ES-2.
\2981\ Payroll employment increased by 2.6 million jobs in 2018,
an average of 216,667 per month. ``The Employment Situation--
December 2018,'' Bureau of Labor Statistics, available at: https://www.bls.gov/news.release/archives/empsit_01042019.pdf.
---------------------------------------------------------------------------
(b) Application Rates for New Technologies as Part of Economic
Practicability
The sales analysis for this final rule also does not account for
the potential consumer acceptance issue of more stringent standards
effectively requiring the application of technologies not yet ready for
widespread deployment. As widely noted, the 2012 rule assumed extremely
high penetration of dual-clutch transmissions in response to standards.
While the agencies stated throughout that final rule that the analysis
was not meant to represent the expected response to the standards, Ford
did apply DCTs to a number of vehicle models in its fleet, that
resulted in major customer satisfaction issues and ultimately caused
extensive buyback campaigns, customer service programs, and class-
action litigation.\2982\ Sales can be impacted as a result of standards
if technologies applied in response to those standards have
operational, maintenance, or customer acceptance problems, or if
consumers are unwilling to pay for it. Manufacturer capital to develop
and add new technologies and manage these rollout issues is finite, as
discussed. Insufficient capital can also cause quality problems. The
cost effects modeled in this final rule analysis, that drive the sales
and scrappage analyses, only include technology costs and RPE--they do
not include the cost of stranded capital or lost consumer surplus,
which are things that could drive up costs, drive down benefits, and
therefore impact sales and scrappage beyond what today's analysis
shows.
---------------------------------------------------------------------------
\2982\ See https://www.autonews.com/technology/dual-clutch-gearbox-complaints-haunt-ford.
---------------------------------------------------------------------------
As Section IV above notes, a great deal of fuel economy-improving
technology has already been added to the fleet since 2012, which means
that the amount of fuel economy-improving technology left to be added
in response to higher standards is less than it was assumed to be in
2012. Looking at the technology penetration rates modeled in today's
analysis, it appears that the augural standards are projected to
require nearly 20 percent total electrification in MY 2029, while the
proposal would have required nearly 7 percent and the final standards
would require nearly 8 percent. Table VIII-11 below shows projected
electrification rates by 2029 for the regulatory alternatives--
electrification refers to all models with strong hybrids, PHEVs, or
full EVs:
[GRAPHIC] [TIFF OMITTED] TR30AP20.741
[[Page 25180]]
As the table shows, the analysis projects that meeting the augural
standards could require over twice as much electrification as the final
rule standards could require.\2983\ The current market penetration for
all such vehicles is only approximately 4 percent even though the
technology is well-established, with hybrids having been first
introduced with the Honda Insight in 1999 and Toyota Prius in 2000,
plug in hybrids with the Chevrolet Volt in late-2010 and electric
vehicles with the Tesla Roadster in 2008 and Nissan Leaf in late 2010.
As Mr. Kreucher commented, and as Figure VIII-2 shows, consumers appear
to be driven by fuel price. Given anticipated fuel prices during this
timeframe and evidence in the market today of cannibalization within
these vehicle segments (not to mention the continued phasing out of
government incentives for these vehicles),\2984\ NHTSA is concerned
that there could be consumer acceptance problems associated with
further electrification under more stringent alternatives, which could
have sales impacts.
---------------------------------------------------------------------------
\2983\ While NHTSA is prohibited by statute from considering
battery electric vehicles as a compliance mechanism, we are aware
that many OEMs will likely opt to produce a smaller number of fully
electric vehicles rather than a large number of strong hybrid
models.
\2984\ 26 U.S.C. Section 30D provides for tax credits ranging
from $2,500 to $7,500 for purchasers of qualifying plug-in hybrid
(PHEV) and battery electric (BEV) vehicles, with a phaseout applying
to vehicle manufactured by an automaker once they sell 200,000
qualifying vehicles. Both Tesla and General Motors have reached this
threshold and the tax credit applicable to purchasers of new PHEV
and BEV vehicles from those manufacturers has been reduced gradually
and will phase out completely on January 1, 2020 for Tesla, and
April 1, 2020 for General Motors.
The California Clean Vehicle Rebate Project was launched in 2010
to provide incentives of up to $5,000 for purchasers or lessees of
qualifying PHEV, BEV, and certain other alternative fuel vehicles.
Since then, the program has undergone significant changes, including
the addition of income eligibility criteria for certain incentives,
and excluding eligibility toward the purchase or lease of a vehicle
with an MSRP exceeding $60,000.
Separately, in 2005, California passed a law allowing hybrid
electric vehicle (HEV), plug in hybrid electric vehicle (PHEV), and
battery electric vehicle (BEV), and other qualifying alternative
fuel vehicle owners to apply for a sticker allowing single-occupant
access to High Occupancy Vehicle (HOV) lanes. HEV access was phased
out in 2011, with eligibility being limited to PHEV, BEV and other
qualifying alternative fuel vehicle owners. Access is now limited to
a four-year period, and only to individuals who do not receive a
rebate under the California Clean Vehicle Rebate Project (unless
meeting income eligibility requirements).
---------------------------------------------------------------------------
We underscore that the table above simply shows the analytical
results of the modeling for today's final rule based upon the most
cost-effective means of achieving a given standard--it does not show
how manufacturers would, or could, comply with the CAFE standards
represented by the different regulatory alternatives. The discussion
below covers the topic of manufacturer compliance shortfalls, and this
discussion and that one are connected: The final rule analysis does not
show significant compliance shortfalls under any regulatory
alternative, but NHTSA believes that this is in large part because the
CAFE model is not programmed with assumptions about consumer acceptance
of strong hybrid technologies. In effect, the model lets manufacturers
lean on hybridization to achieve compliance at a lower cost than if
manufacturers instead pursued, for example, more advanced engine
technologies. If cost-effectiveness is the only concern, that may be a
valid compliance choice. If consumer acceptance of hybrid vehicles is
accounted for, especially in a time of foreseeably low fuel prices, it
may not be a valid compliance choice.
[GRAPHIC] [TIFF OMITTED] TR30AP20.742
[[Page 25181]]
As Figure VIII-2 illustrates, the market share of strong hybrids in
the new vehicle market has mostly tracked fuel prices. The bars
represent the market share (left axis) and the line tracks the price of
fuel (on the right axis). The light numbers inside of each bar
represent the number of unique strong hybrid models offered for sale in
that year. Initially, we see rapid growth that continues during the
fuel price increases of the mid-2000s and peaking at around 3.5 percent
market share. The figure shows that neither the passage of time, where
consumers become more familiar with the technology over successive
vehicle purchases, nor the number of models offered for sale have much
of an impact on the market share for strong hybrids. Despite a doubling
of the number of models offered for sale in subsequent years, market
share continued to track fuel price closely, and fell dramatically as
prices fell in 2015 and 2016. At fuel prices at or above $3.50/gallon,
strong hybrids were able to capture additional market share. However,
the current projection does not show prices returning to those levels
for quite some time--leaving manufacturers uncertain about their
ability to sell strong hybrids in the numbers estimated to be needed to
comply with CAFE and CO2 standards before MY 2026.
The agencies conducted a sensitivity analysis to evaluate the
impact of compliance pathways that did not rely on strong hybrids (see
Chapter 7 of the Final RIA). As we discuss in the sensitivity analysis,
in the absence of strong hybrids, compliance pathways tend toward a
greater reliance on advanced engines and transmissions, and more
aggressive exploitation of opportunities to reduce vehicles' mass.
These alternative technology pathways carry with them additional
technology costs that increase compliance costs in the baseline and
increase the savings associated with the preferred alternative.
Under the CAFE program, where battery electric vehicles are not a
compliance option (due to statutory restrictions on their consideration
for rulemaking), the additional cost of advanced engine technology in
the baseline increases baseline technology cost by about $800 per
vehicle, and increases the cost savings under the preferred
alternative, which has a much smaller reliance on strong hybrids to
achieve compliance, by about $600 per vehicle. This difference is
sufficient to change the sign on net social benefits for the preferred
alternative to being slightly negative, to being very positive (nearly
$80 billion at a 3 percent discount rate). The magnitude of this impact
is comparable to the impact of varying fuel price projections.
As shown in, Figure VIII-2 even the preferred alternative requires
levels of strong hybridization (and PHEV share) that would be about
twice what has been observed at the market, even at its peak. Both the
baseline and the 2%/3% alternative have even greater reliance on
hybridization--more than twice as much in the baseline. The compliance
costs associated with each alternative in today's rule depend upon the
estimated levels of hybridization in the compliance scenarios being
possible to achieve in the new vehicle market. The sensitivity analysis
shows that manufacturers can still reach comparable levels of fuel
economy without additional reliance on hybridization, but at
significantly higher per-vehicle costs. Those higher costs have
implications for the sales response, vehicle retirement rates in the
existing vehicle population, and the penetration rate of emerging
safety features.
(c) Consumer Demand as Part of Economic Practicability
As discussed above, NHTSA's consideration of consumer demand as
relevant to economic practicability has been upheld by the D.C. Circuit
in Center for Auto Safety v. NHTSA. A number of commenters argued that
consumers do, in fact, demand more fuel economy than the NPRM analysis
assumed; that consumers will appreciate more widespread application of
fuel economy-improving technologies that NHTSA appears to believe they
will tolerate; that NHTSA was wrong to assume that fuel prices will
remain relatively low in the future and continue to dampen consumer
demand for fuel economy; and that vehicle manufacturers will not make
tradeoffs between investments in fuel economy improvements and
investments in other vehicle characteristics which consumers also
demand, such that requiring manufacturers to meet more stringent
standards will not impair consumer demand for new vehicles because less
of those other characteristics will be available. Those commenters also
often highlighted the CAS language stating that consideration of
consumer demand may not undermine EPCA's goal of energy conservation.
NHTSA agrees with commenters that some consumers seek out vehicle
models with higher fuel efficiency, and notes that those consumers have
increasing numbers of relatively high-efficiency vehicle models to
choose from in the current new-vehicle market, as shown in the previous
section. CAFE does not affect fuel economy improvements that are
supported by consumer demand--market forces will take care of that.
Instead, it specifically addresses fuel economy improvements that are
not preferred by consumers, and the agency sets standards that require
manufacturers to make fuel economy improvements that consumers are not
otherwise seeking. Section IV.B.3 discusses at some length the fact
that alternative powertrains and higher fuel-efficiency vehicle models
have proliferated widely since 2011--consumers no longer lack for
choice if fuel economy is what they want. NHTSA's concern regarding
consumer demand is that in an era of relatively low gasoline prices--as
EIA currently projects and NHTSA has no basis to second-guess, and
which may be even lower than currently projected--it does not appear
likely that the market for higher fuel-economy vehicles and alternative
powertrains in particular will increase significantly in the rulemaking
timeframe, beyond the 30-month payback period that the agencies
currently use as a proxy for market demand for fuel economy. It is
worth citing the CAS case at greater length here in light of its
parallels: As the D.C. Circuit stated in that case,
[T]he petitioners do not challenge the consideration of consumer
demand per se, but rather the weight the agency has given the factor
in downgrading standards when, they argue, the principal
impracticability is paying a civil penalty [note: today, using or
purchasing credits]. Until the model years at issue here, there has
been little tension between consumer demand and the fuel
conservation goals of EPCA. The agency now relies on market
projections in a setting in which falling gas prices have relaxed
consumer demand for fuel efficiency. Earlier consideration of
consumer demand in setting standards could not have alerted Congress
to the agency's current application of this factor. Because Congress
has not spoken clearly on the issue before us, it must be determined
whether the agency's interpretation represents a reasonable
accommodation of the policies embodied in the statute.
. . .
The agency concluded that if manufacturers had to restrict the
availability of larger trucks and engines in order to adhere to CAFE
standards, the effects ``would go beyond the realm of `economic
practicability' as contemplated in the Act.'' [Citation omitted.]
The original projections of technological feasibility for the 1985
model year standards were based on the assumption that gasoline
prices would remain high and consumer demand for fuel-efficient
vehicles would remain strong. No one disputes that actual
circumstances have deviated from these assumptions. NHTSA acted
within the reasonable range of interpretations of the statute in
correcting the 1985 standards to
[[Page 25182]]
account for these changed conditions. Consideration of product mix
effects was also reasonable in setting the standards for 1986, as
there is no evidence that the same trends in consumer demand will
not continue.
. . .
In short, while it may be disheartening to witness the erosion
of fuel conservation measures in the face of changes in consumer
priorities, this court is nonetheless compelled to uphold the
agency's standards. They are the result of a balancing process
specifically committed to the agency by Congress, and, in this case,
the weight given to consumer demand was not outside the range
permitted by EPCA.
CAS, 793 F.2d 1322, 1340-41 (D.C. Cir. 1986). As in the situation
presented in the CAS case, the agencies believed in 2012 based on the
evidence then before them that fuel prices would be significantly
higher than the fuel prices currently projected today. Using the fuel
prices currently projected, which are lower because of the structural
changes to the global oil market described at length above, Figure
VIII-3 shows the difference in annual fuel consumption for a typical
driver under the augural standards, proposed standards, and final
standards. As the figure shows, the difference in annual consumption
(for a user that drives 14K miles per year) \2985\ is fewer than 40
gallons by MY 2030--the largest difference between the alternatives.
Rising fuel prices over time increase the value of those forty gallons,
but the diminishing returns to successive increases in fuel economy are
nonetheless evident.\2986\
---------------------------------------------------------------------------
\2985\ Parts of the central analysis assume a typical new
vehicle is driven 14,000 miles per year, for each of the first three
years it is owned. In practice, the average is slightly higher,
through affected by a smaller number of users that drive much more
than average. There is no single value that is representative of all
households, and the National Household Travel Survey has shown lower
annual usage estimates than 14,000 miles per year for a typical new
vehicle.
\2986\ In general, because fuel savings are subject to
diminishing returns as CAFE standards become more stringent, and
per-vehicle costs increase as CAFE standards become more stringent,
the relationship between per-vehicle costs and the value of fuel
savings is more of a curve than a line, although the slope of the
curve is reduced by the fact that we rely on EIA's forecast of
rising fuel prices over time.
[GRAPHIC] [TIFF OMITTED] TR30AP20.743
Thus, on the supply side, greater and more stable global oil
supply, which reduces projected fuel prices, means that the benefits of
more stringent CAFE standards are lower than they appeared to be in
2012 when the agencies believed oil supply would be scarcer and less
stable, and projected fuel prices were consequently higher.
On the demand side, as already explained, while NHTSA agrees that
some consumers do seek out higher fuel economy, those consumers have
vastly more higher fuel-economy-vehicle options than they did when the
agencies wrote the 2012 final rule, as shown in Section IV above. For
the other consumers who are driven more by the economics of their
vehicle-purchasing decisions, NHTSA believes that they are likely
making reasonably informed decisions about the new vehicle attributes
they want in light of expectations about future fuel costs. This can be
illustrated by examining estimated payback periods under the different
regulatory alternatives, because payback period directly compares
estimated future fuel savings with estimated vehicle purchase and
ownership costs. A number of commenters suggested that per-vehicle cost
was not a meaningful metric in isolation, because consumers would also
be saving money on fuel under more stringent standards. The agencies
discuss affordability issues further below, but the rulemaking presents
Table VIII-12 here as a comparison of per-vehicle costs to lifetime
fuel savings to illustrate the point raised by commenters:
[[Page 25183]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.744
Table VIII-12 shows the differences in regulatory costs, other
registration costs (taxes and financing, though the cost of insurance
also increases to cover more expensive vehicles), lifetime fuel
savings, and the payback relative to a MY 2017 vehicle. It is important
to compare apples to apples, so in this case, because the agencies are
considering fuel costs over a vehicle's full lifetime, this rulemaking
needs to compare that against a broader lifetime cost of ownership,
instead of comparing it simply to the estimated increase in initial
purchase price. Under the augural standards, the analysis projects that
it would take a full five years for the undiscounted value of fuel
savings to offset the estimated upfront increase in purchase cost
(relative to a MY 2017 vehicle). For reference, the average new car
buyer holds on to that car for about six or seven years.\2987\
Naturally, this payback period, and the fuel savings on which it is
based, depend upon fuel prices. Higher fuel prices shorten payback
periods, while declining fuel prices lengthen them. For this analysis,
the agencies have employed fuel prices estimated using the version of
NEMS used to produce AEO 2019, as discussed in Section VI.
---------------------------------------------------------------------------
\2987\ IHS Markit estimates the average length of new vehicle
ownership at about 79 months, see https://www.forbes.com/sites/jimgorzelany/2018/01/12/the-long-haul-15-vehicles-owners-keep-for-at-least-15-years/#4e971b576237.
---------------------------------------------------------------------------
Thus, all of the regulatory alternatives considered in today's
analysis result in significantly longer payback periods than the 2.5
years assumed by the agencies, the industry, and the NAS--i.e., while
fuel economy would foreseeably improve in the rulemaking timeframe in
the absence of regulation, it would do so at a rate slower even than
the proposal would have required.\2988\ NHTSA thus does not expect that
consumer demand for fuel-efficient vehicles will grow significantly in
the rulemaking timeframe without regulation to prop up manufacturer
sales of significantly larger volumes of more fuel-efficient models.
This increases the economic practicability of regulatory alternatives
that represent less stringent standards, as compared to those that
represent more stringent standards.
---------------------------------------------------------------------------
\2988\ While presented at the industry level, technology
application and compliance simulation occur at the level of each
individual manufacturer's respective fleets. Some OEMs and fleets
are able to increase CAFE more easily than others--starting from
more favorable positions and adding less expensive technology, or
taking advantage of credit provisions, to improve the fuel economy
of their fleets. However, for several OEMs, even the proposed
standards are binding, and the costs associated with bringing their
fleets into compliance are significant. At the level of the industry
average, the cost of compliance with the proposal--and as a
corollary, with the other alternatives--exceeds the 2.5 year payback
for fuel economy technology, even while a small amount of
overcompliance occurs at the industry level.
---------------------------------------------------------------------------
(d) Manufacturer Compliance Shortfalls as Part of Economic
Practicability
Manufacturer compliance shortfalls given the pace of increase in
standard stringency over time are also relevant to economic
practicability, and were considered as part of the 2012 final rule.
Some commenters argued that it was not reasonable for NHTSA to
interpret automakers' fuel economy improvements over time as evidence
that less stringent standards might be maximum feasible, suggesting
that evidence of improvements means that improvements are possible, and
that automakers' stated difficulties with meeting more stringent
standards may be overstated. Fleet fuel economy improvements over time
have been possible, NHTSA agrees. NHTSA does not agree, however, that
improvements thus far constitute de facto evidence of automakers'
ability to meet rapidly increasing standards indefinitely into the
future. Section IV above illustrates this clearly--many more very fuel-
efficient models are available now than in 2012, while fuel prices have
been trending downward on an absolute basis over the same time period.
Simultaneously and relatedly, the rate at which various manufacturer
fleets have been falling short of their standards has been increasing
steadily. As Section IV explains, at the time of the 2012 analysis,
most manufacturers were in reasonable shape in terms of compliance. The
total fleet outperformed CAFE standards by a full mile per gallon--
reflecting the historical trend that the full fleet always exceeds
[[Page 25184]]
the average fuel economy target.\2989\ Of the then 45 import passenger
car, domestic passenger car, and light truck compliance fleets in the
2012 model year, 26 of the fleets exceeded their fuel economy targets,
while 19 failed to meet their standard.\2990\ Of those 19 fleets that
failed to meet their standard, the total shortfall was 41,033,802
credits--the equivalent of $225,685,911 in penalties.\2991\ That is no
longer the case. 2016 marked the first model year in CAFE history that
the entire light duty fleet failed to meet its target.\2992\ This
continued in the 2017 model year (the most recent full model year of
compliance data).\2993\ In the 2017 model year, of the now 42
compliance fleets, only 14 fleets exceeded their targets.\2994\ 25
failed to meet their target, with a total shortfall of 166,715,863
credits--the equivalent of $1,133,430,584 in penalties.\2995\ Required
manufacturer reporting data shows the situation continuing to get worse
in the 2018 and 2019 model years,\2996\ despite manufacturers'
increasing ability to utilize generous credit provisions related to
alternative fueled vehicles and A/C efficiency and off-cycle
adjustments.
---------------------------------------------------------------------------
\2989\ Data from CAFE Public Information Center (PIC), https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm, last accessed Dec. 27,
2019.
\2990\ NHTSA MY 2011-2019 Industry CAFE Compliance, https://one.nhtsa.gov/cafe_pic/MY%202011-MY_2019_Credit_Shortfall_Report_v08.pdf.
\2991\ Id. While we denominate shortfalls in terms of credits,
that is simply for convenience, and any given manufacturer's
shortfall is measured in tenths of a mile per gallon for compliance
purposes.
\2992\ Data from CAFE Public Information Center (PIC), https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm, last accessed Dec. 27,
2019.
\2993\ Id.
\2994\ NHTSA MY 2011-2019 Industry CAFE Compliance, https://one.nhtsa.gov/cafe_pic/MY%202011-MY_2019_Credit_Shortfall_Report_v08.pdf.
\2995\ Id.
\2996\ Id.
---------------------------------------------------------------------------
Although each year has continued to see improvements in fuel
economy performance, each successive increase in stringency requires
many fleets not only to achieve the new level from the resulting
increase, but to resolve deficits from the prior year as well. The
problem is particularly marked in the light truck fleet, where sales of
lower fuel-economy vehicles have proliferated over this time period,
despite availability of higher fuel-economy models. But the passenger
car fleet is facing compliance challenges as well, as more consumers
have shifted away from sedans and into crossover utility vehicles that
are considered passenger cars for compliance purposes. While the
agencies' move toward footprint based standards account for vehicle
length and track width--which certainly affect fuel economy as
described above--they do not account for mass-intensive increases in
vehicle ride height that crossover purchasers value, the additional
frontal area and higher drag at highway speeds, or the additional power
required to achieve similar performance as the equivalent sedan. These
issues are further exacerbated by the fact that consumers are demanding
more powerful engines than the baseline efficient four cylinder
versions the agencies assumed consumers would find acceptable, instead
opting to upgrade to more powerful powertrains.\2997\ If the augural
standards were finalized and energy prices remain as currently
projected, the shortfall situation could well erase large portions of
assumed fuel savings/emissions reduction benefits from higher
standards.
---------------------------------------------------------------------------
\2997\ Mr. Rykowski's comments for EDF, for example, stated that
EPA's recent Fuel Economy and CO2 Trends Reports show
clearly that manufacturers have been improving vehicle performance
at the expense of fuel economy. See NHTSA-2018-0067-12018, at 31.
---------------------------------------------------------------------------
In the current analysis, gasoline prices are projected to rise
steadily from about $2.50/gallon in 2017 to $3.5/gallon by 2035. While
CAFE can provide some insurance against unexpected and sudden price
increases, in the case of sustained, consistent increases in gasoline
prices, market demand for fuel economy would outpace the standards over
time. In an earlier analysis, the agencies considered the impact of a
sudden gasoline price shock in a single year, where the price of
gasoline jumped from $3.50/gallon to $6/gallon for most of a year. If
instead of that one-year spike, the price of gasoline rose steadily
from current levels to $6/gallon by 2040, the response of both
consumers and manufacturers in the marketplace would cause the industry
to consistently over-comply with even the augural standards.\2998\ The
payback assumption in this analysis, where consumers are willing to pay
for any fuel economy improvement that pays for itself in the first 2.5
years of vehicle usage, would likely be too short in a world with $6/
gallon gasoline, where the cost of operating a vehicle consumed a
larger share of a household's budget and even longer payback periods
could be seen as sound investments. Thus, if it turns out that fuel
prices rise steadily over the next decade, at a significantly faster
rate than currently projected, the market will end up demanding more
efficient vehicles and the gap between the baseline and the preferred
alternative will shrink further. However, the agencies do not currently
have information that projects $6/gallon fuel in 2040 is likely, for
the reasons discussed at length above.
---------------------------------------------------------------------------
\2998\ We simulated this response in the CAFE Model, where all
other inputs were identical to the central analysis.
---------------------------------------------------------------------------
As also discussed above, while the analysis for this final rule
does not show significant shortfalls under any regulatory alternative,
that appearance of compliance is predicated on the assumption that
automakers will be able to sell the hybrids that we simulate them
producing in response to the standards. Again, given foreseeably low
fuel prices going forward, it is also foreseeable that selling greater
volumes of hybrid vehicles will be even more difficult than at present.
It is very possible that manufacturer compliance shortfalls could end
up being worse than the agency's analysis currently forecasts for the
more stringent alternatives.
Given the ongoing shortfall problem illustrated above, and given
the payback period estimates, the proposal might appear to be the
correct answer in the absence of other considerations. NHTSA believes
that the bubble concerns may be significant, and the diminishing
returns of higher standards identified in Section IV above calls into
question the value of pushing that bubble. Compliance shortfalls
represent a growing problem with the current standards and will
continue to be a problem if stringency does not converge at least
somewhat more closely with what the market appears willing to bear. If
industry is unable to comply with standards, that non-compliance means
that the standards are not achieving what they set out to achieve in
terms of fuel savings or emissions reductions, or at least they are not
achieving what NHTSA estimated they would achieve. The NPRM disagreed
with the idea that ``if you build it, they will come''--that
manufacturers would find a way to market higher fuel-economy vehicles,
and consumers would eventually buy them. Comments on that topic were
mixed: some commenters agreed with the NPRM's sentiment, while other
commenters argued that manufacturers' past ability to exceed standards
combined with consumers' growing interest in fuel economy/lower
emissions meant that concerns about the market's ability to bear
further increases were misplaced. The shortfall discussion above and in
Section IV suggests that the NPRM's sentiment may be accurate, but this
difference in perspective highlights the core philosophical question of
the CAFE program--whether consumers should choose for themselves how
much fuel
[[Page 25185]]
economy they want, or whether the government should choose for them.
(4) Considering Safety Along With the Other Factors in Determining
Maximum Feasible Standards
In addition to the above, as explained in the NPRM and as discussed
extensively by commenters, NHTSA considers safety effects in
determining maximum feasible CAFE standards. A number of commenters
objected to aspects of the safety analysis, as discussed in Section VI
above, and some made suggestions for improvement. In response to those
comments, NHTSA took a very conservative approach in making a number of
changes to the safety analysis for this final rule:
Commenters disagreed with certain aspects of the sales and
scrappage effects on the safety analysis; in response to those
comments, changes have been made and the scrappage effect on fatalities
is lower now than it was in the NPRM;
Commenters disagreed with certain aspects of mass
reduction; in response to those comments, changes have been made there;
Commenters argued that additional technologies should be
accounted for; in response to those comments, many of those
technologies have been added;
Commenters argued that the NPRM did not account for crash
avoidance technologies; in response to those comments, the final rule
accounts for the effects of crash avoidance technologies;
Commenters argued that the NPRM did not account for the
mortality/morbidity effects of criteria pollution differences between
the alternatives; in response, the final rule accounts for these
effects explicitly in these values.
Overall, the final rule analysis suggests that fatalities may be
lower than the NPRM analysis showed; injuries may be greater; and the
safety effects overall are less than the NPRM suggested, but they are
still significant. Less-stringent standards remain better for safety
and are projected to save thousands of lives and prevent tens of
thousands of hospitalizations, even if the amount by which they are
better is lower than previously estimated.
EPCA/EISA directs NHTSA to conserve energy and consider the need of
the U.S. to conserve energy, while simultaneously directing NHTSA to
set attribute-based standards whose outcome varies depending on what
consumers choose to buy, and directing NHTSA to consider economic
practicability. The greater the need of the U.S. to conserve energy,
the more the government should decide for consumers how much fuel
economy will be in their new vehicles. Based on the information before
NHTSA in this final rule, NHTSA agrees with the commenters who
suggested that increasing CAFE stringency can function as ``insurance''
against future oil price volatility, although as illustrated above, the
short-term effects of that insurance may be relatively minor and the
longer-term effects may be too uncertain to consider meaningfully.
NHTSA also agrees that environmental considerations necessitate energy
conservation, though the long-term benefits of emissions reductions
(even accounting for the increased costs of delayed action) require
consideration of the immediate costs to consumers, the industry, and
the environment.
Balancing all of the factors and issues identified above, NHTSA
concludes that standards that increase at 1.5% per year are the maximum
feasible for passenger cars and light trucks for MYs 2021-2026, based
on the information currently before the agency. We recognize that more
stringent standards, including the baseline/augural standards, could
conserve more energy and might be technologically feasible (in the
narrowest sense), but the additional incremental fuel savings,
emissions reductions, and environmental benefits of higher standards is
not significant enough to outweigh the immediate economic costs. There
is still risk to the U.S. from circumstances outside our control that
the CAFE program may be able to mitigate, but there must also be
recognition of the limited extent to which this program can address
that risk, certainly without exacerbating considerable challenges
currently being faced by automakers, dealers, and consumers. Economic
practicability would be best served by slower increases, as discussed
above. And while these two factors weigh in different directions, NHTSA
has discretion to accommodate conflicting statutory priorities in a
reasonable manner. Beginning with MY 2021, the first MY addressed by
this rule, Congress eliminated the obligation to increase FE standards
ratably.\2999\ Thus, the appropriateness of an increase, if any, is
within NHTSA's discretion based on its balancing of statutory
factors.\3000\
---------------------------------------------------------------------------
\2999\ Previously applied for MYs 2011-2020.
\3000\ NHTSA also notes that it was expressly anticipated in the
2012 final rule that the current rulemaking could determine that the
augural standards were not maximum feasible. NHTSA stated that
``Whether different alternatives may be maximum feasible can also be
influenced by differences and uncertainties in the way in which key
economic factors (e.g., the price of fuel and the social cost of
carbon) and technological inputs could be assessed and valued. While
NHTSA believes that our analysis for this final rule uses the best
and most transparent technology-related inputs and economic
assumption inputs that the agencies could derive for MYs 2017-2025,
we recognize that there is uncertainty in these inputs, and the
balancing could be different if the inputs were different. When the
agency undertakes the future rulemaking to develop final standards
for MYs 2022-2025, for example, we expect that much new information
will inform that future analysis, which may potentially lead us to
choose different standards than the augural ones presented today.''
(emphasis added) 77 FR at 63037 (Oct. 15, 2012).
---------------------------------------------------------------------------
In past rulemakings, as discussed above, NHTSA has expressly
considered the point at which net benefits appear to be maximized as
potentially relevant to determining maximum feasible CAFE
standards.\3001\ Whether the standards maximize net benefits has thus
been a significant, but not dispositive, factor in the past for NHTSA's
consideration of economic practicability. Executive Order 12866, as
amended by Executive Order 13563, states that agencies should ``select,
in choosing among alternative regulatory approaches, those approaches
that maximize net benefits . . .'' In practice, however, NHTSA must
consider that the modeling of net benefits does not capture all
considerations relevant to the EPCA statutory factors. Additionally,
nothing in EPCA or EISA mandates that NHTSA set standards at the point
at which net benefits are maximized, and case law confirms that whether
to maximize net benefits in determining maximum feasible standards is
within NHTSA's discretion.\3002\ As explained extensively in prior
rulemakings, even if the agency believed it could quantify enough
relevant factors to determine the CAFE
[[Page 25186]]
levels at which net benefits were maximized with reasonable accuracy,
there may be other considerations which lead the agency to conclude
that maximum feasible CAFE standards are not the ones that maximize net
benefits. For example, in 2012, NHTSA rejected the regulatory
alternative that appeared to maximize net benefits (and all
alternatives more stringent than that one) based on the conclusion that
even though net benefits were maximized, the ``resultant technology
application and cost'' were simply too high, and thus made those
standards economically impracticable, and thus beyond maximum
feasible.\3003\
---------------------------------------------------------------------------
\3001\ See, e.g., the 2006 final rule, which concluded that the
point at which net benefits were maximized was the maximum feasible
CAFE level (71 FR 17566 (Apr. 6, 2006)); the 2010 final rule, which
considered among the regulatory alternatives one that maximized net
benefits, but explained that nothing in EPCA or EISA mandated that
NHTSA choose CAFE standards that maximize net benefits (75 FR 25324,
at 25606, 25167 (May 7, 2010)); and the 2012 final rule, which also
considered among the regulatory alternatives one that maximized net
benefits, and also explained that nothing in EPCA or EISA mandated
that NHTSA choose CAFE standards that maximize net benefits, in
fact, directly rejecting the regulatory alternative that maximized
net benefits as beyond maximum feasible for the MYs 2017-2025
timeframe (77 FR 62624 (Oct. 15, 2012)).
\3002\ The Ninth Circuit has agreed with NHTSA that ``EPCA
neither requires nor prohibits the setting of standards at the level
at which net benefits are maximized,'' stating further that ``The
statute is silent on the precise question of whether a marginal
cost-benefit analysis may be used. See Chevron, 467 U.S. at 843, 104
S.Ct. 2778. Public Citizen and Center for Auto Safety persuade us
that NHTSA has discretion to balance the oft-conflicting factors in
49 U.S.C. 32902(f) when determining ``maximum feasible'' CAFE
standards under 49 U.S.C. 32902(a).'' CBD v. NHTSA, 538 F.3d 1172,
1188 (9th Cir. 2008).
\3003\ 77 FR at 63050 (Oct. 15, 2012).
---------------------------------------------------------------------------
Table VII-95 and Table VII-96, above, appear to suggest that net
benefits would be maximized under a 3 percent discount rate by choosing
the 2%/3% alternative, and under a 7 percent discount rate by choosing
the 0% (proposed) alternative. Across all alternatives under either
discount rate, the variation in net benefits is within $20 billion over
the lifetimes of vehicles produced during the rulemaking timeframe.
While $20 billion may seem like a large amount of money, it must be
understood within context--the auto industry accounted for
approximately $89 billion of U.S. GDP in 2018 alone,\3004\ and
Americans spent approximately $370 billion on gasoline in 2019
alone.\3005\ For a program this large, if the difference between the
net benefits created by different regulatory alternatives is within $20
billion (over the full lifetimes of six model years), the net benefits
are relatively small. Furthermore, given how close together the net
benefits are across the range of regulatory alternatives considered,
NHTSA does not believe that the point at which net benefits are
maximized is meaningful for determining maximum feasible CAFE standards
in this final rule.
---------------------------------------------------------------------------
\3004\ See Bureau of Economic Analysis, GDP by Industry, ``Value
Added by Industry,'' Oct. 29, 2019, https://apps.bea.gov/iTable/iTable.cfm?ReqID=51&step=1 (accessed Mar. 18, 2020)
\3005\ Using EIA estimates of an average of $2.60/gallon
gasoline cost in 2019 (https://www.eia.gov/todayinenergy/detail.php?id=42435) and EIA estimates of about 142 billion gallons
total gasoline consumed (https://www.eia.gov/tools/faqs/faq.php?id=23&t=10).
---------------------------------------------------------------------------
Important to that conclusion is the fact that the net benefits
estimates produced by the analysis depend heavily on EIA's future
forecasts of fuel prices, which were made prior to the recent collapse
of oil prices. If the former OPEC+ members continue to pursue market
share, fuel prices will likely continue to drop. If, instead of
pursuing market share, they try to control prices by restricting
supply, U.S. shale production can ramp back up and exert downward
pressure on price. If fuel prices end up even lower than our analysis
assumes, benefits from saving additional fuel will be worth even less
to consumers. Our analysis captures none of these effects. Depending
upon future fuel prices, net benefits estimates described above could
foreseeably be overstated, possibly by a significant amount. It is
possible, depending on future fuel prices, that the final rule 1.5
percent annual increase standards could end up being more stringent
than standards that would maximize net benefits. Moreover, sustained
low oil prices can be expected to have real effects on consumer demand
for additional fuel economy, which will have real effects on sales,
jobs, and many other things relevant to NHTSA's consideration of what
standards would be maximum feasible. Choosing a regulatory alternative
more stringent than the final rule's 1.5 percent annual increases could
foreseeably either lead to more hybridization than the market is likely
to bear given foreseeably low fuel prices, or lead to significantly
more cost than the analysis currently suggests. Neither of those
outcomes would be beneficial for consumers or for industry, even
considering the additional fuel savings for consumers.\3006\
---------------------------------------------------------------------------
\3006\ It is within NHTSA's discretion to adopt an alternative
based on unquantified/unquantifiable benefits. See, e.g., Inv. Co.
Inst. v. Commodity Futures Trading Comm'n, 720 F.3d 370, 379 (D.C.
Cir. 2013) (``The appellants further complain that CFTC failed to
put a precise number on the benefit of data collection in preventing
future financial crises. But the law does not require agencies to
measure the immeasurable. CFTC's discussion of unquantifiable
benefits fulfills its statutory obligation to consider and evaluate
potential costs and benefits. See Fox, 556 U.S. at 519, 129 S.Ct.
1800 (holding that agencies are not required to `adduce empirical
data that' cannot be obtained). Where Congress has required
`rigorous, quantitative economic analysis,' it has made that
requirement clear in the agency's statute, but it imposed no such
requirement here. American Financial Services Ass'n v. FTC, 767 F.2d
957, 986 (DCCir.1985); cf., e.g., 2 U.S.C. 1532(a) (requiring the
agency to `prepare a written statement containing . . . a
qualitative and quantitative assessment of the anticipated costs and
benefits' that includes, among other things, `estimates by the
agency of the [rule's] effect on the national economy').'');
BellSouth Corp. v. FCC, 162 F.3d 1215, 1221 (D.C. Cir.1999) (`When .
. . an agency is obliged to make policy judgments where no factual
certainties exist or where facts alone do not provide the answer,
our role is more limited; we require only that the agency so state
and go on to identify the considerations it found persuasive').''
---------------------------------------------------------------------------
NHTSA concludes that steady increases at 1.5 percent annually, with
the same rate for cars and trucks as suggested by several commenters,
are the optimal way to move the needle forward on fuel economy, fuel
savings, and emissions reductions without imposing excessive cost on
automakers and consumers and overly reducing vehicle sales. Requiring
demand changes (through CAFE standards) much faster than what the
market will bear creates a substantial likelihood of a mis-match
between what companies produce and what consumers buy. While companies
can manage that mis-match for short periods through incentivization and
cross-subsidization, we have seen that over time automakers begin to
fall short on fuel economy performance relative to the standards. Over
time, if swaths of the industry continually fall short of fuel economy
targets, and consumer demand for fuel economy does not significantly
increase, then continuing to force technology into the fleet does not
achieve the program's objectives (i.e., energy conservation). This is
the case regardless of how much manufacturers spend manufacturing
vehicles that consumers do not purchase (implicating concerns with
economic practicability) to reduce their compliance liability. This is
one part of why NHTSA believes that the 1.5 percent alternative is
maximum feasible during the rulemaking timeframe.
While the 1.5 percent alternative being finalized is new for the
final rule, it is responsive to comments requesting steady increases at
the same rate for both cars and trucks, and it is within the range of
rates of increase considered in the NPRM. As both the NPRM analysis and
the final rule analysis show, after MY 2020 the proposed (0%) standards
are not binding at the industry level (though some manufacturers, and
fleets, remain below their standard after that model year) as a
consequence of market demand for fuel economy at projected gasoline
prices. However, the preferred (1.5% percent) alternative, while
producing slightly higher achieved CAFE levels, tracks closely to the
level produced by the combination of existing CAFE standards (through
MY 2020) and subsequent market demand for fuel economy represented by
the proposal. It is also likely close to the point at which net
benefits will be maximized, even if it remains unclear exactly where
that point will end up.
As a kind of insurance policy against future fuel price volatility,
standards that increase at 1.5 percent per year for cars and trucks
will help to keep fleet fuel economy higher than they would be
otherwise when fuel prices are low, which is not improbable over the
next several years.\3007\ These standards will
[[Page 25187]]
also enable industry to choose how to spend the capital that would
otherwise be spent meeting more stringent standards on more of what
consumers are demanding, which could also include more fuel economy if
the market heads unexpectedly in that direction. As explained above,
even if more stringent standards might be technologically feasible in a
narrow sense, and even if the effect of other motor vehicle standards
of the Government does not vary significantly between regulatory
alternatives, economic practicability concerns still counsel against
more stringent standards, and the need of the U.S to conserve energy
does not, at present, appear to counsel toward higher stringency.
Standards that increase at 1.5 percent per year represent a reasonable
balance of additional technology and required per-vehicle costs,
consumer demand for fuel economy, fuel savings and emissions avoided
given the foreseeable state of the global oil market and the minimal
effect on climate between finalizing 1.5 percent standards versus more
stringent standards. The final standards will also result in year-over-
year improvements in fleetwide fuel economy, resulting in energy
conservation that helps address environmental concerns, including
criteria pollutant, air toxic pollutant, and carbon emissions. All
things considered, NHTSA determines that an increase of 1.5 percent per
year is maximum feasible for both passenger cars and light trucks for
MYs 2021-2026.
---------------------------------------------------------------------------
\3007\ For example, EIA currently expects U.S. retail gasoline
prices to average $2.14/gallon in 2020, compared to $2.69/gallon in
2019 (see https://www.eia.gov/outlooks/steo/archives/mar20.pdf), and
$3.68/gallon in 2012 (see https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=EMM_EPM0_PTE_NUS_DPG&f=A). While gasoline
prices may foreseeably rise over the rulemaking time frame, it is
also very foreseeable that they will not rise to the $4-5/gallon
that many American saw over the 2008-2009 time frame, that caused
the largest shift seen toward smaller and higher-fuel-economy
vehicles. See, e.g., Figure VIII-2 above.
---------------------------------------------------------------------------
Compliance and Enforcement
A. Introduction
1. Overview
The CAFE and CO2 emissions standards are both fleet-
average standards, and for both programs, determining compliance begins
by testing vehicles on dynamometers in a laboratory over pre-defined
test cycles under controlled conditions.\3008\ A machine is connected
to the vehicle's tailpipe while it performs the test cycle, which
collects and analyzes the resulting exhaust gases; a vehicle that has
no tailpipe emissions has its performance measured differently, as
discussed below. CO2 quantities, as one of the exhaust
gases, can be evaluated for vehicles that produce CO2
emissions directly. Fuel economy is determined from the amount of
CO2 emissions, because the two are directly mathematically
related.\3009\ Manufacturers generally perform their own testing, and
EPA confirms and validates those results by testing a sample of
vehicles at the National Vehicle and Fuel Emissions Laboratory (NVFEL)
in Ann Arbor, Michigan. The results of this testing form the basis for
determining a manufacturer's compliance in a given model year, through
the following steps:
---------------------------------------------------------------------------
\3008\ For readers unfamiliar with this process, it is similar
to running a car on a treadmill following a program--or more
specifically, two programs. 49 U.S.C. 32904(c) states that, in
testing for fuel economy, EPA must ``use the same procedures for
passenger automobiles [that EPA] used for model year 1975 (weighted
55 percent urban cycle and 45 percent highway cycle), or procedures
that give comparable results.'' Thus, the ``programs'' are the
``urban cycle,'' or Federal Test Procedure (abbreviated as ``FTP'')
and the ``highway cycle,'' or Highway Fuel Economy Test (abbreviated
as ``HFET''), and they have not changed substantively since 1975.
Each cycle is a designated speed trace (of vehicle speed versus
time) that vehicles must follow during testing--the FTP is meant
roughly to simulate stop and go city driving, and the HFET is meant
roughly to simulate steady flowing highway driving at about 50 mph.
The 2-cycle dynamometer test results differ somewhat from what
consumers will experience in the real world driving environment
because of the lack of high speeds, rapid accelerations, and hot and
cold temperatures evaluations with the A/C operation. These added
conditions are more so reflected in the EPA 5-cycle test results
listed on each vehicle's fuel economy label and on the
fueleconomy.gov website.
\3009\ Technically, for the CAFE program, carbon-based tailpipe
emissions (including CO2, CH4, and CO) are
measured, and fuel economy is calculated using a carbon balance
equation. EPA uses carbon-based emissions (CO2,
CH4, and CO, the same as for CAFE) to calculate the
tailpipe CO2 equivalent for the tailpipe portion of its
standards.
---------------------------------------------------------------------------
Each vehicle model's performance on the test cycles is
calculated;
The number of vehicles of that model that were produced is
divided by the performance;
That number, in turn is summed for all the manufacturer's
model types;
The manufacturer's total product volume is then divided by
the summed value of all the model types; and
That number represents the manufacturer's fleet harmonic
average performance.
That performance is then compared to the manufacturer's unique
compliance obligation (standard). This compliance obligation is
calculated using the same approach that is used to determine
performance, except that the fuel economy or CO2 target
value (based on the footprint of each vehicle model) is used instead of
the model's measured performance value. The fuel economy or
CO2 target values for each of the vehicle models in the
manufacturer's fleet and production volumes are used to derive the
manufacturer's fleet harmonic average standard. Using fuel economy
targets to illustrate the concept, the following figure shows two
vehicle models produced in a model year for which passenger cars are
subject to a fuel economy target function that extends from about 30
mpg for the largest cars to about 41 mpg for the smallest cars:
[[Page 25188]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.745
If these are the only two vehicle models the manufacturer produces,
the manufacturer's required CAFE obligation is determined by
calculating the production-weighted harmonic average of the fuel
economy target values applicable at the hatchback and sedan footprints
(from the curve, about 41 mpg for the hatchback and about 33 mpg for
the sedan). The manufacturer's achieved CAFE level is determined by
calculating the production-weighted harmonic average of the hatchback
and sedan fuel economy levels (in this example the values shown in the
boxes in Figure IX-1, 48 mpg for the hatchback and 25 mpg for the
sedan). Depending on the relative mix of hatchbacks and sedans the
manufacturer produces, the manufacturer's fleet may meet the standard,
or perform better than the standard (if required CAFE is less than
achieved CAFE) and thereby earn credits or perform worse than the
standard (if required CAFE is greater than achieved CAFE) and thereby
have a shortfall that may be made up, in whole or in part, using CAFE
credits, discussed below, or be subject to civil penalties. Although
the arithmetic is different for CO2 standards (which do not
involve harmonic averaging), the underlying concept is the same.
There are thus two parts to the foundation of compliance with CAFE
and CO2 emissions standards: First, how well any given
vehicle model performs relative to its target, and second, how many of
each vehicle model a manufacturer produces. While no given model need
precisely meet its target (and virtually no model exactly meets its
target in the real-world), if a manufacturer finds itself producing
large numbers of vehicles that fall well short of their targets, it
will have to find a way of offsetting that shortfall, either by
increasing production of vehicles that exceed their targets, or by
taking advantage of compliance flexibilities and incentives, or the
manufacturer will be subject to civil penalties. Given that
manufacturers typically need to produce for sale vehicles that
consumers want to buy, and not all consumers value fuel economy, their
options for pursuing the former approach can often be limited.
The CAFE and CO2 programs both offer a number of
compliance flexibilities and incentives, discussed in more detail
below. For example, starting in model year 2017, manufactures have
flexibility to account for efficiency improvements in air conditioning
(A/C) systems and/or for the application fuel economy improving
technologies that increase fuel economy in the real-world, but that
are, in whole or in part, not accounted for (e.g., stop-start
technology, or high efficiency alternators) using the 1975-based 2-
cycle compliance dynamometer test procedures.\3010\ These fuel economy
improvements are added to the 2-cycle performance results and are
included in the calculation of a manufacturer's fuel economy in
determining compliance relative to standards. In addition, for MYs
2017--2021, there are also two levels of compliance incentives for
full-size pickup trucks with mild-HEV or strong-HEV technology or that
overperform standards by 15 percent or more, or by 20 percent or
more.\3011\ This final rule removes this incentive starting in MY 2022,
as discussed in more detail below. These fuel economy improvements are
also included, for those model years and as earned, in the
[[Page 25189]]
calculation of a manufacturer's fuel economy.\3012\
---------------------------------------------------------------------------
\3010\ EPA regulations provided an equivalent program beginning
in MY 2012.
\3011\ Manufacturers also must apply the technology to a minimum
percentage of their full-size pickup truck production.
\3012\ NHTSA characterizes any programmatic benefit
manufacturers can use to comply with CAFE standards that fully
accounts for fuel use as a ``flexibility'' (e.g., credit trading)
and any benefit that counts less than the full fuel use as an
``incentive'' (e.g., adjustment of alternative fuel vehicle fuel
economy). NHTSA flexibilities and incentives are discussed further
in Section IX.D.
---------------------------------------------------------------------------
Some flexibilities and incentives are expressly provided for by
statute, and some have been implemented by the agencies through
regulations, consistent with the statutory scheme. Compliance
flexibilities and incentives for the CAFE and CO2 programs
have a great deal of theoretical attractiveness: If designed properly,
they can help to reduce overall regulatory costs, while maintaining or
improving programmatic benefits. If designed poorly, they may create
significant potential for market distortion (for instance, when
manufacturers--in response to an incentive to deploy a particular type
of technology--produce vehicles for which there is no natural market,
such vehicles must be discounted in order to sell).\3013\
Manufacturers' use of compliance flexibilities and incentives requires
proper governmental and industry collaboration for manufacturers to
achieve the most effective pathways to compliance.\3014\ Overly-
complicated flexibility and incentive programs can result in greater
expenditure of both private sector and government resources to track,
account for, and manage. Moreover, flexibilities or incentives that
tend to favor specific technologies could distort the market. By these
means, compliance flexibilities or incentives could create an
environment in which entities are encouraged to invest in such favored
technologies and, unless those technologies are independently supported
by market forces, encourage rent seeking in order to protect, preserve,
and enhance profits of companies that seek to take advantage of the
distortions created by government mandate. Further, to the extent that
there is a market demand for vehicles with lower CO2
emissions and higher fuel economy, compliance flexibilities and
incentives may cause some manufacturers to fall behind the industry's
pace if they become overly reliant on them rather than simply improving
the efficiency of their vehicles to meet that market demand.
---------------------------------------------------------------------------
\3013\ While many manufacturers publicly discuss their
commitment to certain technologies that reduce CO2
emissions, consumer interest in them thus far remains low, despite
often-large financial incentives from both manufacturers and the
Federal and State governments in the form of tax credits (i.e.,
natural gas or fuel-cell vehicles). It is questionable whether
continuing to provide significant compliance incentives for
technologies that consumers appear not to want is an efficient means
to achieve either compliance or national goals (see, e.g., Congress'
phase-out of the AMFA dual-fueled vehicle incentive in EISA, 49
U.S.C. 32906).
\3014\ For these reasons, in this final rule, NHTSA is asking
manufacturers to provide more detailed information on the new
incentives allowed for A/C and off-cycle technologies and on credit
trades for better collaboration in understanding the economic impact
of these flexibilities and incentives and for the government to
provide better oversight of the CAFE program.
---------------------------------------------------------------------------
If standards are maximum feasible levels, as required by statute,
then the need for extensive compliance flexibilities and incentives
should be low. The agencies sought comments in the NPRM on whether and
how each agency's existing flexibilities and incentives might be
amended, revised, or deleted to avoid the inefficiencies and market
distortions discussed above. Specifically, comments were sought on the
appropriate level of compliance flexibility, including credit trading,
in a program that is correctly designed to be maximum feasible, in
accordance with the statute. Comments were also sought on whether to
allow all incentive-based adjustments, except those that are mandated
by statute, to expire, in addition to other possible simplifications to
reduce market distortion, improve program transparency and
accountability, and improve overall performance of the compliance
programs. The agencies considered comments on those issues in preparing
the final rule. A summary of all the flexibilities for the CAFE and
CO2 programs finalized as a part of this final rule is
provided in Table IX-1 though Table IX-4.
BILLING CODE 4910-59-P
[[Page 25190]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.746
[[Page 25191]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.747
[[Page 25192]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.748
[[Page 25193]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.749
[[Page 25194]]
BILLING CODE 4910-59-C
2. Light-Duty CAFE Compliance Data for MYs 2011-2019
To understand manufacturers' potential approaches to using
compliance flexibilities and incentives, CAFE compliance data for MYs
2011 through 2019 is discussed in this section. NHTSA believes that
providing these data is important because it gives the public a better
understanding of current compliance trends and the potential impacts
that increasing CAFE standards have had on those model years and future
model years addressed by this rulemaking.
NHTSA uses data from CAFE reports submitted by manufacturers to EPA
or directly to NHTSA to evaluate compliance with the CAFE program. The
data for MYs 2011 through 2017 include manufacturers' final compliance
data that have been verified by EPA.\3015\ The data for MYs 2018 and
2019 include the most recent projections from manufacturers' mid-model
year and final-model year reports submitted to EPA and NHTSA, as
required by 49 CFR part 537 and 40 CFR 600.512-12.\3016\ Because the
projections do not reflect final vehicle production levels, the EPA
verified final CAFE values may be slightly different than the
manufacturers' projections. MY 2011 was selected as the start of the
data because it represents the first compliance model year for which
manufacturers were permitted to trade and transfer credits.\3017\ MY
2019 is also important because it shows the projected performance of
the fleet two years after manufacturers were allowed to use new
flexibilities and incentives starting in MY 2017 to address increasing
CAFE standards.
---------------------------------------------------------------------------
\3015\ The data contain the latest information available from
manufacturers except certain low volume manufacturers complying with
standards under 49 CFR part 525.
\3016\ MY 2018 data come from information received in
manufacturers' final reports submitted to EPA according to 40 CFR
600.512-12 and MY 2019 data come from information received in
manufacturers' mid-model year CAFE reports submitted to NHTSA
according to 49 CFR part 537.
\3017\ 49 CFR 535.6(c).
---------------------------------------------------------------------------
Figure IX-2 through Figure IX-5 provide a graphical overview of
fuel economy performance and standards. Fuel economy performance
includes three parts: (1) Measured performance, on the 2-cycle
dynamometer test; (2) performance increases for alternative fueled
vehicles, under the Alternative Motor Fuels Act of 1988 (AMFA); and (3)
performance adjustments for improved A/C systems and off-cycle
technologies.3018 3019 3020 These Figures do not account for
credits earned or expected to be earned from overcompliance in prior or
future model years that were used or are available for complying with
CAFE standards. Graphs are included for the total fuel economy
performance (the combination of all passenger cars and light trucks
produced for sale during the model year) as a single fleet, and for
each of the three CAFE compliance fleets: Domestic passenger car,
import passenger car, and light truck fleets.
---------------------------------------------------------------------------
\3018\ In the Figures, the label ``CAFE with Capped AMFA''
represents the maximum increase each year in the average fuel
economy set to the limitation ``cap'' for manufacturers attributable
to dual-fueled automobiles as prescribed in 49 U.S.C. 32906. The
labels ``A/C'' and ``off-cycle'' represents the increase in the
average fuel economy adjusted for A/C and off-cycle fuel consumption
improvement values as prescribed by 40 CFR 600.510-12.
\3019\ The Alternative Motor Fuels Act (AMFA) allows
manufacturers to increase their fleet fuel economy performance
values by producing dual-fueled vehicles. Incentives are available
for building advanced technology vehicles such as hybrids and
electric vehicles, compressed natural gas vehicles and for building
vehicles able to run on dual-fuels such as E85 and gasoline. For MYs
1993 through 2014, the maximum possible increase in CAFE performance
is ``capped'' for a manufacturer attributable to dual-fueled
vehicles at 1.2 miles per gallon for each model year and thereafter
decreases by 0.2 miles per gallon each model year through MY 2019.
49 U.S.C. 32906.
\3020\ Consistent with applicable law, NHTSA established
provisions starting in MY 2017 allowing manufacturers to increase
fuel economy performance-based on fuel consumption benefits gained
by technologies not accounted for during normal 2-cycle EPA
compliance testing (called ``off-cycle technologies'' for
technologies such as stop-start systems) as well as for A/C systems
with improved efficiencies and for hybrid or electric full-size
pickup trucks.
---------------------------------------------------------------------------
BILLING CODE 4910-59-P
[[Page 25195]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.750
[[Page 25196]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.751
As shown in Figure IX-2, manufacturers' fuel economy performance
for the total fleet was better than the overall CAFE standard through
MY 2015. On average, the total fleet exceeded the overall CAFE
standards by approximately 0.9 mpg for MYs 2011 to 2015. Comparatively,
as shown in Figure IX-3 through Figure IX-5, for these same model
years, domestic and import passenger cars exceeded standards on average
by 2.1 mpg and 2.3 mpg, respectively. By contrast, for light trucks,
manufacturers on average fell below standards by 0.3 mpg.
For MYs 2016 through 2019, as shown in the Figures, NHTSA has
determined that the combined CAFE performance, including all
flexibilities and incentives, of the total fleet has or is expected to
be worse than the applicable CAFE standards, and increasingly so. The
domestic passenger car fleet is the only compliance category expected
to continue to be better than CAFE standards through MY 2018. But even
the overall domestic passenger car fleet
[[Page 25197]]
is expected to be worse than standards in MY 2019. The data show MYs
2016 through 2019 standards involve significant compliance challenges
for many vehicle manufacturers. This is evident in the fact that the
total fleet falls below the applicable CAFE standards on average by 0.6
mpg for these model years. Compliance challenges become even more
substantial when observing individual compliance fleets. The largest
individual performance shortfalls (i.e. the difference between CAFE
performance values and standards) exist for import passenger car
manufacturers, with an expected shortfall of 2.5 mpg in MY 2019,
followed by light truck manufacturers, with a shortfall of 1.4 mpg in
MY 2016.
Table IX-5 provides the numerical final CAFE performance values and
standards for MYs 2004 to 2017. Notably, there was an increase in total
fleet fuel economy of only 0.1 mpg for MY 2014, and no increase for MY
2016. In MY 2016, the total fleet's performance fell below the CAFE
standard by 0.5 mpg. An increase in the total fleet's CAFE performance
for MY 2017 was largely due to manufacturers gaining benefits from A/C
and off-cycle technologies. For MY 2017, the total fleet's CAFE
performance without A/C and off-cycle allowances increased by 0.1 mpg
compared to MY 2016. However, even combined with new flexibilities, the
total fleet's CAFE performance, for MY 2017, still falls below the CAFE
standard by 0.4 mpg.
[GRAPHIC] [TIFF OMITTED] TR30AP20.752
BILLING CODE 4910-59-C
Figure IX-6 provides a historical overview of the industry's use of
CAFE compliance flexibilities for addressing performance
shortfalls.\3021\ MY 2016 is the latest model year for which CAFE
compliance determinations are complete, and credit application and
civil penalty payment determinations made by the manufacturer.
Historically, manufacturers have generally resolved credit shortfalls
first by carrying forward any earned credits and then applying traded
credits. In MYs 2014 and 2015, the amount of credit shortfalls is
almost the same as the amount of carry-forward and traded credits.
Manufacturers occasionally carryback credits or opt to transfer earned
credits between their fleets to resolve performance shortfalls. Trading
credits from another manufacturer and transferring them across fleets
occurs far more frequently. Also, credit trading has generally taken
the place of civil penalty payments for resolving performance
shortfalls. Only a handful of manufacturers have made civil penalty
payments since the implementation of the credit trading program.\3022\
NHTSA expects there may be sufficient credits in manufacturers' credit
accounts to resolve all import passenger car and light truck
performance shortfalls expected through MY 2019. By statute,
manufacturers cannot use traded or transferred credits to address
performance shortfalls for failing to meet the minimum domestic
passenger car standards.\3023\ One domestic passenger car manufacturer
paid civil penalties for failing to comply with the minimum domestic
passenger car standards for MYs 2016 and 2017.\3024\ Additional
manufacturers are
[[Page 25198]]
expected to pay civil penalty payments for failing to comply with the
minimum domestic passenger cars standards for MYs 2018 through 2019.
---------------------------------------------------------------------------
\3021\ The Figure includes all credits manufacturers have used
in credit transactions to date. Credits contained in carryback plans
yet to be executed or in pending enforcement actions are not
included in the Figure.
\3022\ Six manufacturers have paid CAFE civil penalties since
credit trading began in 2011. Fiat Chrysler paid the largest civil
penalty total over the period, followed by Jaguar Land Rover and
then Volvo. See Summary of CAFE Civil Penalties Collected, CAFE
Public Information Center, https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Fines_LIVE.html.
\3023\ Congress prescribed minimum domestic passenger car
standards for domestic passenger car manufacturers and unique
compliance requirements for these standards in 49 U.S.C. 32902(b)(4)
and 32903(f)(2).
\3024\ Fiat Chrysler paid $77,268,702.50 in civil penalties for
MY 2016 and $79,376,643.50 for MY 2017 for failing to comply with
the minimum domestic passenger car standards for those MYs.
[GRAPHIC] [TIFF OMITTED] TR30AP20.753
The compliance data show that the rate at which industry has been
increasing fuel economy, as shown by the actual fuel economy of the
overall fleet, has not kept pace with the year-over-year increases in
the stringency of the standards since MY 2010. The margin of CAFE
overcompliance diminished steadily through MY 2015. In MY 2016, the
fuel economy of the fleet was worse than standards, and the margin of
the shortfall has or is projected to become worse through MY 2019.
Manufacturers have increasingly used CAFE compliance flexibilities and
paid more in civil penalties to address the growing CAFE shortfalls.
The data show use of these flexibilities is likely to increase at least
through 2019.
3. Shift in Sales Production From Passenger Cars to Light Trucks
The notable trend in the stagnant growth in the automotive
industry's CAFE performance is likely related to an increase in the
purchase of light trucks beginning with MY 2013. Light trucks had a
sharp spike in sales, increasing by a total of 5 percent from MYs 2013
to 2014. In MY 2014, light trucks comprised approximately 41 percent of
the total sales production volume of automobiles and has continued to
grow ever since. In comparison, for model year 2014, domestic passenger
cars represented 36 percent of the total fleet and import passenger
cars represented 23 percent. Both domestic and import passenger car
sales have continued to fall every year since MY 2013. Figure IX-7
shows the sales production volumes of light trucks and domestic and
import passenger cars for MYs 2004 to 2017. The proportion of light
trucks in the fleet, being driven by consumer demand and lower fuel
prices, raises some concern for the ability of that fleet to comply
with future CAFE standards. Historically, light truck fleets have
fallen below their associated CAFE standards and have had larger
performance shortages than either import and domestic passenger car
fleets. This trend is expected to continue, even with allowance for A/C
and off-cycle flexibilities. For MY 2019, NHTSA expects even greater
CAFE performance shortages in the light truck and import passenger car
fleets than in prior model years, based upon manufacturer's MMY
reports. The combined effect of these fuel economy shortages will
require manufacturers to rely heavily on compliance flexibilities or
pay civil penalties.
Another important factor in automobile sales production impacting
CAFE performance values involves increasing trends in the volume of
small SUVs and pickup trucks. These vehicles as a percentage of total
fleet increased from approximately 52 percent in MY 2012 to 63 percent
in MY 2017. As shown in Figure IX-8, small SUVs, with 4WD and 2WD
drivetrains, in particular have surpassed the sales production volumes
of all other vehicle classes over these the given model years. The
number of small and standard SUVs sold in the U.S. for MY 2017 nearly
doubled compared to sales in the U.S. for MY 2012. During that same
period, passenger car sales production as a total of vehicle sales
production decreased by approximately 11 percent. The combination of
low gas prices and the increased utility that SUVs provide may explain
the shift in sales production. Nonetheless, if the sales of these small
SUVs and pickup trucks continue to increase, NHTSA expects there will
be continued stagnation in the CAFE performance of the overall fleet.
BILLING CODE 4910-59-P
[[Page 25199]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.754
[[Page 25200]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.755
BILLING CODE 4910-59-C
4. Vehicle Classification
Before manufactures can comply with CAFE and CO2
standards, they must first determine how a vehicle is classified in
accordance with 49 CFR part 523, ``Vehicle Classification.'' In EPCA,
Congress designated some vehicles as passenger automobiles and some as
non-passenger automobiles. Vehicle classification, for purposes of the
light-duty CAFE and CO2 programs, refers to whether a
vehicle is classified as a passenger automobile (car) or a non-
passenger automobile (light truck).3025 3026 As discussed
previously, passenger cars and light trucks are subject to different
fuel economy and CO2 standards, and light trucks have less
stringent standards to accommodate their utility usage.
---------------------------------------------------------------------------
\3025\ See 40 CFR 86.1803-01. For the MYs 2012-2016 standards,
the MYs 2017-2025 standards, and this rule, EPA uses NHTSA's
regulatory definitions for determining which vehicles would be
subject to which CO2 standards.
\3026\ EPCA uses the terms ``passenger automobile'' and ``non-
passenger automobile;'' NHTSA's regulation on vehicle
classification, 49 CFR part 523, further clarifies the EPCA
definitions and introduces the term ``light truck'' as a plainer
language alternative for ``non-passenger automobile.''
---------------------------------------------------------------------------
Under EPCA and NHTSA's current regulations, vehicles are classified
as light trucks either on the basis of off-highway capability or on the
basis of having truck-like (utility)
characteristics.3027 3028 3029 Determining whether a vehicle
is capable of ``off-highway operation'' is a two-part determination:
First, does the vehicle either have 4-wheel drive or a gross vehicle
weight rating (GVWR) over 6,000 pounds, and second, does the vehicle
(that has either 4-wheel drive or over 6,000 pounds GVWR) also have ``a
significant feature . . . designed for off-highway operation.'' \3030\
NHTSA's current regulations specify that this ``significant feature''
requires the vehicle to meet at least four out of five ground clearance
dimensions.\3031\ Further, to be classified as a light truck on the
basis of having truck-like characteristics instead, NHTSA regulations
also require the vehicle to perform at least one of the following
[[Page 25201]]
functions: Carry more than 10 persons, provide temporary living
quarters, have an open bed (i.e., a pickup truck), provide more cargo-
carrying volume than passenger-carrying volume, or permit expanded
cargo volume capacity by the removal or stowing of rear seats.\3032\
---------------------------------------------------------------------------
\3027\ 49 U.S.C. 32901(a)(18); 49 CFR part 523.
\3028\ 49 CFR 523.5(b).
\3029\ 49 CFR 523.5(a).
\3030\ 49 U.S.C. 32901(a)(18).
\3031\ The ground clearance dimensions are: (i) Approach angle
of not less than 28 degrees; (ii) breakover angle of not less than
14 degrees; (iii) departure angle of not less than 20 degrees; (iv)
running clearance of not less than 20 centimeters; and/or (v) front
and rear axle clearances of not less than 18 centimeters each.
\3032\ By statute, vehicles that NHTSA, on behalf of the
Secretary of DOT, ``decides by regulation [are] manufactured
primarily for transporting not more than 10 individuals'' are
passenger automobiles. 49 U.S.C. 32901(a)(18).
---------------------------------------------------------------------------
Over time, NHTSA has revised its light truck vehicle classification
regulations and issued legal interpretations to address changes in
vehicle designs. Based upon agency observations of current vehicle
design trends, compliance testing and evaluation, and discussions with
stakeholders, NHTSA has become aware of certain additional design
changes that further complicate light truck classification
determinations for the CAFE and CO2 programs. NHTSA
discussed several classification issues in the NPRM and sought comments
on potential resolutions. Only a few comments were received, primarily
from vehicle manufacturers, and they were aimed generally at requesting
flexibility in how NHTSA applies the existing classification criteria.
A summary of the comments received and NHTSA's responses for the final
rule are explained in the following sections.
a) Classification Based on ``Truck-Like Characteristics''
One of the ``truck-like characteristics'' that allows manufacturers
to classify vehicles as light trucks is having at least three rows of
seats as standard equipment, as long as the design also ``permit[s]
expanded use of the automobile for cargo-carrying purposes or other
non-passenger-carrying purposes through the removal or stowing of
foldable or pivoting seats so as to create a flat, leveled cargo
surface extending from the forwardmost point of installation of those
seats to the rear of the automobile's interior.'' \3033\ Typically,
most minivans qualify under the provision by expanding the cargo area
through removable or stowable seats, and a small percentage of sports
utility vehicles qualify through folding seats that use the seat backs
to form a secondary ``raised'' cargo floor.\3034\ NHTSA identified two
issues with this criterion that various manufacturers appear to be
approaching differently. Both relate to how expanded cargo area is
provided when seats are removed or stowed in the vehicle.
---------------------------------------------------------------------------
\3033\ 49 CFR 523.5(a)(5)(ii).
\3034\ All minivans and a small percentage of sports utility
vehicles that qualify as light trucks do so by meeting the
characteristic for third row seats. As more advanced seating designs
are introduced in minivans, manufacturers that wish to retain this
status will need to avoid losing the expanded cargo characteristics
that are the basis for the allowing minivans to be qualified as
light trucks.
---------------------------------------------------------------------------
The first issue is how to identify the ``forwardmost point of
installation'' and how the location impacts the available cargo floor
area and volume behind the seats. Seating configurations have evolved
considerably over the last twenty years, as minivan seats are now very
complex in design, providing far more ergonomic functionality. For
example, the market demand for increased rear seat leg room has
resulted in adjustable second row seats mounted to sliding tracks.
Earlier seating designs had fixed attachment points on the vehicle
floor, and it was easy to identify the ``forwardmost point of
installation'' because it was readily observable and did not change.
When seats move forward and backward on sliding tracks, however, the
``forwardmost point of installation'' is less readily identifiable. To
avoid this complication, most manufacturers maintain light truck
qualification by using adjustable seats that can be removed from the
vehicle and having a flat floor rearward of the front seats.\3035\ For
others, the qualification is not as apparent because new adjustable
seats have been introduced that remain within vehicle to accommodate
side airbags. Manufacturers designate various positions for the
forwardmost point of installation in vehicles where the seat in the
sliding track can be moved far enough forward to allow the entire seat
to compress against the back of the front seat where it can be stowed
beyond the forwardmost point of installation, while the seat cushion
bottom folds towards the seatback. In some cases, manufacturers
designate the forwardmost point of installation at a location in the
sliding track where the seat is positioned at its rearmost position in
the track. In others, the initial point of installation is designated
at a location in the sliding track accommodating the seating position
of a 75-percentile male test dummy. The amount of the flat floor
surface area and cargo volume behind the seats can vary depending on
which approach a manufacturer adopts.
---------------------------------------------------------------------------
\3035\ NHTSA notes that to qualify as a light truck, a vehicle
still requires a flat floor from the forwardmost point of
installation of removable second row seats to the rear of the
vehicle.
---------------------------------------------------------------------------
NHTSA sought public comments in the NPRM to explore potential
options for establishing the forwardmost point of installation for
adjustable second row seats and to evaluate whether an additional
classification criteria could be required, specifying a minimum amount
of cargo volume behind the seats. Comments were received from the Auto
Alliance and Fiat Chrysler.\3036\ Both the Auto Alliance and Fiat
Chrysler commented that some flexibility is needed in determining the
forwardmost point of installation that allows manufacturers to set the
location of the seat attachment point to the sliding track in any
manufacturer-designated position that allows for customer-ergonomics
and safety, while still meeting the spirit of the expanded cargo-
carrying requirement.\3037\ The Auto Alliance further commented that
the forwardmost attachment point of the seat structure to the floor is
still a viable method of measurement, even when there is a sliding
track between the floor attachment point and the seat.\3038\
---------------------------------------------------------------------------
\3036\ The National Automobile Dealers Association commented
generally that it does not support any substantial modifications to
the existing passenger car and light truck fleet definitions.
\3037\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3038\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
NHTSA did not propose any vehicle reclassifications and is not
adopting a regulatory change at this time. Based on its review of the
comments, NHTSA agrees that flexibility is warranted to accommodate
safety and customer demand but clarifies that the regulation requires
seats that are not removed to be stowed--that is, moved so as to form a
cargo area behind the seats. Manufacturers can freely designate the
seating location in the sliding track to establish the forwardmost
point of installation. At that seat location, the forwardmost point of
installation is the forwardmost attachment point of the seat structure
(including any carriage structures) to the floor in the sliding track.
Vehicles will be considered to meet the characteristic provided the
rear of the seats can be moved forward beyond that point and the seats
articulate to an unusable stowed position either in the floor of the
vehicle or at the front perimeter of expanded cargo area.\3039\
---------------------------------------------------------------------------
\3039\ The front perimeter of the cargo area is the plane formed
behind the front seats and extending from one side of the vehicle to
the other.
---------------------------------------------------------------------------
The second issue concerns the ``flatness'' and ``levelness'' of
folded rear seats that use the seat backs to form a raised cargo
surface and whether the seats must form a continuous flat, leveled
surface. Many SUVs have three rows of designated seating positions,
where the second row has ``captain's seats'' (i.e., two independent
bucket
[[Page 25202]]
seats), rather than the traditional bench-style seating more common
when the provision was added to NHTSA's regulation. When captain's
seats are folded down, the seatback can form a flat surface for
expanded cargo-carrying purposes, but the surface of the seatbacks may
be angled (i.e., at some angle slightly greater than 0[deg]), or may be
at a different level with the rest of the cargo area (i.e., horizontal
surface of folded seats is 0[deg] at a different height from horizontal
surface of cargo area behind the seats). Captain's seats, when folded
flat, may also leave significant gaps around and between the seats.
Some manufacturers have opted to use plastic panels to level the
surface and to covers the gaps between seats, while others have left
the space open and the surface angled or at different levels. NHTSA
sought comments in the NPRM on the following questions related to the
requirement for a flat, leveled cargo surface:
Does the cargo surface need to be flat and level in
exactly the same plane, or does it fulfill the intent of the criterion
and provide appropriate cargo-carrying functionality for the cargo
surface to be other than flat and level in the same plane?
Does the cargo surface need to be flat and level across
the entire surface, or are (potentially large) gaps in that surface
consistent with the intent of the criterion and providing appropriate
cargo-carrying functionality? Should panels to fill gaps be required?
Certain third row seats are located on top the rear axle
causing them to sit higher and closer to the vehicle roof. When these
seats fold flat the available cargo-carrying volume is reduced. Is
cargo-carrying functionality better ensured by setting a minimum amount
of useable cargo-carrying volume in a vehicle when seats fold flat?
The Auto Alliance, Fiat Chrysler, Hyundai, Kia, and one individual,
Walter Kreucher, commented on these seating issues. The Auto Alliance,
Fiat Chrysler, and Walter Kreucher believed that the criteria for a
``flat, leveled cargo surface'' should not be interpreted to mean that
a cargo surface must be flat and level in exactly the same plane.\3040\
The comments noted that a surface that is not exactly flat and level in
the same plane can still provide substantial cargo-carrying capacity,
while allowing manufacturers to provide ergonomically comfortable seats
that meet safety requirements.\3041\ The comments stated that NHTSA
should not establish a minimum amount of cargo surface area for seats
that remain within the vehicle.\3042\ Instead, they preferred that
manufacturers should be allowed to determine the methodology for
providing appropriate cargo-carrying functionality without NHTSA
stipulating additional requirements for flat and level surfaces or gaps
and gap-filling panels.\3043\
---------------------------------------------------------------------------
\3040\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Walter
Kreucher, Detailed Comments, NHTSA-2018-0067-0444.
\3041\ See, e.g., Fiat Chrysler, Detailed Comments, NHTSA-2018-
0067-11943.
\3042\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
\3043\ See, e.g., Fiat Chrysler, Detailed Comments, NHTSA-2018-
0067-11943.
---------------------------------------------------------------------------
The Auto Alliance and Fiat Chrysler argued that area or volume
requirements are not needed, as those attributes speak to overall
vehicle size and shape, which should remain a consumer choice.\3044\
The requirements for expanded cargo- or other non-passenger-carrying
purposes are fully met in the existing regulation, which requires a
flat, leveled cargo surface with two rows of seats that are folded or
stowed. Fiat Chrysler also commented that potential new requirements
would likely be interpreted and executed differently across
manufacturers and could narrow the choice of engineering solutions and
negatively affect other important vehicle attributes.\3045\
---------------------------------------------------------------------------
\3044\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3045\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
Hyundai and Kia commented that instead of requiring panels, NHTSA
could limit the size of the gaps around and between folded seats.\3046\
In that case, manufacturers would have flexibility to use panels if
they wish but could take other measures to narrow gaps. On the other
hand, Walter Kreucher stated that NHTSA should allow gaps of any size
and not require the use of panels to cover them.\3047\
---------------------------------------------------------------------------
\3046\ Hyundai, Detailed Comments, EPA-HQ-OAR-2018-0283-4411;
Kia, Detailed Comments, EPA-HQ-OAR-2018-0283-4195.
\3047\ Kreucher, Detailed Comments, NHTSA-2018-0067-0444.
---------------------------------------------------------------------------
NHTSA is not adopting a regulatory change at this time. NHTSA
agrees with commenters that it should not require a minimum amount of
cargo surface area or volume for seats that remain within the vehicle,
which could be difficult to meet for certain vehicle sizes and shapes
that would otherwise be considered non-passenger vehicles. NHTSA agrees
that the amount of cargo volume should be a consumer choice. Setting a
minimum amount of cargo area or volume could have an adverse effect on
potential new car buyers.
NHTSA notes that there may also be safety considerations involved
with the requirement to have a flat, leveled cargo surface area formed
by seat backs. A flat, leveled cargo surface area could prevent objects
from having a ramp-like surface to gain momentum in rolling backwards
into the tailgate's interior surface, potentially causing stress or
damage on the tailgate's latching mechanism. For these reasons, several
standards exist in the industry for preventing objects from sliding,
such as standards from the American Disability Act (ADA) that specify
floor and ground design requirements for protecting wheelchair seated
occupants. In addition, objects resting on the tailgate could become a
hazard or source of injury for individuals opening the tailgate. At
this time, NHTSA accepts the commenters' position that having a cargo
surface area that is exactly flat and level in the same plane may not
be necessary. Comments did not provide enough information for NHTSA to
identify any changes to the existing requirements. Therefore, at this
time, NHTSA will retain its existing provisions for the stowing of
foldable or pivoting seats to create a flat, leveled cargo surface, but
NHTSA may consider conducting research in the future regarding these
issues. NHTSA has also determined that it should set not a limit on the
size of the gaps between folded seats at this time, although it may
consider adopting such limits in the future. NHTSA continues to
encourage manufacturers to consider the safety implications of all
aspects of their vehicle designs, including any angling of the seat
back cargo surface and whether it is appropriate to offer panels as
optional equipment for covering any large gap openings.
b) Issues That NHTSA Has Observed Regarding Classification Based on
``Off-Road Capability''
(1) Measuring Vehicle Characteristics for Off-Highway Capability
For a vehicle to qualify as off-highway capable, in addition to
either having 4WD or a GVWR more than 6,000 pounds, the vehicle must
have four out of five characteristics indicative of off-highway
operation.\3048\ These characteristics are:
---------------------------------------------------------------------------
\3048\ 49 CFR 523.5(b)(2).
---------------------------------------------------------------------------
An approach angle of not less than 28 degrees
A breakover angle of not less than 14 degrees
A departure angle of not less than 20 degrees
[[Page 25203]]
A running clearance of not less than 20 centimeters
Front and rear axle clearances of not less than 18
centimeters each
NHTSA's regulations require manufacturers to measure these
characteristics when a vehicle is at its curb weight, on a level
surface, with the front wheels parallel to the automobile's
longitudinal centerline, and the tires inflated to the manufacturer's
recommended cold inflation pressure.\3049\ Given that the regulations
describe the vehicle's physical position and characteristics at time of
measurement, NHTSA previously assumed that manufacturers would use
physical measurements of vehicles. In practice, NHTSA has instead
received from manufacturers a mixture of angles and dimensions from
design models (i.e., the vehicle as designed, not as actually produced)
and/or physical vehicle measurements.\3050\ When appropriate, the
agency will verify reported values by measuring production vehicles in
the field. NHTSA currently requires that manufacturers use physical
vehicle measurements as the basis for values reported to the agency for
purposes of vehicle classification. NHTSA sought comment on whether
regulatory changes are needed with respect to this issue.
---------------------------------------------------------------------------
\3049\ Id.
\3050\ NHTSA previously encountered a similar issue when
manufacturers reported CAFE footprint information. In the October
2012 final rule, NHTSA clarified manufacturers must submit footprint
measurements based upon production values. 77 FR 63138 (October 15,
2012).
---------------------------------------------------------------------------
(2) Approach, Breakover, and Departure Angles
Approach angle, breakover angle, and departure angle are relevant
to determining off-highway capability. Large approach and departure
angles ensure the front and rear bumpers and valance panels have
sufficient clearance for obstacle avoidance while driving off-road. The
breakover angle ensures sufficient body clearance from rocks and other
objects located between the front and rear wheels while traversing
rough terrain. Both the approach and departure angles are derived from
a line tangent to the front (or rear) tire static loaded radius arc
extending from the ground near the center of the tire patch to the
lowest contact point on the front or rear of the vehicle. The term
``static loaded radius arc'' is based upon the definitions in SAE J1100
and J1544. The term is defined as the distance from wheel axis of
rotation to the supporting surface (ground) at a given load of the
vehicle and stated inflation pressure of the tire (manufacturer's
recommended cold inflation pressure).\3051\
---------------------------------------------------------------------------
\3051\ 49 CFR 523.2.
---------------------------------------------------------------------------
The static loaded radius arc is easy to measure, but the imaginary
line tangent to the static loaded radius arc is difficult to ascertain
in the field. The approach and departure angles are the angles between
the line tangent to the static loaded radius arc and the level ground
on which the test vehicle rests. Simpler measurements that provide good
approximations for the approach and departure angles involve using
either a line tangent to the outside diameter or perimeter of the tire
or a line that originates at the geometric center of the tire contact
patch and extends to the lowest contact point on the front or rear of
the vehicle. The first method provides an angle slightly greater than,
and the second method provides an angle slightly less than, the angle
derived from the true static loaded radius arc. Both approaches can be
used to measure angles in the field to verify data submitted by the
manufacturers used to determine light truck classification decisions.
NHTSA sought comment on what the effect would be if it replaced
reference to the ``static loaded arc radius'' with a different term
like ``outside perimeter of the tire'' or ``geometric center of the
tire contact patch.'' The Auto Alliance and Fiat Chrysler offered
comments. The Auto Alliance and Fiat Chrysler commented that only a
measurement using the static loaded arc radius reasonably reflects the
tire condition during off-road events that approach, breakover, and
departure angles are quantifying. They also stated the static loaded
arc radius best reflects the actual condition that exists versus the
outside tire diameter.\3052\ Finally, the Auto Alliance commented the
static loaded arc radius is easy to measure; therefore, the off-road
criteria should remain tied to the static loaded arc radius.\3053\
---------------------------------------------------------------------------
\3052\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3053\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
After reviewing the comments, NHTSA agrees that the static loaded
arc radius is the most accurate way to account for the condition of the
tire and the vehicle-to-ground interaction during off-road events.
NHTSA has decided to accept the Auto Alliance's and Fiat Chrysler's
views and will retain the existing definitions for off-road angles
based upon the static loaded arc radius.
(3) Running Clearance
NHTSA regulations define ``running clearance'' as ``the distance
from the surface on which an automobile is standing to the lowest point
on the automobile, excluding unsprung weight.'' \3054\ Unsprung weight
includes the components (e.g., suspension, wheels, axles, and other
components directly connected to the wheels and axles) that are
connected and translate with the wheels. Sprung weight, on the other
hand, includes all components fixed underneath the vehicle and
translate with the vehicle body (e.g., mufflers and subframes). To
clarify these requirements, NHTSA previously issued a letter of
interpretation stating that certain parts of a vehicle--such as tire
aero deflectors that are made of flexible plastic, bend without
breaking, and return to their original position--would not count
against the 20-centimeter running clearance requirement.\3055\ The
agency explained that this does not mean a vehicle with less than 20-
centimeters running clearance could be elevated by an upward force that
bends the deflectors and still be considered compliant with the running
clearance criterion, as it would be inconsistent with the conditions
listed in the introductory paragraph of 49 CFR 523.5(b)(2). Further,
NHTSA explained that without a flexible component installed, the
vehicle must meet the 20-centimeter running clearance along its entire
underside. This 20-centimeter clearance is required for all sprung
weight components.
---------------------------------------------------------------------------
\3054\ Id.
\3055\ See letter to Mark D. Edie, Ford Motor Company, July 30,
2012, available at https://isearch.nhtsa.gov/files/11-000612%20M.Edie%20(Part%20523).htm.
---------------------------------------------------------------------------
The agency is aware of vehicle designs that incorporate rigid
(i.e., inflexible) air dams, valance panels, exhaust pipes, and other
components, equipped as manufacturers' standard or optional equipment
(e.g., running boards and towing hitches), that likely do not meet the
20-centimeter running clearance requirement. Despite these rigid
features, it appears manufacturers are not taking these components into
consideration when making measurements. Additionally, NHTSA believes
some manufacturers may provide dimensions for their base vehicles
without considering optional or various trim level components that may
reduce the vehicle's ground clearance. Consistent with our approach to
other measurements, NHTSA believes that ground clearance, as well as
all the other off-highway criteria for a light truck determination,
should use the measurements from vehicles with all standard and
optional equipment
[[Page 25204]]
installed, at the time of the first retail sale.\3056\ The agency
reiterates that the characteristics listed in 49 CFR 523.5(b)(2) are
characteristics indicative of off-highway capability. A fixed feature--
such as an air dam that does not flex and return to its original state
or an exhaust that could detach--inherently interferes with the off-
highway capability of these vehicles. If manufacturers seek to classify
these vehicles as light trucks under 49 CFR 523.5(b)(2) and the
vehicles do not meet the four remaining characteristics to demonstrate
off-highway capability, they must be classified as passenger cars.
---------------------------------------------------------------------------
\3056\ See NHTSA's footprint test procedure for verifying CAFE
standards uses vehicles equipped at time of first retail sale. See
TP-537-01 located at https://www.nhtsa.gov/vehicle-manufacturers/test-procedures.
---------------------------------------------------------------------------
In the NPRM, NHTSA sought public comments on how to consider
components such as air dams, exhaust pipes, and other hanging component
features--especially those that are inflexible--as relates to running
clearance and whether the agency should consider amending its
definition in Part 523 to account for these components. The Auto
Alliance and three automobile manufacturers--Fiat Chrysler, Hyundai,
and Kia--commented on the questions. The Auto Alliance and Fiat
Chrysler commented that no change is needed for the 20-centimeter
running clearance requirement for fixed features of the vehicle; all
fixed components must have 20-centimeter of running clearance.\3057\
They agreed that flexible components that bend without breaking and
return to their original position do not count against the 20-
centimeter running clearance requirement.\3058\ They disagreed with
NHTSA's position that these requirements should apply to all vehicles
with standard and optional equipment installed at the time of the first
retail sale and proposed instead that the requirement should be ``as
shipped to the dealer.'' \3059\ Additionally, the Auto Alliance asked
NHTSA to make a specific allowance for vehicles that have adjustable
ride height, such as air suspension, and permit the running clearance
and other off-road clearance measurements to be made in the lifted or
off- road mode.\3060\ Hyundai and Kia urged NHTSA not to modify the
definition of ``running clearance,'' which currently is defined as
``the distance from the surface on which an automobile is standing to
the lowest point on the automobile, excluding unsprung weight.'' \3061\
---------------------------------------------------------------------------
\3057\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3058\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3059\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3060\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3061\ Hyundai, Detailed Comments, EPA-HQ-OAR-2018-0283-4411;
Kia, Detailed Comments, EPA-HQ-OAR-2018-0283-4195.
---------------------------------------------------------------------------
Based upon the comments above, NHTSA has decided to retain its
running clearance requirements for qualifying light trucks without
change. First, running clearance means the distance from the surface on
which an automobile is standing to all fixed components under the
vehicle, excluding unsprung components, axle clearance components and
flexible components that bend without breaking and returning to their
original position as explained in NHTSA's previous interpretation.
Second, NHTSA acknowledges that at this time, during validation testing
for running clearance, a vehicle with optional equipment installed will
only be tested ``as shipped to the dealer.'' NHTSA has found that
optional equipment can impact a vehicle's ability to comply with
running clearance requirements, while optional equipment must be
considered for other light truck agency validation tests unless the
equipment has no impact on the outcome of the test.
(4) Front and Rear Axle Clearance
NHTSA regulations state that front and rear axle clearances of not
less than 18 centimeters are another criterion that can be used for
designating a vehicle as off-highway capable.\3062\ The agency defines
``axle clearance'' as the vertical distance from the level surface on
which an automobile is standing to the lowest point on the axle
differential of the automobile.\3063\
---------------------------------------------------------------------------
\3062\ 49 CFR 523.5(b)(2)(v).
\3063\ 49 CFR 523.2.
---------------------------------------------------------------------------
The agency believes this definition may be outdated because of
vehicle design changes, including axle system components and
independent front and rear suspension components. In the past,
traditional light trucks with and without 4WD systems had solid rear
axles with center- mounted differentials on the axle. For these trucks,
the rear axle differential was closer to the ground than any other axle
or suspension system component. This traditional axle design still
exists today for some trucks with a solid chassis (also known as body-
on-frame configuration). Today, however, many SUVs and CUVs that
qualify as light trucks are constructed with a unibody frame and have
unsprung (e.g., control arms, tie rods, ball joints, struts, shocks,
etc.) and sprung components (e.g., the axle subframes) connected
together as a part of the axle assembly.\3064\ These unsprung and
sprung components are located under the axles, making them lower to the
ground than the axles and the differential, and were not contemplated
when NHTSA established the definition and the allowable clearance for
axles. The definition also did not originally account for 2WD vehicles
with GVWRs greater than 6,000 pounds that had one axle without a
differential, such as the model year 2018 Ford Expedition. Vehicles
with axle components that are low enough to interfere with the
vehicle's ability to perform off-road would seem inconsistent with the
regulation's intent of ensuring off-highway capability, as Congress
required.\3065\
---------------------------------------------------------------------------
\3064\ Unibody frames integrate the frame and body components
into a combined structure.
\3065\ 49 U.S.C. 32901(a)(18)(A).
---------------------------------------------------------------------------
In light of these issues, comments were sought in the NPRM on
whether (and if so, how) to revise the definition of axle clearance.
NHTSA sought comments on what unsprung axle components should be
considered when determining a vehicle's axle clearance. The agency
questioned whether the definition for axle clearance should be modified
to account for axles without differentials. NHTSA also sought comment
on whether the axle subframes surrounding the axle components but
affixed directly to the vehicle unibody as sprung mass (lower to the
ground than the axles) should be considered in the allowable running
clearance discussed above. Finally, NHTSA sought comments on whether it
should consider replacing both the running and axle clearance criteria
with a single ground clearance criterion that considers all components
underneath the vehicle that impact a vehicle's off-road capability.
Comments were received from the Auto Alliance, Fiat Chrysler,
Hyundai, and Kia. All the manufacturers that commented claimed no
change is needed to the current definition, regardless of whether the
axle components are sprung or unsprung masses, as the bottom of the
differential is the vulnerable component.\3066\ The Auto Alliance also
stated there is no
[[Page 25205]]
need to further modify the definition to account for axles without
differentials. Further, the Auto Alliance does not think a single
criterion that considers all components under the axle is needed and
prefers to keep the existing regulation.\3067\ Fiat Chrysler and the
Auto Alliance also recommended that 2WD SUVs and CUVs be reclassified
back into the truck fleet, where they had been placed prior to the 2011
MY. Their position is that 2WD SUVs are designed to meet the ``off-
road-capable'' definition in NHTSA's rules by having the required
running and/or axle clearances as well as meeting other off-road
dimensional criteria.\3068\ Hyundai stated that changing the point of
measurement now would have significant development and economic
impacts.\3069\ Kia stated that it has designed its vehicles and
developed product plans in reliance on the current definitions, and
those designs and product plans cannot be modified cheaply or
quickly.\3070\
---------------------------------------------------------------------------
\3066\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Hyundai,
Detailed Comments, EPA-HQ-OAR-2018-0283-4411; Kia, Detailed
Comments, EPA-HQ-OAR-2018-0283-4195.
\3067\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3068\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943;
Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3069\ Hyundai, Detailed Comments, EPA-HQ-OAR-2018-0283-4411.
\3070\ Kia, Detailed Comments, EPA-HQ-OAR-2018-0283-4195.
---------------------------------------------------------------------------
NHTSA already addressed the comments on 2WD SUVs in a previous
rulemaking, and NHTSA has no additional response at this time.\3071\
Upon review of other comments, manufacturers did not clearly
distinguish which parts of the axle sub-frames should be considered as
sprung masses in order for NHTSA to understand if modifications are
needed to its axle clearance requirements. Therefore, at this time,
NHTSA is retaining its axle clearance requirements as currently
specified. However, NHTSA still believes it is beneficial to continue
efforts at defining those axle components that are sprung or unsprung
masses before considering any changes to its regulatory provisions. In
addition, NHTSA needs to understand any significant developmental and
economic impacts that might be associated with any possible changes to
its requirements. Therefore, NHTSA will consider collecting further
information on these issues and may take further action related to this
issue in the future.
---------------------------------------------------------------------------
\3071\ No new arguments have been raised beyond those already
considered in the April 6, 2006, final rule (see 71 FR 17566).
---------------------------------------------------------------------------
B. EPA Compliance and Enforcement
1. Overview of the EPA Compliance Process
EPA established comprehensive vehicle certification, compliance,
and enforcement provisions for the GHG standards as part of the
rulemaking establishing the initial GHG standards for MY 2012-2016
vehicles.\3072\ Manufacturers have been following these provisions
since MY 2012 and EPA did not propose or seek comments on changing its
compliance and enforcement program.
---------------------------------------------------------------------------
\3072\ See 75 FR 25468-25488 and 77 FR 62884-62887 for a
description of these provisions. See also ``The 2018 EPA Automotive
Trends Report, Greenhouse Gas Emissions, Fuel Economy, and
Technology since 1975,'' EPA-420-R-19-002 March 2019 for additional
information regarding EPA compliance determinations.
---------------------------------------------------------------------------
a) What Compliance Flexibilities and Incentives are Currently Available
Under the CO2 Program and How Do Manufacturers Use Them?
Under EPA's regulations, manufacturers can use credit flexibilities
to comply with CO2 standards for passenger car or light
truck compliance fleets. Similar to the CAFE program, manufacturers
gain credits when the performance of a fleet exceeds its required
CO2 fleet average standard which can be carried forward for
five years. EPA also allows a one-time credit carry-forward exceeding 5
years, allowing MY 2010-2015 to be carried forward through MY2021. A
manufacturer's fleet performance that does not meet the fleet average
standard generates a credit deficit. Manufacturers can carry credit
deficits forward up to three model years before having to resolve the
shortfall.
NHTSA's program continues the 5-year carry-forward and 3-year
carryback, as required by statute. Credit ``transfer'' means the
ability of manufacturers to move credits from their passenger car fleet
to their light truck fleet, or vice versa. As part of the EISA
amendments to EPCA, NHTSA was required to establish by regulation a
CAFE credit transferring program, now codified at 49 CFR part 536, to
allow a manufacturer to transfer credits between its car and truck
fleets to achieve compliance with the standards. For example, credits
earned by over-compliance with a manufacturer's car fleet average
standard could be used to offset debits incurred because the
manufacturer did not meet the truck fleet average standard in a given
year.
Under Section 202(a) of the CAA, there is no statutory limitation
on car/truck credit transfers, and EPA's CO2 program allows
unlimited credit transfers across a manufacturer's car and light truck
fleets to meet CO2 standards.
EPA requested comment on a variety of ``enhanced flexibilities''
whereby EPA could make adjustments to current incentives and credit
provisions and potentially add new flexibility opportunities to expand
the means by which manufacturers may satisfy standards. Some of these
additional flexibilities would not result in a reduction in program
stringency, while others would incentivize technologies that could
realize greater CO2 emissions reductions over a longer term,
but would result in a loss of emission benefits in the short-term, as
discussed below. EPA requested comments on these topics to support the
increased application of technologies that the automotive industry is
developing and deploying that could potentially lead to further long-
term emissions reductions and allow manufacturers to comply with
standards while reducing costs.
EPA explained that one category of flexibilities, such as off-cycle
credits and credit banking, involve credits that are based on real
world emissions reductions and do not represent a loss of overall
emissions benefits or a reduction in program stringency, yet offer
manufacturers potentially lower-cost or more efficient path to
compliance. Another category of flexibilities, such as incentives for
battery electric vehicles, hybrid technologies, and alternative fuels,
do result in a loss of emissions benefit and represent a reduction in
the effective stringency of the standards to the extent the incentives
are used by manufacturers. These incentives would help manufacturers
meet a numerically more stringent standard, but would not reduce real-
world CO2 emissions in the short term compared to a lower
stringency option with fewer such incentives. EPA's policy rationale
for providing such incentives, as articulated in the 2012 rulemaking,
was that such programs could incentivize the development and deployment
of advanced technologies with the potential to lead to greater
CO2 emissions reductions in the longer-term, where such
technologies today are limited by higher costs, market barriers,
infrastructure, and consumer awareness.\3073\ Such incentive approaches
would also result in rewarding automakers who invest in certain
technological pathways, rather than being technology neutral.
---------------------------------------------------------------------------
\3073\ See 77 FR 62810-62826 (Oct. 15, 2012).
---------------------------------------------------------------------------
Prior to the proposal, automakers and other stakeholders expressed
support for
[[Page 25206]]
this type of compliance flexibility. For example, in March 2018, Ford
stated, ``We support increasing clean car standards through 2025 and
are not asking for a rollback. We want one set of standards nationally,
along with additional flexibility to help us provide more affordable
options for our customers.'' \3074\ Honda, in April 2018, also
expressed its support for an approach that retained the existing
standards while extending the advanced technology multipliers for
electrified vehicles, eliminated automakers' responsibility for the
impact of upstream emissions from the electric grid, and accommodated
more off-cycle technologies.\3075\
---------------------------------------------------------------------------
\3074\ ``A Measure of Progress'' Bill Ford, Executive Chairman,
Ford Motor Company, and Jim Hackett, President and CEO, Ford Motor
Company, March 27, 2018, https://medium.com/cityoftomorrow/a-measure-of-progress-bc34ad2b0ed.
\3075\ Honda Release ``Our Perspective--Vehicle Greenhouse Gas
and Fuel Economy Standards,'' April 20, 2018, http://news.honda.com/newsandviews/pov.aspx?id=10275-en.
---------------------------------------------------------------------------
EPA's request for comments was largely based on its consideration
of input from automakers and other stakeholders, including suppliers
and alternative fuels industries, supporting a variety of program
flexibilities.\3076\ The following provides an overview of EPA's
request for comments on several flexibility concepts, the comments EPA
received, and the agency's response to those comments. After
considering comments, EPA is not adopting new incentives in the areas
of credit multipliers (with the exception of multipliers for natural
gas vehicles), new incentives for hybrid vehicles, incentives for
autonomous or connected vehicles, or alternative fueled vehicles other
than natural gas, as part of this final rule. EPA is finalizing program
changes for the treatment of upstream emissions for electric vehicles,
the treatment of natural gas vehicles, the treatment of hybrid and
target-beating full-size pickup trucks, and off-cycle credits, as
discussed below.
---------------------------------------------------------------------------
\3076\ Memorandum to docket EPA-HQ-OAR-2018-0283 regarding
meetings with the Alliance of Automobile Manufacturers on April 16,
2018 and Global Automakers on April 17, 2018. EPA-HQ-OAR-2018-0283-
0022.
---------------------------------------------------------------------------
(1) Credit Flexibilities
Under the EPA program, CO2 credits may be carried
forward, or banked, for a period of five years, with the exception that
MY 2010-2015 credits may be carried forward and used through MY 2021.
CO2 credits may also be traded between manufacturers and
transferred between passenger car and light truck fleets similar to the
CAFE program, but without any adjustment for fuel savings. Under
Section 202(a) of the CAA, there is no statutory limitation on credit
transfers between a manufacturer's passenger car and light truck
fleets, and EPA's CO2 program allows unlimited credit
transfers across a manufacturer's passenger car and light truck fleets
to comply with CO2 standards. This flexibility is based on
the expectation that it will help facilitate manufacturer compliance
with CO2 standards in the lead time provided, and allow
CO2 emissions reductions to be achieved in the most cost
effective way.
Automakers suggested, prior to the NPRM proposal, a variety of ways
in which CO2 credit life could be extended under the CAA,
like allowing automakers to carry-forward MY 2010 and later banked
credits to MY 2025, extending the life of credits beyond five years, or
even unlimited credit life where credits would not expire. EPA
requested comments in the NPRM on extending credit carry-forward under
the CO2 program beyond the current five years, including
unlimited credit life.
General comments were received in response to the NPRM from the
National Automobile Dealers Association and Volkswagen. They commented
that credit carry-forward and carryback options help with annual
compliance with the CO2 program.\3077\ They stated that
these mechanisms allow manufacturers to become compliant over the
course of the time a credit is usable in the market.\3078\ Toyota,
General Motors, Fiat Chrysler, the Auto Alliance, and the Global
Automakers each commented that CO2 credits earned by
manufacturers need a longer life so they may be carried forward further
than the current five-year limitation.\3079\ They asked for an
unlimited period for using CO2 credits without restrictions,
since they argue that automakers have earned those credits and should
be allowed to use them however they see fit.\3080\ They also stated
that this would incentivize manufacturers to make early reductions in
CO2 emissions.\3081\ Furthermore, it was noted that credits
are earned when manufacturers achieve lower CO2 fleet
average emissions than otherwise required by regulation in any given
model year. They stated that this typically results from actions taken
by a manufacturer to deploy specific models or more efficient
technology than required, often at a higher cost. Such technologies
reduce the amount of CO2 emissions released into the
atmosphere over the life of the vehicle, which could be over several
decades. Therefore, the resulting credit earned by a manufacturer for
having made the product or technology investment that resulted in the
reduced emissions should not be limited to five years.
---------------------------------------------------------------------------
\3077\ National Automobile Dealers Association, Detailed
Comments, NHTSA-2018-0067-12064; Volkswagen, Detailed Comments,
NHTSA-2017-0069-0583.
\3078\ See, e.g., National Automobile Dealers Association,
NHTSA-2018-0067-12064.
\3079\ Toyota, Detailed Comments, NHTSA-2018-0067-12150; General
Motors, Detailed Comments, NHTSA-2018-0067-11858; Fiat Chrysler,
Detailed Comments, NHTSA-2018-0067-11943; Auto Alliance, Detailed
Comments, NHTSA-2018-0067-12073; Global Automakers, Detailed
Comments, NHTSA-2018-0067-12032.
\3080\ See, e.g., Global Automakers, Detailed Comments, NHTSA-
2018-0067-12032.
\3081\ See, e.g., General Motors, Detailed Comments, NHTSA-2018-
0067-11858.
---------------------------------------------------------------------------
Global Automakers, the Auto Alliance, Fiat Chrysler, and Toyota
requested a one-time expiration date extension through 2026 for
CO2 credits earned in MYs 2010-2015.\3082\ They asserted
that earned credits represent actual CO2 reductions and
increasing their lifespan will allow for better compliance. Conversely,
Honda disagreed with the extension of MY 2010-2015 credits through 2026
because they have been selling their credits under the assumption that
they would expire.\3083\ Honda stated that shorter life (soon to
expire) credits are worth less than longer life credits, leading to a
disadvantage for manufacturers who have already sold these credits at a
lower price. Honda asserted that the one-time extension would benefit
only a few automakers.\3084\ However, Honda did agree that a one-time
extension through 2026 for MYs 2016-2020 CO2 credits would
assist with compliance because these credits have yet to be involved in
trades.\3085\
---------------------------------------------------------------------------
\3082\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032; Alliance, Detailed Comments, NHTSA-2018-0067-12073; Fiat
Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Toyota Detailed
Comments, NHTSA-2018-0067-12150.
\3083\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3084\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3085\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
---------------------------------------------------------------------------
In sum, commenters requested either unlimited allowances to carry-
forward surplus credits without any expiration date, a one-time
expiration date extension through 2026 for CO2 credits
earned from MY 2010 and later, or consideration for extending credit
life longer than the current five-year provision. After considering the
comments received, EPA has decided not to change its credit carry-
forward provisions at this time, and will retain the credit carry-
forward period under the CO2 program at five years for
credits
[[Page 25207]]
generated in MYs 2016 and later. EPA does not believe any changes to
its credit carry-forward provisions are warranted. EPA notes that
NHTSA's CAFE program is constrained by statute to a five-year carry-
forward so if EPA adopted a longer carry-forward period, it might be of
limited use since the level of stringency of the CO2 and
CAFE standards is similar across the programs. Also, the analysis on
which the tailpipe CO2 emissions standards finalized today
are based, assumed a five-year carry-forward period for credits.
Another reason for denying manufacturers' requests is the potential
inequitable advantage a longer credit life could have for manufacturers
with surplus credits, especially those with significant amounts of
credits currently banked for multiple model years. Manufacturers
without credits, or manufacturers who have already sold their credits
at current market values based on the present five-year carry-forward
credit lifespan, as Honda discussed, will be significantly
disadvantaged.\3086\ These manufacturers are unlikely to be able to
renegotiate the price of credit trades already made. Manufacturers with
large amounts of credits would clearly be advantaged and able to
distort the market in ways unfavorable to the goal of reducing
emissions. EPA is concerned that these manufacturers will be able to
create uncertainties in the market by being able to infuse large
volumes of credits into future model years where it may even be
possible to delay some cost-effective technologies from entering
production because manufacturers are relying upon these credits as an
alternative pathway to compliance.
---------------------------------------------------------------------------
\3086\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
---------------------------------------------------------------------------
(2) Advanced Technology Incentives
The existing EPA CO2 program provides incentives for
electric vehicles, fuel-cell vehicles, plug-in hybrid vehicles, and
natural gas vehicles. The 2012 rulemaking allowed manufacturers to use
a 0 grams/mile emissions factor for all electric powered vehicles
rather than having to account for the CO2 emissions
associated with upstream electricity generation, up to a per-
manufacturer cumulative production cap for MYs 2022-2025. The program
also includes multiplier incentives that allow manufacturers to count
advanced technology vehicles as more than one vehicle in the compliance
calculations. The multipliers began with MY 2017 and end after MY
2021.\3087\ Prior to the proposal, stakeholders suggested that these
incentives should be expanded to support further the production of
advanced technologies by allowing manufacturers to continue to use the
0 grams/mile emissions factor for electric powered vehicles rather than
having to account for upstream electricity generation emissions and by
extending and potentially increasing the multiplier incentives.
---------------------------------------------------------------------------
\3087\ The multipliers are for EV/FCVs: 2017-2019--2.0, 2020--
1.75, 2021--1.5; for PHEVs and dedicated and dual-fuel CNG vehicles:
2017-2019--1.6, 2020--1.45, 2021--1.3.
---------------------------------------------------------------------------
First, EPA requested comments on extending the use of 0 grams/mile
emissions factor for electric powered vehicles.
The Auto Alliance, Global Automakers, and several manufacturers
commented that upstream utility emissions come from power plants, not
vehicle tailpipes, and manufacturers have no control over the feedstock
used by those power plants and should not be held responsible for their
upstream electricity emissions.\3088\ The Auto Alliance further
commented that removing upstream accounting is not an incentive for
advanced technology vehicles; rather, it should be seen as a correction
to remove responsibility for emissions over which the automakers have
no control.\3089\ Fiat Chrysler commented that ``requiring upstream
accounting could impede development of BEVs or PHEVs, as accounting of
upstream emissions degrades the CO2 performance of BEVs to
the level of PHEVs, and PHEVs to the level of a conventional hybrid
electric vehicle. This, in effect, disincentivizes the technology.''
\3090\
---------------------------------------------------------------------------
\3088\ See, e.g., Volkswagen, Detailed Comments, NHTSA-2017-
0069-0583.
\3089\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3090\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
Several other commenters also supported not counting upstream
emissions and instead only counting electric powered vehicle tailpipe
emissions of 0 grams/mile.\3091\ These commenters included NCAT, SAFE,
BorgWarner, CALSTART, Eaton, and Edison Electric Institute.
---------------------------------------------------------------------------
\3091\ See, e.g., NCAT, NHTSA-2018-0067-11969.
---------------------------------------------------------------------------
API did not support continuing the 0 grams/mile emission factor for
electricity use, commenting that by failing to factor the real
contribution of upstream CO2 emissions from electric
generation, the regulatory agencies would distort the market for
developing transportation fuel alternatives.\3092\ API commented that
EPA should not ignore the environmental burden of upstream emissions in
granting production incentives to automakers.
---------------------------------------------------------------------------
\3092\ API, Detailed Comments, EPA-HQ-OAR-2018-0283-5458.
---------------------------------------------------------------------------
Manufacturers of Emission Controls Association (MECA) commented
that ``with the growing emphasis on real-world emission reductions, it
becomes increasingly important to consider all emissions to the
environment, including upstream emissions. Numerous studies have shown
that in many parts of the country, the temporary 0 grams/mile upstream
emissions factor is not delivered in the real-world . . . MECA believes
that EPA should continue to set performance-based standards that assess
technology pathways based on delivering the intended emission
reductions over the full well-to-wheels vehicle life cycle in the real-
world.'' \3093\ Motor & Equipment Manufacturers Association (MEMA) also
supported a well-to-wheel fuel lifecycle approach, commenting that
without this type of comprehensive assessment on the fuel impacts and
comprehensive CO2 costs, policies improperly ``slant toward
preferred technologies.'' \3094\ Nonetheless, MEMA commented that it is
not opposed to continuing to allow 0 grams/mile emissions factor for
electric powered vehicles through 2026.
---------------------------------------------------------------------------
\3093\ MECA, Detailed Comments, NHTSA-2018-0067-11994.
\3094\ MEMA, EPA-HQ-OAR-2018-0283-5692. See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
The Union of Concerned Scientists (UCS) commented that not
accounting for upstream emissions combined with the multipliers has a
significant impact on the efficacy of the standard, and extending these
regulatory incentives is more likely to result in a credit giveaway
than to drive additional deployment of electric vehicles.\3095\ UCS
further commented that, to date, more than half of the electric
vehicles sold have been in California and the states that have adopted
California's ZEV standards; however, UCS asserted, federal standards
ignore the upstream emissions for all vehicles sold.
---------------------------------------------------------------------------
\3095\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
After carefully considering the wide range of comments on whether
to include upstream emissions associated with electricity use in the
compliance calculations for electrified vehicles, EPA has decided to
allow the continued use of the 0 grams/mile emissions factor with no
per-manufacturer production caps or other limitations. EPA is revising
its regulations to remove the production caps and related provisions.
When EPA initially adopted a production cap for manufacturers that
[[Page 25208]]
use the 0 grams/mile emissions factor, in the rulemaking to establish
CO2 standards for MY 2012-2016 vehicles, there were no
controls in place for CO2 emissions from electricity
production.\3096\ This was also the case when EPA extended the 0 grams/
mile upstream provision and revised the production caps in the rule
establishing MY 2017-2025 standards.\3097\ However, since then, EPA has
adopted a program to control CO2 emissions from power
plants.\3098\ Emissions from the power sector have been declining and
that trend is projected to continue.\3099\ For these reasons, EPA no
longer views the upstream emissions factor as an incentive in the same
way it views a multiplier incentive which provides bonus credits. EPA
agrees that, at this time, manufacturers should not account for
upstream utility emissions. Therefore, EPA is adopting regulatory
changes consistent with its historical practice of basing compliance
with vehicle emissions standards on tailpipe emissions through model
year 2026. EPA may choose to reconsider this decision in a future
CO2 rulemaking, and will reexamine the issue when
establishing standards commencing with the 2027 model year.\3100\
---------------------------------------------------------------------------
\3096\ 75 FR 25341, May 7, 2010.
\3097\ 77 FR 62816, October 15, 2012.
\3098\ 84 FR 32520, July 8, 2019.
\3099\ 84 FR 32561.
\3100\ By comparison, the CAFE program uses an energy efficiency
metric instead of an emissions metric, and standards that are
expressed in miles per gallon. For PHEVs and BEVs, to determine
gasoline the equivalent fuel economy for operation on electricity, a
Petroleum Equivalency Factor (PEF) is applied to the measured
electrical consumption. The PEF for electricity was established by
the Department of Energy, as required by statute, and includes an
accounting for upstream energy associated with the production and
distribution for electricity relative to gasoline. Therefore, the
CAFE program includes upstream accounting based on the metric that
is consistent with the fuel economy metric. The PEF for electricity
also includes an incentive that effectively counts only 15 percent
of the electrical energy consumed.
---------------------------------------------------------------------------
Second, EPA requested comments on extending or increasing advanced
technology incentives, including multiplier incentives, with
multipliers in the range of 2.0-4.5. EPA received a wide range of
comments both for and against increasing the multiplier incentives. The
MY 2017-2025 CO2 program finalized in 2012 included
incentive multipliers for certain advanced technologies for MY 2017-
2021 vehicles.
The Auto Alliance, Global Automakers, and several individual
manufacturers commented in support of continued and increased
multipliers. The Auto Alliance commented that EPA should extend and
significantly expand multipliers ``to encourage a transition to these
technologies while cost, range, and infrastructure challenges are
addressed to encourage ongoing investments in advanced technologies.''
\3101\ Global Automakers commented that multipliers should be included
through MY 2026, set at values that encourage ongoing investment in
advanced technologies, without diluting overall efficiency improvements
in the program.\3102\ NCAT, Eaton, Plug-in America, Alliance to Save
Energy, SAFE, and MEMA also supported additional multiplier incentives
to encourage further the production and sale of advanced technology
vehicles.\3103\
---------------------------------------------------------------------------
\3101\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3102\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
\3103\ NCAT, Detailed Comments, NHTSA-2018-0067-11969; Eaton,
Detailed Comments, EPA-HQ-OAR-2018-0283-5068; Plug-In America,
Detailed Comments, NHTSA-2018-0067-12028; Alliance to Save Energy,
Detailed Comments, NHTSA-2018-0067-11837; SAFE, Detailed Comments,
NHTSA-2018-0067-11981; see https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
EPA also received comments against extending the multiplier
credits. UCS commented that reducing the stringency of the standards
lessens the need for the adoption of these vehicles and undermines the
initial rationale for these credits, resulting in a significant bank of
credits which would further erode the benefits of these
standards.\3104\ American Council for an Energy-Efficient Economy
(ACEEE) commented that providing multiplier incentives for any longer
period, or at a greater rate than those currently in place, would
create windfall credits for manufacturers given the industry's current
product plans.\3105\ Fiat Chrysler commented generally in support of a
multiplier incentive, but noted that since multipliers are a
CO2--only flexibility not present in the CAFE program,
greater use of multipliers would result in further disharmonizing the
programs.\3106\ API commented against multipliers, stating that the
program should be technology neutral and that regulatory agencies
should not incentivize either producer or consumer investments in
government-selected technologies applied to government-selected vehicle
categories.\3107\
---------------------------------------------------------------------------
\3104\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3105\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122.
\3106\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3107\ API, Detailed Comments, EPA-HQ-OAR-2018-0283-5458.
---------------------------------------------------------------------------
In this final rule, EPA is neither adopting any additional EV or
FCV multipliers nor extending the existing multipliers scheduled to
phase out after MY 2021 for EVs, PHEVs, and FCVs. EPA is concerned that
additional multiplier incentives beyond those already in place for
these vehicles which are currently available to consumers would reduce
the emissions benefits associated with the program. As discussed below
in section IX.B.1.a.(3)(b), EPA is providing an additional multiplier
for dedicated and dual-fuel NGVs, which are not currently produced by
auto manufacturers, for MYs 2022-2026. The CO2 program
already provides a significant incentive for PHEVs, EVs, and FCVs by
only counting tailpipe emissions (not accounting for upstream
emissions).
(3) Special Considerations
(a) Incentives for Connected or Automated Vehicles
Connected and automated (including autonomous) vehicles have the
potential to impact significantly vehicle emissions in the future, with
their aggregate impact being either positive or negative, depending on
a large number of vehicle-specific and system-wide factors. EPA noted
in the proposal that connected or automated vehicles would be eligible
for credits under the off-cycle program if a manufacturer provides data
sufficient to demonstrate the real-world emissions benefits of such
technology applied to its vehicles. However, demonstrating the
incremental real-world benefits of these emerging technologies will be
challenging. Prior to the proposal, stakeholders suggested that EPA
should consider an incentive for these technologies without requiring
individual manufacturers to demonstrate real-world emissions benefits
of the technologies. A number of stakeholders also requested that EPA
consider credits for automated and connected vehicles that are placed
in ridesharing or other high mileage applications, where any potential
environmental benefits could be multiplied due to the high utilization
of these vehicles. EPA requested comment on such incentives as a way to
facilitate increased use of these technologies, including some level of
assurance that they will lead to future additional emissions
reductions. For example, EPA stated in the proposal that any near-term
incentive program should include some demonstration that the
technologies will be both truly new and have some connection to overall
environmental benefits. EPA further outlined and sought comment on
several approaches
[[Page 25209]]
to incentivize automated and connected vehicle technologies.
EPA received comments supporting and opposing incentives for
automated and connected vehicles. The Auto Alliance commented that the
agencies should incentivize the adoption of these technologies and
provide for possibly additional credit once the benefits beyond the
credit values have been confirmed.\3108\ It further commented that a
growing body of modeling results, as well as real-world driving
statistics, show that current automated driving technologies improve
real-world fuel efficiency and reduce CO2 emissions. SAFE
commented that connected automated vehicles have tremendous potential
to save lives, and when combined with ride-sharing and electric
powertrains, they can also increase efficiencies and save fuel.\3109\
SAFE argued that an initial review of the literature shows the
potential for these technologies to improve fuel economy by up to 25
percent when they are optimized and aggregated alongside other
traditional efficiency technologies. Toyota commented that automated
vehicles, and possibly new mobility models such as ridesharing, can
help attain societal goals concerning climate change, energy security,
traffic congestion, and safety.\3110\ Ford commented that it is
supportive of credits for future connected and automated vehicles and
that autonomous vehicles are considered the future of personal
mobility, with many manufacturers announcing plans to release
autonomous-capable vehicles in the near term.\3111\ Ford added that
these vehicles have the potential to not only provide meaningful real-
world CO2 and fuel economy benefits, but also add true
societal benefit for the public good by providing transportation to
those who would otherwise not have access. General Motors and Jaguar
Land Rover commented in favor of additional credits for vehicles placed
in ride-sharing or high mileage applications.\3112\
---------------------------------------------------------------------------
\3108\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3109\ SAFE, Detailed Comments, NHTSA-2018-0067-11981.
\3110\ Toyota, Detailed Comments, NHTSA-2018-0067-12150.
\3111\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
\3112\ General Motors, Detailed Comments, NHTSA-2018-0067-11858;
Jaguar Land Rover, Detailed Comments, NHTSA-2018-0067-11916.
---------------------------------------------------------------------------
SAFE commented that autonomous vehicles will lead to new jobs and
better worker productivity. It stated that these vehicles will also
reduce congestion and lead to safer travel.\3113\
---------------------------------------------------------------------------
\3113\ SAFE, Detailed Comments, NHTSA-2018-0067-11981.
---------------------------------------------------------------------------
Other commenters opposed incentives for automated and connected
vehicles, generally commenting that while the technologies are
promising, the impacts of the technologies remain highly uncertain and
therefore incentives are not appropriate. ACEEE commented that EPA
should not incentivize technologies such as automated vehicle
technology or ridesharing services, unless and until it can be
demonstrated that such an incentive will result in emissions reduction
benefits and will not undermine the existing standards.\3114\ ACEEE
believes that there currently exists no real-world data to justify
granting of off-cycle credits for automated vehicle technologies, and
that providing automakers credits for deploying technologies which are
driven by demands other than fuel savings and emissions reduction only
allows them to make fewer real-world emissions reductions elsewhere.
ACEEE further stated that while automated vehicles promise all-new
possibilities and efficiencies in transportation and the use of
infrastructure, the net impact on transportation sector energy use and
emissions is unknown.
---------------------------------------------------------------------------
\3114\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122.
---------------------------------------------------------------------------
UCS commented that the ``evidence to-date does not warrant
incentivizing such technologies--there is no provable environmental
benefit of such technologies, and the agencies have previously
correctly acknowledged that any such potential impacts would be related
to indirect benefits, which raise serious concerns about compliance and
enforcement to ensure the integrity of the program.'' \3115\ Honda
commented that there remains considerable uncertainty in the literature
regarding the energy and environmental benefits (or negative benefits)
of connected/automated vehicle technology.\3116\ Honda commented that
if technology benefits can be verified under robust, repeatable
conditions, they should warrant off-cycle credits under the existing
off-cycle program. Honda does not believe credits should be granted for
application of technology alone.
---------------------------------------------------------------------------
\3115\ U.S.C., Detailed Comments, NHTSA-2018-0067-12039.
\3116\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
---------------------------------------------------------------------------
CARB commented that new compliance flexibilities (or off-cycle
credit categories) for automated vehicles are not appropriate at this
time.\3117\ CARB believes that, although the technology is widely
expected to provide safety and mobility benefits, automakers are
expected to bring the technology to market regardless, so incentives
are unnecessary, and it is not established that these technologies will
reduce emissions given their potential for high annual mileage.
Resources for the Future commented they do not see a rationale for
providing special credits to automated vehicles since such vehicles
could increase or decrease emissions.\3118\ Competitive Enterprise
Institute (CEI) commented that some connected and/or automated vehicle
technology applications--namely platooning--may improve fuel efficiency
through improved aerodynamics and thus reduce CO2 emissions;
however, such applications to date are limited to heavy-vehicle
prototypes beyond the scope of this rulemaking and in any event should
be subject to verification prior to any award of off-cycle
credits.\3119\ CEI commented further: ``We urge EPA to preserve the
existing off-cycle program requirement that manufacturers demonstrate
CO2 emissions reductions prior to the award of credits,
rather than picking technology winners and losers that have nothing to
do with fuel economy or emissions.'' National Association of Truck Stop
Operators (NATSO) commented against incentives, stating that although
automated vehicles have the potential positively to transform
transportation (and indeed day-to-day life) in the U.S., there are also
a number of complexities and potential costs associated with
them.\3120\
---------------------------------------------------------------------------
\3117\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
\3118\ Resources for the Future, Detailed Comments, NHTSA-2018-
0067-11789.
\3119\ CEI, Detailed Comments, EPA-HQ-OAR-2018-0283-4166.
\3120\ NATSO, Detailed Comments, EPA-HQ-OAR-2018-0283-5484.
---------------------------------------------------------------------------
EPA is not adopting new incentives for automated and connected
vehicles. While EPA agrees there may be potential for such technologies
to reduce emissions long-term, depending on how the technologies are
developed, implemented, and used, EPA remains concerned about the high
degree of uncertainty regarding the impacts of the technologies and
potential loss of emissions reductions associated with such incentives.
EPA agrees with the comments that, at this time, it is more appropriate
for manufacturers to seek credits through the existing off-cycle
credits program where manufacturers would be required to provide data
demonstrating direct emissions improvements for the technologies.
[[Page 25210]]
(b) Natural Gas Vehicle (NGV) Credits
Vehicles that are able to run on compressed natural gas (CNG) are
eligible for an advanced technology multiplier credit for MYs 2017-
2021, as discussed in the Advanced Technology Incentives section above.
Dual-fueled natural gas vehicles, which can run either on natural gas
or on gasoline, also may use utility factors higher than 0.5 when
weighting tailpipe emissions measured over the test procedures while
operating on natural gas and gasoline test fuels if the vehicles meet
minimum design criteria, including minimum CNG range requirements.
Prior to the proposal, EPA received input from several industry
stakeholders that supported expanding these incentives to stimulate
production of vehicles capable of operating on natural gas, including
treating incentives for natural gas vehicles on par with those for
electric vehicles and other advanced technologies, and adjusting or
removing the minimum range requirements for dual-fueled CNG vehicles.
EPA requested comments on these potential additional incentives for
natural gas fueled vehicles.
Among comments received regarding incentives for NGVs, Ariel
Corporation and VNG together commented that NGVs can be effectively
promoted by providing a level playing field and regulatory parity with
EVs.\3121\ They stated, ``an effective alternative compliance pathway
for NGVs can be established with a few simple changes to the
regulations including applying the '0.15 divisor' to emissions
calculations, which would harmonize EPA's regulations with the
statutory CAFE program, and recognize the real-world emissions benefits
of RNG [renewable natural gas], and provide NGVs with reasonable parity
with EVs.'' Ariel and VNG commented also that EPA should offer advanced
technology production multipliers for NGVs on par with EVs and FCVs,
with NGVs receiving these incentives at the same level and for the same
duration as electric and fuel-cell vehicles. These commenters believe
that while NGVs have lower technology hurdles than these vehicles, they
face similar infrastructure challenges and offer similar or superior
emissions benefits through the use of RNG.
---------------------------------------------------------------------------
\3121\ Joint Submission from Ariel Corp. and VNG.co, Detailed
Comments, NHTSA-2018-0067-7573.
---------------------------------------------------------------------------
Coalition for Renewable Natural Gas, NGVAmerica, the American Gas
Association, and the American Public Gas Association commented in a
joint submission that NHTSA and EPA should use this rulemaking
opportunity to expand incentives for NGVs and thereby increase the
availability of NGVs in the light-duty sector, particularly for pickup
trucks, work vans, and sport utility vehicles.\3122\ These commenters
also submitted comments supporting additional incentives for full-size
pickup NGVs and incentives for vehicles equipped to be converted to
operate on natural gas. Coalition for Renewable Natural Gas, et al.,
commented that allowing 0 grams/mile accounting for electricity use is
favorable to electric vehicles because it allows electric vehicle
manufacturers to take credit for anticipated improvements in emissions
associated with the electric grid resulting from increased use of
natural gas and renewable energy.\3123\ It further commented that given
the significant amount of renewable natural gas currently being used
and projected to be used in future years, using a factor of 0.15 or
even greater to offset NGV emissions is warranted because RNG use
reduces carbon dioxide emissions by 85 percent or more in most cases.
Ingevity similarly commented in support of EPA including a 0.15
multiplier incentive for purposes of CO2 compliance parity
between natural gas and electric dual-fuel vehicles as necessary and
critical to promote the commercialization of light-duty natural gas
vehicles and stimulate the increased utilization of RNG. Ingevity added
that growth in the natural gas vehicle market is necessary to meet
future RFS obligations.\3124\
---------------------------------------------------------------------------
\3122\ Joint Submission from the Coalition for Renewable Natural
Gas, NGVAmerica, the American Gas Association, and the American
Public Gas Association, Detailed Comments, NHTSA-2018-0067-11967.
\3123\ Joint Submission from the Coalition for Renewable Natural
Gas, NGVAmerica, the American Gas Association, and the American
Public Gas Association, Detailed Comments, NHTSA-2018-0067-11967.
\3124\ Ingevity, Detailed Comments, NHTSA-2018-0067-8666.
---------------------------------------------------------------------------
United States Senator James M. Inhofe commented that ``even if all
current incentives for EVs are eliminated, EVs still have a compliance
advantage going forward. This is because the policy and technical
approaches underlying the [CO2] regulations embedded
preferential treatment for the previous administration's favored
technology. I respectfully ask you not to give NGVs preferential
treatment, but to level the playing field to allow the marketplace to
determine the future of NGV adoption and not the federal bureaucracy.
To achieve this parity, reinstating the 0.15 [CO2]
multiplier is essential.'' \3125\
---------------------------------------------------------------------------
\3125\ James M. Inhofe, Detailed Comments, EPA-HQ-OAR-2018-0283-
7456.
---------------------------------------------------------------------------
In addition to supporting the application of a 0.15 factor, some in
the natual gas industry also commented in support of production
multipliers for NGVs. Ariel and VNG commented that EPA should offer
advanced technology production multipliers for NGVs on par with EVs and
FCVs, with NGVs receiving these incentives at the same level and for
the same duration as electric and fuel cell vehicles. Ingevity
commented that dual-fuel and dedicated NGV multipliers should be
extended through 2025 as an effective way to promote the
commercialization of these kinds of vehicles by the automakers. NGV
America et al. commented that ``NGVs, both dedicated and dual-fuel,
should be provided with the same vehicle production multiplier credits
as have previously been, and continue to be, provided to EVs and FCVs.
Given that the expected and likely range capabilities of NGVs will
generally exceed EV ranges (including natural gas dual-fuel vehicles
that significantly outperform the range capabilities of PHEVs which
justifiably enjoy a lower multiplier as compared to EVs), the vehicle
production multipliers that are used for EVs should be applied to NGVs,
including dual fuel NGVs. Specifically, dedicated and dual-fuel NGVs
(or all covered advanced technology vehicles) should receive a base
multiplier of 2.0 (or any such higher multiplier afforded to EVs/FCVs)
for at least model years 2019 through 2021 and the same multipliers
afforded to EVs/FCVs thereafter through 2025.''
National Association of Convenience Stores (NACS) and the Society
of Independent Gasoline Marketers of America (SIGMA) commented, ``the
Associations urge you to treat all fuels and technologies equally,
including NGVs, EVs, and petroleum-based motor fuels. It is the role of
the Agencies to set performance specifications via notice-and-comment
rulemaking to ensure that they are appropriate. Once the specifications
are set, however, it should be up to the market to determine how best
to meet them.'' \3126\
---------------------------------------------------------------------------
\3126\ Joint submission on behalf of NACS and SIGMA, Detailed
Comments, EPA-HQ-OAR-2018-0283-5824.
---------------------------------------------------------------------------
UCS commented that natural gas is a potent greenhouse gas, and any
direct emissions of methane pose a significant threat to any effort to
limit climate change.\3127\ UCS stated, ``these direct emissions
upstream significantly
[[Page 25211]]
undermine any potential benefit that could come from the pump-to-wheel
benefits of displacing gasoline or diesel with natural gas.'' UCS also
commented, ``furthermore, the technology underpinning any natural gas-
powered vehicle is exceptionally mundane--natural gas has been deployed
previously in vehicles like the Honda Civic, and aftermarket CNG
conversions have long been available on the market. Again, there is no
critical hurdle to overcome with CNG powered vehicles, and there is
little if any benefit to any such incentives. We strongly recommend
that EPA eliminate all incentives for natural gas vehicles and instead
ensure such vehicles are credited commensurate with their impact on the
environment.'' CARB also commented that new compliance flexibilities
for NGVs are not appropriate at this time.\3128\
---------------------------------------------------------------------------
\3127\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3128\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
The Natural Gas Vehicles of America (NGVAmerica) commented that
there is no incentive under existing EPA and NHTSA regulations for an
automaker to sell vehicles equipped to be converted to operate on
natural gas (so-called ``gaseous-prep vehicles''), even though selling
such vehicles often results in the increased availability of
alternative fuel vehicles. Today, most alternative fuel conversions are
performed on newly manufactured gaseous-prep vehicles or vehicles that
have been equipped by the original equipment manufacturers with
hardened valves, valve seats, pistons, and piston rings. As an example,
most of Ford's commercial truck line-up is available as gaseous-prep,
and many such vehicles are converted to natural gas or propane by
qualified vehicle manufacturers. Converting these vehicles, producing
an assembly-line gaseous-prep vehicle, and sharing diagnostic
information are critical to ensuring that aftermarket conversions
perform well in-use and do not degrade the vehicle's emission control
equipment. Given the complexity of today's automobiles, it is virtually
impossible to legally convert new vehicles without this level of
cooperation from vehicle manufacturers.
NGVAmerica further commented that providing a regulatory incentive
for automakers to sell these vehicles would expand the availability of
gaseous-prep vehicles and increase consumer choice for alternative fuel
vehicles. EPA, therefore, should provide a credit for selling such
vehicles if the automaker can verify that the vehicles were
subsequently upfitted or converted using an EPA certified alternative
fuel system. Given the significant cost associated with certifying
vehicles and installing natural gas tanks, there is very little
likelihood that such an incentive would be abused by automakers. As
with credits for original equipment manufactured vehicles, the utility
factor for these vehicles would be based on the range of the vehicle
when operating on natural gas. In this way, vehicles with larger range
would earn more credit and vehicles with reduced range would earn less
credit.
Regarding comments that EPA should provide additional credits to
auto manufacturers for the potential use of RNG due to upstream
benefits associated with the production of RNG by applying a 0.15
factor, EPA disagrees because auto manufacturers would not be required
to ensure such fuels are used in the vehicles they produce over the
life of those vehicles. Commenters provided a rationale for why they
believe all NGVs produced in the future will be fueled with RNG, but
EPA believes there is no assurance that this would be the case. If
fossil fuel-based natural gas is used in the vehicles, the
environmental benefits asserted by the commenters would not exist and
the substantial vehicle incentives recommended by the commenters would
result in a loss of environmental benefits. EPA does not believe it is
appropriate to attribute most or all of the potential benefits of the
production and use of RNG to the vehicle manufacturer. EPA's Renewable
Fuel Standards (RFS) already appropriately credit RNG use as compared
to fossil fuel-based natural gas. The RFS program provides a
substantial incentive for RNG production, and those incentives may lead
to even lower fuel pricing and greater demand for RNG as vehicle fuel,
and for NGVs in the future. The RFS program also can provide incentives
for liquid cellulosic fuels, advanced bio-diesel, and other types of
renewable transportation fuels. Consistent with EPA's decision not to
include upstream emissions associated with electricity use for EVs and
PHEVs discussed above, EPA believes it is appropriate at this time to
maintain the focus of the light-duty vehicle GHG standards on the
capabilities of the vehicle to control emissions, and not rely on
lifecycle fuel characteristics as a basis for developing specific
vehicle incentives, particularly where those fuels are already
incentivized by the RFS program.
After considering comments regarding incentive multipliers for NGVs
and the current lack of light-duty NGV offerings by OEMs in the market,
EPA has decided to include a multiplier incentive of 2.0 for MY 2022-
2026 dedicated and dual-fuel NGVs. This multiplier will go into effect
when the previously established multipliers expire, thus extending the
mulipler for NGVs for 5 years beyond those previously established for
NGVs. While other alternative fuel vehicles that were provided
multiplier incentives are increasingly available in the light-duty
marketplace, no OEM is currently offering light-duty NGVs. Since Honda
ended production of the CNG version of the Honda Civic at the end of MY
2015, there have been no OEM NGV offerings available to consumers. EPA
continues to believe that NGVs could be an important part of the
overall light-duty vehicle fleet mix, and such offerings would enhance
the diversity of potentially cleaner alternative fueled vehicles
available to consumers.\3129\ EPA believes it is appropriate to extend
the availability of a production multiplier through MY 2026 for both
dual-fuel and dedicated NGVs to potentially help spur their re-
introduction by OEMs in the light-duty vehicle market.
---------------------------------------------------------------------------
\3129\ The CNG Honda Civic had approximately 20 percent lower
CO2 than the gasoline Civic in MY 2015.
---------------------------------------------------------------------------
EPA also received comments on the application of the regulatory
utility factor. For dual-fuel vehicles, emissions are measured on both
fuels (e.g., gasoline and natural gas) and weighted using a factor
referred to in the regulations as a utility factor. To use a utility
factor for natural gas greater than 0.5, a dual-fuel NGV must meet
design criteria requiring the vehicle to have a natural gas to gasoline
driving range of 2:1. The vehicle must also preferentially operate on
natural gas until the natural gas tank is empty. EPA adopted these
design criteria as part of the 2012 final rule to help ensure vehicles
using a utility factor of higher than 0.5 would likely be fueled with
and use natural gas most of the time on the road. At that time, EPA was
concerned that natural gas refueling may be much more inconvenient for
drivers relative to electric charging for PHEVs due to a lack of CNG
refueling stations (or home refueling, compared to the availability of
home chargers for many PHEVs) and, therefore, dual-fuel vehicles with
limited driving range on natural gas would likely frequently operate on
gasoline.
EPA received comments regarding the design criteria. Ingevity
commented that it has developed a low-pressure (900 psi) adsorbed
natural gas (ANG) fuel storage technology that allows vehicles to be
refueled using an affordable and reliable low-pressure natural gas
fueling
[[Page 25212]]
appliance.\3130\ Ingevity commented that ANG will allow for a
distributed refueling network at users' homes and businesses, just like
electrical recharging equipment has been installed for PHEVs over the
last several years. Ingevity commented that the design criteria for
dual-fuel NGVs that were established in the MYs 2017-2025 final rule
``make it impossible to reasonably and affordably manufacture a dual-
fuel NGV that can fully utilize the utility factor (UF) approach for
determining fuel economy and [CO2] emissions.'' Ingevity
recommended that the design criteria for dual-fuel NGVs be removed and
that the utility factor be based only on the range of the NGV on
natural gas, equivalent to the treatment of PHEVs. MECA submitted
similar comments regarding ANG technology.\3131\
---------------------------------------------------------------------------
\3130\ Ingevity, Detailed Comments, NHTSA-2018-0067-8666.
\3131\ See MECA, Detailed Comments, NHTSA-2018-0067-11999.
---------------------------------------------------------------------------
Ariel and VNG also commented that design criteria imposed on dual-
fuel NGVs add unnecessary costs and complexity, and currently are
arbitrarily applied only to dual-fuel NGVs, and not to their dual-fuel
hybrid counterparts.\3132\ NACS, SIGMA, and NATSO also recommended that
EPA remove eligibility requirements associated with the utility
factor.\3133\
---------------------------------------------------------------------------
\3132\ Joint Submission from Ariel Corp. and VNG, Detailed
Comments, NHTSA-2018-0067-7573.
\3133\ Joint submission on behalf of NACS and SIGMA, Detailed
Comments, EPA-HQ-OAR-2018-0283-5824; NATSO, Detailed Comment, EPA-
HQ-OAR-2018-0283-5484.
---------------------------------------------------------------------------
After considering the comments, EPA is removing the design criteria
from the regulations and thereby allowing higher utility factors to be
used for dual-fuel natural gas vehicles based solely on driving range
on natural gas, as is the case for PHEVs. The utility factor represents
a reasonable way of weighting the emissions of a dual-fuel vehicle on
each fuel to derive a single emissions value when including the dual-
fuel vehicles in a manufacturer's fleet average compliance
determination. Ideally, the utility factor would match the use of each
fuel in real-world vehicle operation. The utility factor is not meant
to incentivize the adoption of a particular technology, so it differs
fundamentally from incentives such as multipliers. With the development
of low-pressure natural gas vehicle fueling system technology since the
2012 final rule, EPA's concerns regarding limited fueling
infrastructure that led the agency to adopt the design criteria in the
2012 rule are significantly diminished. EPA believes that low-pressure
fueling is a new advancement that offers the potential for more
convenient refueling for individuals or businesses similar to that for
PHEVs. EPA expects owners of dual-fuel CNG vehicles preferentially to
seek to refuel and operate on CNG fuel as much as possible, both
because the owner would have to pay a higher vehicle price for the
dual-fuel capability, and because CNG fuel is considerably cheaper than
gasoline. With the opportunity for relatively low-cost on-site
refueling at homes or businesses, EPA expects such vehicles to be
refueled with natural gas similar to how people refuel PHEVs. Vehicle
purchasers that choose high pressure vehicle systems over low pressure
systems would likely do so only if they have ready access to a high
pressure refueling system, for example, at a fleet's central fueling
location. Removing the design criteria for dual-fuel natural gas
vehicles also addresses the concerns of some commenters regarding the
differing treatment of PHEVs and dual-fuel NGVs.
EPA believes that with the advancement of technology offering the
potential for more flexible refueling of NGVs, removing the design
criteria is a reasonable change to the regulations. This regulatory
change will apply starting with MY 2021. MY 2021 will provide
sufficient time for orderly implementation and EPA is not aware of any
dual-fuel NGVs emissions certified for MYs 2019-2020 that would
otherwise be affected if this change were to be implemented sooner.
EPA received comments that vehicle conversions and ``gaseous-prep''
vehicles should be eligible for credits. In response to comments on
vehicle conversions, alternative fuel converters are not required to
meet fleet average standards but instead may comply with 40 CFR part 85
subpart F regulations providing a tampering exemption. Fleet average
standards are generally not appropriate for fuel conversion
manufacturers because the ``fleet'' of vehicles to which a conversion
system may be applied has already been accounted for under the OEM's
fleet average standard. Alternative fuel converters are not
manufacturing new vehicles, but are converting existing vehicles that
have already been certified by the OEM. CO2 credits are
available to OEMs based on fleet emissions performance compared to the
fleet average standards and therefore conversions are not eligible for
these credits. EPA did not propose to change and is not changing the
exemption process promulgated in 40 CFR part 85 subpart F. Because fuel
conversions are not required to meet the fleet average standards,
credits generated under those standards are not available. Regarding
gaseous-prep vehicles, these vehicles are not NGVs at initial sale and
therefore are not eligible for NGV incentives. Instead, they are
included in the OEM's fleet as gasoline-only vehicles. EPA disagrees
with the commenters that such vehicles should be eligible for NGV
incentives at time of initial sale if the vehicle is later converted to
natural gas since the OEM does not measure the emissions of the vehicle
on natural gas at time of certification and is not responsible for the
emissions performance of the vehicle on natural gas over the life of
the vehicle.
C. NHTSA Compliance and Enforcement
1. Overview of the NHTSA Compliance Process
Consumer choice drives the mixture of automobiles on the road.
Manufacturers largely produce a mixture of vehicles to meet consumer
demand and address compliance with CAFE standards though the
application of fuel economy improving technologies to those vehicles,
and by using compliance flexibilities and incentives that are available
in the CAFE program. As discussed earlier in this notice, each vehicle
manufacturer is subject to separate CAFE standards for passenger cars
and light trucks, and for the passenger car standards, a manufacturer's
domestically-manufactured and imported passenger car fleets are
required to comply separately.\3134\ Additionally, domestically-
manufactured passenger cars are subject to a statutory minimum
standard.\3135\ CAFE program flexibilities are largely provided for in
statute. Credits for air conditioning efficiency, off-cycle, and pickup
truck advanced technologies are not expressly specified by CAFE
statute, but are ``implemented consistent with EPCA's provisions
regarding calculation of fuel economy'' as discussed in section C.2
below.
---------------------------------------------------------------------------
\3134\ 49 U.S.C. 32904(b).
\3135\ 49 U.S.C. 32902(b)(4).
---------------------------------------------------------------------------
Compliance with the CAFE program begins with manufacturers
submitting required reports to NHTSA in advance and during the model
year that contain information, specifications, data, and projections
about their fleets.\3136\ Manufacturers report early product
projections to NHTSA describing their efforts to comply with CAFE
standards per EPCA's reporting requirements.\3137\ Manufacturers' early
projections are required to identify any of the
[[Page 25213]]
flexibilities and incentives manufacturers plan to use for air-
conditioning (A/C) efficiency, off-cycle and, through MY 2021, full-
size pickup truck advanced technologies. EPA consults with NHTSA when
reviewing and considering manufacturers' requests for fuel consumption
improvement values for A/C and off-cycle technologies that improve fuel
economy. NHTSA evaluates and monitors the performance of the industry
using the information provided. NHTSA also audits manufacturers'
projected data for conformance and verifies vehicle design data through
testing to ensure manufacturers are complying as projected. After the
model year ends, manufacturers submit final reports to EPA, including
final information on all the flexibilities and incentives allowed or
approved for the given model year.\3138\ EPA then calculates the fuel
economy level of each fleet produced by each manufacturer, and
transmits that information to NHTSA.\3139\
---------------------------------------------------------------------------
\3136\ 49 U.S.C. 32907(a); 49 CFR 537.7.
\3137\ 49 U.S.C. 32907(a).
\3138\ For example, alternative fueled vehicles get special
calculations under EPCA (49 U.S.C. 32905-06), and fuel economy
levels can also be adjusted to reflect air conditioning efficiency
and ``off-cycle'' improvements, as discussed below.
\3139\ 49 U.S.C. 32904(c)-(e). EPCA granted EPA authority to
establish fuel economy testing and calculation procedures; EPA uses
a two-year early certification process to qualify manufacturers to
start selling vehicles, coordinates manufacturer testing throughout
the model year, and validates manufacturer-submitted final test
results after the close of the model year.
---------------------------------------------------------------------------
NHTSA notes that some manufacturers have submitted and/or
resubmitted requests for A/C and off-cycle benefits after EPA final
reports are completed or nearly completed and, in those cases, such
submissions are causing considerable delays in EPA's ability to
finalize CAFE reports. Late and revised submissions can place
significant burdens on the government in order to reassess a
manufacturer's CAFE performances and standards and can also cause
significant impacts on previous compliance model years. In the
following sections, EPA and NHTSA are incorporating regulatory
modifications or providing guidance to help manufacturers expedite
approvals and to facilitate the governments processing of the
flexibilities and incentives.
NHTSA determines each manufacturer's obligation to comply with
applicable model year's CAFE standards and notifies the manufacturer if
any of its fleet performances fall below standards. Manufacturers must
submit plans detailing the compliance flexibilities to be used to
resolve any possible noncompliances or may pay civil penalties to
address any deficits for falling below standards. NHTSA periodically
releases data and reports to the public through its CAFE Public
Information Center (PIC) based on information in the EPA final reports
for the given compliance model year, and based on the projections
manufacturers provide to NHTSA for the next two model years.\3140\
---------------------------------------------------------------------------
\3140\ NHTSA CAFE Public Information Center, https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm.
---------------------------------------------------------------------------
2. NHTSA's CAFE Program Compliance
EPCA and EISA specify several flexibilities and incentives that are
available to help manufacturers comply with CAFE standards. Some
flexibilities are defined, and sometimes limited by statute--for
example, while Congress allowed manufacturers to transfer credits
earned for over-compliance from their car fleet to their truck fleet
and vice versa, Congress also limited the amount by which manufacturers
could increase their CAFE levels using those transfers.\3141\
Consistent with the limits Congress placed on certain statutory
flexibilities and incentives, NHTSA crafted and implements the credit
transfer and trading regulations authorized by EISA to help ensure that
total fuel savings are preserved when manufacturers exercise statutory
compliance flexibilities.
---------------------------------------------------------------------------
\3141\ See 49 U.S.C. 32903(g).
---------------------------------------------------------------------------
NHTSA and EPA have previously developed other compliance
flexibilities and incentives for the CAFE program consistent with the
statutory provisions regarding EPA's calculation of manufacturers' fuel
economy levels. As discussed previously, NHTSA finalized in the 2012
final rule, for MYs 2017 and later, an approach for manufacturers'
``credits'' under EPA's program to be applied as fuel economy
``adjustments'' or ``improvement values'' under NHTSA's program for:
(1) Technologies that cannot be measured or cannot be fully measured on
the 2-cycle test procedure, i.e., ``off-cycle'' technologies; and (2)
A/C efficiency improvements that also improve fuel economy but cannot
be measured on the 2-cycle test procedure. Additionally, both agencies'
programs give manufacturers compliance incentives through MY 2021 for
utilizing specified technologies on pickup trucks, such as pickup truck
hybridization.
The following sections outline how NHTSA determines whether
manufacturers are in compliance with the CAFE standards for each model
year, and how manufacturers may use compliance flexibilities, or
address noncompliance by paying civil penalties. As addressed above,
some compliance flexibilities are expressly prescribed in statute and
some are implemented consistent with EPCA's provisions regarding
calculation of fuel economy. NHTSA proposed new language updating and
clarifying existing regulatory text in this area as part of the NPRM.
NHTSA also sought comments in the NPRM on these changes, as well as on
the general efficacy of these flexibilities in the fuel economy and
CO2 programs.
Moreover, the following sections explain how manufacturers submit
data and information to the agency. As part of the NPRM, NHTSA proposed
to implement a new standardized template for manufacturers to use to
submit CAFE data to the agency, as well as a standardized template for
reporting credit transactions. Additionally, NHTSA proposed adding
requirements that specify the precision of the fuel savings adjustment
factor in 49 CFR 536.4. These new requirements are intended to
streamline reporting and data collection from manufacturers, in
addition to helping the agency use the best available data to inform
CAFE program decision makers. The comments received to these proposals
are included in Section IX.C.2.a)(2)(d) along with NHTSA's responses to
the comments and final resolutions established in the final rule.
NHTSA also sought comments on removing or modifying certain CAFE
program flexibilities. The comments received and NHTSA's responses to
those comments are discussed below.
a) How does NHTSA determine compliance?
(1) Manufacturers Submit Data to NHTSA and EPA and the Agencies
Validate Results
EPCA, as amended by EISA, requires a manufacturer to submit reports
to the Secretary of Transportation explaining whether the manufacturer
will comply with an applicable CAFE standard for the model year for
which the report is made; the actions a manufacturer has taken or
intends to take to comply with the standard; and other information the
Secretary requires by regulation.\3142\ A manufacturer must submit a
report containing the above information during the 30-day period before
the beginning of each model year, and during the 30-day period
beginning the 180th day of the model year.\3143\ When a manufacturer
determines it is unlikely to comply with a CAFE standard, the
manufacturer must report additional
[[Page 25214]]
actions it intends to take to comply and include a statement about
whether those actions are sufficient to ensure compliance.\3144\
---------------------------------------------------------------------------
\3142\ 49 U.S.C. 32907(a).
\3143\ Id.
\3144\ Id.
---------------------------------------------------------------------------
To implement these reporting requirements, NHTSA issued 49 CFR part
537, ``Automotive Fuel Economy Reports,'' which specifies three types
of CAFE reports that manufacturers must submit. A manufacturer must
first submit a pre-model year (PMY) report containing the
manufacturer's projected compliance information for that upcoming model
year. By regulation, the PMY report must be submitted in December of
the calendar year prior to the corresponding model year.\3145\
Manufacturers must then submit a mid-model year (MMY) report containing
updated information from manufacturers based upon actual and projected
information known midway through the model year. By regulation, the MMY
report must be submitted by the end of July for the applicable model
year.\3146\ Finally, manufacturers must submit a supplementary report
to supplement or correct previously submitted information, as specified
in NHTSA's regulation.\3147\
---------------------------------------------------------------------------
\3145\ 49 CFR 537.5(b).
\3146\ Id.
\3147\ 49 CFR 537.8.
---------------------------------------------------------------------------
If a manufacturer wishes to request confidential treatment for a
CAFE report, it must submit both a confidential and redacted version of
the report to NHTSA. CAFE reports submitted to NHTSA contain estimated
sales production information, which may be protected as confidential
until the termination of the production period for that model
year.\3148\ NHTSA temporarily protects each manufacturer's competitive
sales production strategies, but does not permanently exclude sales
production information from public disclosure. Sales production volumes
are part of the information NHTSA routinely makes publicly available
through the CAFE PIC.
---------------------------------------------------------------------------
\3148\ 49 CFR part 512, appx. B(2).
---------------------------------------------------------------------------
The manufacturer reports provide information on light-duty
automobiles such as projected and actual fuel economy standards, fuel
economy performance values, and production volumes, as well as
information on vehicle design features (e.g., engine displacement and
transmission class) and other vehicle attribute characteristics (e.g.,
track width, wheelbase, and other off-road features for light trucks).
Beginning with MY 2017, to obtain credit for fuel economy improvement
values attributable to additional technologies, manufacturers must also
provide information regarding A/C systems with improved efficiency,
off-cycle technologies (e.g., stop-start systems, high-efficiency
lighting, active engine warm-up), and full-size pickup trucks with
hybrid technologies or with emissions/fuel economy performance that is
better than footprint-based targets by specified amounts. This includes
identifying the makes and model types equipped with each technology,
the compliance category those vehicles belong to, and the associated
fuel economy improvement value for each technology.\3149\ In some
cases, NHTSA may require manufacturers to provide supplementary
information to justify or explain the benefits of these technologies
and their impact on fuel consumption or to evaluate the safety
implication of the technologies. These details are necessary to
facilitate NHTSA's technical analyses and to ensure the agency can
perform enforcement audits as appropriate.
---------------------------------------------------------------------------
\3149\ NHTSA collects model type information based upon the EPA
definition for ``model type'' in 40 CFR 600.002.
---------------------------------------------------------------------------
NHTSA uses manufacturer-submitted PMY, MMY, and supplementary
reports to assist in auditing manufacturer compliance data and
identifying potential compliance issues as early as possible.
Additionally, as part of its footprint validation program, NHTSA
conducts vehicle testing throughout the model year to confirm the
accuracy of the track width and wheelbase measurements submitted in the
reports.\3150\ These tests help the agency better understand how
manufacturers may adjust vehicle characteristics to change a vehicle's
footprint measurement, and ultimately its fuel economy target. NHTSA
also includes a summary of manufacturers' PMY and MMY data in an annual
fuel economy performance report made publicly available on its PIC.
---------------------------------------------------------------------------
\3150\ U.S. Department of Transportation, NHTSA, Laboratory Test
Procedure for 49 CFR part 537, Automobile Fuel Economy Attribute
Measurements (Mar. 30, 2009), available at http://www.nhtsa.gov/DOT/NHTSA/Vehicle%20Safety/Test%20Procedures/Associated%20Files/TP-537-01.pdf.
---------------------------------------------------------------------------
NHTSA uses EPA-verified final-model year (FMY) data to evaluate
manufacturers' compliance with CAFE program requirements, and draws
conclusions about the performance of the industry. After manufacturers
submit their FMY data, EPA verifies the information, accounting for
NHTSA and EPA testing, and subsequently forwards the final verified
data to NHTSA.
(2) Changes to CAFE Reporting Requirements Made by This Final Rule
NHTSA proposed changes to its CAFE reporting requirements with the
intent of streamlining data collection and reporting for manufacturers
while helping the agency obtain the best available data to inform CAFE
program decision-makers. The agency developed two new standardized
reporting templates for manufacturers and proposed to start using the
templates beginning in the 2019 compliance model year. In the NPRM,
NHTSA sought comments on the templates. NHTSA's responses to the
comments received and the changes to the templates for the final rule
are presented below.
(a) Standardized CAFE Reporting Template
When NHTSA received and reviewed manufacturers' projection reports
for MYs 2013--2015, the agency observed that most did not conform to
the requirements specified in 49 CFR part 537. For example, NHTSA
identified several instances where manufacturers' CAFE reports included
a ``yes'' or ``no'' response to a request for a vehicle's numerical
ground clearance values. In a 2015 notice of proposed rulemaking, NHTSA
proposed to amend 49 CFR part 537 to require a new data format for
manufacturers' light-duty vehicle CAFE projection reports.\3151\ In
response to the proposal, some manufacturers commented that the
previous changes in reporting requirements generated confusion and led
to reporting errors. NHTSA recognized that the modification to the base
tire definition in the 2012 final rule for MYs 2017 and later seemed to
make some manufacturers uncertain about what footprint data was
required in the reports.\3152\ Specifically, certain manufacturers did
not understand that the modified base tire definition required them to
provide estimated attribute-based target standards for each unique
model type/footprint combination beginning with MY 2013. NHTSA
discovered cases where manufacturers only provided target or vehicle
data for certified vehicle configurations, and did not report
information for each of the unique model type/footprint combinations
for their available production vehicles in the market. However, NHTSA
did not adopt the proposed data format from the 2015 proposed rule
after receiving
[[Page 25215]]
adverse comments from manufacturers.\3153\
---------------------------------------------------------------------------
\3151\ 80 FR 40540 (Jul. 13, 2015).
\3152\ 49 CFR 523.2.
\3153\ 81 FR 73958 (Oct. 25, 2016).
---------------------------------------------------------------------------
Since the issuance of the final rule in 2016, NHTSA has continued
to receive projection reports that contain inaccurate and/or missing
data. These noncompliant reports impede NHTSA's ability to audit
manufacturer compliance data, identify potential compliance issues, and
analyze industry trends. Problems with inaccurate or missing data has
become an even greater issue for manufacturers reporting on the new MY
2017 incentives for efficient A/C systems, off-cycle technologies, and
full-size pickup trucks with hybrid technologies/improved exhaust
emission performance.\3154\ These incentives are explained in Section
IX.C.2.c). Manufacturers seeking to take advantage of these new
benefits must provide information at the model-type level; however,
many manufacturers did not submit the required information in their PMY
reports for MYs 2017, 2018, and 2019. This caused NHTSA's Office of
Enforcement to send letters reminding manufacturers of their obligation
to submit accurate and complete CAFE reports. NHTSA will continue to
monitor the accuracy, completeness, and timeliness of manufacturers'
CAFE reports and may take additional action as appropriate.
---------------------------------------------------------------------------
\3154\ NHTSA allows manufacturers to use these flexibilities and
incentives for complying with standards starting in MY 2017; the
FCIV for full-size pickup trucks with hybrid technologies/improved
exhaust emission performance applies only through MY 2021, as
discussed further below.
---------------------------------------------------------------------------
In the NPRM, NHTSA proposed a new standardized template for
reporting PMY and MMY information, as specified in 49 CFR 537.7(b) and
(c), as well as supplementary information required by 49 CFR 537.8. The
template allows manufacturers to build out the required confidential
versions of CAFE reports specified in 49 CFR part 537 and to produce
automatically the required non-confidential versions by clicking a
button within the template. While NHTSA recognizes that modifications
to the reporting requirements may initially be a slight inconvenience
to manufacturers, the number of noncompliant reports the agency
continues to receive justifies development of a uniform reporting
method to help ensure compliance with CAFE regulations. Adopting a
standardized template will assist manufacturers in providing the agency
with all necessary data, thereby helping manufacturers to ensure they
are complying with CAFE regulations. The template organizes the
required data in a manner consistent with NHTSA and EPA regulations and
simplifies the reporting process by incorporating standardized
responses consistent with those provided to EPA. The template collects
the relevant data, calculates intermediate and final values in
accordance with EPA and NHTSA methodologies, and aggregates all the
final values required by NHTSA regulations in a single summary
worksheet. Thus, NHTSA believes that the standardized templates will
benefit both the agency and manufacturers by helping to avoid reporting
errors, such as data omissions and miscalculations, and will ultimately
simplify and streamline reporting.
NHTSA proposed to require that manufacturers use the standardized
template for all PMY, MMY, and supplementary CAFE reports. NHTSA
observed that a significant number of manufacturers submit their MMY
reports as updated PMY reports--using the same amount of information,
despite fewer data requirements. To conform with this method, NHTSA
designed the template based on one standardized format that uses the
same data requirements for all CAFE reports. This approach will further
simplify CAFE projection reporting for manufacturers. The template
contains a few additional data fields for certain vehicle
characteristics; however, the inclusion of model type indexes will
limit the number of required entries by populating a number of pre-
entered data fields based on one value.
The standardized template will also allow NHTSA to modify its
existing compliance database to accept and import uniform data and
automatically aggregate manufacturers' data. This will allow NHTSA to
execute its regulatory obligations more efficiently and effectively.
Overall, the template will help to ensure compliance with data
requirements under EPCA/EISA and drastically reduce the industry and
government's burden for reporting in accordance with the Paperwork
Reduction Act.\3155\ NHTSA made the template available through its
docket as well as its PIC, and sought comment on the regulatory changes
to the reporting process.
---------------------------------------------------------------------------
\3155\ 44 U.S.C. 3501 et seq.
---------------------------------------------------------------------------
Comments on the template were received from the Auto Alliance,
Global Automakers, Ford, Mercedes-Benz, Toyota, Volvo and Volkswagen.
The Auto Alliance, Toyota, and Volkswagen opposed adopting the proposed
template; however, Global Automakers agreed with the appropriateness of
a standardized template that combines credit trading information with a
data reporting template.\3156\ Global Automakers also made two
recommendations: (1) Combine EPA's AB&T template with NHTSA's CAFE
Projections Reporting Template to streamline reporting and reduce
burden; and (2) add an FMY report requirement as an update to the MMY
report submission.\3157\
---------------------------------------------------------------------------
\3156\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Toyota, Detailed Comments, NHTSA-2018-0067-12150; Volkswagen,
Detailed Comments, NHTSA-2017-0069-0583.
\3157\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
---------------------------------------------------------------------------
Mercedes-Benz, Ford, and Volkswagen commented about data fields
they believed were outdated, or not relevant to fuel economy testing or
projecting fuel economy performance.\3158\ Mercedes-Benz stated that
some required data fields are not currently collected as a part of the
fuel economy testing process, and their capture would require
additional burden.\3159\ Mercedes-Benz believes those data fields
should be an optional requirement. Additionally, Mercedes-Benz
recommended that NHTSA omit certain data fields, and stated that it
would be helpful if NHTSA clarified its intention for the information
in others.\3160\ The specific data fields mentioned by Mercedes-Benz
are in Table IX-6. Ford stated that many of the data fields are
outdated, have no bearing on compliance assessments, and are misaligned
with the current reporting structure, which is dictated by model type
index.\3161\ Similarly, Volkswagen stated that the proposed reporting
template is populated with many fields that do not immediately appear
relevant to projecting CAFE performance, align with the existing
requirements in 49 CFR 537.7, or seem relevant in the space of
automotive technology.\3162\
---------------------------------------------------------------------------
\3158\ Daimler Mercedes, Detailed Comments, EPA-HQ-OAR-2018-
0283-4182; Ford, Detailed Comments, NHTSA-2018-0067-11928;
Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
\3159\ Daimler Mercedes, Detailed Comments, EPA-HQ-OAR-2018-
0283-4182.
\3160\ Daimler Mercedes, Detailed Comments, EPA-HQ-OAR-2018-
0283-4182.
\3161\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
\3162\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
[[Page 25216]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.756
The Auto Alliance and Mercedes-Benz noted the differences in how
NHTSA and EPA request data on A/C efficiency and off-cycle
technologies. Mercedes-Benz highlighted the difficulty in predicting
the projected sales production of the technologies, and the Auto
Alliance cautioned that the number of reporting entries would increase
by a factor of ten or more.\3163\ The Auto Alliance stated its belief
that the change in reporting requirements would cost its members more
than $1 million in information technology changes and that the changes
could not be completed prior to MY 2021.\3164\ Likewise, Ford contended
that an implementation date for MY 2019 is aggressive and does not
provide manufacturers with adequate lead time.\3165\
---------------------------------------------------------------------------
\3163\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Daimler Mercedes, Detailed Comments, EPA-HQ-OAR-2018-0283-4182.
\3164\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3165\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
The Auto Alliance emphasized that the templates lack common
reporting standardization with submissions to EPA.\3166\ The Auto
Alliance, Global Automakers, Toyota, and Volvo all requested that NHTSA
and EPA accept a single, common reporting format to satisfy reporting
for both agencies.\3167\ Mercedes-Benz and Volkswagen requested
stakeholder workshops to review the template with agency staff, with
the former recommending that NHTSA host the workshops in partnership
with EPA.\3168\
---------------------------------------------------------------------------
\3166\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3167\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Global Automakers, Detailed Comments, NHTSA-2018-0067-12032; Toyota,
Detailed Comments, NHTSA-2018-0067-12150; Volvo, Detailed Comments,
NHTSA-2018-0067-12036.
\3168\ Daimler Mercedes, Detailed Comments, EPA-HQ-OAR-2018-
0283-4182; Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
Ford requests that NHTSA re-examine the proposed required
submission methods and reconsider current electronic submission
methods.\3169\ Ford expressed concern about the efficiency and security
issues involved in submitting data on a CD through the mail containing
confidential business information.\3170\ Ford identified what it
believes are better available avenues for submission, such as secured
email or online portals like EPA's Central Data Exchange.\3171\
---------------------------------------------------------------------------
\3169\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
\3170\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
\3171\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
NHTSA disagrees with many of the manufacturers' assertions.
Differences in EPA and NHTSA regulations prevent establishing a single
reporting format for CAFE purposes. For example, EPA only needs early
model year information for manufacturers' applications for
certification required under 40 CFR 86.1843-01. Manufacturers submit a
single application with extensive details for each certified vehicle
within a test group (i.e., the certified vehicle represents all the
vehicles within the test group with similar technologies and
performance characteristics). In comparison, NHTSA's required early
model year information is far less detailed and is aggregated for model
types and compliance categories. However, NHTSA and EPA already share
all the relevant CAFE FMY information pursuant to an interagency
agreement. This arrangement not only benefits manufacturers but also
reduces the burden on the Federal government. Since much of the
required data in NHTSA's projections template is already contained in
EPA final reports, manufacturers would not be required to generate
additional information but simply to provide estimates along the way to
finalizing the data. NHTSA plans to release a data matrix that maps
data elements between the CAFE template and the EPA final CAFE reports.
NHTSA will notify the public when the matrix will be available on its
website. Consequently, there is no need to create an additional final
report as an updated version of NHTSA's MMY report, as suggested by
Global Automakers. Once NHTSA configures its CAFE database to accept
the reporting template via file upload, the agency will be able to use
the model type index data field to connect data values from the
template to corresponding values in EPA's final CAFE report.
Manufacturers should note that CAFE reports are estimated projections
of the EPA final CAFE compliance data. Contrary to Mercedes concerns
about the difficulty in predicting the projected sales production of
the technologies, NHTSA only expects manufacturers to provide the most
up-to-date information available 30 days before a report is required to
be submitted to the
[[Page 25217]]
Administrator as specified in 49 CFR part 537.5(d). While manufacturer
PMY reports may be limited in certain instances (excluding vehicles
already in sales distribution), the MMY reports should be more
inclusive and closer to the final values reported to EPA. Manufacturers
should also be submitting supplementary reports to NHTSA if they
believe there will be significant differences between CAFE MMY reports
and the EPA final reports.
Commenters also stated that the A/C and off-cycle information
reported in the NHTSA template is inconsistent with the EPA EV-
CIS.\3172\ NHTSA notes that the inconsistency between the agencies is
intentional and necessary. NHTSA's off-cycle and A/C information must
be collected in greater detail than that reported to the EPA EV-CIS.
NHTSA collects detailed information on A/C and off-cycle technologies
for determining penetration rates of specific technologies in the
market, as well as analyzing the types of technologies as equipped on
specific model types. In comparison, EPA aggregates the data for
calculating credits, which allows for combining the benefits for all
the technologies equipped on a model type. NHTSA also will use the
detailed information for public disclosure and for auditing purposes.
However, NHTSA acknowledges the Auto Alliance's concerns about the
burden placed on the industry for providing more detailed data and
therefore will not require manufacturers to start using the templates
for reporting until MY 2023. NHTSA also agrees with Ford that it is
important to consider the issues of security and efficiency with
respect to the submission of confidential information to the agency,
and the agency will consider possible changes to its procedures
relating to the receipt and handling of confidential information to
ensure streamlined, secure, and efficient submission of confidential
information, including CAFE reports.\3173\
---------------------------------------------------------------------------
\3172\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
\3173\ See 49 CFR part 512, 537.5.
---------------------------------------------------------------------------
Secondly, NHTSA agrees with Mercedes-Benz and Volkswagen that
workshops will aid in implementing the templates by providing
instruction on how to complete them. NHTSA plans to host a workshop for
manufacturers to discuss the implementation process. NHTSA believes
finalizing the template in this rulemaking is important to address
continuing concerns with reporting noncompliance (i.e., missing,
incomplete, or inaccurate submissions) with the existing provisions in
Part 537. Ultimately, establishing the new templates and holding
educational workshops will be more effective in achieving industry
compliance than imposing penalties on a case-by-case basis for failure
to comply with reporting provisions.
Finally, NHTSA is also adopting changes to the proposed template in
response to comments from Mercedes-Benz, Ford, and Volkswagen. NHTSA
made changes to several of the data fields discussed by Mercedes-Benz.
NHTSA does not agree with Mercedes-Benz's recommendation to omit the
``Type of Overdrive'' or ``Type of Torque Converter'' data fields;
however, the agency does believe the proposed data to be inserted into
those fields may be too specific for CAFE purposes. Therefore, the
agency is finalizing a requirement that manufacturers identify whether
vehicles are equipped with overdrive or a torque converter by selecting
``Yes'' or ``No'' from a dropdown list. The agency has also changed the
``Calibration'' field to ``Other Calibration'' to clarify the data
being requested, and changed the ``Auxiliary Emission Control Device''
in the ``Fuel Economy'' worksheets to a dropdown that allows users to
select multiple emission control systems. NHTSA believes that adding
dropdown lists in the template creates uniformity in the reported
information and makes the information more relevant to current
vehicles.
The agency agrees with the essence of Volkswagen's assertion that
some of the required data fields may no longer be as common on
contemporary vehicles, and therefore, may not apply to all
manufacturers. As suggested by Mercedes-Benz, NHTSA has decided to make
the ``Catalyst Usage,'' ``Distributor Calibration,'' ``Choke
Calibration,'' and ``Other Calibration'' data fields optional with a
default value of ``N/A.'' NHTSA does not agree with Mercedes-Benz's
recommendation that NHTSA provide a better understanding of its
intention for the information in certain data fields. ``Electric
Traction Motor, Motor Controller,'' ``Battery Configuration,''
``Electrical Charging System,'' and ``Energy Storage Device'' are the
data fields that characterize the basic powerplant for electric
vehicles. Basic Engine, along with Carline and Transmission Class, make
up a model type for light-duty vehicles. Therefore, those five fields
are used to group vehicles by model type in accordance with EPA
regulations. Fuel economy performance is calculated by
Subconfiguration, which is a subset of a model type. As such, those
five data fields are an integral part of grouping vehicles for fuel
economy testing purposes in accordance with EPA regulations. NHTSA also
does not agree with Volkswagen's assertion that the template is
populated with many fields that do not appear relevant to projecting
CAFE performance. As previously mentioned, many of the data fields are
used to arrange vehicles into groups for calculating fuel economy
performance in accordance with 49 CFR 537.7.
Furthermore, NHTSA has re-engineered the template in a few areas to
include additional supporting data elements used in calculating other
data fields required by Part 537. These fields may not directly align
with the existing requirements in Part 537 but are necessary for
validation purposes. For this reason, NHTSA is also finalizing its
proposal in the NPRM to remove the optional provisions for reporting
the data fields for determining the CAFE model type target standards,
making the information mandatory in the template. Additional changes
have been made to the template to improve fuel economy calculations.
NHTSA edited the template to include the calculation procedure for
alternative-fuel vehicles and corrected the test procedure adjustment
(TPA) calculation to align the fleet average fuel economy calculation
methodology with 40 CFR 600.510-12. Several expanded worksheets and
functional features were also added to the template to improve the
usability of the templates for manufacturers. These changes include
modifications such as adding the estimated credits and a minimum
domestic passenger shortfall calculator as the last fields to the
``Summary'' worksheet. Other functional changes include protecting
users from changing the formatting or data validation in each cell and
allowing columns to be widened by users.
(b) Standardized Credit Documents
A credit ``[t]rade'' is defined in 49 CFR 536.3 as ``the receipt by
NHTSA of an instruction from a credit holder to place its credits in
the account of another credit holder.'' \3174\ ``Traded credits are
moved from one credit holder to the recipient credit holder within the
same compliance category for which the credits were originally earned.
If a credit has been traded to another credit holder and is
subsequently traded back to the originating manufacturer, it will be
deemed not to have been traded for compliance purposes.'' \3175\ NHTSA
does not administer trade negotiations between manufacturers and when a
[[Page 25218]]
trade document is received the agreement must be issued jointly by the
current credit holder and the receiving party.\3176\ NHTSA does not
settle contractual or payment issues between trading manufacturers.
---------------------------------------------------------------------------
\3174\ 49 CFR 536.3(b).
\3175\ Id.
\3176\ See 49 CFR 536.8(a).
---------------------------------------------------------------------------
NHTSA created its CAFE database to maintain credit accounts for
manufacturers and to track all credit transactions. A credit account
consists of a balance of credits in each compliance category and
vintage held by the holder. While maintaining accurate credit records
is essential, it has become a challenging task for the agency given the
recent increase in credit transactions. Manufacturers have requested
that NHTSA approve trade or transfer requests not only in response to
end-of-model year shortfalls, but also, during the model year, when
purchasing credits to bank.
To reduce the burden on all parties, encourage compliance, and
facilitate quicker NHTSA credit transaction approval, the agency
proposed in the NPRM to add a required template to standardize the
information parties submit to NHTSA in reporting a credit transaction.
Presently, manufacturers are inconsistent in submitting the information
required by 49 CFR 536.8, creating difficulty for NHTSA in processing
transactions. The template NHTSA proposed is a simple spreadsheet that
trading parties fill out. When completed, parties will be able to click
a button on the spreadsheet to generate a credit transaction summary
and if applicable credit trade confirmation, the latter of which shall
be signed by both trading entities. The credit trade confirmation
serves as an acknowledgement that the parties have agreed to trade
credits. The completed credit trade summary and a PDF copy of the
signed trade confirmation must be submitted to NHTSA. Using the
template simplifies CAFE compliance aspects of the credit trading
process, and helps to ensure that trading parties follow the
requirements for a credit transaction in 49 CFR 536.8(a).\3177\
---------------------------------------------------------------------------
\3177\ Submitting a properly completed template and accompanying
transaction letter will satisfy the trading requirements in 49 CFR
part 536.
---------------------------------------------------------------------------
Additionally, the credit trade confirmation includes an
acknowledgement of the ``error or fraud'' provisions in 49 CFR
536.8(f)-(g), and the finality provision of 49 CFR 536.8(g). NHTSA
sought comment on this approach, as well as on any changes to the
template that may be necessary to facilitate manufacturer credit
transaction requests. The agency uploaded the proposed template to the
NHTSA's docket and the CAFE PIC site for manufacturers to download and
review.
Only Global Automakers commented on the proposed credit transaction
template, and Global Automakers supported adopting a uniform template.
Global Automakers stated that, in theory, it agrees that a standardized
template with credit trading information is appropriate, and a similar
template is already in use for these types of reporting requirements by
its members that could be integrated into the end of the year EPA final
report. Global Automakers believes the use of similar templates have
been well-established, and such a template could be implemented across
multiple agencies (i.e. NHTSA and EPA) with very little lag time in
learning.\3178\ No comments were received on the transaction letter
generated by the template.
---------------------------------------------------------------------------
\3178\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
---------------------------------------------------------------------------
For the final rule, NHTSA is finalizing the proposed requirements
for its credit templates to be incorporated into provisions for Part
536. NHTSA understands that manufacturers may be using similar credit
reporting templates as part of their current business processes but has
decided to adopt the template proposed in the NPRM. The NHTSA credit
templates are an integral part of a long-range technology deployment
that is already underway and will automate the NHTSA's CAFE database
and web portal systems. When complete, the systems and portals will
receive information directly from manufacturers and enable
manufacturers, independently, to confirm credit trades and receive
real-time credit balances. For this reason, diverging from the proposed
templates for the final rule would impose unnecessary costs upon NHTSA.
In the interest of accommodating the transition by manufacturers from
other standardized templates, the agency will delay mandatory use of
the CAFE credit template until January 1, 2021. Manufacturers may
deviate from the generated language in the NHTSA credit trade
confirmation by adding additional qualifications but, at a minimum,
must include the core information generated by the template.
(c) Credit Transaction Information
Credit trading among entities commenced in the CAFE program
starting in MY 2011.\3179\ To date, NHTSA has received numerous credit
trades from manufacturers but has only made limited information
publicly available.\3180\ As discussed earlier, NHTSA maintains an
online CAFE database with manufacturer and fleetwide compliance
information that includes year-by-year accounting of credit balances
for each credit holder. While NHTSA maintains this database, the
agency's regulations currently state that it does not publish
information on individual transactions, and NHTSA has not previously
required trading entities to submit information regarding the
compensation (whether financial, or other items of value) manufacturers
receive in exchange for credits.3181 3182 Thus, NHTSA's PIC
offers sparse information to those looking to determine the value of a
credit.
---------------------------------------------------------------------------
\3179\ 49 CFR 536.6(c).
\3180\ Manufacturers may generate credits, but non-manufacturers
may also hold or trade credits. Thus, the word ``entities'' is used
to refer to those that may be a party to a credit transaction.
\3181\ 49 CFR 536.5(e)(1).
\3182\ NHTSA understands that not all credits are exchanged for
monetary compensation. The proposal that NHTSA is adopting in this
final rule requires entities to report compensation exchanged for
credits, and is not limited to reporting monetary compensation.
---------------------------------------------------------------------------
The lack of information regarding credit transactions means
entities wishing to trade credits have little, if any, information to
determine the value of the credits they seek to buy or sell. It is
widely assumed that the civil penalty for noncompliance with CAFE
standards largely determines the upper value of a credit, because it is
logical to assume that manufacturers would not purchase credits if it
cost less to pay civil penalties instead, but it is unknown how other
factors affect the value. For example, a credit nearing the end of its
five-model-year lifespan would theoretically be worth less than a
credit within its full five-model-year lifespan. In the latter case,
the credit holder would likely value the credit more, as it can be used
for compliance purposes for a longer period of time.
In the interest of facilitating a transparent and efficient credit
trading market, NHTSA stated in the NPRM that consideration is being
given to modifying its regulations for credit trade information. NHTSA
sought comment in the NPRM about the feasibility of requiring more
information disclosure around trades, including price information,
noting that neither the public, shareholders, competitors, nor even the
agencies themselves know the price of credit transactions. More
specifically, NHTSA proposed requiring trading parties to submit
information disclosing the identities of the parties to credit trades,
the number of credits traded, and the amount of compensation exchanged
for credits. Furthermore, NHTSA proposed that regulations
[[Page 25219]]
would also permit the agency to publish information about specific
transactions on the PIC.
NHTSA received comments from Volkswagen, Honda, Fiat Chrysler,
Toyota, Global Automakers, the Auto Alliance, UCS, and from one private
citizen, Mr. Jason Schwartz, regarding the scope of available credit
information. All auto associations and manufacturers requested that
NHTSA maintain the confidentiality of credit trades and transactions.
The remaining commenters felt increased transparency would benefit the
market.
Global Automakers, the Auto Alliance, Fiat Chrysler, and Volkswagen
stated that credit trades are business-to-business, contain internal
information and can involve both financial and non-financial
compensation between parties.\3183\ They stated credit transactions
should be viewed as being similar to other competitive purchase
agreements, which include non-disclosure terms and strict
confidentiality with regard to cost and compensation.\3184\ They
contended that negotiations must remain confidential to protect the
sensitive business practices for both the buyer and seller, and that
revealing purchasing terms could result in a competitive disadvantage
for both.\3185\ Further, it was stated that certain transactions may
not happen if they are publicized for fear of public criticism, making
the program less efficient.\3186\
---------------------------------------------------------------------------
\3183\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032; Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073; Fiat
Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Volkswagen,
Detailed Comments, NHTSA-2017-0069-0583.
\3184\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
\3185\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
\3186\ See, e.g., Fiat Chrysler, Detailed Comments, NHTSA-2018-
0067-11943.
---------------------------------------------------------------------------
Honda added that disclosing trading terms may not be as simple as a
spot purchase at a given price.\3187\ Honda explained that it has
undertaken a number of transactions for both CAFE and CO2
credits, and there has been a range of complexity in these transactions
due to numerous factors that are reflective of the marketplace, such as
the volume of credits, compliance category, credit expiration date, a
seller's compliance strategy, and even the CAFE penalty rate in effect
at that time.\3188\ In addition, Honda stated that automakers have a
range of partnerships and cooperative agreements with their own
competitors.\3189\ Honda commented that credit transactions can be an
offshoot of these broader relationships, and difficult to price
separately and independently.\3190\ Thus, Honda believes there may not
be a reasonable, or even meaningful, presentation of ``market''
information in a transaction ``price.'' \3191\ Finally, Honda concluded
by stating that information on pricing terms and business partner
pairings is highly competitive and, if made public, could divulge to
competitors a buyer's and/or seller's future compliance strategy.\3192\
For these reasons, Honda believes it is appropriate to maintain the
confidentiality of trade terms, pricing information, and of trading
partner identification.\3193\
---------------------------------------------------------------------------
\3187\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3188\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3189\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3190\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3191\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3192\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3193\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
---------------------------------------------------------------------------
Fiat Chrysler stated that revealing credit transaction information
would reveal highly confidential business information.\3194\ It stated
that credit transaction information may reveal the technology that is
most valued by a company and the value of putting certain technology
into a vehicle.\3195\ It believed that credit trades are complex
business transactions made at arm's length.\3196\ As such, they may
include monetary and non-monetary compensation, non-disclosure
provisions, and other sensitive terms.\3197\ Fiat Chrysler commented
that publicizing such sensitive information could stifle the credit
market and potentially result in uncompetitive outcomes, and could also
decrease the efficiency in the credit trading marketplace.\3198\ Fiat
Chrysler further stated that the NPRM's justifications for requiring
the disclosure of credit transaction information is unfounded and the
government has no need of this information in the regular course of
doing business.\3199\
---------------------------------------------------------------------------
\3194\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3195\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3196\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3197\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3198\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3199\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
The Auto Alliance, Honda, Toyota, and Volkswagen argued against
NHTSA publishing credit movements each model year on its PIC. They
stated that detailed credit banks by account holder are available to
the public or entities wishing to engage in the credit market and that
information is already sufficient.\3200\ Global Automakers further
contended that the agencies know which companies are trading and how
those credits are being used, which is all that should be required for
administering the program.\3201\ The Auto Alliance argued that in
private markets, trades and prices often are not made public; this
privacy does not mean that the markets operate any less effectively,
nor that the public at large does not benefit from the transactions
that lower costs for all parties.\3202\
---------------------------------------------------------------------------
\3200\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Honda, Detailed Comments, NHTSA-2018-0067-11818; Toyota, Detailed
Comments, NHTSA-2018-0067-12150; Volkswagen, Detailed Comments,
NHTSA-2017-0069-0583.
\3201\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
\3202\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Volkswagen further commented that revealing confidential purchase
terms has no precedent in the automotive industry. Volkswagen's
position is that it does not disclose contract pricing for purchasing
fuel saving technologies from suppliers, such as for turbochargers or
battery packs. Therefore, Volkswagen does not believe it is appropriate
to disclose the purchase price for CAFE credits.\3203\
---------------------------------------------------------------------------
\3203\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
Opposite views from those expressed by automobile manufacturers
were received in the comments from UCS and Jason Schwartz. Both
commenters strongly supported an increase in information regarding
credit trading in the CAFE program.\3204\ They argued that more
information will allow manufacturers to make better informed decisions
and lead to greater industry efficiency in general.\3205\ UCS added
that while the PIC does have some information, it is difficult to
discern how the manufacturers are dividing credits to offset
shortfalls.\3206\ It requested NHTSA disclose at least as much
information as EPA provides from its program, if not providing more
information on transaction price and
[[Page 25220]]
compliance category.\3207\ Jason Schwartz had similar arguments for
more transparency. Mr. Schwartz added that the agencies can assume that
credits may be traded at prices similar to the civil penalty rate for
noncompliance under the CAFE standards, but not knowing the actual
prices greatly complicates the agencies' estimations of the costs of
complying with the standards.\3208\ Schwartz used several examples to
explain and justify the need for making data on credit transactions,
prices, and holdings publicly available to help the agency and the
public assess the efficacy of the program.\3209\ He also explained that
such information will enable the smooth operation of the credit market
by enabling credit buyers to better evaluate the value of credits and
placing all players on equal informational footing which facilitates
price discovery, and assists buyers and sellers in reaching
terms.\3210\ He added that regulators should require greater
transparency to facilitate oversight.\3211\ He asserted his belief that
greater transparency in tracking transactions and credits helps
regulators detect fraud, manipulation, market power, abuse, and to
enforce compliance.\3212\
---------------------------------------------------------------------------
\3204\ UCS, Detailed Comments, NHTSA-2018-0067-12039; Jason
Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
\3205\ See, e.g., UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3206\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3207\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3208\ Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
\3209\ Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
\3210\ Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
\3211\ Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
\3212\ Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162.
---------------------------------------------------------------------------
In response to these comments, NHTSA has decided not to share
detailed information on credit transactions or the cost of individual
credit transactions with the public. NHTSA agrees with manufacturers
that revealing confidential purchase terms could result in a
competitive disadvantage for both credit buyers and sellers, as well as
harm to companies revealing highly confidential business materials.
However, NHTSA believes that greater government oversight is needed
over the CAFE credit market. NHTSA needs to understand more information
surrounding trades, including costing information. As Honda recognized
in its comments, NHTSA needs to understand the full range of complexity
in transactions, monetary and non-monetary, in addition to the range of
partnerships and cooperative agreements between credit account
holders--which may impact the price of credit trades.\3213\ NHTSA also
believes, as mentioned by commenters, that disclosure of information
concerning credit trades is important for facilitating government
oversight for protecting against fraud, manipulation, market power, and
abuse which may occur in the credit market.
---------------------------------------------------------------------------
\3213\ Honda, Detailed Comments, NHTSA-2018-0067-11819.
---------------------------------------------------------------------------
NHTSA is adopting new reporting provisions in this final rule.
Starting January 1, 2021, manufacturers will be required to submit all
credit trade contracts, including costing and transactional
information, to the agency. This information may be submitted
confidentially, in accordance with 49 CFR part 512.\3214\ NHTSA will
use this information to determine the true cost of compliance for all
manufacturers. This information will allow NHTSA to assess better the
impact of its regulations on the industry, and provide more insightful
information to use in developing future rulemakings. This confidential
information will be held by secure electronic means in NHTSA's database
systems. As for public information, NHTSA will include more information
on the PIC on aggregated credit transactions, such as the combined
flexibilities all manufacturers used for compliance as shown in Figure
IX-6, or information comparable to the credit information EPA makes
available to the public. In the future, NHTSA will consider what
information, if any, can be meaningfully shared with the public on
credit transactional details or costs, while accounting for the
concerns raised by the automotive industry.
---------------------------------------------------------------------------
\3214\ See also 49 U.S.C. 32910(c).
---------------------------------------------------------------------------
(d) Precision of the CAFE Credit Adjustment Factor
EPCA, as amended by EISA, required the Secretary of Transportation
to establish an adjustment factor to ensure total oil savings are
preserved when manufacturers trade credits.\3215\ The adjustment factor
applies to credits traded between manufacturers and to credits
transferred across a manufacturer's compliance fleets.
---------------------------------------------------------------------------
\3215\ 49 U.S.C. 32903(f)(1).
---------------------------------------------------------------------------
In establishing the adjustment factor, NHTSA did not specify the
exact precision of the output of the equation in 49 CFR 536.4(b).
NHTSA's standard practice has been round to the nearest four decimal
places (e.g., 0.0001) for the adjustment factor. However, in the
absence of a regulatory requirement, many manufacturers have contacted
NHTSA for guidance, and NHTSA has had to correct several credit
transaction requests. In some instances, manufacturers have had to
revise signed credit trade documents and submit additional trade
agreements to properly address credit shortfalls.
NHTSA proposed in the NPRM to add requirements to 49 CFR 536.4
specifying the precision of the adjustment factor by rounding to four
decimal places (e.g., 0.0001). NHTSA has also included equations for
the adjustment factor in its proposed credit transaction report
template, mentioned above, with the same level of precision. NHTSA
sought comment on this approach but received no comments, and therefore
is finalizing this approach in this final rule.
(3) NHTSA Then Analyzes EPA-Certified CAFE Values for Compliance
After manufacturers complete certification testing and submit their
final compliance values to EPA, EPA verifies the data and issues final
CAFE reports to manufacturers and NHTSA. NHTSA then evaluates whether
the manufacturers' compliance categories (i.e., domestic passenger car,
imported passenger car, and light truck fleets) meet the applicable
CAFE standards. NHTSA uses EPA-verified data to compare fleet average
standards with actual fleet performance values in each compliance
category. Each vehicle a manufacturer produces has a fuel economy
target based on its footprint (footprint curves are discussed above in
Section II.C), and each compliance category has a CAFE standard
measured in miles per gallon (mpg). The manufacturer's fleet average
CAFE standard is calculated based on the fuel economy target value and
production volume of each vehicle model. The CAFE performance is
calculated based on the compliance value and production volume of each
vehicle model. A manufacturer complies with the CAFE standard if its
fleet average performance is greater than or equal to its required
standard, or if it is able to use available compliance flexibilities,
described below in Section IX.C.2.c. to resolve any shortfall.
If the average fuel economy level of the vehicles in a compliance
category falls below the applicable fuel economy standard, NHTSA
provides written notification to the manufacturer that it has not met
that standard. The manufacturer is then required to confirm the
shortfall and either submit a plan indicating how it will allocate
existing credits, or if it does not have sufficient credits available
in that fleet, how it will earn, transfer, and/or acquire credits, or
pay the appropriate civil penalty. The manufacturer must submit a
credit allocation plan or payment within 60 days of receiving agency
notification.
[[Page 25221]]
NHTSA approves a credit allocation plan unless it finds the
proposed credits are unavailable or that it is unlikely that the plan
will result in the manufacturer earning sufficient credits to offset
the projected shortfall. If a plan is approved, NHTSA revises the
manufacturer's credit account accordingly. If a plan is rejected, NHTSA
notifies the manufacturer and requests a revised plan or payment of the
appropriate civil penalty. Similarly, if the manufacturer is delinquent
in submitting a response within 60 days, NHTSA takes action to collect
a civil penalty. If NHTSA receives and approves a manufacturer's plan
to carryback future earned credits within the following three years in
order to comply with current regulatory obligations, NHTSA will defer
levying civil penalties for noncompliance until the date(s) when the
manufacturer's approved plan indicates that the credits will be earned
or acquired to achieve compliance. If the manufacturer fails to acquire
or earn sufficient credits by the plan dates, NHTSA will initiate
noncompliance proceedings to collect civil penalties.\3216\
---------------------------------------------------------------------------
\3216\ See generally 49 CFR part 536.
---------------------------------------------------------------------------
(4) Civil Penalties for Noncompliance
In the event that a manufacturer does not comply with a CAFE
standard, EPCA provides that the manufacturer is potentially liable for
a civil penalty.\3217\ The manufacturer determines whether to use
available credits to reduce or offset its potential penalty.\3218\ This
penalty rate is $5.50 for each tenth of a mpg that a manufacturer's
average fuel economy falls short of the standard for a given model year
multiplied by the total volume of those vehicles in the affected
compliance category manufactured for that model year.\3219\ A person
(or manufacturer) that violates 49 U.S.C. 32911(a), including general
CAFE violations other than those for failing to comply with CAFE
standards (i.e., fuel economy labeling violations), is also liable to
the United States Government for a civil penalty of not more than
$42,530 for each violation. A separate violation occurs for each day
the violation continues. All penalties are paid to the U.S. Treasury
and not to NHTSA.\3220\
---------------------------------------------------------------------------
\3217\ 49 U.S.C. 32911-12.
\3218\ See 49 U.S.C. 32912.
\3219\ NHTSA finalized a retaining the $5.50 civil penalty rate
in an April 2018 NPRM. See 83 FR 13904 (Apr. 2, 2018).
\3220\ 49 U.S.C. 32912(e) allows for fiscal year 2008 and each
fiscal year thereafter, the total amount deposited in the general
fund of the Treasury during the preceding fiscal year from fines,
penalties, and other funds obtained through enforcement actions
conducted pursuant to EISA and EPCA (including funds obtained under
consent decrees), the Secretary of the Treasury, subject to the
availability of appropriations, shall: (1) transfer 50 percent of
such total amount to the account providing appropriations to the
Secretary of Transportation for the administration of this chapter,
which shall be used by the Secretary to support rulemaking under
this chapter; and (2) transfer 50 percent of such total amount to
the account providing appropriations to the Secretary of
Transportation for the administration of this chapter, which shall
be used by the Secretary to carry out a program to make grants to
manufacturers for retooling, reequipping, or expanding existing
manufacturing facilities in the United States to produce advanced
technology vehicles and components.
Potential Civil Penalty = $5.50 x (Avg. FE Performance-Avg. FE
---------------------------------------------------------------------------
Standard) x 10 x Total Production
Since the inception of the CAFE program, the U.S. Treasury has
collected a total of $1,049,355,116 in CAFE civil penalty payments.
Generally, import manufacturers have paid significantly more in civil
penalties than domestic manufacturers, with the majority of payments
made by import manufacturers for passenger cars and not light trucks.
Over the total program lifetime, import manufacturers paid a total of
$1,048,896,676 in CAFE penalties while domestic manufacturers paid a
total of $458,440.\3221\
---------------------------------------------------------------------------
\3221\ These totals include penalties associated with all fleets
for these manufacturers. For example, the total penalties paid by
import manufacturers includes penalties associated with shortfalls
in those manufacturers' domestic passenger car fleets.
---------------------------------------------------------------------------
Prior to the CAFE credit trade and transfer program, several
manufacturers opted to pay civil penalties instead of complying with
CAFE standards. Since NHTSA introduced trading and transferring,
manufacturers have largely traded or transferred credits to achieve
compliance, rather than paying civil penalties for noncompliance. NHTSA
therefore assumes that buying and selling credits is a more cost-
effective strategy for manufacturers than paying civil penalties, in
part, because it seems logical that the price of a credit is directly
related to the civil penalty rate and decreases as a credit's life
diminishes.\3222\ Prior to trading and transferring, on average,
manufacturers paid $28,073,281.93 in civil penalty payments annually (a
total of $814,125,176 from MYs 1982 to 2010). Since trading and
transferring began, manufacturers now pay an average of $26,136,660
each model year. The agency notes that six manufacturers have paid
civil penalties since 2011 totaling $235,229,940; Fiat Chrysler paid a
civil penalty in MY 2016 equal to $77,268,720.50 and in MY 2017 equal
to $79,376,643.50 for for failing to meet the minimum domestic
passenger car standards for those MYs. NHTSA expects that, over the
next several years, manufacturers will face challenges in avoiding
paying further civil penalties as standards increase in stringency.
Compared to the current $5.50 CAFE civil penalty rate, a rate of $14
would cause manufacturers that do not comply with CAFE to pay
significantly higher civil penalties, potentially in the magnitude of
hundreds of millions of dollars annually beyond current projections.
Additionally, although NHTSA has not historically been privy to the
monetary terms of credit trades, NHTSA expects that the price of
credits would increase in line with any increase in the CAFE civil
penalty rate.
---------------------------------------------------------------------------
\3222\ See 49 CFR 536.4 for NHTSA's regulations regarding CAFE
credits.
---------------------------------------------------------------------------
b) What Exemptions and Exclusions Does NHTSA Allow?
(1) Emergency and Law Enforcement Vehicles
Under EPCA, manufacturers are allowed to exclude emergency
vehicles, which include law enforcement vehicles, from their CAFE
fleet.\3223\ All manufacturers that produce emergency vehicles have
historically done so. NHTSA did not propose any changes to this
exclusion and therefore is retaining the provision without change for
the final rule.
---------------------------------------------------------------------------
\3223\ 49 U.S.C. 32902(e).
---------------------------------------------------------------------------
(2) Small Volume Manufacturers
Per 49 U.S.C. 32902(d), NHTSA established requirements for exempted
small volume manufacturers in 49 CFR part 525, ``Exemptions from
Average Fuel Economy Standards.'' The small volume manufacturer
exemption is available for any manufacturer whose projected or actual
combined sales (whether in the U.S. or not) are fewer than 10,000
passenger automobiles in the model year two years before the model year
for which the manufacturer seeks an exemption.\3224\ The manufacturer
must submit a petition with information stating that the applicable
CAFE standard is more stringent than the maximum feasible average fuel
economy level that the manufacturer can achieve.\3225\ NHTSA must then
issue by Federal Register notice, a proposed decision granting or
denying the petition and inviting public comment.\3226\ If the agency
proposed to grant the petition, the notice includes an alternative
average fuel economy standard for the passenger automobiles
manufactured by the manufacturer.\3227\ After conclusion of the public
comment period, the agency publishes a final
[[Page 25222]]
decision in the Federal Register.\3228\ If the agency grants the
petition, it establishes an alternative standard, which is the maximum
feasible average fuel economy level for the manufacturers to which the
alternative standard applies.\3229\ NHTSA did not propose and is not
making any changes to the small volume manufacturer provision or
alternative standards regulations in this rulemaking.
---------------------------------------------------------------------------
\3224\ 49 CFR 525.5.
\3225\ 49 CFR 525.7(h).
\3226\ 49 CFR 525.8(c).
\3227\ Id.
\3228\ 49 CFR 525.8(e).
\3229\ 49 U.S.C. 32902(d)(2); 49 CFR 525.8(e).
---------------------------------------------------------------------------
c) What Compliance Flexibilities and Incentives Are Currently Available
Under the CAFE Program and How Do Manufacturers Use Them?
There are several compliance flexibilities and incentives that
manufacturers can use to achieve compliance with CAFE standards beyond
applying fuel economy-improving technologies. Some compliance
flexibilities and incentives are statutorily mandated by Congress
through EPCA and EISA. These specifically include program credits
generated from overcompliance, including the ability to carry-forward,
carryback, trade and transfer credits, and special fuel economy
calculations for dual- and alternative-fueled vehicles (discussed in
turn, below). However, 49 U.S.C. 32902(h) expressly prohibits NHTSA
from considering the availability of statutorily established credits
(either for building dual- or alternative-fueled vehicles or from
accumulated transfers or traders) in setting the level of the
standards. Thus, NHTSA may not raise CAFE standards because
manufacturers have enough credits to meet higher standards, or because
alternative fuel vehicles (including electric vehicles) are available
to help manufacturers achieve compliance. This is an important
difference from EPA's authority under the CAA, which does not contain
such a restriction, and which flexibility EPA has utilized in the past
in determining appropriate levels of stringency for its program.
Generating, trading, transferring, and applying CAFE credits is
governed by statute.\3230\ Program credits are generated when a vehicle
manufacturer's fleet over-complies with its standard for a given model
year, meaning its vehicle fleet achieved a higher corporate average
fuel economy value than the amount required by the CAFE program for
that fleet in that model year. Conversely, if the fleet average CAFE
level does not meet the standard, the fleet would incur debits (also
referred to as a shortfall). A manufacturer whose fleet generates a
credit shortfall in a given model year can resolve its shortfall using
any one or combination of several credits flexibilities, including
credit carryback, credit carry-forward, credit transfers, and credit
trades.
---------------------------------------------------------------------------
\3230\ 49 U.S.C. 32903.
---------------------------------------------------------------------------
NHTSA also has promulgated compliance flexibilities and incentives
consistent with EPCA's provisions regarding calculation of fuel economy
levels for individual vehicles and for fleets.\3231\ These compliance
flexibilities and incentives, which were first adopted in the 2012 rule
for MYs 2017 and later, include A/C efficiency improvement and off-
cycle adjustments, and adjustments for advanced technologies in full-
size pickup trucks, including adjustments for mild and strong hybrid
electric full-size pickup trucks and performance-based incentives in
full-size pickup trucks. The fuel consumption improvement benefits of
these technologies measured by various testing methods can be used by
manufacturers to increase the CAFE performance of their fleets. As
discussed below, the adjustments for advanced technologies in full-size
pickup trucks will no longer be available beginning in MY 2022.
---------------------------------------------------------------------------
\3231\ 49 U.S.C. 32904.
---------------------------------------------------------------------------
Under NHTSA regulations, credit holders (including, but not limited
to manufacturers) have credit accounts with NHTSA where they can, as
outlined above, hold credits, and use them to achieve compliance with
CAFE standards, by carrying forward, carrying back, or transferring
credits across compliance categories. Manufacturers with excess credits
in their accounts can also trade credits to other manufacturers, who
may use those credits to resolve a shortfall currently or in a future
model year. A credit may also be cancelled before its expiration date
if the credit holder so chooses. Traded and transferred credits are
subject to an ``adjustment factor'' to ensure total oil savings are
preserved.\3232\ Credits earned before MY 2011 may not be traded or
transferred.\3233\
---------------------------------------------------------------------------
\3232\ 49 CFR 536.4(c).
\3233\ 49 CFR 536.6(c).
---------------------------------------------------------------------------
Credit ``carryback'' means that manufacturers are able to use
credits to offset a deficit that had accrued in a prior model year,
while credit ``carry-forward'' means that manufacturers can bank
credits and use them towards compliance in future model years. EPCA, as
amended by EISA allows manufacturers to carryback credits for up to
three model years, and to carry-forward credits for up to five model
years.\3234\ Credits expire the model year after which the credits may
no longer be used to achieve compliance with fuel economy
regulations.\3235\ Manufacturers seeking to use carryback credits must
have an approved carryback plan from NHTSA demonstrating their ability
to earn sufficient credits in future MYs that can be carried back to
resolve the current MY's credit shortfall.
---------------------------------------------------------------------------
\3234\ 49 U.S.C. 32903(a).
\3235\ 49 CFR 536.3(b).
---------------------------------------------------------------------------
Credit ``trading'' refers to the ability of manufacturers or
persons to sell credits to, or purchase credits from, one another. EISA
gave NHTSA discretion to establish by regulation a CAFE credit trading
program, to allow credits to be traded between vehicle manufacturers,
now codified at 49 CFR part 536.\3236\ EISA prohibited manufacturers
from using traded credits to meet the minimum domestic passenger car
CAFE standard.\3237\
---------------------------------------------------------------------------
\3236\ 49 U.S.C. 32903(f).
\3237\ 49 U.S.C. 32903(f)(2).
---------------------------------------------------------------------------
As mentioned previously, the agencies sought comments in the NPRM
on whether and how each agency's existing flexibilities and incentives
might be amended, revised, or deleted to avoid the inefficiencies and
market distortions as discussed earlier. NHTSA was concerned with the
potential for unintended consequences. Specifically, comments were
sought on the appropriate level of compliance flexibilities, including
credit trading, in a program that is correctly designed to follow
statutory direction to create maximum feasible fuel economy standards.
Given that the credit trading program is discretionary under EISA,
NHTSA also sought comments on whether the credit trading provisions in
49 CFR part 536 should cease to apply beginning in MY 2022. Comments
were sought on whether to allow all incentive-based adjustments, except
those that are mandated by statute, to expire, in addition to other
possible simplifications to reduce market distortion, improve program
transparency and accountability, and improve overall performance of the
compliance programs.
The comments received from the public and NHTSA's responses to
those comments are discussed below. A summary of all the flexibilities
and incentives, and information on whether they were either retained or
modified for the final rule, is presented in Table IX-1 through Table
IX-4.
[[Page 25223]]
(1) Credit Carry-Forward and Back
Under the CAFE program, when the average fuel economy of a
compliance fleet manufactured in a particular model year exceeds its
applicable average fuel economy standard, the manufacturer earns
credits.\3238\ The credits may be applied to: (1) Any of the 3
consecutive model years immediately before the model year for which the
credits are earned; and (2) any of the 5 consecutive model years
immediately after the model year for which the credits are earned. For
example, a credit earned for exceeding model year 2017 standards will
be usable for compliance purposes through and including the 2022
compliance model year. NHTSA did not seek comment on or propose changes
to any of the aspects of its lifespan for CAFE credits because of the
existing statutory limitation set forth by Congress. The public offered
no comments on such flexibilities under the CAFE program.
---------------------------------------------------------------------------
\3238\ 49 U.S.C. 32903 and 49 CFR 536.
---------------------------------------------------------------------------
(2) Credit Trading
All commenters responding to the NPRM on this issue favored
retaining the existing CAFE credit trading program. Comments on credit
trading were received from Volkswagen, Honda, General Motors, CARB,
BorgWarner, Jaguar Land Rover, Fiat Chrysler, Global Automakers, the
Auto Alliance, the Institute for Policy Integrity, Toyota, and academic
commenters, Jeremy Michalek and Jason Schwartz. No comments were
received supporting the idea of changing the existing credit trading
program.
In general, manufacturers' comments centered around problems in
predicting whether consumers will purchase the fuel efficient vehicles
necessary for manufacturers to meet their compliance obligations. They
stated that continuing the credit trading program allows manufacturers
to address uncertainty in the market better.\3239\ The Auto Alliance,
Volkswagen, Fiat Chrysler, and Honda commented that credit
flexibilities allow manufacturers to comply with the program even when
faced with market uncertainties.\3240\ Honda stated that credit trading
allows the government to set reasonable standards without fear of
having to cater to the least-capable manufacturer.\3241\ Jaguar Land
Rover stated the removal of NHTSA's credit trading programs would
increase and intensify the dis-harmonization between the CO2
and CAFE programs.\3242\
---------------------------------------------------------------------------
\3239\ See, e.g., Fiat Chrysler, Detailed Comments, NHTSA-2018-
0067-11943.
\3240\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Volkswagen, Detailed Comments, NHTSA-2017-0069-0583-22; Fiat
Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Honda, Detailed
Comments, NHTSA-2018-0067-11818.
\3241\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3242\ Jaguar Land Rover, Detailed Comments, NHTSA-2018-0067-
11916-9.
---------------------------------------------------------------------------
Global Automakers, Fiat Chrysler, Jason Schwartz, and Jeremy
Michalek each commented that the credit trading program allows for a
more efficient compliance process given that more fuel-efficient
manufacturers can sell their credits to manufacturers who fall
short.\3243\ These commenters and BorgWarner stated that the program
lowers the overall cost of reducing fuel consumption.\3244\ Likewise,
Jaguar Land Rover, Fiat Chrysler, and General Motors argued compliance
flexibilities, like trading, increase the ability to achieve higher
fuel economy and reduced CO2 emissions. They found that the
credit trading flexibility allows them to invest more money in
technologies that will lead to future increases in their fuel
economy.\3245\ Similarly, CARB argued credit flexibilities have been
shown to be successful in reducing emissions and spurring innovation.
It saw no reason to remove a successful program.\3246\
---------------------------------------------------------------------------
\3243\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032; Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943;
Jason Schwartz, Detailed Comments, NHTSA-2018-0067-12162; Jeremy
Michalek, Detailed Comments, NHTSA-2018-0067-11903.
\3244\ BorgWarner, Detailed Comments, NHTSA-2018-0067-11895.
\3245\ Jaguar Land Rover, Detailed Comments, NHTSA-2018-0067-
11916; Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943;
General Motors, Detailed Comments, NHTSA-2018-0067-11858.
\3246\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
Fiat Chrysler stated that credit trading allows manufacturers to
provide more choices for consumers since manufacturers are not required
to meet the standard exactly, but rather, they can purchase traded
credits and then provide vehicles the public is demanding while still
complying with fleet average standards.\3247\ They stated that this
leads to the overall compliance of the U.S. fleet while allowing for
more consumer choices. They further added that if the program is
removed, manufacturers that currently generate credits from their fuel-
efficient fleet may find it more profitable to begin producing less
fuel-efficient vehicles, perhaps even halting the current improvements
in fuel efficiency across the industry.\3248\
---------------------------------------------------------------------------
\3247\ General Motors, Detailed Comments, NHTSA-2018-0067-11943.
\3248\ General Motors, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
Honda commented that regulatory flexibilities, such as credit
trading, built into the CO2 and CAFE programs have become
critical elements to the programs' success, especially in the face of
product cadences with uneven sales that do not always match compliance
obligations.\3249\ General Motors stated its belief that program
flexibilities will continue to play an increasingly important role in
reducing CO2 emissions and increasing fuel economy through
technologies and innovations.\3250\ CARB stated that existing
flexibilities create consistency in compliance planning for automakers
for model years in the existing program.\3251\ Fiat Chrysler added that
each of the CAFE and CO2 programmatic tools and
flexibilities should be retained, improved and strengthened. Fiat
Chrysler opined that this is a chance for the agencies to make better
policies that work more efficiently and as intended, and cautioned that
eliminating them now could have the serious negative impact of making
the standards more stringent and costlier for manufacturers.\3252\
---------------------------------------------------------------------------
\3249\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3250\ General Motors, Detailed Comments, NHTSA-2018-0067-11858.
\3251\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
\3252\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
NHTSA is not making changes to its credit trading provisions in the
final rule. NHTSA sought comments on removing the optional credit
trading program to explore public views on market distortions or
windfalls that occur as a result of the credit trading program.
However, commenters consistently opined that removing existing
flexibilities might result in manufacturers not building certain types
of vehicles. This could adversely impact compliance plans over multiple
model years. NHTSA concurs with those views, and since this final rule
adopts CAFE standards that continuously increase through MY 2026,
understands the importance of allowing for credit trading to provide
additional means of achieving compliance for manufacturers who face
varying degrees of difficulty in achieving the standards the agencies
are finalizing today. With increasing standards, credit trading
flexibilities help to compensate for the possibility of an uneven sales
mix of vehicle types and to aid with compliance planning.
[[Page 25224]]
Final sales volumes, as presented earlier, show a shift over the past
several years in consumers purchasing more small SUVs subject to
passenger car standards, and these vehicles are less fuel efficient
than the compact and mid-sized passenger cars that previously dominated
the market. The need to ensure consumer choice is adequately considered
drives the need for NHTSA to provide credit trading flexibility to
manufacturers. For example, even with increasing standards, a
manufacturer could continue to sell certain types of vehicles with
lower mpg performance over a longer period of time to satisfy its
consumers by purchasing credits or carrying credits back from future
model years to address the mpg fleet shortages caused by these
vehicles, before ultimately having to introduce more fuel-efficient
technologies. NHTSA believes that these types of scenarios are
consistent with the purpose of the CAFE credit program, as adopted by
Congress.
(3) Credit Transferring
Credit ``transfer'' means the ability of manufacturers to move
credits from their passenger car fleet to their light truck fleet, or
vice versa. As part of the EISA amendments to EPCA, NHTSA was required
to establish by regulation a CAFE credit transferring program, now
codified at 49 CFR part 536, to allow a manufacturer to transfer
credits between its car and truck fleets to achieve compliance with the
standards.\3253\ For example, credits earned by overcompliance with a
manufacturer's car fleet average standard may be used to offset debits
incurred because of that manufacturer's failed to meet the truck fleet
average standard in a given year. However, EISA imposed a cap on the
amount by which a manufacturer could raise its CAFE performance through
transferred credits: 1 mpg for MYs 2011-2013; 1.5 mpg for MYs 2014-
2017; and 2 mpg for MYs 2018 and beyond.\3254\ These statutory limits
will continue to apply to the determination of compliance with CAFE
standards. EISA also prohibits the use of transferred credits to meet
the minimum domestic passenger car fleet CAFE standard.\3255\
---------------------------------------------------------------------------
\3253\ See 49 U.S.C. 32903(g)(1).
\3254\ 49 U.S.C. 32903(g)(3).
\3255\ 49 U.S.C. 32903(g)(4).
---------------------------------------------------------------------------
In the NPRM, NHTSA responded to the 2016 petition for rulemaking
from the Auto Alliance and Global Automakers (Alliance/Global or
Petitioners) asking to amend the regulatory definition of ``transfer''
as it pertains to compliance flexibilities.\3256\ In particular,
Alliance/Global requested that NHTSA add text to the definition of
``transfer'' stating that the statutory transfer cap in 49 U.S.C.
32903(g)(3) applies when the credits are transferred. Alliance/Global
assert that adding this text to the definition is consistent with
NHTSA's prior position on this issue in the MYs 2012-2016 final rule,
in which NHTSA stated:
---------------------------------------------------------------------------
\3256\ Auto Alliance and Global Automakers Petition for Direct
Final Rule with Regard to Various Aspects of the Corporate Average
Fuel Economy Program and the Greenhouse Gas Program (June 20, 2016)
at 13, available at https://www.epa.gov/sites/production/files/2016-09/documents/petition_to_epa_from_auto_alliance_and_global_automakers.pdf
[hereinafter Alliance/Global Petition].
NHTSA interprets EISA not to prohibit the banking of transferred
credits for use in later model years. Thus, NHTSA believes that the
language of EISA may be read to allow manufacturers to transfer
credits from one fleet that has an excess number of credits, within
the limits specified, to another fleet that may also have excess
credits instead of transferring only to a fleet that has a credit
shortfall. This would mean that a manufacturer could transfer a
certain number of credits each year and bank them, and then the
credits could be carried forward or back `without limit' later if
and when a shortfall ever occurred in that same fleet.\3257\
---------------------------------------------------------------------------
\3257\ 75 FR 25666 (May 7, 2010).
NHTSA clarified in the NPRM, based upon a previous interpretation,
that the transfer cap from EISA does not limit how many credits may be
transferred in a given model year, but it does limit the application of
transferred credits to a compliance category in a model year.\3258\ The
interpretation concludes by stating that, ``Thus, manufacturers may
transfer as many credits into a compliance category as they wish, but
transferred credits may not increase a manufacturer's CAFE level beyond
the statutory limits.'' \3259\
---------------------------------------------------------------------------
\3258\ See, letter from O. Kevin Vincent, Chief Counsel, NHTSA
to Tom Stricker, Toyota (July 5, 2011), available at https://isearch.nhtsa.gov/files/10-004142%20-%20Toyota%20CAFE%20credit%20transfer%20banking%20-%205%20Jul%2011%20final%20for%20signature.htm (last accessed Apr.
18, 2018).
\3259\ Id.
---------------------------------------------------------------------------
NHTSA maintains its views that the transfer caps in 49 U.S.C.
32903(g)(3) are properly read to apply to the application of credits.
As NHTSA explained in the NPRM, it understands that the language in the
MYs 2012-2016 final rule could be read to suggest that the transfer cap
applies at the time credits are transferred. However, NHTSA believes
its existing interpretation--that the transfer cap applies at the time
the credits are used--is a more appropriate, plain language reading of
the statute. While manufacturers have approached NHTSA with various
interpretations that would essentially allow them to circumvent the
EISA transfer cap, NHTSA believes such interpretations are improper
because they would not give effect to the statutory transfer cap.
Therefore, NHTSA proposed in the NPRM to deny Alliance/Global's
petition to revise the definition of ``transfer'' in 49 CFR 536.3, and
is now finalizing that denial.
In response to the tentative denial of the petition above in the
NPRM, comments were received from the Global Automakers and Toyota
asking NHTSA to reconsider applying the transfer cap of 2.0 mpg per
year when credits are transferred rather than when they are
applied.\3260\ They reiterated that imposing the cap when applying the
credits is overly burdensome, but did not provide any new information
that has persuaded NHTSA to change its view that the petition should be
denied. The Auto Alliance also stated that NHTSA should revise its
definition of ``transfer'' to be more consistent with EPA.\3261\
---------------------------------------------------------------------------
\3260\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032; Toyota, Detailed Comments, NHTSA-2018-0067-12150.
\3261\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Other more general comments to the NPRM were also received from
Walter Kreucher, Jeremy Michalek, Global Automakers, the Auto Alliance,
and Toyota, regarding the use of the credit transfer flexibility. These
commenters generally appreciated the transfer flexibility for its
ability to reduce compliance costs.\3262\ More specifically, Walter
Kreucher commented that the ability to transfer credits between
compliance categories was beneficial for manufacturers and allowed for
efficiency in the markets and reduce compliance costs.\3263\
---------------------------------------------------------------------------
\3262\ See, e.g., Global Automakers, Detailed Comments, NHTSA-
2018-0067-12032.
\3263\ Walter Kreucher, Detailed Comments, NHTSA-2018-0067-0444.
---------------------------------------------------------------------------
For the final rule, NHTSA is not making any changes to the existing
provisions regarding transferring credits. NHTSA's position remains
unchanged that the transfer cap in 49 U.S.C. 32903(g)(1) clearly limits
the amount of performance increase for a manufacturer's fleet that
fails to achieve the prescribed standards. The same statutory provision
prevents NHTSA from changing its definition for transfer to be
consistent with EPA. Consequently, NHTSA is not changing its definition
or its previous interpretation that the application of transfer caps
applies at the time the credits are used and not when
[[Page 25225]]
transferred. Therefore, NHTSA is finalizing its decision to deny the
Auto Alliance and Global Automakers petition.
(4) Minimum Domestic Passenger Car Standard
EPCA, as amended by EISA, addresses the minimum domestic passenger
car standard (MDPCS), clearly stating that any manufacturer's
domestically-manufactured passenger car fleet must meet the greater of
either 27.5 mpg on average, or 92 percent of the average fuel economy
projected by the Secretary for the combined domestic and non-domestic
passenger automobile fleets manufactured for sale in the U.S. by all
manufacturers in the model year, which projection shall be published in
the Federal Register when the standard for that model year is
promulgated in accordance with 49 U.S.C. 32902(b).\3264\ Since that
requirement was added to the statute, NHTSA has always calculated the
``92 percent'' as greater than 27.5 mpg. NHTSA published the 92 percent
MDPCS for MYs 2017-2025 at 49 CFR 531.5(d) as part of the 2012 final
rule. 49 CFR 531.5(e) explains that the published MDPCS for MYs 2022-
2025 are not final and may change when NHTSA sets standards for those
model years. This is consistent with the statutory requirement that the
92 percent standards must be determined at the time an overall
passenger car standard is promulgated and published in the Federal
Register.\3265\ Any time NHTSA establishes or changes a passenger car
standard for a model year, the MDPCS for that model year must also be
evaluated or re-evaluated and established accordingly. Thus, this final
rule establishes the applicable MDPCS for MYs 2021-2026.
---------------------------------------------------------------------------
\3264\ 49 U.S.C. 32902(b)(4).
\3265\ 49 U.S.C. 32904(b)(4)(B).
---------------------------------------------------------------------------
NHTSA considered comments received about the MDPCS, and discusses
the comments and the agency's assessment in Section VIII.B.1.b).
Table IX-7 lists the minimum domestic passenger car standards and
compares them to standards that would correspond to each of the other
regulatory alternatives considered. NHTSA has updated these to reflect
its overall analysis and resultant projection for the CAFE standards
finalized today, highlighted below as ``Preferred (Alternative 3),''
and has calculated what those standards would be under the no action
alternative (as issued in 2012, as updated for the NPRM, and as further
updated by today's analysis) and under the other alternatives described
and discussed further in Section V, above.
[GRAPHIC] [TIFF OMITTED] TR30AP20.757
(5) Fuel Savings Adjustment Factor
Under NHTSA's credit trading regulations, a fuel savings adjustment
factor is applied when trading occurs between manufacturers or when a
manufacturer transfers credits between its fleets, but not when a
manufacturer carries credits forward or carries back credits within the
same fleet.\3266\ The Alliance/Global requested in their 2016 petition
that NHTSA require manufacturers to apply the fuel savings adjustment
factor when credits are carried forward or carried back within the same
fleet, including for existing, unused credits.
---------------------------------------------------------------------------
\3266\ See 49 CFR 536.4(c).
---------------------------------------------------------------------------
Per EISA, total oil savings must be preserved in NHTSA's credit
trading program.\3267\ The statutory provisions for credit transferring
within a manufacturer's fleet do not explicitly include the same
requirement; however, NHTSA prescribed a fuel savings adjustment factor
that applies to both credit trades between manufacturers and credit
transfers between a manufacturer's compliance fleets.
3268 3269
---------------------------------------------------------------------------
\3267\ 49 U.S.C. 32903(f)(1).
\3268\ 49 U.S.C. 32903(g).
\3269\ See 49 CFR 536.5; see also 74 FR 14430 (Mar. 30, 2009)
(Per NHTSA's final rule for MY 2011 Average Fuel Economy Standards
for Passenger Cars and Light Trucks, ``There is no other clear
expression of congressional intent in the text of the statute
suggesting that NHTSA would have authority to adjust transferred
credits, even in the interest of preserving oil savings. However,
the goal of the CAFE program is energy conservation; ultimately, the
U.S. would reap a greater benefit from ensuring that fuel oil
savings are preserved for both trades and transfers. Furthermore,
accounting for traded credits differently than for transferred
credits does add unnecessary burden on program enforcement. Thus,
NHTSA will adjust credits both when they are traded and when they
are transferred so that no loss in fuel savings occurs.'').
---------------------------------------------------------------------------
When NHTSA initially considered the preservation of oil savings,
the agency
[[Page 25226]]
explained how one credit is not necessarily equal to another. For
example, the fuel savings lost if the average fuel economy of a
manufacturer falls one-tenth of an mpg below the level of a relatively
low standard are greater than the average fuel savings gained by
raising the average fuel economy of a manufacturer one-tenth of a mpg
above the level of a relatively high CAFE standard.\3270\ The effect of
applying the adjustment factor is to increase the numerical value of
credits for compliance accounting that are earned for exceeding a CAFE
standard, that are applied to a compliance category with a higher CAFE
standard. Likewise, the adjustment factor has the effect of decreasing
the numerical value of credits for compliance accounting that are
earned for exceeding a CAFE standard, that are applied to a compliance
category with a lower CAFE standard. While applying the adjustment
factor impacts the compliance accounting value of credits which are
denominated in miles per gallon, the adjustment maintains the real
world value of credits from the perspective of the actual amount of
fuel consumed or saved.
---------------------------------------------------------------------------
\3270\ 74 FR 14432 (Mar. 30, 2009).
---------------------------------------------------------------------------
Alliance/Global stated, in its 2016 petition, that while carry-
forward and carryback credits have been used for many years, the CAFE
standards did not change during the Congressional CAFE freeze, meaning
credits earned during those years were associated with the same amount
of fuel savings from year to year.\3271\ Alliance/Global suggest that
because there is no longer a Congressional CAFE freeze, NHTSA should
apply the adjustment factor when moving credits within a manufacturer's
fleet (i.e. carry-forward or carryback) beginning retroactively in MY
2011.\3272\
---------------------------------------------------------------------------
\3271\ Alliance/Global Petition at 10.
\3272\ Alliance/Global Petition at 4.
---------------------------------------------------------------------------
In the NPRM, NHTSA tentatively denied Alliance/Global's request to
apply the fuel savings adjustment factor to credits that are carried
forward or carried back within the same fleet to the extent that the
request would impact credits carried forward or back retroactively
within manufacturers' compliance fleets (i.e., credits that were
generated prior to MY 2021 when the standards set by this rule first
apply). NHTSA tentatively determined that applying the adjustment
factor to credits earned in prior model years would be inequitable to
apply retroactively. There would be an advantage for manufacturers
carrying credits into future model years with higher CAFE standards.
Manufacturers have historically planned compliance strategies based, at
least in part, on the existing rules for how credits could be carried
forward and back, including the lack of an adjustment factor when
credits are carried forward or back within the same fleet. Thus,
retroactively requiring an adjustment factor could disadvantage certain
manufacturers without credits, and result in windfalls for other
manufacturers.
To explore the impact on future model years, NHTSA sought
additional comments in the NPRM on the feasibility of applying the fuel
savings adjustment factor to credits carried forwards or back starting
in MY 2021. Global Automakers submitted new comments arguing that the
application of fuel savings adjustment factors to credits carried
forward or back would not result in a credit windfall. They believed
this practice would ensure that credits have a consistent value over
time.\3273\
---------------------------------------------------------------------------
\3273\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
---------------------------------------------------------------------------
Comments from Global Automakers provided no further justification
that would persuade NHTSA to consider changing its position on denying
the application of the adjustment factor to carry-forward and carryback
credits beginning with MY 2011. NHTSA continues to be concerned about
the inequitable outcome retroactive adjustments would have on the
credit market. Therefore, NHTSA is finalizing its decision to deny the
Alliance/Global request to apply the adjustment factor to credits
carried forward or carried back within a compliance category
retroactively beginning as early as MY 2011.
Congress expressly required that DOT establish a credit
``transferring'' regulation, to allow individual manufacturers to move
credits from one of their fleets to another (e.g., using a credit
earned for exceeding the light truck standard for compliance with the
domestic passenger car standard). Congress also gave DOT discretion to
establish a credit ``trading'' regulation so that credits may be bought
and sold between manufacturers.\3274\ Congress specified that trading
was for earned credits ``to be sold to manufacturers whose automobiles
fail to achieve the prescribed standards such that the total oil
savings associated with manufacturers that exceed the prescribed
standards are preserved.'' \3275\ NHTSA established 49 CFR part 536
believing it was consistent with the statute for transferred credits to
be subject to the same ``adjustment factor'' to ensure total oil
savings are preserved.\3276\ NHTSA believed that no further application
of the adjustment factor to other credit flexibilities would be
appropriate at that time. NHTSA sought comments in the NPRM to explore
the consequences associated with applying the adjustment factor to
credits carried forward and back starting in MY 2021, but no further
insight was gained from the comments received. Therefore, NHTSA is
retaining its existing requirements for the adjustment factor to be
applied to transferred and traded credits only. NHTSA will continue
considering potential application of the adjustment factor for all
types of credit flexibilities in the future, and may consider
regulatory changes in subsequent rulemakings.
---------------------------------------------------------------------------
\3274\ 49 U.S.C. 32903(f).
\3275\ 49 U.S.C. 32903(f)(1).
\3276\ 74 FR 14196, 14434 (Mar. 30, 2009).
---------------------------------------------------------------------------
(6) VMT Estimates for Fuel Savings Adjustment Factor
NHTSA uses the vehicle miles traveled (VMT) estimate as part of its
fuel savings adjustment equation to ensure that when traded or
transferred credits are used, fuel economy credits are adjusted to
ensure fuel oil savings is preserved.\3277\ For MYs 2017-2025, NHTSA
finalized VMT values of 195,264 miles for passenger car credits, and
225,865 miles for light truck credits.\3278\ These VMT estimates
harmonized with those used in EPA's CO2 program. For MYs
2011-2016, NHTSA estimated different VMTs by model year.
---------------------------------------------------------------------------
\3277\ See 49 CFR 536.4(c).
\3278\ 77 FR 63130 (Oct. 15, 2012).
---------------------------------------------------------------------------
In the NPRM, NHTSA explained that Alliance/Global requested in
their 2016 petition that NHTSA apply fixed VMT estimates to the fuel
savings adjustment factor for MYs 2011-2016 similar to how NHTSA
handled VMT values for MYs 2017-2025.\3279\ NHTSA rejected a similar
request from the Auto Alliance in the MY 2017 and later rulemaking,
citing lack of scope, and expressing concern about the potential loss
of fuel savings.\3280\
---------------------------------------------------------------------------
\3279\ Alliance/Global Petition at 5, 11.
\3280\ Id.
---------------------------------------------------------------------------
The Alliance/Global argued that data from MYs 2011-2016 demonstrate
that no fuel savings would have been lost, as was NHTSA's
concern.\3281\ Alliance/Global asserted that by not revising the MY
2012-2016 VMT estimates, credits earned during that timeframe were
undervalued.\3282\ Therefore, Alliance/
[[Page 25227]]
Global argued that NHTSA should retroactively revise its VMT estimates
to ``reflect better the real-world fuel economy results.'' \3283\
---------------------------------------------------------------------------
\3281\ Alliance/Global Petition at 11.
\3282\ Id.
\3283\ Alliance/Global Petition at 11.
---------------------------------------------------------------------------
Such retroactive adjustments could have unfair adverse effects upon
manufacturers for decisions they made based on the regulations as they
existed at the time. As Alliance/Global acknowledged, adjusting VMT
estimates would disproportionately affect manufacturers that have a
credit deficit and were part of EPA's Temporary Lead-time Allowance
Alternative Standards (TLAAS). The TLAAS program sunsets for MYs 2021
and later. Given that some manufacturers would be disproportionately
affected were NHTSA to adopt Alliance/Global's proposal, in the NPRM,
NHTSA tentatively denied Alliance/Global's request to change the
agency's VMT schedules for MYs 2011-2016 retroactively. Alliance/
Global's suggestion that a TLAAS manufacturer should be allowed to
elect either approach does not change the fact that manufacturers in
the TLAAS program made production decisions based on the regulations as
understood at the time.\3284\ NHTSA sought comments on the Alliance/
Global requests in the NPRM.
---------------------------------------------------------------------------
\3284\ See id. at 11-12, n.12.
---------------------------------------------------------------------------
However, no further comments were received on this issue in
response to the NPRM. Therefore, NHTSA is finalizing its decision to
deny the Alliance/Global request to modify the VMT schedules for MYs
2011-2016.
(7) Special Fuel Economy Calculations for Dual and Alternative Fueled
Vehicles
As discussed at length in prior rulemakings, EPCA, as amended by
EISA, encouraged manufacturers to build alternative-fueled and dual-
(or flexible-) fueled vehicles by providing special fuel economy
calculations for ``dedicated'' (that is, 100 percent) alternative
fueled vehicles and ``dual-fueled'' (that is, capable of running on
either the alternative fuel or gasoline/diesel) vehicles.
Dedicated alternative-fuel automobiles include electric, fuel cell,
and compressed natural gas vehicles, among others. The statutory
provisions for dedicated alternative fuel vehicles in 49 U.S.C.
32905(a) state that the fuel economy of any dedicated automobile
manufactured after MY 1992 shall be measured ``based on the fuel
content of the alternative fuel used to operate the automobile. A
gallon of liquid alternative fuel used to operate a dedicated
automobile is deemed to contain 0.15 gallon of fuel.'' Under EPCA, for
dedicated alternative fuel vehicles, there are no limits or phase-out
for this special fuel economy calculation, unlike for duel-fueled
vehicles, as discussed below.
EPCA's statutory incentive for dual-fueled vehicles at 49 U.S.C.
32906 and the measurement methodology for dual-fueled vehicles at 49
U.S.C. 32905(b) and (d) expire after MY 2019; therefore, NHTSA had to
examine the future of these provisions in the MY 2017 and later CAFE
rulemaking. NHTSA and EPA concluded that it would be inappropriate to
measure duel-fueled vehicles' fuel economy like that of conventional
gasoline vehicles with no recognition of their alternative fuel
capability, which would be contrary to the intent of EPCA/EISA. The
agencies determined that for MY 2020 and later vehicles, the general
statutory provisions authorizing EPA to establish testing and
calculation procedures provide discretion to set the CAFE calculation
procedures for those vehicles. The methodology for EPA's approach is
outlined in the 2012 final rule for MYs 2017 and later at 77 FR 63128
(Oct. 15, 2012). In the NPRM, NHTSA sought comments on that current
approach.
NHTSA received comments from the Coalition for Renewable Natural
Gas, NGV America, the American Gas Association, the American Public Gas
Association, CARB, Ingevity Corporation, Fuel Freedom Foundation, UCS,
National Farmers Union, Indiana Corn Growers Association, Volkswagen,
and a joint submission from Ariel Corp. and VNG.co.
Fuel Freedom Foundation and the National Farmers Union asserted
that the agencies should continue offering incentives for emerging
technology vehicles including natural gas vehicles, internal combustion
engine (ICE) vehicles that encourage renewable fuel use, electric and
hydrogen fuel cell vehicles, flex-fuel vehicles (FFVs), and dedicated
high-octane vehicles designed for compatibility with mid-level ethanol
blends.\3285\
---------------------------------------------------------------------------
\3285\ Fuel Freedom Foundation, Detailed Comments, NHTSA-2018-
0067-12016; National Farmers Union, Detailed Comments, NHTSA-2018-
0067-11972.
---------------------------------------------------------------------------
Indiana Corn Growers Association and Fuel Freedom Foundation
specified that FFVs, as well as vehicles that run on mid-level ethanol
blends, should receive credit for the petroleum reduction value.\3286\
For vehicles using higher-ethanol blends, these commenters stated that
the agencies should establish more accurate petroleum equivalency
factors for the proportion of ethanol versus gas.\3287\ Clean Fuels
Development Coalition requested credits for producing ``Engines
Optimized for High-Octane'' be reinstated.\3288\ Volkswagen made the
same request and added that a pathway to higher-octane fuel is
important to it.\3289\
---------------------------------------------------------------------------
\3286\ Indiana Corn Growers Association, Detailed Comments,
NHTSA-2018-0067-12003; Fuel Freedom Foundation, Detailed Comments,
NHTSA-2018-0067-12016.
\3287\ Fuel Freedom Foundation, Detailed Comments, NHTSA-2018-
0067-12016.
\3288\ Clean Fuels Development Coalition, Detailed Comments,
NHTSA-2018-0067-12031.
\3289\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
Ariel Corp. and VNG.co, the Coalition for Renewable Natural Gas,
NGVAmerica, the American Gas Association, and the American Public Gas
Association commented that the agencies should expand incentives for
natural gas vehicles in the light-duty sector especially for pick-up
trucks, work vans, and sport utility vehicles.\3290\ They argued that
current incentives are not strong enough to induce manufacturers to
produce natural gas vehicles. They further requested that the market
penetration rates be removed for light-duty trucks.\3291\
---------------------------------------------------------------------------
\3290\ Joint submission from Ariel Corp and VNG.co LLC, Detailed
Comments, NHTSA-2018-0067-7573; Joint submission from the Coalition
for Renewable Natural Gas, NVG America, the American Gas
Association, and American Public Gas Association, Detailed Comments,
NHTSA-2018-0067-11967.
\3291\ See, e.g., joint submission from the Coalition for
Renewable Natural Gas, NGVAmerica, the American Gas Association, and
the American Public Gas Association, Detailed Comments, NHTSA-2018-
0067-11967.
---------------------------------------------------------------------------
The Coalition for Renewable Natural Gas, NGVAmerica, the American
Gas Association, and the American Public Gas Association argued that an
AMFA factor of 0.15 is low and because some natural gas vehicles can
operate at 100 percent natural gas, a higher fuel economy credit is
justified. They further supported a permanent use of the 0.15 factor
for dual-fuel vehicles.\3292\ Similarly, Ingevity Corporation, and
Ariel Corp. and VNG.co argued that natural gas vehicle emissions should
return to the 0.15 divisor.\3293\
---------------------------------------------------------------------------
\3292\ Joint submission from the Coalition for Renewable Natural
Gas, NGVAmerica, the American Gas Association, and the American
Public Gas Association, Detailed Comments, NHTSA-2018-0067-11967.
\3293\ Ingevity Corporation, Detailed Comments, NHTSA-2018-0067-
8666; Joint submission from Ariel Corp. and VNG.co LLC, Detailed
Comments, NHTSA-2018-0067-7573.
---------------------------------------------------------------------------
Ingevity Corporation, Ariel Corp. and VNG.co, the Coalition for
Renewable
[[Page 25228]]
Natural Gas, NGVAmerica, the American Gas Association, and the American
Public Gas Association requested that the agencies remove the minimum
driving range of natural gas compared to gasoline and ``drive to
empty'' design requirements for dual-fueled natural gas vehicles and
allow higher utility factors based on driving range only, so that dual-
fuel NGVs are treated similarly to PHEVs. They stated a belief that the
design constraints for dual-fuel NGVshold NGVs to an unfairly higher
standard.\3294\ As discussed above in Section IX.B, EPA is removing
these design constraints for dual-fuel NGVs.
---------------------------------------------------------------------------
\3294\ Ingevity, Detailed Comments, NHTSA-2018-0067-8666; Joint
submission from Ariel Corp. and VNG.co LLC, Detailed Comments,
NHTSA-2018-0067-7573; Joint submission from The Coalition for
Renewable Natural Gas, NGVAmerica, the American Gas Association, the
American Public Gas Association, Detailed Comments, NHTSA-2018-0067-
11967.
---------------------------------------------------------------------------
CARB argued that flexibilities for natural gas vehicles and high-
octane blend vehicles are not yet warranted.\3295\ Similarly, UCS
argued that natural gas is a greenhouse gas and benefits from natural
gas vehicles are undermined by their costs. UCS further commented that
natural gas vehicle technology does not need any incentives since it
has already been deployed and in the market.\3296\
---------------------------------------------------------------------------
\3295\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
\3296\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
In response to comments, NHTSA has determined that EPCA and EISA
prescribe the incentive that is used for dedicated liquid and gaseous
alternative fuel vehicles, and the CAFE program will continue to use
those statutory incentives. For dedicated alternative fuel vehicles,
the statute provides a significant incentive that only counts 15
percent of the actual energy used.\3297\ For dual fuel vehicles, NHTSA
has determined that, for the portion of operation that occurs on an
alternative fuel, it is consistent to use the same incentive that is
specified by EPCA and EISA for dedicated fuel vehicles. For example,
for the hypothetical case of a vehicle that operates 99 percent of the
time on an alternative fuel, it would be appropriate for that vehicle
to receive nearly the same incentive as a dedicated alternative fuel
vehicle that operates 100 percent of the time on alternative fuel.
Applying the same 15 percent of energy used incentive for both
dedicated and duel fuel vehicles remains appropriate. NHTSA therefore
is not adopting any new incentives for any alternative fueled vehicles.
---------------------------------------------------------------------------
\3297\ 32905(a) ``. . . A gallon of a liquid alternative fuel
used to operate a dedicated automobile is deemed to contain .15
gallon of fuel.'' 32905(c) ``. . . One hundred cubic feet of natural
gas is deemed to contain .823 gallon equivalent of natural gas. The
Secretary of Transportation shall determine the appropriate gallon
equivalent of other gaseous fuels. A gallon equivalent of gaseous
fuel is deemed to have a fuel content of .15 gallon of fuel.''
---------------------------------------------------------------------------
D. Compliance Issues That Affect Both the CO2 and CAFE
Programs
Because the real world CO2 emissions reduction benefits
of certain technologies cannot be measured or fully measured using 2-
cycle test procedures, EPA established new compliance flexibilities
under its CAA authority, starting in MY 2012, that allow manufacturers
credit for emission compliance for installing these technologies. Those
flexibilities are designed to recognize improvements in A/C systems
with greater efficiency and other ``off-cycle'' technologies that
reduce real world tailpipe CO2 emissions. More specifically,
real world improvements that cannot be measured or fully measured on 2-
cycle tests are determined and used to calculate additional
CO2 credits (in Megagrams (Mg)) for each model type that has
the technologies. Because these tailpipe CO2 improving
technologies also impact fuel economy, NHTSA adopted the same
flexibilities and incentives beginning in MY 2017. EPA and NHTSA also
established incentives for both the CO2 and CAFE programs
that give added compliance credits and fuel consumption improvement
values for the production of strong and mild hybrid full-size pickup
trucks beginning in MY 2017.\3298\ EPA adjusts manufacturers' CAFE
performance values using the emissions benefits or incentives provided
for these technologies. EPA developed a methodology for manufacturers
to increase their passenger car and light truck fuel economy
performance in accordance with procedures set forth by EPA in 40 CFR
part 600. For the NHTSA CAFE program, the CO2 reductions (in
grams per mile) are converted to fuel consumption improved values
(FCIVs, gallons per mile) and then the benefits are summed for all the
model types in the manufacturer's fleets. The total FCIVs are used to
adjust and increase manufacturers' CAFE (mpg) performance values.
---------------------------------------------------------------------------
\3298\ See 40 CFR 86.1867-86.1868, 86.1870.
---------------------------------------------------------------------------
It is important to note that while these flexibilities and
incentives have similar value for compliance in the CAFE and
CO2 programs, there are differences in how they are
accounted for in each of the programs due to differences in the
structure of the programs. The CAFE program accounts for A/C efficiency
and off-cycle improvements through EPA measurement procedures that
determine fuel consumption improvement values (FCIVs). The CAFE A/C
efficiency and off-cycle provisions do not involve manufacturer
credits.\3299\ There are no bankable, tradable, or transferrable
credits earned by a manufacturer for implementing more efficient A/C
systems or installing an off-cycle technology. In fact, the only
credits provided for in NHTSA's CAFE program are those earned by
overcompliance with a standard.\3300\ As discussed above, EPA adjusts
CAFE performance values based on the FCIVs generated through the use of
these technologies. Off-cycle technologies and A/C efficiency
improvements represent adjustments to individual vehicle compliance
values based on the fuel consumption improvement values of these
technologies.
---------------------------------------------------------------------------
\3299\ This is not to be confused with EPA's parallel program,
which refers to the GHG's consideration of A/C improvements and off-
cycle technologies as ``credits.''
\3300\ 49 U.S.C. 32903.
---------------------------------------------------------------------------
Illustrative of this confusion, in the 2016 Alliance/Global
petition, the petitioners asked NHTSA to avoid imposing unnecessary
restrictions on the use of credits. Alliance/Global referenced language
from an EPA report that stated compliance is assessed by measuring the
tailpipe emissions of a manufacturer's vehicles, and then reducing
vehicle CO2 compliance values depending on A/C efficiency
improvements and off-cycle technologies.\3301\ This language is
consistent with NHTSA's statement in the MY 2017 and later final rule,
which explained how the agencies coordinate and apply off-cycle and A/C
adjustments. ``There will be separate improvement values for each type
of credit, calculated separately for cars and for trucks. These
improvement values are subtracted from the manufacturer's 2-cycle-based
fleet fuel consumption value to yield a final new fleet fuel
consumption value, which would be inverted to determine a final fleet
fuel CAFE value.'' \3302\
---------------------------------------------------------------------------
\3301\ See Alliance/Global Petition at 15.
\3302\ 77 FR 62726 (Oct. 15, 2012).
---------------------------------------------------------------------------
In the NPRM, NHTSA proposed to deny Alliance/Global's request
because what the petitioners refer to as ``technology credits'' are
actually FCIVs applied to the fuel economy performance of individual
vehicles.\3303\
[[Page 25229]]
Thus, these adjustments are not actually ``credits,'' per the usage of
``credit'' in EPCA/EISA and are not subject to the ``carry-forward''
and ``carryback'' provisions in 49 U.S.C. 32903. To alleviate
confusion, and to ensure consistency in nomenclature, NHTSA proposed to
update language in its regulations to reflect that the use of the term
``credits'' to refer to A/C efficiency and off-cycle technology
adjustments should actually be termed fuel consumption improvement
values (FCIVs). No further comments were received on this issue in
response to the NPRM. For the final rule, NHTSA is finalizing the
proposed changes in its regulations to remove the term ``credits'' and
to replace it with the term ``adjustments'' for the FCIV benefit for A/
C and off-cycle technologies in the CAFE program.
---------------------------------------------------------------------------
\3303\ The agencies also refer to A/C and off-cycle technology
improvement values as ``credits'' sporadically throughout their
regulations. NHTSA is amending its regulations to reflect these are
adjustments and not actual credits that can be carried forward or
back. For a further discussion, see above.
---------------------------------------------------------------------------
Manufacturers seeking to use these flexibilities and incentives
start the process each model year by submitting information to EPA and
seeking any necessary approvals, as appropriate. The use of certain
technologies only requires submitting information to EPA, whereas
others require a formal request process for approval. The differences
are explained in the following sections. The compliance information
manufacturers must submit to EPA describes the technologies, the
flexibilities or incentives being used, and the testing approach for
deriving benefits. Initial information is required as a part of the EPA
certification process, as specified by 40 CFR 86.1843-01 in advance of
each model year. For technologies requiring approvals, EPA must confirm
the manufacturer's testing approach, receive test results to assess the
benefit of the technology, and then where applicable issue a Federal
Register notice that invites public comment. EPA review and
determination usually occurs before the end of the compliance model
year, if manufacturers provide information to EPA on a timely basis. To
receive the benefit under the CAFE program for technologies that
require approvals, manufacturers must concurrently submit to NHTSA the
same information that is sent to EPA. EPA consults with NHTSA in
reviewing A/C efficiency and off-cycle adjustments to fuel economy
performance values that require approval. NHTSA provides EPA its
assessment of the suitability of a technology considering: (1) Whether
the technology has a direct impact upon improving fuel economy
performance; (2) whether the technology is related to crash-avoidance
technologies, safety critical systems or systems affecting safety-
critical functions, or technologies designed for the purpose of
reducing the frequency of vehicle crashes; (3) information from any
assessments conducted by EPA related to the application, the
technology, and/or related technologies; and (4) any other relevant
factors.
EPA and NHTSA sought comments on several aspects of the shared
flexibilities and incentives in the NPRM. Presented in the following
sections is a summary of the comments received and the agencies final
decisions for the final rule.
1. Incentives for Advanced Technologies in Full-Size Pickup Trucks
In the 2012 rulemaking for MYs 2017 and beyond, EPA and NHTSA
created incentives to encourage implementation of hybrid electric full
size pickup trucks for both the CO2 and CAFE programs.
CO2 credits and CAFE FCIVs were made available for
manufacturers that produce full-size pickup trucks with Mild HEV or
Strong HEV technology, provided the percentage of production with the
technology is greater than specified percentages.\3304\ In addition,
CO2 credits and CAFE FCIVs were made available for
manufacturers that produce full-size pickups with other technologies
that enables full size pickup trucks to exceed performance of their
CO2 or CAFE targets based on footprints by specified
amounts.\3305\ These performance-based incentives created a technology-
neutral path (as opposed to the other technology-encouraging path) to
achieve the CO2 credits and CAFE FCIVs, which would
encourage the development and application of new technological
approaches.
---------------------------------------------------------------------------
\3304\ 77 FR 62651 (Oct. 15, 2012).
\3305\ Id.
---------------------------------------------------------------------------
EPA and NHTSA established limits on the vehicles eligible to
qualify for these incentives; a truck must meet minimum criteria for
bed size and towing or payload capacity, and meet minimum production
thresholds (in terms of a percentage of a manufacturer's full-size
pickup truck fleet) in order to qualify for the incentives. As
designed, the strong hybrid credit is 20 grams/mile per vehicle,
available through MY 2025, if installed on at least 10 percent of the
manufacturer's full-size pickup truck fleet in the model year. The
program also included an incentive for mild hybrids of 10 grams/mile
per vehicle during MYs 2017-2021. To be eligible the manufacturer would
have to show that the mild hybrid technology is utilized in a specified
portion of its truck fleet beginning with at least 20 percent of a
company's full-size pickup production in MY 2017 and ramping up to at
least 80 percent in MY 2021.\3306\
---------------------------------------------------------------------------
\3306\ 77 FR 62651-2 (Oct. 15, 2012).
---------------------------------------------------------------------------
At present, no manufacturer has qualified to use the full-size
pickup truck incentives. One vehicle manufacturer introduced a mild
hybrid pickup truck for MY 2019 but did not meet the minimum production
threshold. Others have announced potential collaborations, or have
already started production on future hybrid or electric models.\3307\
---------------------------------------------------------------------------
\3307\ Chrysler released the 2019 Dodge Ram 1500 ``eTorque''
(see https://www.fueleconomy.gov/feg/Find.do?action=sbs&id=40736&id=40737&id=40394&id=40397) which
qualifies as a mild hybrid pickup truck by replacing the traditional
alternator on the engine with a 48-volt Li-on battery-powered, belt-
driven motor generator that improves performance, efficiency,
payload, towing capabilities and drivability. The production volume
of these vehicles did not qualify for the full-size pickup truck
electric/hybrid incentive for MY 2019. Other vehicle models are
currently in research or in development for future years but it is
uncertain whether they will reach the required sales volumes to
qualify for incentives. For example, the hybrid and battery-electric
versions of the F-150 pickup, see https://www.trucks.com/2019/09/18/ford-truck-engineer-explains-electric-f-150-pickup-plans (September
18, 2019), or the new electric pickup truck manufactured by Rivian,
https://www.trucks.com/2019/04/24/ford-plans-new-electric-truck-rivian-invests-500-million/ (April 24, 2019); or the Tesla all
electric pickup truck (https://www.cnn.com/2019/11/08/success/tesla-pickup-reveal/index.html) (November 8, 2019).
---------------------------------------------------------------------------
Prior to the NPRM, the agencies received input from automakers that
these incentives should be extended and available to all light-duty
trucks (e.g., cross-over vehicles, minivans, sport utility vehicles,
and smaller-sized pickups) and not only full-size pickup trucks.\3308\
Automakers also recommended that the program's eligibility production
thresholds should be removed because they discourage the application of
technology since manufacturers cannot be confident of achieving the
thresholds. Some stakeholders have also suggested an additional
incentive for strong and mild hybrid passenger cars. In the proposal,
the agencies sought comment on whether these incentives should be
expanded along the lines suggested by stakeholders, on the basis that
perhaps these incentives could lead to additional product offerings of
strong hybrids, and technologies that offer similar emissions
reductions, which could enable manufacturers to achieve additional
long-term CO2 emissions reductions. In addition, the
agencies sought comment on whether to extend either the incentive for
hybrid full-size pickup trucks or the performance-based incentive past
the dates that EPA specified in the 2012 final rule for MY
[[Page 25230]]
2017 and later. The agencies also sought comment on eliminating
incentive programs, as discussed above.
---------------------------------------------------------------------------
\3308\ 83 FR 43461 (Aug. 24, 2018).
---------------------------------------------------------------------------
The agencies received a variety of comments on the full-size pickup
truck incentives. Comments were received from General Motors,
Volkswagen, Honda, BorgWarner, Fiat Chrysler, Toyota, DENSO
International, Ford, CARB, Global Automakers, UCS, Electric Drive
Transportation Association, the Auto Alliance, Ariel Corp. and VNG.co,
ACEEE, the Coalition for Renewable Natural Gas, NGVAmerica, the
American Gas Association, and the American Public Gas Association.
The Auto Alliance, Toyota, General Motors, BorgWarner, Global
Automakers, and Volkswagen advocated to expand the full-size pickup
truck hybrid incentives to all hybrid vehicles.\3309\ They argued that
prices for all hybrid-drive technologies are projected to remain high
and consumer demand for these vehicles is still slow to increase.\3310\
They asserted that expanding the full-size pickup truck hybrid
incentive to all hybrid vehicles will help encourage investments in
hybrid technology and continue to help manufacturers address their
compliance challenges.\3311\ Similarly, these commenters reported that
the current market, fueled by consumer demand for SUVs and lower than
expected gas prices, is not conducive to consumer acceptance of or
demand for electric vehicles.\3312\ For these reasons, they stated
their belief that it is important to support adjustments and expansion
of the current incentives to promote hybrid technologies.
---------------------------------------------------------------------------
\3309\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Toyota, Detailed Comments, NHTSA-2018-0067-12150; General Motors,
Detailed Comments, NHTSA-2018-0067-11858; BorgWarner, Detailed
Comments, NHTSA-2018-0067-11895; Global Automakers, Detailed
Comments, NHTSA-2018-0067-12032; Volkswagen, Detailed Comments,
NHTSA-2017-0069-0583.
\3310\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
\3311\ See, e.g., General Motors, Detailed Comments, NHTSA-2018-
0067-11858.
\3312\ See, e.g., Toyota, Detailed Comments, NHTSA-2018-0067-
12150.
---------------------------------------------------------------------------
The Auto Alliance, DENSO International, Global Automakers, Fiat
Chrysler, and Honda also argued for alternative pathways for the
agencies to consider allowing the full-size pickup truck hybrid
incentives to be expanded to the light-duty truck segment, but not to
all passenger vehicles. They argued that hybrid technology has been
slow to be applied in the light-duty truck segment, but has been
broadly applied to passenger cars.\3313\
---------------------------------------------------------------------------
\3313\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
DENSO, Detailed Comments, NHTSA-2018-0067-11880; Global Automakers,
Detailed Comments, NHTSA-2018-0067-12032; Fiat Chrysler, Detailed
Comments, NHTSA-2018-0067-11943; Honda, Detailed Comments, NHTSA-
2018-0067-11818.
---------------------------------------------------------------------------
Toyota, Global Automakers, and the Auto Alliance suggested the
incentives for light-duty trucks should amount to 20 grams/mile.\3314\
Global Automakers added that in addition to expanding full-size pickup
truck hybrid incentives to light trucks, the agency should consider a
smaller incentive for hybrid electric passenger vehicles as well.\3315\
The Auto Alliance and Toyota suggested a 10 grams/mile credit for
passenger cars.\3316\ Volkswagen further requested the hybrid pickup
credit to be expanded to all hybrid cars and trucks.\3317\
---------------------------------------------------------------------------
\3314\ Toyota, Detailed Comments, NHTSA-2018-0067-12150; Global
Automakers, Detailed Comments, NHTSA-2018-0067-12032; Auto Alliance,
Detailed Comments, NHTSA-2018-0067-12073.
\3315\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
\3316\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Toyota, Detailed Comments, NHTSA-2018-0067-12150.
\3317\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
Toyota, the Auto Alliance, Electric Drive Transportation
Association, Ford, DENSO International, Global Automakers, Fiat
Chrysler, and BorgWarner commented that having minimum production
percentages for hybrid pickup trucks discourages manufacturers from
investing in hybrid technologies. They requested that the agencies
consider eliminating the percentage of production requirement and
provide incentives in proportion to the value of the technology.\3318\
Ford stated that the minimum production percentages unfairly penalize
larger manufacturers who must produce more pickup trucks to claim the
incentives than a smaller volume manufacturer.\3319\
---------------------------------------------------------------------------
\3318\ Toyota, Detailed Comments, NHTSA-2018-0067-12150; Auto
Alliance, Detailed Comments, NHTSA-2018-0067-12073; Electric Drive
Transportation Association, Detailed Comments, NHTSA-2018-0067-1201;
Ford, Detailed Comments, NHTSA-2018-0067-11928; DENSO, Detailed
Comments, NHTSA-2018-0067-11880; Global Automakers, Detailed
Comments, NHTSA-2018-0067-12032; Fiat Chrysler, Detailed Comments,
NHTSA-2018-0067-11943; BorgWarner, Detailed Comments, NHTSA-2018-
0067-11895.
\3319\ Ford, Detailed Comments, NHTSA-2018-0067-11928.
---------------------------------------------------------------------------
Ariel Corp. and VNG.co, the Coalition for Renewable Natural Gas,
NGVAmerica, the American Gas Association, and the American Public Gas
Association commented the pickup truck incentives should be expanded to
include natural gas vehicles.\3320\ They suggested a ``Natural Gas
Pickup'' incentive like the hybrid-electric and performance-based
pickup credits, but no minimum production requirement.\3321\
---------------------------------------------------------------------------
\3320\ Joint submission from Ariel Corp. and VNG.co, Detailed
Comments, NHTSA-2018-0067-7573; Joint submission from The Coalition
for Renewable Natural Gas, NGVAmerica, the American Gas Association,
and the American Public Gas Association, Detailed Comments, NHTSA-
2018-0067-11967.
\3321\ See, e.g., Joint submission from Ariel Corp. and VNG.co,
Detailed Comments, NHTSA-2018-0067-7573.
---------------------------------------------------------------------------
ACEEE and UCS commented that hybrid technology has been around for
quite a while and has been applied in every vehicle class. They
discouraged the agencies from applying more incentives to these
vehicles.\3322\ Specifically, UCS stated that incentives for electric
vehicles are mostly driven by state regulation, and EPA and NHTSA
policies are rewarding manufacturers for meeting standards they were
already required to meet.\3323\ UCS commented that hybrids are not
innovators or game-changing vehicles--they are simply one of many
strategies by which manufacturers can reduce emissions and should not
receive special treatment.\3324\
---------------------------------------------------------------------------
\3322\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122-29; UCS,
Detailed Comments, NHTSA-2018-0067-12039.
\3323\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3324\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
CARB commented that incentives for full-size hybrid pickup trucks
should remain limited in their scope and that increasing or expanding
those incentives can erode emissions benefits.\3325\ CARB further
commented that hybrid electric vehicles (HEVs) are widely available at
varying levels of power and performance across vehicle sizes, and CARB
does not believe HEVs deserve special treatment in the CO2
vehicle regulations.
---------------------------------------------------------------------------
\3325\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
---------------------------------------------------------------------------
After carefully considering the comments received, EPA and NHTSA
are not adopting any new or expanded incentives for hybrid vehicles or
full-size pickup trucks, and are removing these incentives beginning in
MY 2022 (the incentive for mild hybrids expires after MY 2021
regardless, so that does not change). The agencies believe any new or
expanded incentives would likely not result in any further emissions
benefits or fuel economy improvements since an increase in sales volume
would not be expected. The agencies agree with CARB and ACEEE, and UCS
that hybrids are a well-
[[Page 25231]]
established technology that has already been applied to a wide range of
vehicles and, as such, no further incentives are warranted at this
time. Further, the agencies believe that incentivizing manufacturers to
implement specific technologies is inappropriate, as manufacturer fuel
economy performance should represent actual fuel consumption. The
agencies believe any new or expanded incentives for hybrids would
likely not result in any further emissions benefits or fuel economy
improvements beyond those measured during testing; to the extent that
manufacturers choose to build full-size pickup trucks that exceed their
targets, those will reap the benefits of target exceedance in the
overall fleet averaging. Manufacturers did not provide sufficient
evidence to support their position in a manner that leads the agencies
to conclude otherwise, and there does not appear to be any likelihood
that manufacturers will be able to take advantage of these
flexibilities beyond MY 2021 that makes it necessary to retain them.
Therefore, the agencies are removing these flexibilities from the
program starting with MY 2022.
2. Flexibilities for Air Conditioning Efficiency
A/C systems are virtually standard automotive accessories, and more
than 95 percent of new cars and light trucks sold in the U.S. are
equipped with mobile A/C systems. A/C system usage places a load on an
engine, which results in additional tailpipe CO2 emissions
and fuel consumption; the high penetration rate of A/C systems
throughout the light-duty vehicle fleet means that efficient systems
can significantly impact the total energy consumed and CO2
emissions. A/C systems also have non-CO2 emissions
associated with refrigerant leakage.\3326\ Manufacturers can improve
the efficiency of A/C systems though redesigned and refined A/C system
components and controls.\3327\ That said, such improvements are not
measurable or recognized using 2-cycle test procedures, since A/C is
turned off during 2-cycle testing. Any A/C system efficiency
improvements that reduce load on the engine and improve fuel economy is
therefore not measurable on those tests.
---------------------------------------------------------------------------
\3326\ See Section V for further details. Notably, manufacturers
cannot claim CAFE-related benefits for reducing A/C leakage or
switching to an A/C refrigerant with a lower global warming
potential. While these improvements reduce GHG emissions consistent
with the purpose of the CAA, they generally do not impact fuel
economy and, thus, are not relevant to the CAFE program.
\3327\ The approach for recognizing potential A/C efficiency
gains is to utilize, in most cases, existing vehicle technology/
componentry, but with improved energy efficiency of the technology
designs and operation. For example, most of the additional A/C-
related load on an engine is because of the compressor, which pumps
the refrigerant around the system loop. The less the compressor
operates, the less load the compressor places on the engine
resulting in less fuel consumption and CO2 emissions.
Thus, optimizing compressor operation with cabin demand using more
sophisticated sensors, controls, and control strategies is one path
to improving the efficiency of the A/C system. For further
discussion of A/C efficiency technologies, see Section II.D of the
NPRM and Chapter 6 of the accompanying PRIA.
---------------------------------------------------------------------------
The CO2 and CAFE programs include flexibilities to
account for the real world CO2 emissions and fuel economy
improvements associated with improved A/C systems and to include the
improvements for compliance.\3328\ The total of A/C efficiency credits
is calculated by summing the individual credit values for each
efficiency improving technology used on a vehicle, as specified in the
A/C credit menu. The total A/C efficiency credit sum for each vehicle
is capped at 5.0 grams/mile for cars and 7.2 grams/mile for trucks.
Additionally, the off-cycle credit program contains credit earning
opportunities for technologies that reduce the thermal loads on a
vehicle from environmental conditions (solar loads or parked interior
air temperature).\3329\ These technologies are listed on a thermal
control menu that provides a predefined improvement value for each
technology. If a vehicle has more than one thermal load improvement
technology, the improvement values are added together, but subject to a
cap of 3.0 grams/mile for cars and 4.3 grams/mile for trucks.
---------------------------------------------------------------------------
\3328\ See 40 CFR 86.1868-12.
\3329\ See 40 CFR 86.1869-12(b).
---------------------------------------------------------------------------
EPA requested comment on the A/C caps and on whether A/C efficiency
technologies and off-cycle thermal control technologies should be
combined under a single cap, since the technologies directly interact
with each other. That is, improved thermal control results in reduced
A/C loads for the more efficient A/C technologies. If the thermal
credits were removed from the off-cycle menu, they would no longer be
counted against the 10 grams/mile menu cap discussed above,
representing a way to provide more room under the menu cap for other
off-cycle technologies. Specifically, EPA sought comment on replacing
the current off-cycle thermal efficiency capped value of 10 grams/mile,
with separate caps of 8 grams/mile for cars and 11.5 grams/mile for
trucks.
Comments concerning the A/C caps were received from the Auto
Alliance, DENSO, Fiat Chrysler, and Volkswagen. DENSO commented that A/
C efficiency credits earned through the off-cycle petition process
should not count toward the A/C credit cap. If A/C credits granted
through the off-cycle petition process are no longer counted toward the
A/C credit cap, it stated that manufacturers would be significantly
incentivized to develop new and innovative technologies.\3330\ Fiat
Chrysler requested that certain A/C credits for electrical technologies
(i.e., A/C blower motor controls that limit wasted electrical energy)
be transferred to the off-cycle credit list.\3331\ Volkswagen further
supported the removal of the thermal control technology credit caps and
suggested that implementing caps at the fleet average level, rather
than per-vehicle, could be less constraining.\3332\ DENSO pointed to an
NREL study which found that A/C improvements were greater than
previously thought possible. Therefore, it requested the agencies
consider increasing the A/C credit cap.\3333\
---------------------------------------------------------------------------
\3330\ DENSO, Detailed Comments, NHTSA-2018-0067-11880.
\3331\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3332\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
\3333\ DENSO, Detailed Comments, NHTSA-2018-0067-11880.
---------------------------------------------------------------------------
Similarly, the Auto Alliance and Fiat Chrysler suggested raising
the cap on A/C efficiency and thermal control technology by 64 percent
and combine them under a single cap.\3334\ Additionally, they proposed
increasing A/C efficiency and thermal control technology credits by up
to 64 percent.\3335\ They also proposed that the agencies create new
regulatory provisions to handle additional new A/C and thermal
technologies.\3336\
---------------------------------------------------------------------------
\3334\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
\3335\ See, e.g., Fiat Chrysler, Detailed Comments, NHTSA-2018-
0067-11943.
\3336\ See, e.g., Auto Alliance, Detailed Comments, NHTSA-2018-
0067-12073.
---------------------------------------------------------------------------
As with increasing the credit caps, manufacturers and suppliers
were generally supportive of higher credit caps, or no caps at all, for
this combined technology group. However, EPA has decided not to adopt
any changes to the caps, including combining the A/C efficiency and
thermal controls menu, due to the uncertainty regarding the menu credit
values. Additional uncertainty exists for these technology groups
because there are likely synergistic effects between A/C efficiency and
thermal technologies that would need to be further considered in
determining appropriate credit levels if
[[Page 25232]]
the two groups of technologies are combined under a single cap. Data is
not currently available to consider these effects. Therefore, the
agencies are not making any changes to the flexibilities for A/C
efficiency improvements in the CO2 or CAFE program, but may
perform research to understand better the relationship between A/C
efficiency and thermal technologies for consideration in future
rulemakings.
3. Flexibilities for Off-Cycle Technologies
``Off-cycle'' technologies are those that reduce vehicle fuel
consumption and CO2 emissions in the real world, but for
which the fuel consumption reduction benefits cannot be measured or
cannot be fully measured under the 2-cycle test procedures (city,
highway or correspondingly FTP, HFET) used to determine compliance with
the fleet average standards. The CAFE city and highway test cycles,
collectively referred to as the 2-cycle laboratory compliance tests (or
2-cycle tests), were developed in the early 1970s. The city test
simulates city driving in the Los Angeles area at that time. The
highway test simulates driving on secondary roads (not expressways).
The cycles are effective in measuring improvements in most fuel economy
improving technologies; however, they are unable to measure or
underrepresent certain fuel economy improving technologies because of
limitations in the test cycles. For example, off-cycle technologies
that improve emissions and fuel economy at idle (such as ``stop start''
systems) and those technologies that improve fuel economy to the
greatest extent at expressway speeds (such as active grille shutters
which improve aerodynamics) receive less than their real-world benefits
in the 2-cycle compliance tests.
Starting with MY 2008, EPA began employing a ``five-cycle'' test
methodology to measure fuel economy for the purpose of improving new
car window stickers (labels) and giving consumers better information
about the fuel economy they could expect under real-world driving
conditions.\3337\ However, for CO2 and CAFE compliance, EPA
continues to use the established ``two-cycle'' test methodology.\3338\
As learned through development of the ``five-cycle'' methodology and
prior rulemakings, there are technologies that provide real-world
CO2 emissions and fuel consumption improvements, but those
improvements are not fully reflected on the ``two-cycle'' test. EPA
established the off-cycle credit program to provide an appropriate
level of CO2 credit for technologies that achieve
CO2 reductions, but are normally not chosen as a
CO2 control strategy because their CO2 benefits
are not measured on the specified 2-cycle test.
---------------------------------------------------------------------------
\3337\ https://www.epa.gov/vehicle-and-fuel-emissions-testing/dynamometer-drive-schedules.
\3338\ The city and highway test cycles, commonly referred to
together as the 2-cycle tests are laboratory compliance tests
required by law for CAFE and are also used for determining
compliance with the GHG standards.
---------------------------------------------------------------------------
Currently, EPA has three compliance pathways. The first approach
allows manufacturers to gain credits without having to prove the
benefits of the technologies on a case-by-case basis. A predetermined
list or ``menu'' of credit values for specific off-cycle technologies
exists and became effective starting in MY 2014.\3339\ This pathway
allows manufacturers to use credit values established by EPA for a wide
range of off-cycle technologies, with minimal or no data submittal or
testing requirements.\3340\ Specifically, EPA established a menu with a
number of technologies that have real-world CO2 and fuel
consumption benefits not measured, or not fully measured, by the two-
cycle test procedures, and those benefits were reasonably quantified by
the agencies at that time. For each of the pre-approved technologies on
the menu, EPA established a quantified default value that is available
without additional testing. Manufacturers must demonstrate that they
were in fact using the menu technology, but not required to conduct
testing to quantify the technology's effects, unless they wish to
receive a credit larger than the default value. The default values for
these off-cycle credits were largely determined from research,
analysis, and simulations, rather than from full vehicle testing, which
would have been both cost and time prohibitive. EPA generally used
conservative predefined estimates to avoid any potential credit
windfall.\3341\
---------------------------------------------------------------------------
\3339\ See 40 CFR 86.1869-12(b).
\3340\ The Technical Support Document (TSD) for the 2012 final
rule for MYs 2017 and beyond provides technology examples and
guidance with respect to the potential pathways to achieve the
desired physical impact of a specific off-cycle technology from the
menu and provides the foundation for the analysis justifying the
credits provided by the menu. The expectation is that manufacturers
will use the information in the TSD to design and implement off-
cycle technologies that meet or exceed those expectations in order
to achieve the real-world benefits of off-cycle technologies from
the menu.
\3341\ While many of the assumptions made for the analysis were
conservative, others were ``central.'' For example, in some cases,
an average vehicle was selected on which the analysis was conducted.
In that case, a smaller vehicle may presumably deserve fewer credits
whereas a larger vehicle may deserve more. Where the estimates are
central, it would be inappropriate for the agencies to grant greater
credit for larger vehicles, since this value is already balanced by
smaller vehicles in the fleet. The agencies take these matters into
consideration when applications are submitted for credits beyond
those provided on the menu.
---------------------------------------------------------------------------
For off-cycle technologies not on the pre-defined technology list,
or obtained through petitioning, EPA created a second pathway which
allows manufacturers to use 5-cycle testing to demonstrate and justify
off-cycle CO2 credits.\3342\ EPA established this
alternative for a manufacturer to demonstrate the benefits of the
technology using 5-cycle testing. The additional emissions tests allow
emission benefits to be demonstrated over some elements of real-world
driving not captured by the CO2 compliance tests, including
high speeds, rapid accelerations, and cold temperatures. Under this
pathway, manufacturers submit test data to EPA, and EPA determines
whether there is sufficient technical basis to approve the off-cycle
credits. No public comment period is required for manufacturers seeking
credits using the EPA menu or using 5-cycle testing.
---------------------------------------------------------------------------
\3342\ See 40 CFR 86.1869-12(c). EPA proposed a correction for
the 5-cycle pathway in a separate technical amendments rulemaking.
See 83 FR 49344 (Oct. 1, 2019). EPA is not approving credits based
on the 5-cycle pathway pending the finalization of the technical
amendments rule.
---------------------------------------------------------------------------
The third pathway allows manufacturers to seek EPA approval,
through a notice and comment process, to use an alternative methodology
other than the menu or 5-cycle methodology for determining the off-
cycle technology CO2 credits.\3343\ Manufacturers must
provide supporting data on a case-by-case basis demonstrating the
benefits of the off-cycle technology on their vehicle models.
Manufacturers may also use the third pathway to apply for credits and
FCIVs for menu technologies where the manufacturer is able to
demonstrate credits and FCIVs greater than those provided by the menu.
---------------------------------------------------------------------------
\3343\ See 40 CFR 86.1869-12(d).
---------------------------------------------------------------------------
Due to the uncertainties associated with combining menu
technologies and the fact that some uncertainty is introduced because
off-cycle credits are provided based on a general assessment of off-
cycle performance, as opposed to testing on the individual vehicle
models, EPA established caps that limit the amount of credits a
manufacturer may generate using the EPA menu. Off-cycle technology is
capped at 10 grams/mile per year on a combined car and truck fleet-wide
average basis. No caps were established for technologies gaining
credits through the petitioning or 5-cycle approval methodologies.
[[Page 25233]]
a) Consideration of Eliminating A/C and Off-Cycle Adjustments in the
CO2 and CAFE Programs
The agencies sought comments in the NPRM on whether to remove the
A/C and off-cycle flexibilities from the CAFE program and adjust the
stringency levels accordingly based upon concern that the flexibilities
might distort the market. Several commenters provided responses
concerning the feasibility of removing any of these flexibilities.
Commenters included the Auto Alliance, the National Automobile Dealers
Association, Global Automakers, the Alliance for Vehicle Efficiency,
ACEEE, BorgWarner, Fiat Chrysler, General Motors, International Council
on Clean Transportation, Toyota, and UCS. Other comments were received
requesting that the agencies look into expanding the flexibilities by
including more technologies.
There was widespread support from commenters for retaining these
flexibilities for A/C and off-cycle technologies in the CO2
and CAFE programs. Commenters preferred that the agencies continue to
include the flexibilities, believing them to enable real world fuel
economy improvements and compliance with CO2 and CAFE
standards with a more cost effective combination of technologies. The
agencies agree that these programs achieve real world fuel economy
improvements and that keeping the flexibilities may enable more cost
effective technology combinations to achieve those real world fuel
economy improvements. For MY 2017, manufacturers introduced a wide
variety of low-cost technologies through the A/C and off-cycle
flexibilities that increased the overall industry's CAFE performance by
1.1 mpg. The agencies also acknowledge that the continued use of these
flexibilities under the EPA program since 2012 warrants consideration
due to automakers' and suppliers' significant investments in developing
the technologies, which could result in stranded capital should the
agencies discontinue them and manufacturers choose to remove the
technologies. For these reasons, the agencies have decided to continue
allowing manufacturers to use the existing flexibilities for A/C
efficiency and off-cycle technologies for future model years.
b) Final Decisions in Response to Manufacturers' and Suppliers'
Requests
Automakers, trade associations, and auto suppliers recommended
several changes to the current off-cycle credit program.\3344\ Prior to
the NPRM, automakers and suppliers suggested changes to the off-cycle
program, including:
---------------------------------------------------------------------------
\3344\ See generally Alliance/Global Petition.
---------------------------------------------------------------------------
Streamlining the program in ways that would give auto
manufacturers more certainty and make it easier for manufacturers to
earn credits;
Expanding the current pre-defined off-cycle credit menu to
include additional technologies and increasing credit levels where
appropriate;
Eliminating or increasing the credit cap on the pre-
defined list of off-cycle technologies and revising the thermal
technology credit cap; and
Creating a role for suppliers directly to seek approval of
their technologies.
EPA requested comments on several aspects of the off-cycle credits
program and, as discussed below, both EPA and NHTSA are adopting some
modest changes, primarily to help streamline and clarify their
programs, and to ease the implementation burden for manufacturers and
the government. The agencies are not adopting a significant expansion
of the programs in this rule, as also discussed below. EPA and NHTSA
are taking this relatively conservative approach for their off-cycle
programs due to the uncertainty that remains in estimating off-cycle
benefits of technologies and the need to remain cautious to help ensure
that emissions and fuel economy benefits expected through the off-cycle
flexibility are realized in the real-world.
(1) Program Streamlining
EPA requested comments on changes to the off-cycle process that
would streamline the program. Currently, under the third pathway,
manufacturers submit an application that includes the methodology they
used to determine the off-cycle credit value and data, which then
undergoes a public notice and comment process prior to an EPA decision
regarding the application. Each manufacturer separately submits an
application to EPA that must undergo a public notice and comment
process even if the manufacturer uses a methodology previously approved
by EPA for another manufacturer. For example, under the current
program, multiple manufacturers have separately submitted applications
for high-efficiency alternators and advanced A/C compressors using
similar methodologies and producing similar levels of credits. If
manufacturers also seek fuel economy improvement values for the CAFE
program, they are also required to send the submissions to NHTSA, as
EPA consults with NHTSA in its determinations for the CAFE program.
NHTSA's involvement is discussed in more detail in Section IX.D.3.b).
EPA requested comment on revising the regulations to allow all auto
manufacturers to make use of a methodology once it has been approved by
EPA under the public process, without subsequent applications from
other manufacturers having to undergo the same process. This would
reduce redundancy in the current program. Manufacturers would need to
provide EPA with at least the same level of data and detail for the
technology and methodology as the manufacturer that went through the
initial public notice and comment process.
EPA received supportive comments for streamlining the approval
process from auto manufacturers and suppliers. The Auto Alliance
commented that it supports all actions that would shorten the time it
takes EPA to evaluate and reach decisions on applications through the
off-cycle alternative methodology pathway, and that manufacturers
should be allowed to use common data from applications that have
already been approved.\3345\ Such common data would include ambient
conditions, general consumer behavior data, and general operating and
performance data for the same off-cycle technologies. Global Automakers
also commented that EPA should streamline efforts to avoid
reduplication of applications in situations where multiple automakers
have submitted petitions for the same technology and recommended
blanket approval for applications using the same specific technologies
and calculation and measurement procedures.\3346\ General Motors
commented that when a credit for a new technology is approved for one
manufacturer, the EPA decision document announcing that approval can
serve as a guidance document that assigns a credit value or calculation
methodology for the technology for all manufacturers without requiring
duplicative testing.\3347\ MEMA commented that it would be sufficient
to uphold the integrity of the off-cycle program to require the next
vehicle manufacturer's application to provide at least the same level
of data and details as the original vehicle manufacturer application
and to validate the level of credit the next vehicle manufacturer is
[[Page 25234]]
applying for based on how the technology is applied in its fleet.\3348\
---------------------------------------------------------------------------
\3345\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3346\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
\3347\ General Motors, Detailed Comments, NHTSA-2018-0067-11858-
21.
\3348\ MEMA, EPA-HQ-OAR-2018-0283-5692. See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
ACEEE commented that any streamlining of the process by which
automakers petition for off-cycle credits must maintain the requirement
that a thorough methodology show real-world benefits and ensure
adequate opportunity for public review.\3349\ International Council on
Clean Transportation (ICCT), while not commenting on this specific
request for comment, commented that the program should remain unchanged
until potential changes can be further analyzed.\3350\
---------------------------------------------------------------------------
\3349\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122.
\3350\ International Council on Clean Transportation, Detailed
Comments, NHTSA-2018-0067-11741.
---------------------------------------------------------------------------
After considering the comments, consistent with its request for
comment, EPA is streamlining the approval process as follows: Once a
methodology for a specific off-cycle technology has gone through the
public notice and comment process and is approved for one manufacturer,
other manufacturers may follow the same methodology to collect data on
which to base their off-cycle credits. Once a methodology is approved,
other manufacturers may submit applications citing the approved
methodology, but those manufacturers must provide their own necessary
test data, modeling, and calculations of credit value specific to their
vehicles, and any other vehicle-specific details pursuant to that
methodology, to assess an appropriate credit value. This is similar to
what occurred, for example, with the advanced A/C compressor, where one
manufacturer applied for credits with data collected through bench
testing and vehicle testing and subsequent manufacturers applied for
credits following the same methodology, but by submitting test data
specific to their vehicle models. However, those subsequent
applications previously required a public notice and comment process.
For future applications, as long as the testing is conducted using the
previously-approved methodology, EPA will evaluate the credit
application and issue a decision with no additional notice and comment,
since the first application that established the methodology was
subject to notice and comment.
EPA is not providing blanket approval for a specific credit value,
nor amending the requirement that manufacturers collect necessary data
or perform modeling or other analyses on their specific vehicle models
as the basis for the credit. However, once a methodology has been fully
vetted and approved through the public process, EPA believes additional
public review of the identical methodology is unnecessarily
duplicative. In EPA's experience thus far (for example with high-
efficiency alternators and advanced A/C compressors for which EPA has
received applications from several manufacturers based on the same
methodology), additional public review has yielded no additional
substantive public comments. EPA believes this change in the program
will help reduce the time necessary for review of applications. EPA
will maintain the option to seek additional public comment in cases
where the agency believes a new application deviates from a previously
approved methodology or raises new issues on which the agency believes
it is prudent to seek comment.
EPA also requested comment on revising the regulations to allow EPA
to, in effect, add technologies to the pre-approved credit menu without
going through a subsequent rulemaking. For example, if one or more
manufacturers submit applications with sufficient supporting data for
the same or similar technology, the data from that application(s) could
potentially be used by EPA as the basis for adding technologies to the
menu. EPA requested comment on revising the regulations to allow EPA to
establish through a decision document a credit value, or scalable value
as appropriate, and technology definitions or other criteria to be used
for determining whether a technology qualifies for the new menu credit.
As envisioned in the NPRM, this streamlined process of adding a
technology to the menu would involve an opportunity for public review
but not a formal rulemaking to revise the regulations, allowing EPA to
add technologies to the menu in a timely manner, where EPA believes
that sufficient data exist to estimate an appropriate credit level for
that technology across the fleet.
EPA received supportive comments regarding this request for
comments from auto manufacturers and suppliers who believe that the
change would help streamline the program. EPA also received comments
from environmental NGOs suggesting that the program should not be
changed at this time. After consideration of these comments, the
agencies are not revising the regulations to allow technologies to be
added to the menu without a rulemaking because EPA believes that menu-
based off-cycle credits should be based on a robust demonstration of
the technology, consistent with the regulations. The agencies will
retain the option to add technologies to the menu through a rulemaking,
similar to the approach being taken for high-efficiency alternators and
advanced A/C compressors as discussed below, where sufficient data has
been collected from multiple manufacturers and vehicle models on which
to base a menu credit. The menu credits are meant to be conservative.
The agencies are concerned that basing a menu credit on data from only
one or a few manufacturers does not guarantee a robust and accurate
credit level representing vehicles across the fleet. At this time, the
agencies continue to believe a rulemaking process with full opportunity
for public comment remains the best approach for adding technologies to
the menu. A rulemaking ensures that all stakeholders including
automakers have an opportunity to provide data to support an
appropriate and conservative credit level for the fleet. This approach
also provides an incentive for manufacturers to, in the meantime,
continue to perform testing and provide actual data that could
eventually be used to inform a rulemaking process to add a technology
to the menu. The agencies want to preserve that element of the program
to maintain the integrity of off-cycle credits representing real-world
reductions.
(2) A/C and Off-Cycle Application Process
The agencies received several comments, in addition to those
received in the petitions from the Auto Alliance and Global Automakers,
discussed below, on the application process for approving additional A/
C and off-cycle credits. Commenters included the Global Automakers, the
Auto Alliance, Volkswagen, Edison Electric Institute, Ford, Fiat
Chrysler, NCAT, Toyota, General Motors, and DENSO International.
Fiat Chrysler, Ford, Volkswagen, DENSO International, Global
Automakers, and the Auto Alliance requested that the agencies respond
more quickly to applications for A/C and off-cycle technologies.\3351\
They
[[Page 25235]]
prefer that petitions be addressed before the close of a model year so
manufacturers can have a better idea of what credits they will earn.
---------------------------------------------------------------------------
\3351\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943-
50; Ford, Detailed Comments, NHTSA-2018-0067-11928-15; Volkswagen,
Detailed Comments, NHTSA-2017-0069-0583-13; DENSO, Detailed
Comments, NHTSA-2018-0067-11880-5; Global Automakers, Detailed
Comments, NHTSA-2018-0067-12032-50; Auto Alliance, Detailed
Comments, NHTSA-2018-0067-12073-120.
---------------------------------------------------------------------------
The agencies agree that responding to petitions before the end of a
model year is beneficial to manufacturers and the government.
Manufacturers would have a better idea of the approved credits, and the
government could carry-out its compliance processes more efficiently.
EPA structured the A/C and off-cycle programs to make it possible to
complete the processes by the end of the model year so manufacturers
could submit their final reports within the required deadline, 90 days
after the calendar year. However, delays currently exist due to the
timing needed to review and approve technologies for the first time and
issue Federal Register notices seeking public comments, where
applicable. The agencies anticipate these problems will resolve
themselves as the off-cycle program reaches maturity and EPA initiates
the new streamlining approaches adopted in this final rule, discussed
in the previous section.
The agencies are also aware that delays exist because manufacturers
frequently submit late applications, new applications, and ask for
retroactive credits or FCIVs for off-cycle technologies equipped on
previously-manufactured vehicles after the model year has ended. As
required under both the CO2 and CAFE programs, manufacturers
are to submit applications for off-cycle credits and FCIVs before the
beginning of each compliance model year, to enable the agencies to make
better informed final decisions before the model year ends.
To expedite the process of approvals, the agencies will enforce
existing EPA and NHTSA regulations requiring manufacturers to notify
and report information on the technologies before the beginning of the
model year. Presently, manufacturers must notify EPA in their pre-model
year reports, and in their applications for certification, of their
intention to generate any A/C and off-cycle credits before the model
year, regardless of the methodology for generating credits.\3352\
Manufacturers choosing to generate credits using the alternative EPA-
approval methodology are required to submit a detailed analytical plan
to EPA prior to a model year in which a manufacturer intends to seek
these credits. The manufacturer may seek EPA input on the proposed
methodology prior to conducting testing or analytical work, and EPA
will provide input on the manufacturer's analytical plan. The
alternative demonstration program must be approved in advance by the
Administrator. NHTSA has similar provisions for its projections reports
in which detailed information on the technologies must be included in
those submissions during the month of December before the model
year.\3353\ NHTSA's provisions also require manufacturers to submit
information to NHTSA at the same time as to EPA. Consequently, the
eligibility of a manufacturer to gain off-cycle CO2 credits
or CAFE adjustments for a given compliance model year requires
appropriate submissions to the agencies. The agencies intend to enforce
these provisions starting with the 2020 compliance model year.
Manufacturers may resubmit MY 2020 information until May 1, 2020. After
that time, the agencies will deny any manufacturers' late submissions
requesting retroactive credits. However, manufacturers who properly
submit information ahead of time will be allowed to make corrections to
resolve inadvertent errors during or after the model year. The agencies
believe that enforcing the existing submission requirements will be the
most efficient approach to expedite approvals until new regulatory
deadlines or additional requirements can be adopted.
---------------------------------------------------------------------------
\3352\ See 40 CFR 86.1869(a) and 40 CFR 1843-01.
\3353\ See 49 CFR part 537.7(c)(7) and 49 CFR part 531.6 and
533.6.
---------------------------------------------------------------------------
Fiat Chrysler, Volkswagen, Global Automakers, and the Auto Alliance
further suggested the EPA issue a Federal Register notice for submitted
off-cycle applications within 30 days and issue a final decision within
90 days.\3354\
---------------------------------------------------------------------------
\3354\ Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943;
Volkswagen, Detailed Comments, NHTSA-2017-0069-0583; Global
Automakers, Detailed Comments, NHTSA-2018-0067-12032; Auto Alliance,
Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
As mentioned, EPA is addressing the issues raised by commenters by
streamlining its required regulatory processes to eliminate the need to
submit multiple Federal Register notices concerning requests from
different manufacturers for the same technology. Under this streamlined
process, after a technology is approved for the initial
manufacturer(s), EPA will approve any subsequent manufacturer requests
for the same technology upon receipt of data submissions validating the
benefit specific to their model types.
General Motors, Toyota, NCAT, Fiat Chrysler, Ford, Volkswagen,
DENSO, Edison Electric Institute, Global Automakers, and the Auto
Alliance further suggested that technologies approved for multiple
manufacturers, to the extent additional automakers will have the same
requests, be added to the menu to encourage additional implementation
of the technology. Doing so would reduce duplicative efforts for the
agencies, as well as manufacturers.\3355\
---------------------------------------------------------------------------
\3355\ General Motors, Detailed Comments, NHTSA-2018-0067-11858;
Toyota, Detailed Comments, NHTSA-2018-0067-12150; NCAT, Detailed
Comments, NHTSA-2018-0067-11969; Fiat Chrysler, Detailed Comments,
NHTSA-2018-0067-11943; Ford, Detailed Comments, NHTSA-2018-0067-
11928; Volkswagen, Detailed Comments, NHTSA-2017-0069-0583; DENSO,
Detailed Comments, NHTSA-2018-0067-11880; Edison Electric Institute,
Detailed Comments, NHTSA-2018-0067-11918; Global Automakers,
Detailed Comments, NHTSA-2018-0067-12032; Auto Alliance, Detailed
Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
As mentioned previously, the agencies have decided to allow only
new technologies to be added to the menu through the regular rulemaking
processes including the opportunity for notice and public comment.
General Motors, DENSO, Global Automakers, and the Auto Alliance
further suggested that suppliers should be allowed to request a ``grams
per mile'' value for their off-cycle technologies. They asserted that
this will provide certainty to manufacturers before they buy that
technology.\3356\ Toyota and the Auto Alliance suggested that the
agencies could improve efficiency and reduce burdens by creating a
``toolbox,'' methodologies that manufacturers can apply to the analysis
of off-cycle credit opportunities.\3357\ They stated it would
additionally help manufacturers if the agency would issue guidance
letters and decision documents for off-cycle credit approvals.\3358\
---------------------------------------------------------------------------
\3356\ General Motors, Detailed Comments, NHTSA-2018-0067-11858;
DENSO, Detailed Comments, NHTSA-2018-0067-11880; Global Automakers,
Detailed Comments, NHTSA-2018-0067-12032; Auto Alliance, Detailed
Comments, NHTSA-2018-0067-12073.
\3357\ Toyota, Detailed Comments, NHTSA-2018-0067-12150; Auto
Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3358\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
The agencies believe that developing a ``toolbox'' may not be
possible due to the development of new and emerging technologies, and
manufacturers' different approaches for evaluating the benefits of the
technologies. The agencies may consider additional guidance, if
feasible, as the programs further matures in the approval process of
technologies and if the agencies can identify consistent methodologies
that may help manufacturers analyze off-cycle technologies.
[[Page 25236]]
NCAT and General Motors requested more transparency in the A/C and
off-cycle approval process. They suggested that the agencies could
provide reports including off-cycle credits approved by vehicle make
and model and provide further clarification of data requirements that
influenced the decision process.\3359\
---------------------------------------------------------------------------
\3359\ NCAT, Detailed Comments, NHTSA-2018-0067-11969; General
Motors, Detailed Comments, NHTSA-2018-0067-11858.
---------------------------------------------------------------------------
EPA and NHTSA have separate approaches for sharing information on
these flexibilities, to provide public transparency. EPA already
provides detailed information on manufacturers generation of A/C and
off-cycle credits for each model year in its end of the year compliance
report, including the magnitude of credits by manufacturer and by
credit type, the credits generated by technology type, and the
penetration of off-cycle technologies in each manufacturer's
fleet.\3360\ NHTSA plans to share similar information on its PIC and to
provide projected data on the market penetration rates of the
technologies as soon as it starts receiving information through its new
reporting templates for the 2023 compliance model year.
---------------------------------------------------------------------------
\3360\ ``The 2018 EPA Automotive Trends Report: Greenhouse Gas
Emissions, Fuel Economy, and Technology since 1975,'' EPA-420-R-19-
002. March 2019; Figures 5.8 through 5.12, and Tables 5.3 and 5.4.
---------------------------------------------------------------------------
(3) High Efficiency Alternators and Advanced Air Conditioning (A/C)
Compressors
EPA sought comments on modifying the off-cycle menu to add certain
technologies for which EPA has collected sufficient data to set an
appropriate credit level. More specifically, EPA received data from
multiple manufacturers on high-efficiency alternators and advanced air
conditioning (A/C) compressors that could serve as the basis for new
menu credits for these technologies.\3361\ EPA requested comments on
adding these two technologies to the menu including comments on credit
level and appropriate definitions. EPA also requested comments on other
off-cycle technologies that EPA could consider adding to the menu
including supporting data that could serve as the basis for the credit.
---------------------------------------------------------------------------
\3361\ https://www.epa.gov/vehicle-and-engine-certification/compliance-information-light-duty-greenhouse-gas-ghg-standards.
---------------------------------------------------------------------------
EPA received only supportive comments on its specific request for
comments regarding adding high efficiency alternators and advanced A/C
compressors to the menu. Toyota, General Motors, BorgWarner, Fiat
Chrysler, the Auto Alliance, Global Automakers, MECA, DENSO, SAFE, and
Volkswagen submitted responses on the off-cycle menu. General Motors,
Volkswagen, Fiat Chrysler, Global Automakers, and the Auto Alliance all
supported adding high-efficiency alternators and advanced A/C
compressors to the menu.\3362\ They commented that these technologies
have already been approved for off-cycle credits through the petition
process multiple times. They contend that it would be less burdensome
if the technologies would be added to the pre-approved off-cycle credit
list. That said, they were concerned about being constrained by the
off-cycle caps.\3363\
---------------------------------------------------------------------------
\3362\ General Motors, Detailed Comments, NHTSA-2018-0067-11858;
Volkswagen, Detailed Comments, NHTSA-2017-0069-0583; Fiat Chrysler,
Detailed Comments, NHTSA-2018-0067-11943; Global Automakers,
Detailed Comments, NHTSA-2018-0067-12032; Auto Alliance, Detailed
Comments, NHTSA-2018-0067-12073.
\3363\ See, e.g., General Motors, Detailed Comments, NHTSA-2018-
0067-11858.
---------------------------------------------------------------------------
The agencies believe that adding high-efficiency alternators and
advanced A/C compressors to the menu is a reasonable step to help
streamline the program by allowing manufacturers to select the menu
credit rather than continuing to seek credits through the public
approval process. Therefore, EPA is revising the regulations to add
these two technologies to the menus. The high-efficiency alternator is
being added to the off-cycle credits menu, and the advanced A/C
compressor with a variable crankcase valve is being added to the menu
for A/C efficiency credits. The credit levels are based on data
previously submitted by multiple manufacturers through the off-cycle
credits application process, and discussed in the NPRM. The high
efficiency alternator credit is scalable with efficiency, providing an
increasing credit value of 0.16 grams/mile CO2 per percent
improvement as the efficiency of the alternator increases above a
baseline level of 67 percent efficiency. The advanced A/C compressor
credit value is 1.1 grams/mile for both cars and light trucks.\3364\
---------------------------------------------------------------------------
\3364\ For additional details regarding the derivation of these
credits see EPA's Memorandum to Docket EPA-HQ-OAR-2018-0283
(``Potential Off-cycle Menu Credit Levels and Definitions for High
Efficiency Alternators and Advanced Air Conditioning Compressors'').
---------------------------------------------------------------------------
EPA also received comments from the Auto Alliance, Fiat Chrysler,
General Motors, Mitsubishi, Gentherm, ITB, and MEMA on a variety of
individual technologies that they suggest adding to the menu.\3365\
These commenters provided little data to support their recommended
credit levels. The Auto Alliance and Alliance for Vehicle Efficiency
further asserted that flexibility mechanisms are increasingly important
and there is a need to develop unconventional and non-traditional fuel
economy technologies to meet standards.\3366\ They requested additional
pre-defined and pre-approved technologies to be included in this
regulation.\3367\
---------------------------------------------------------------------------
\3365\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073-
48; Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943; General
Motors, Detailed Comments, NHTSA-2018-0067-11858; Mitsubishi,
Detailed Comments, NHTSA-2018-0067-12056; MEMA, Detailed Comments,
MEMA, EPA-HQ-OAR-2018-0283-5692 (See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20
Appendices%20Oct%2026%202018.pdf); ITB, Detailed Comments, EPA-HQ-
OAR-2018-0283-5469; Gentherm, Detailed Comments, EPA-HQ-OAR-2018-
0283-5058.
\3366\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Alliance for Vehicle Efficiency, Detailed Comments, NHTSA-2018-0067-
11696.
\3367\ NHTSA-2018-0067-12073-48.
---------------------------------------------------------------------------
The agencies have reviewed manufacturers' requests for adding
additional technologies to the picklist and concluded that there is
insufficient data in the record at this time on which to base an
appropriate menu credit value for the technologies. Therefore, none of
these technologies are being added to the menu at this time. Given the
limited data and uncertainty, EPA also does not believe it would be
appropriate to add any of the technologies to the menu without an
opportunity for public review and comment. Although the agencies are
not adding these technologies to the menu at this time, manufacturers
may seek off-cycle credits for these technologies through the other
program pathways.
(4) Stop-Start Technology
In 2014, EPA approved additional credits for the Mercedes-Benz's
stop-start system through the off-cycle credit process based on data
submitted by Mercedes-Benz on fleet idle time and its system's real-
world effectiveness (i.e., how much of the time the system turns off
the engine when the vehicle is stopped).\3368\ Prior to proposal,
multiple auto manufacturers requested that EPA revise the table menu
value for stop-start technology based solely on one input value EPA
considered, idle time, in the context of the Mercedes-Benz stop-start
system. No manufacturers provided additional data on any of the other
factors evaluated during consideration of a conservative credit value
for stop-start systems. Stop-start systems vary significantly in
hardware,
[[Page 25237]]
design, and calibration, leading to wide variations in the amount of
idle time during which the engine is actually turned off in real-world
driving. EPA has learned that some stop-start systems may be less
effective in the real-world than the agency estimated in its 2012
rulemaking analysis, for example, due to systems having a disable
switch available to the driver, or because stop-start systems can be
disabled under certain temperature conditions or auxiliary loads, which
would offset the benefits of the higher idle time estimates. EPA
requested additional data from manufacturers, suppliers, and other
stakeholders regarding a comprehensive update to the stop-start off-
cycle credit table value. EPA did not receive any additional real-world
system effectiveness data from commenters on which to base an adjusted
credit level. MEMA commented that EPA should base an increase in the
credit on the agencies' updated estimated effectiveness of stop-start
technology in the Draft Technical Assessment Report (TAR), which shows
a 67 percent increase in effectiveness.\3369 3370\ However, EPA notes
that this estimate is for system effectiveness over the 2-cycle test
procedures and, therefore, is not an appropriate basis to adjust the
off-cycle credits. The agencies are not adjusting the menu credits for
stop-start systems at this time. Manufacturers may apply for additional
credits if they are able to collect data demonstrating a system
effectiveness that would serve as the basis for those credits.
---------------------------------------------------------------------------
\3368\ ``EPA Decision Document: Mercedes-Benz Off-cycle Credits
for MY 2012-2016,'' EPA-420-R-14-025 (Sept. 2014).
\3369\ Draft Technical Assessment Report: Midterm Evaluation of
Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate
Average Fuel Economy Standards for Model Years 2022-2025, EPA-420-D-
16-900 (July 2016).
\3370\ MEMA, EPA-HQ-OAR-2018-0283-5692. See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
(5) Menu Credit Cap
The off-cycle menu currently includes a fleetwide cap on credits of
10 grams/mile to address the uncertainty surrounding the data and
analysis used as the basis of the menu credits.\3371\ Prior to
proposal, some stakeholders expressed concern that the current cap may
constrain manufacturers' future ability to fully utilize the menu
especially if the menu is expanded to include additional technologies,
as described above. For example, Global Automakers suggested raising
the cap from 10 grams/mile to 15 grams/mile.\3372\ EPA requested
comments on increasing the current cap, for example, from the current
10 grams/mile to 15 grams/mile to accommodate increased use of the
menu. EPA also requested comment on a concept that would replace the
current menu cap with an individual manufacturer cap that would scale
with the manufacturer's average fleetwide target levels. The cap would
be based on a percentage of the manufacturer's fleetwide 2-cycle
emissions performance, for example at five to ten percent of
CO2 of a manufacturer's emissions fleet-wide target. With a
cap of five percent for a manufacturer with a 2-cycle fleetwide average
CO2 level of 200 grams/mile, for example, the cap would be
10 grams/mile.
---------------------------------------------------------------------------
\3371\ 40 CFR 86.1869-12(b)(2).
\3372\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
---------------------------------------------------------------------------
There was widespread support from automakers and suppliers for
removing the cap entirely or raising the cap from 10 grams/mile to 15-
20 grams/mile. Toyota, General Motors, BorgWarner, Fiat Chrysler, the
Auto Alliance, Global Automakers, MECA, DENSO, SAFE, and Volkswagen
submitted responses on the off-cycle cap to EPA.\3373\ They argued that
the 2-cycle test does not always account for all the benefits a
technology provides.\3374\ General Motors, Fiat Chrysler, the Auto
Alliance, Global Automakers, and Volkswagen agreed that EPA should
remove the 10 grams/mile cap and, if they must keep the cap, increasing
it to 15 grams/mile.\3375\
---------------------------------------------------------------------------
\3373\ Toyota, Detailed Comments, NHTSA-2018-0067-12150; General
Motors, Detailed Comments, NHTSA-2018-0067-11858; BorgWarner,
Detailed Comments, NHTSA-2018-0067-11895; Fiat Chrysler, Detailed
Comments, NHTSA-2018-0067-11943; Auto Alliance, Detailed Comments,
NHTSA-2018-0067-12073; Global Automakers, Detailed Comments, NHTSA-
2018-0067-12032; MECA, Detailed Comments, NHTSA-2018-0067-11994;
DENSO, Detailed Comments, NHTSA-2018-0067-11880; SAFE, Detailed
Comments, NHTSA-2018-0067-11981; Volkswagen, Detailed Comments,
NHTSA-2017-0069-0583.
\3374\ See, e.g., DENSO, Detailed Comments, NHTSA-2018-0067-
11880.
\3375\ General Motors, Detailed Comments, NHTSA-2018-0067-11858;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943; Auto
Alliance, Detailed Comments, NHTSA-2018-0067-12073; Global
Automakers, Detailed Comments, NHTSA-2018-0067-12032; Volkswagen,
Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
Global Automakers commented that, as more technology receives off-
cycle credit values, the cap will restrict innovation and therefore EPA
should lift the cap now in anticipation of increased use of
technologies.\3376\ General Motors similarly commented that the cap was
an arbitrary limit without any technical justification and that, if the
agency was to add emission reduction technologies to the menu these
devices could not be effectively incentivized if the 10 grams/mile cap
remains in place, since there would be no room under the cap.\3377\
General Motors suggested that as the program continues, manufacturers
will continue to find new technologies and will be limited by the cap.
They stated that the cap will stifle additional investments for
technologies. MEMA commented that if EPA expands the off-cycle
technologies menu and continually adds off-cycle technologies to the
menu, it is critical that EPA increase or eliminate the cap on the
credits gained from the off-cycle menu.\3378\
---------------------------------------------------------------------------
\3376\ Global Automakers, Detailed Comments, NHTSA-2018-0067-
12032.
\3377\ General Motors, Detailed Comments, NHTSA-2018-0067-11858.
\3378\ MEMA, EPA-HQ-OAR-2018-0283-5692. See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
The Auto Alliance argued that putting caps on emerging new
technologies will hinder further vehicle investments and improvements.
The planning cycle is implemented years out and without a guarantee
they will see benefits, the Auto Alliance stated that manufacturers
lack incentivization to work toward large technological advances.\3379\
The Auto Alliance and Alliance for Vehicle Efficiency further asserted
that flexibility mechanisms are increasingly important and there is a
need to develop unconventional and non-traditional fuel economy
technologies.\3380\
---------------------------------------------------------------------------
\3379\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3380\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Alliance for Vehicle Efficiency, Detailed Comments, NHTSA-2018-0067-
11696.
---------------------------------------------------------------------------
ACEEE commented that the off-cycle credit menu cap should not be
increased or modified without the agency first defining any other
changes it might consider making to the off-cycle credit program and
this should be done through a separate NPRM and public review
process.\3381\ ICCT commented that if the agencies allow more use of
off-cycle credits without clear validation of their real-world
benefits, the regulations cannot serve their intended objectives to
reduce CO2 and fuel use.\3382\
---------------------------------------------------------------------------
\3381\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122.
\3382\ ICCT, Detailed Comments, NHTSA-2018-0067-11741-43.
---------------------------------------------------------------------------
EPA also received a few comments warning about the risks of
removing the caps and over incentivizing the CAFE and CO2 programs.
ACEEE pointed out that while expanding and updating the flexibilities
that incentivize innovation
[[Page 25238]]
and research is a great method to increase fuel efficiency, it is
important to put a time limit on those incentives and carefully design
them so manufacturers do not take advantage. ACEEE argued that, if
these flexibilities are not implemented thoughtfully, they can end up
reducing the program benefits. UCS commented that, given the potential
interaction from multiple incentives, it is important to consider the
combined impacts of flexibilities on the overall stringency of the
regulation. UCS stated that given the potential for widespread harm,
credits within the program should be severely limited, and the
agencies' assessment of the impacts of such incentives should be
extremely conservative in order to promote increased environmental
benefits of the fuel economy and carbon dioxide emissions
standards.\3383\
---------------------------------------------------------------------------
\3383\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
---------------------------------------------------------------------------
The agencies are not increasing the 10 grams/mile menu credit cap
at this time. EPA established the 10 grams/mile credit cap to address
the uncertainty surrounding the data and analysis used as the basis of
the menu credits, and agrees with ACEEE, ICCT, and UCS that sufficient
uncertainty remains such that increasing the current cap is not
justified. As noted in the 2012 final rule, EPA included the fleet-wide
cap because the default credit values were based on limited data, and
also because the agencies recognized that some uncertainty is
introduced when credits are provided based on a general assessment of
off-cycle performance as opposed to testing on the individual vehicle
models.\3384\ That uncertainty has not significantly diminished since
the 2012 final rule. Also, over the course of implementing the program,
EPA has encountered issues with the regulatory definitions currently in
place for some technologies. The regulations specify that manufacturers
may claim credits for technologies that meet the regulatory
definitions. However, there have been instances where manufacturers
have claimed credits for a technological approach that they have argued
meets the regulatory definition, but EPA found that the technology was
not implemented consistent with the technological approach envisioned
when the off-cycle program was established. This has raised questions
of whether the credits for the technological approach in question truly
represent real-world reductions, and whether the credits should
ultimately be allowed. These types of issues have resulted in
uncertainty, which can lead to delays in credit calculations,
competitive inequities, as well as increased burden on the agency to
review and resolve issues. The caps continue to serve as an important
measure against the loss of emissions reductions and fuel savings given
the uncertainty in the credit values as the program is implemented.
Since the agencies are not expanding the menu beyond the two
technologies discussed above, the agencies believe there remains enough
room under the cap such that the menu may continue to serve its purpose
as a source of off-cycle credits. Although a few manufacturers
approached the cap limit in MY 2018, the fleet average menu credit was
4.7 grams/mile, less than half the cap value.\3385\ If the agencies
undertake a rulemaking in the future to modify the menu or regulatory
definitions, the agencies may re-evaluate the cap levels at that time.
The agencies note that the cap only applies to credits based on the
menu. Under the current program, manufacturers may apply for credits
beyond the cap through other available pathways based on a
demonstration of off-cycle technology emission reduction data for their
fleets.
---------------------------------------------------------------------------
\3384\ 77 FR 62834 (Oct. 15, 2012).
\3385\ The 2018 EPA Automotive Trends Report, Greenhouse Gas
Emissions, Fuel Economy, and Technology since 1975, EPA-420-R-19-002
(Mar. 2019).
---------------------------------------------------------------------------
As noted above, the agencies have decided to continue the option to
add technologies to the menu only through the rulemaking process and,
for this final rule, have decide to add two new menu items; one for
high-efficiency alternators and another for advanced A/C compressors.
The agencies stated that they will only add technologies when
sufficient data has been collected from multiple manufacturers and
vehicle models on which to base a menu credit. Accordingly, the
agencies believe this approach ensures that conservative, robust and
accurate credit levels are being added representing vehicles ``on
average'' across the fleet.
Finally, NHTSA has been studying how the combination of
flexibilities and incentives may adversely affect the stringency of the
CAFE regulations. NHTSA is aware of an instance in which combining
incentives for alternative fueled vehicles and adjustments for A/C and
off-cycle technologies allowed one manufacturer to increase in CAFE
fleet performance to a combined average of 516.8 mpg for MY 2017, a
curious result. NHTSA iscontinuing to evaluate the issue of combining
incentives and flexibilities and may address this issue further in the
future.
(6) Eligibility
Though, in the NPRM, EPA did not explicitly request comment on the
eligibility criteria for determining what technologies are eligible for
off-cycle credits, EPA received comments on this topic. UCS commented
that regulations should be clarified so that the program does not
result in unwarranted credits for baseline technologies, noting that in
the 2012 final rule EPA stated that technologies integral or inherent
to the basic vehicle design were not eligible for credits and
specifically excluded technologies identified by the agency as
technologies a manufacturer may use to meet the two-cycle
CO2 standards.\3386\ ACEEE commented that off-cycle credits
should be limited to new and innovative technologies and, that to be
eligible for credit, a technology must reduce emissions from the
vehicle receiving the credit (as opposed to other vehicles on the road,
for example, through system effects of technologies designed for crash
avoidance or improving traffic flow).\3387\ The Auto Alliance also
commented in the area of eligibility, suggesting regulatory changes
that would allow off-cycle credits for any technology where the
manufacturer could demonstrate an off-cycle emissions benefit.\3388\
The Auto Alliance commented that the program is intended to provide
credit for technologies that provide more fuel economy and
CO2 emissions reduction benefit in the real-world than is
realized in FTP and HFET on-cycle testing and that a baseline
technology should be eligible for such credits.
---------------------------------------------------------------------------
\3386\ UCS, Detailed Comments, NHTSA-2018-0067-12039.
\3387\ ACEEE, Detailed Comments, NHTSA-2018-0067-12122.
\3388\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Given the various public comments on eligibility of technologies
for off-cycle credits, the agencies are clarifying the regulations
regarding technology eligibility, consistent with the intent and EPA's
interpretation of the 2012 rule, as expressed in the preamble to the
proposed and final rules. The agencies believe that clarifying the
regulations will reduce confusion among manufacturers as to what
technologies are eligible and reduce the overall program burden
associated with EPA staff giving continued guidance to manufacturers
regarding eligibility, as detailed in the 2012 rule preamble.
Eligibility was thoroughly addressed in the 2012 final rule preamble,
but the regulations were not as clear, which has led to confusion on
the part of some manufacturers and delays in reviewing
[[Page 25239]]
credit applications.\3389\ The agencies are not establishing a new
policy regarding eligibility, only amending the language reflecting the
existing policy in the regulations for sake of clarity.
---------------------------------------------------------------------------
\3389\ 77 FR 62726-36, 62835-37.
---------------------------------------------------------------------------
As noted in the 2012 final rule preamble, the goal of the off-cycle
credits program is to provide ``an incentive for the development and
use of additional technologies to achieve real-world reductions in
CO2 emissions.'' \3390\ EPA further stated that the intent
of the program is to ``provide an incentive for CO2 and fuel
consumption reducing off-cycle technologies that would otherwise not be
developed because they do not offer a significant 2-cycle benefit.''
\3391\ The regulation at 40 CFR 86.1869-12(a) provides that
manufacturers may generate credits for CO2 reducing
technologies ``where the CO2 reduction benefit for the
technology is not adequately captured on the Federal Test Procedure
and/or Highway Fuel Economy Test.'' The regulation continues: ``[t]hese
technologies must have a measurable, demonstrable, and verifiable real-
world CO2 reduction that occurs outside the conditions of
the Federal Test Procedure and the Highway Fuel Economy Test.''
---------------------------------------------------------------------------
\3390\ 77 FR 62833.
\3391\ 77 FR 62836.
---------------------------------------------------------------------------
Off-cycle credits are available for technologies that are not
utilized when performing FTP and HFET tests because their operation is
linked to a condition not found during the 2-cycle testing. For
example, heating and cooling systems are not operated during the 2-
cycle test, and therefore, efficiency improvements to these systems are
not captured at all on the 2-cycle tests. As the 2012 rule's language
indicates, off-cycle credits are not necessarily limited to
technologies listed on the menu or off-cycle technologies with no
measurable benefit on the FTP and/or HFET. Off-cycle credits may be
available for some technologies whose performance is measurable to some
extent on the FTP and/or HFET but which perform measurably better off-
cycle. Active aerodynamic and stop-start technologies (menu item) are
examples. However, there are limits on what the agencies would consider
to be an off-cycle technology eligible for credits, as discussed below.
Just as the regulations and preamble to the 2012 final rule listed
technologies that the agencies considered to be off-cycle technologies,
the preamble also discussed technologies that the agency would not
consider off-cycle technologies--i.e., technologies the agencies
consider to be ``adequately captured'' by the FTP and therefore not
eligible for off-cycle credits. The preamble specifically noted that
engine, transmission, mass reduction, passive aerodynamic design, and
base tire technologies are not considered to be off-cycle technologies
eligible for credits.\3392\ These are technologies that are considered
to be ``integral or inherent to basic vehicle design.'' \3393\ In
response to comments in the final rule, the agencies further clarified
that advanced combustion concepts, such as camless engines, variable
compression ratio engines, micro air/hydraulic launch assist devices,
would not be considered to be eligible for credits.\3394\ This
limitation to eligibility further extends to other engine designs,
transmission designs, and electrification systems not specifically
contemplated in the rulemaking, such as Atkinson combustion engines,
and 9 and 10 speed transmissions, as well as to other hybrid systems
such as 48 Volt technologies. Further, the 2012 final rule preamble
stated that technologies included in the agencies' assessment for
purposes of developing the standard would not be allowed to generate
off-cycle credits and cites the technologies described in Chapter 3 of
the 2012 final rule TSD.\3395\ Finally, off-cycle credits are not
available for technologies required to be used by Federal Law or for
crash avoidance systems, safety critical systems, or technologies that
may reduce the frequency of vehicle crashes.\3396\
---------------------------------------------------------------------------
\3392\ 77 FR 62732, 62836.
\3393\ 77 FR 62732, 62836/1; 81 FR 73499.
\3394\ 77 FR 62732.
\3395\ 77 FR 62836.
\3396\ 40 CFR 86.1869-12(a); 77 FR 62836.
---------------------------------------------------------------------------
The preamble to the 2012 final rule provides the rationale for what
the agency considers an off-cycle technology and, therefore, eligible
for credits. Technologies that are integral or inherent to the vehicle
are, by necessity, well represented on the 2-cycle test.\3397\ Examples
provided in the preamble are engine, transmission, mass reduction,
passive aerodynamic design, and base tire technologies. The control
logic for these powertrain components, like the components themselves
(i.e. engine and transmission), are constantly active, fully
functioning, and operating over the entirety of the FTP and HFET.
Similarly, an automatic transmission, regardless of whether it has 6-
speeds or 8-speeds, would still be constantly active, fully functioning
and operating over the entirety of the FTP and HFET.\3398\ This would
also be true for base engine technologies, advanced combustion
concepts, engine components (pistons, valves, camshafts, crankshafts,
oil pumps, etc.), and driveline components (individual components of
the transmission, axle, and differential).\3399\
---------------------------------------------------------------------------
\3397\ 77 FR 62732, 62836.
\3398\ 76 FR 75024 (Dec. 1, 2011).
\3399\ 77 FR 62732/2.
---------------------------------------------------------------------------
Further, even if these technologies have greater benefits on
supplemental test cycles, EPA has explained that it would be difficult
to devise accurate A/B testing (i.e., with and without the technology)
for these technologies.\3400\ The 2012 preamble states that ``EPA is
limiting the off-cycle program to technologies that can be identified
as add-on technologies conducive to A/B testing,'' partly because it
would be very difficult accurately to parse out the off-cycle benefits
for some integral technologies.\3401\ Because the technology is
integral to the vehicle, there would not be an appropriate baseline
(i.e., without the technology) vehicle to use for comparison. Vehicles
are not built without tires, engines, passive aerodynamics or
transmissions.
---------------------------------------------------------------------------
\3400\ 76 FR 75024.
\3401\ 77 FR 62836.
---------------------------------------------------------------------------
Also, because these technologies are inherent to the vehicle
design, their performance is already reflected in the stringency of the
standard and giving credits for these inherent technologies would be a
type of double-counting windfall.\3402\ ``[S]ince these methods are
integral to basic vehicle design, there are fundamental issues as to
whether they would ever warrant off-cycle credits. Being integral,
there is no need to provide an incentive for their use, and (more
importantly), these technologies would be incorporated regardless.
Granting credits would be a windfall.'' \3403\ As such, EPA has laid
out a clear basis that technological improvements to integral and
inherent components are considered to be adequately captured on the FTP
and HFET test.
---------------------------------------------------------------------------
\3402\ 77 FR 62732.
\3403\ See also 76 FR 75024.
---------------------------------------------------------------------------
EPA is clarifying the regulations in a manner that is consistent
with the intent and our interpretation of the 2012 rule, as expressed
in the preambles to the proposed and final rules. The regulations are
revised to specify that technologies used primarily to meet the 2-cycle
standards are not eligible for off-cycle credits and that only
technologies primarily installed for reducing off-cycle emissions would
be eligible. The revised regulations specify that the technologies must
not be integral or inherent to the basic vehicle design, such as, for
example, engine,
[[Page 25240]]
transmission, mass reduction, passive aerodynamic design, and tire
technologies. Exceptions to these general provisions include
technologies already specified on the menu, including engine idle stop-
start, active aerodynamic improvements, and high-efficiency
alternators. These technologies may provide some benefit on the 2-cycle
test, but EPA determined in the 2012 rule that they are eligible for
off-cycle credits because they are technologies that could be added to
vehicles to provide discernable off-cycle reductions.
Regulatory text at 40 CFR 86.1869-12(a) states: ``Manufacturers may
generate credits for CO2 reducing technologies where the
CO2 reduction benefit of the technology is not adequately
captured on the Federal Test Procedure and/or the Highway Fuel Economy
Test,'' to which EPA is adding, ``such that the technology would not be
otherwise installed for purposes of reducing emissions (directly or
indirectly) over those test cycles (i.e., on-cycle) for compliance with
the [CO2] standards.'' EPA is also adding text to this
paragraph of the regulations specifying: ``The technologies must not be
integral or inherent to the basic vehicle design, such as engine,
transmission, mass reduction, passive aerodynamic design, and tire
technologies. Technologies installed for non-off-cycle emissions
related reasons are also not eligible as they would be considered part
of the baseline vehicle design. The technology must not be inherent to
the design of occupant comfort and entertainment features except for
technologies related to reducing passenger A/C demand and improving A/C
system efficiency. Notwithstanding the provisions of this paragraph
(a), off-cycle menu technologies included in paragraph (b) of this
section remain eligible for credits.''
The agencies believe the above regulatory changes will help reduce
confusion over what technologies are eligible for off-cycle credits,
refocusing the program on technologies that manufacturers would install
on vehicles for purposes of reducing off-cycle emissions rather than
obtaining additional credits for technologies installed primarily for
2-cycle emissions reduction or for other reasons not related to
emissions. This approach is consistent with the intent of the program
as stated in the 2012 final rule to provide an incentive to develop and
employ off-cycle technologies not adequately captured on the 2-cycle
test procedure.
Of the technologies recommended by manufacturers to be added to the
menu, cooled EGR is an example of a technology that would not be
eligible because it is an integral 2-cycle technology that EPA noted in
its technology assessment in the MY 2012 rule. Cooled EGR is often an
integral component of turbo charged gasoline direct injection engines
which is a primary CO2 reduction strategy used by
manufacturers to reduce 2-cycle emissions. The technologies are
calibrated to act as a system such that is not possible to separate
them in a way that would allow for a clear indication of the off-cycle
benefit of cooled EGR as a stand-alone technology.
EPA also received comments from the Auto Alliance regarding several
technologies they believe should qualify as active warm-up off-cycle
technologies. The Auto Alliance commented that systems that use waste
heat from the exhaust gas stream should receive additional credits
beyond the menu credits currently established for active engine and
transmission warm-up.\3404\ However, when EPA established the menu
credits for active transmission and engine warm-up in the 2012 rule,
EPA envisioned waste heat from the exhaust as the primary source of
heat to quickly bring the system to operating temperature as the basis
for the warm-up technology credits.\3405\ Therefore, EPA does not
believe additional credits, as suggested by the Auto Alliance, are
warranted. EPA further notes that the definitions for active engine and
transmission warm-up specify that ``waste heat'' be used in active
warm-up technologies in order to qualify for the credits.\3406\ If a
system first directs heat to warm the engine oil or warm the interior
cabin, and only then to the engine or transmission, thereby delaying
active warm-up, EPA would not view that heat as waste heat since it is
serving other purposes during initial vehicle warm-up. EPA would also
not consider this approach to be warming up the engine or transmission
``quickly'' due to the potentially significant delay in warm-up
activation. In developing the active warm-up credits, EPA focused on
systems using heat from the exhaust as a primary source of waste heat
because that heat would be available quickly and also be exhausted by
the vehicle and otherwise unused.
---------------------------------------------------------------------------
\3404\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3405\ See Joint Technical Support Document: Final Rulemaking
for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards
and Corporate Average Fuel Economy Standards, EPA-420-R-12-901,
August 2012, p. 5-96--5-100.
\3406\ 40 CFR 86.1869-12(b)(4)(v) and (vi).
---------------------------------------------------------------------------
EPA allowed for the possible use of other sources of heat such as
coolant as the basis for credits as long as those methods would
``provide similar performance'' as extracting the heat directly from
the exhaust system.\3407\ However, EPA may require manufacturers to
demonstrate that the system is based on ``waste heat'' or heat that is
not being preferentially used by the engine or other systems to warm-up
other areas like engine oil or the interior cabin. Systems using waste
heat from the coolant do not qualify for credits if their operation
depends on, and is delayed by, engine oil temperature or interior cabin
temperature. As the engine and transmission components are warming up,
the engine coolant and transmission oil do not have any `waste' heat
available for warming up anything else on the vehicle. During engine
and transmission warm-up, the only waste heat source in a vehicle with
an internal combustion engine is the engine exhaust as the transmission
and coolant have not reached warmed-up operating temperature and
therefore do not have any heat to share. Conserving heat in a
transmission is not a rapid transmission warm-up using waste heat.
Unless the component with lubricating oil and coolant is operating at
its fully warmed-up design temperature, by EPA's definition, that
component does not have any waste heat available for transfer from the
lubricating oil or coolant to any other device until it has reached its
fully warmed-up operating temperature (i.e. the temperature when the
cooling system is enabled). A qualifying system may involve a second
cooling loop that operates independent of the primary coolant system
and is not dependent on or otherwise delayed by, for example, cabin
temperature. Evaluating whether such systems qualify for menu credits
often requires additional information regarding system design to
understand better how the system uses waste heat. Given the complexity
of these systems and the need to sometimes consider the details of how
a system operates, EPA is not making any changes to the menu regarding
warm-up technologies.
---------------------------------------------------------------------------
\3407\ See Joint Technical Support Document: Final Rulemaking
for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards
and Corporate Average Fuel Economy Standards, p. 5-99, EPA-420-R-12-
901, August 2012.
---------------------------------------------------------------------------
The Auto Alliance further commented that active transmission bypass
valves should qualify for active transmission warm-up credits.\3408\
The Auto Alliance
[[Page 25241]]
commented that traditional transmission oil coolers are always active
and sized for extreme or worst-case hot ambient conditions. The coolers
will, in colder ambient conditions, keep the transmission temperatures
well outside of their most efficient operating range. The bypass valve
circumvents the cooler when the transmission is relatively cold
preserving the transmission heat, so the transmission warms more
quickly. EPA disagrees that this type of approach should be eligible
for active transmission warm-up because it does not use waste heat to
add heat to the transmission. Instead, it prevents useful heat already
present in the transmission from being unnecessarily removed. Also, EPA
does not view this type of bypass valve as an off-cycle technology but
rather as part of a good engineering design of a transmission cooler
system. Many vehicles already are designed with transmission cooler
bypass valves. EPA does not believe existing coolers qualify as warm-up
technologies simply because they are disabled under cold conditions.
This approach does not represent the addition of a new off-cycle warm-
up technology but the disabling of an existing cooling technology.
---------------------------------------------------------------------------
\3408\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
Although the agencies did not consider changes to the program to
allow credits for safety-related technologies and autonomous vehicle
technologies in the proposal, comments were received both in favor of
and not in favor of allowing such credits.\3409\ The agencies note that
the rationale for not allowing off-cycle credits for safety-related or
crash avoidance technologies has not changed since the 2012 rule and,
therefore, in the proposed rule the agencies did not consider making
any changes to allow off-cycle credits for safety-related
technologies.\3410\ The agencies continue to believe that there is a
very significant distinction between technologies providing direct and
reliably quantifiable improvements to fuel economy and CO2
emission reductions, and technologies which provide those improvements
by indirect means, where the improvement is not reliably quantifiable,
and may be speculative (or in many instances, non-existent), or may
provide benefit to other vehicles on the road more than for themselves.
The agencies also continue to believe that the advancement of crash-
related and crash avoidance systems specifically is best left to
NHTSA's exercise of its vehicle safety authority.
---------------------------------------------------------------------------
\3409\ See, e.g., SAFE, Detailed Comments, NHTSA-2018-0067-
11981; AAA, Detailed Comments, NHTSA-2018-0067-11979.
\3410\ 77 FR 62733.
---------------------------------------------------------------------------
Auto manufacturers and suppliers also commented that EPA should
adopt ``eco-innovation'' credits approved in the European Union (EU)
vehicle CO2 reduction program as part of the off-cycle
credits program.\3411\ No data was provided as to why the credits would
be appropriate for the U.S. vehicle fleet. EPA did not consider or
request comment on the EU credits program and does not believe the
credit levels would necessarily be appropriate for the U.S. fleet given
the very different vehicle use and driving patterns between Europe and
the U.S. Thus, there is no assurance that the credits would be based on
real-world emissions reductions.
---------------------------------------------------------------------------
\3411\ See, e.g., Mitsubishi, Detailed Comments, NHTSA-2018-
0067-12056.
---------------------------------------------------------------------------
EPA received comments from the Auto Alliance and Global Automakers
that EPA should automatically award credits if the agency does not take
final action within 90 days of receiving a request for credits.\3412\
Regarding these comments, EPA does not believe such a provision is in
keeping with maintaining the integrity of the off-cycle credits
program. As discussed above, EPA often requires time to sort through
complex issues to determine if the technologies meet the regulatory
requirements for receiving credits and whether the credits have been
quantified appropriately. In some instances, EPA has received public
comments and manufacturer rebuttals to those comments that takes
additional time to consider before making a final decision. EPA's goal
continues to be to evaluate applications for credits in as timely a
manner as is possible given the issues that must be addressed and
within the resources available. While EPA's need carefully to consider
applications may slow down the approval process or result in credits
not being approved, it remains paramount to ensure credits are not
provided to technologies that do not provide actual off-cycle benefits,
and thereby do not meet the regulations. In the past, longer time
frames for EPA review have not caused manufacturers to lose credits
where credits are determined by EPA to be warranted under the
regulations. EPA believes that the changes EPA is making to the program
will help streamline the program and reduce confusion, thus helping to
reduce the time necessary to evaluate applications and provide final
decisions to manufacturers.
---------------------------------------------------------------------------
\3412\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Global Automakers, Detailed Comments, NHTSA-2018-0067-12032.
---------------------------------------------------------------------------
(7) Supplier Role in the Off-Cycle Credits Program
Prior to proposal, EPA heard from many suppliers and their trade
associations about an interest in allowing suppliers to have a formal,
regulatorily defined role in the off-cycle credits program.\3413\ EPA
requested comment on providing a pathway for suppliers, along with at
least one auto manufacturer partner, to submit off-cycle applications
for EPA approval. As described in the proposal, under such an approach,
an application submitted by a supplier and vehicle manufacturer would
establish a credit and/or methodology for demonstrating credits that
all auto manufacturers could then use in their subsequent applications.
EPA requested comment on requiring that the supplier be partnered in a
substantive way with one or more auto manufacturers to ensure that
there is a practical interest in the technology prior to EPA investing
resources in the approval process. The supplier application would be
subject to public review and comment prior to an EPA decision. However,
once approved, subsequent auto manufacturer applications requesting
credits based on the supplier methodology would not be subject to
public review. Under this concept, the credits would be available
provisionally for a limited period of time, allowing manufacturers to
implement the technology and collect data on their vehicles in order to
support a continuation of credits for the technology in the longer
term. Also, as envisioned by EPA in its request for comment, the
provisional credits could be included under the menu credit cap since
they would be based on a general analysis of the technology rather than
manufacturer-specific data.
---------------------------------------------------------------------------
\3413\ 83 FR 43461.
---------------------------------------------------------------------------
Auto manufacturers' and suppliers' comments were generally
supportive of an expanded role for suppliers in the off-cycle credit
program. The Auto Alliance supported allowing a supplier to lead the
application process but did not support the provisional credit concept
since the follow-up testing conducted by manufacturers may not support
the level of credits initially claimed by the supplier, resulting in a
lower than anticipated credit.\3414\ Instead, the Auto Alliance
suggested a separate cap for supplier-based credits and noted that
manufacturers could submit their own data if they wanted to pursue
credits levels that exceeded the cap. General Motors similarly
disagreed with the provisional credits that might
[[Page 25242]]
be rescinded if subsequent testing does not fully validate the value of
the technology.\3415\ MEMA supported the request for comments regarding
a supplier-led process but did not support requiring that suppliers
have an auto manufacturer partner.\3416\ MEMA commented that there
would be no incentive for a supplier to go through the product/
technology development process, collect the necessary data, and
undertake the full application process for a product/technology that
would not generate manufacturer interest.
---------------------------------------------------------------------------
\3414\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3415\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
\3416\ MEMA, EPA-HQ-OAR-2018-0283-5692. See https://www.mema.org/sites/default/files/resource/MEMA%20CAFE%20and%20GHG%20Vehicle%20Comments%20FINAL%20with%20Appendices%20Oct%2026%202018.pdf.
---------------------------------------------------------------------------
At this time, EPA believes additional discussions with interested
parties and an opportunity for public comment, both of which are beyond
the scope of this rulemaking, are needed. EPA continues to believe such
an approach could encourage the further development of off-cycle
technologies, but must be done in a reasonable way that ensures the
credits are based on real-world emissions reductions.
Under the approach suggested by the Auto Alliance, manufacturers
could claim supplier-based credits indefinitely and EPA might never
receive any manufacturer data substantiating the credits unless that
data supported a credit that exceeded the level established through the
supplier process. EPA is concerned such a one-way ratchet approach
could result in the loss of emissions benefits and undermine the
integrity of the off-cycle credit program. EPA also remains concerned
about the potential for a significantly increased volume of credit
applications, including the potential for applications for proposed
technologies that manufacturers might in reality have no interest in
adopting. EPA understands MEMA's perspective on the issue of requiring
a manufacturer partner, but a supplier-only process would potentially
open the door to many requests such that the agency would need to
expend considerable additional resources. EPA notes that nothing in the
current regulations prevents collaboration between manufacturers and
suppliers. Suppliers can initiate this process; manufacturer
participation will be necessary to complete an application. EPA will
provide additional clarity about this process through a subsequent
technical amendments rulemaking.
(8) Other Considerations
Avista Oil commented that EPA should provide an opportunity for
credits based on the use of recycled engine oil. Avista Oil commented
that there are CO2 emissions reductions associated with the
use of recycled used engine oil and that vehicle manufacturers should
be awarded credits for the use of recycled oil. Avista Oil's comment is
not within the scope of the rulemaking. The off-cycle credits program
focuses on providing credits for technologies that, when applied to the
vehicle, the result is lower quantifiable real-world emissions from the
vehicle. According to Avista Oil's comment, their recycled oil
technology benefits are associated with the recycling process rather
than lowering vehicle emissions on the road. Therefore, EPA would not
view the technology as eligible for off-cycle credits, and EPA did not
propose any other credit specific to the use of recycled engine oil.
Several commenters recommended that EPA raise the credit caps and
credit values for thermal controls based on recent work by the National
Renewable Energy Lab (NREL). Commenters suggested that credit values
should be raised by 64 percent. In response, as discussed in the
preamble, EPA is retaining the current menu credit caps and menu credit
values due to uncertainties involved with the emissions projections and
estimated credit values. Manufacturers may generate additional credits
through the off-cycle credits program using the other two pathways by
providing individual vehicle data. EPA recognizes additional modeling
analysis has been performed by NREL that indicates the potential
benefit of all thermal technologies including glazing. EPA designed the
thermal control program and related caps based on previous NREL work
and applied the thermal caps at the current levels to account for the
wide range of uncertainties--including the uncertainty of the benefit
from the combination of thermal technologies and the uncertainty
highlighted by the different credit levels across the NREL studies. EPA
believes the separate current thermal menu program cap and AC
efficiency program cap continue to be reasonable for application across
the fleet given these uncertainties.
Enhanced Protective Glass Automotive Association (EPGAA) and Vitro
commented that the regulations established by the 2012 rule included an
oversight in defining the baseline Tts (the metric used to evaluate
thermal reflectivity of glass). EPGAA commented that there was an
omission in the case of trucks, where the regulations do allow the use
of privacy glass in locations other than the windshield and the front
doors. The commenter discussed that the reference baseline glass for
trucks, SUVs, and CUVs should have already included privacy glass for
some of the rearward windows. In response, EPA recognized when the
thermal credit program was finalized in 2012 that some of the vehicles
within the reference fleet upon which the credits were based were
already composed of vehicles with this type of thermal reflective
glass. However, the agency found it difficult to estimate what portion
of the fleet contained privacy glass and what the Tts rating was for
privacy glass across the fleet. Because of this lack of specificity in
the fleet composition and glass ratings, the agencies determined that
the most appropriate approach was to allow credit for any glass meeting
the finalized Tts requirements, and the total thermal cap was designed
to account for this and other uncertainties.
Ford and others commented that thermal control technology credit
caps should be implemented on a fleet average basis rather than on a
``per VIN'' basis. These commenters argued that the per VIN basis
creates a reporting burden that is misaligned with the current
reporting structure and creates program complexity and unnecessary
workload. In response, EPA continues to believe that applying the
thermal control credit cap on a per vehicle (per VIN) basis is
appropriate due to the synergistic effects among these technologies.
The CO2 reduction potential of applying thermal control technologies is
limited within any given vehicle. The program has been implemented in
this manner since MY2014, and manufacturers have in fact reported the
necessary information to generate thermal control credits.
Gentherm, GM, MEMA, and The ITB Group commented that cooled seats
should be added to the menu based on the approved GM off-cycle credits
application and NREL study. EPA and NHTSA are not adding cooled seat
technology to the menu because the agencies have received data from
only a single manufacturer. By contrast, for the technologies EPA and
NHTSA are adding to the menu in this final rule, the agencies have
assessed data from multiple manufacturers. EPA notes however that the
streamlining provisions being finalized in this action should
facilitate other manufacturers in being able to apply for off-cycle
credits by using GM's methodology.
Finally, on October 1, 2018, EPA proposed a technical correction
separate from the SAFE Vehicles rulemaking for
[[Page 25243]]
the off-cycle credits pathway based on 5-cycle testing (83 FR 49344).
This proposal would correct an error in the regulations established as
part of the 2012 final rule. Some commenters expressed their support
for the correction as part of their SAFE Vehicles rule comments. EPA
notes that this correction continues to be part of a separate
rulemaking and is not being addressed in the SAFE Vehicles final rule.
c) Final Decisions on the 2016 Alliance/Global Petition
(1) Retroactive A/C and Off-Cycle CAFE Adjustments
In 2016, the Alliance and Global submitted a petition for
rulemaking, which included requests that: (1) NHTSA allow retroactive
credits for A/C and off-cycle incentives for MYs 2012 to 2016; and (2)
NHTSA and EPA revisit the average A/C efficiency benefit calculated by
EPA applicable to MYs 2012 through 2016. The Alliance/Global argued
that A/C efficiency improvements were not properly acknowledged in the
CAFE program, and that manufacturers had exceeded the A/C efficiency
improvements estimated by the agencies. The petitioners requested that
EPA also amend its regulations such that manufacturers would be
entitled to additional A/C efficiency improvement benefits
retroactively. The petitioners also argued that NHTSA incorrectly
stated the agency had taken off-cycle adjustments into consideration
when setting standards for MYs 2017 through 2025, but not for MYs 2010-
2016. The Alliance/Global further contended that because neither NHTSA
nor EPA considered off-cycle adjustments in formulating the stringency
of the MY 2012-2016 standards, NHTSA should retroactively grant
manufacturers off-cycle adjustments for those model years as EPA did.
Doing so, they said, would maintain consistency between the agencies'
programs.
Of the two agencies, EPA was the first to establish an off-cycle
technology program. For MYs 2012 through 2016, EPA allowed
manufacturers to request off-cycle credits for ``technologies that
achieve [CO2] reductions that are not reflected on current
test procedures . . .'' \3417\ In the subsequent MY 2017 and later
rulemaking, NHTSA joined EPA and included an off-cycle program for CAFE
compliance. The Alliance/Global petition cited a statement in the MYs
2012-2016 final rule as affirmation that NHTSA took off-cycle
adjustments into account in formulating the MYs 2012-2016 stringencies,
and therefore should allow manufacturers to earn off-cycle benefits in
model years that have already passed.
---------------------------------------------------------------------------
\3417\ 75 FR 25341, 25344 (May 7, 2010). EPA had also provided
an option for manufacturers to claim ``early'' off-cycle credits in
the 2009-2011 time frame.
---------------------------------------------------------------------------
In the NPRM, NHTSA tentatively decided to retain the structure of
the existing A/C efficiency program and not extend it to MYs 2010
through 2016. For the rulemaking for MYs 2012 through 2016, NHTSA
determined it was unable to consider improvements manufacturers made to
passenger car A/C efficiency in calculating CAFE
compliance.3418 3419 However, EPA did consider passenger car
improvements to A/C efficiency for that timeframe. To allow
manufacturers to build one fleet that complied with both EPA and NHTSA
standards, the CAFE and CO2 standards were offset to account
for the differences borne out of A/C efficiency improvements.
Specifically, the agencies converted EPA's grams/mile standards to
NHTSA mpg (CAFE) standards. EPA then estimated the average amount of
improvement manufacturers were expected to earn via improved A/C
efficiency. From there, NHTSA took EPA's converted mpg standard and
subtracted the average improvement attributable to improvement in A/C
efficiency. NHTSA set its standard at this level to allow manufacturers
to comply with both standards with similar levels of technology.\3420\
---------------------------------------------------------------------------
\3418\ At that time, NHTSA stated ``[m]odernizing the passenger
car test procedures, or even providing similar credits, would not be
possible under EPCA as currently written.'' 75 FR 25557 (May 7,
2010).
\3419\ 74 FR 49700 (Sept. 28, 2009).
\3420\ Id.
---------------------------------------------------------------------------
Likewise, EPA tentatively decided in the NPRM not to modify its
regulations to change the way to account for A/C efficiency
improvements. EPA believed this was appropriate as manufacturers
decided what fuel economy-improving technologies to apply to vehicles
based on the standards as finalized in 2010.\3421\ This included
deciding whether to apply traditional tailpipe technologies, A/C
efficiency improvements, or both. Granting A/C efficiency adjustments
to manufacturers retroactively could result in arbitrarily varying
levels of adjustments granted to manufacturers, similar to the
Alliance/Global request regarding retroactive off-cycle adjustments.
Thus, the existing A/C efficiency improvement structure for MYs 2010
through 2016 would remain unchanged.
---------------------------------------------------------------------------
\3421\ In the MY 2017 and later rulemaking, NHTSA reaffirmed its
position it would not extend A/C efficiency improvement benefits to
earlier model years. 77 FR 62720 (Oct. 15, 2012).
---------------------------------------------------------------------------
NHTSA also tentatively decided manufacturers should not be granted
retroactive off-cycle adjustments for MYs 2010 through 2016, and
presented a number of clarifications to justify the denial. In
particular, Alliance/Global pointed to a general statement where NHTSA,
while discussing consideration of ``the effect of other motor vehicle
standards of the Government on fuel economy,'' stated that that
rulemaking resulted in consistent standards across the program.\3422\
The Alliance/Global petition took this statement as a blanket assertion
that NHTSA's consideration of all ``relevant technologies'' included
off-cycle technologies. To the contrary, as quoted above, NHTSA
explicitly stated it had not considered these off-cycle
technologies.\3423\
---------------------------------------------------------------------------
\3422\ Id.
\3423\ Likewise, EPA stated it had not considered off-cycle
technologies in finalizing the MYs 2012-2016 rule. ``Because these
technologies are not nearly so well developed and understood, EPA is
not prepared to consider them in assessing the stringency of the
CO2 standards.'' Id. at 25438.
---------------------------------------------------------------------------
The fact that NHTSA had not taken off-cycle adjustments into
consideration in setting its MYs 2012-2016 standards makes granting the
Alliance/Global request inappropriate. Doing so could result in a
question as to whether the MY 2012-2016 standards were maximum feasible
under 49 U.S.C. 32902(b)(2)(B). If NHTSA had considered industry's
ability to earn off-cycle adjustments--an incentive that allows
manufacturers to utilize technologies other than those that were being
modeled as part of NHTSA's analysis--the agency might have concluded
more stringent standards were maximum feasible. Additionally, granting
off-cycle adjustments to manufacturers retroactively raises questions
of equity. NHTSA issued its MYs 2012-2016 standards without an off-
cycle program, and manufacturers had no reason to anticipate that NHTSA
would allow the use off-cycle technologies to meet fuel economy
standards. Therefore, manufacturers made fuel economy compliance
decisions with the expectation that they would have to meet fuel
economy standards using on-cycle technologies. Generating off-cycle
adjustments retroactively would arbitrarily reward some (and
potentially disadvantage other) manufacturers for compliance decisions
they made without the knowledge such technologies would be eligible for
NHTSA's off-cycle program. Thus, NHTSA tentatively decided to deny
Alliance/Global's request for retroactive off-cycle adjustments.
[[Page 25244]]
It is worth noting that in the MYs 2017 and later rulemaking, NHTSA
and EPA did include off-cycle technologies in establishing the
stringency of the standards. As Alliance/Global noted, NHTSA and EPA
limited their consideration to stop-start and active aerodynamic
features because of limited technical information on these
technologies.\3424\ At that time, the agencies stated they ``have
virtually no data on the cost, development time necessary,
manufacturability, etc. [sic] of these technologies. The agencies thus
cannot project that some of these technologies are feasible within the
2017-2025 timeframe.'' \3425\
---------------------------------------------------------------------------
\3424\ Alliance/Global Petition at 7.
\3425\ Draft Joint Technical Support Document: Rulemaking for
2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards and
Corporate Average Fuel Economy Standards (November 2011), p. 5-57.
---------------------------------------------------------------------------
As described above, NHTSA first allowed manufacturers to generate
off-cycle technology fuel consumption improvement values equivalent to
CO2 off-cycle credits in MY 2017.\3426\ In finalizing the
rule covering MYs 2017 and later, NHTSA declined to retroactively
extend its off-cycle program to apply to model years 2012 through
2016,\3427\ explaining ``NHTSA did not take [off-cycle credits] into
account when adopting the CAFE standards for those model years. As
such, extending the credit program to the CAFE program for those model
years would not be appropriate.'' \3428\
---------------------------------------------------------------------------
\3426\ 77 FR 62840 (Oct. 15, 2012).
\3427\ See id.; EPA decided to extend provisions from its MY
2017 and later off-cycle program to the 2012-2016 model years.
\3428\ Id.
---------------------------------------------------------------------------
In the NPRM, NHTSA and EPA sought any further comments on the
tentative denials of the retroactive requests in the Alliance/Global.
The Auto Alliance and Fiat Chrysler provided additional comments on the
tentative denial of the petition requests from the Alliance/Global. The
commenters cited that the widening gap between the regulatory standards
and actual industry-wide new vehicle average fuel economy that has
become evident since 2016, despite the growing use of improvement
``credits'' from various flexibility mechanisms, such as off-cycle
technology credits, mobile air conditioner efficiency credits, mobile
air conditioner refrigerant leak reduction credits and credits from
electrified vehicles.\3429\ The commenters believe that applying
retroactive credits for the new flexibilities for MYs 2012 to 2016 can
address the current compliance deficiencies.
---------------------------------------------------------------------------
\3429\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073;
Fiat Chrysler, Detailed Comments, NHTSA-2018-0067-11943.
---------------------------------------------------------------------------
Upon consideration of the issue, NHTSA is finalizing its decision
to deny any retroactive off-cycle adjustments in the CAFE program for
MYs 2012-2016. As mentioned in the NPRM, NHTSA is concerned about the
negative impact of allowing retroactive credits, which could undermine
the stringency of the MYs 2012-2016 standards. EPA is finalizing its
decision not to modify its regulations to change the benefits for A/C
efficiency improvements. As mentioned by EPA, the current approach
creates uniformity and objectivity in determining A/C efficiency
benefits. Consequently, because EPA is maintaining the current A/C
determination methodology and NHTSA already considered those A/C
adjustments in its MYs 2012-2016 CAFE standards, NHTSA is also
finalizing its decisions in this rule to deny any retroactive A/C
adjustments in the CAFE program for MYs 2012-2016.
(2) Petition Requests on A/C Efficiency and Off-Cycle Program
Administration
As discussed above, NHTSA and EPA jointly administer the off-cycle
program. The 2016 Alliance/Global petition requested that EPA and NHTSA
make various adjustments to the off-cycle program; specifically, the
petitioners requested that the agencies should:
re-affirm that technologies meeting the stated definitions
are entitled to the off-cycle credit at the values stated in the
regulation;
re-acknowledge that technologies shown to generate more
emissions reductions than the pre-approved amount are entitled to
additional credit;
confirm that technologies not in the null vehicle set but
which are demonstrated to provide emissions reductions benefits
constitute off-cycle credits; and
modify the off-cycle program to account for unanticipated
delays in the approval process by providing that applications based on
the 5-cycle methodology are to be deemed approved if not acted upon by
the agencies within a specified timeframe (for instance 90 days),
subject to any subsequent review of accuracy and good faith.\3430\
---------------------------------------------------------------------------
\3430\ Alliance/Global Petition at 20.
---------------------------------------------------------------------------
With respect to Alliance/Global's request regarding off-cycle
technologies that demonstrate emissions reductions greater than what is
allowable from the menu, this final rule retains that capability. As
was the case for MYs 2017-2021, a manufacturer may still apply for
FCIVs and CO2 credits beyond the values listed on the menu,
provided the manufacturer demonstrates the CO2 and fuel
economy improvement.\3431\ This includes the two-alternative processes
for demonstrating CO2 reductions and fuel economy
improvement for gaining benefits using either the 5-cycle or
alternative approval methodologies.\3432\
---------------------------------------------------------------------------
\3431\ 77 FR 62837 (Oct. 15, 2012).
\3432\ 40 CFR 86.1869-12.
---------------------------------------------------------------------------
The agencies have considered Alliance/Global's requests to
streamline aspects of the A/C efficiency and off-cycle programs in
response to the issues outlined above. Among other things, Alliance/
Global requested that the agencies consider providing for a default
acceptance of petitions for off-cycle credits after a specified period
of time, provided that all required information has been provided, to
accelerate the processing of off-cycle credit requests. While the
agencies agree with the merits of A/C efficiency and off-cycle
programmatic improvements, there are significant concerns with the
concept of approving petition requests by default because such requests
may not address program issues like uncertainty in quantifying program
benefits, or general program administration.
Based on its consideration of the issues raised by the Alliance/
Global, EPA has adopted in this final rule new processes for
streamlining the compliance mechanisms for approving off-cycle and
applications as discussed in the preceding section.
(3) Other EPA Responses to Alliance Requests
One issue raised in the Alliance/Global Automakers June 2016
petition (item 6 titled ``Refrain from Imposing Unnecessary
Restrictions on the Use of Credits'') for EPA's consideration concerns
how credits are managed within the CO2 program. The Alliance
and Global Automakers suggested that EPA allow more flexibility in
using credits generated under the various credit programs such as air
conditioning or off-cycle credits by allowing them to be carried
forward or back independently. Under this approach, a manufacturer
would be allowed, for example, to carry their air conditioning credits
back to cover a previous deficit while running a deficit in a current
model year. The Alliance referred to this petition request in their
comments, noting they believe the request ``remains pertinent in the
context of this rulemaking.''
In response, EPA did not raise this issue or any related
programmatic changes in the proposal and therefore
[[Page 25245]]
these comments are not within the scope of the rulemaking. EPA notes
the GHG and CAFE programs are harmonized on the aggregation of credits.
The automakers' petition also requested that EPA correct the
multiplier equation in the regulations so that manufacturers may
generate the intended number of credits (item 8, ``Correct the
Multiplier for BEVs, PHEVs, FCVs, and CNGs''). This request concerns an
error in the regulations established in the 2012 Final Rule that
results in manufacturers generating fewer than intended for MY 2017-
2021 vehicles in some cases. In October 2018, in response to this
petition request, EPA issued a proposed rule separate from the SAFE
Vehicles NPRM to correct the error in the previously established
regulations. EPA will continue to address this issue and related
comments in that separate rulemaking. CAFE does not include multiplier
credits and therefore this is not a harmonization issue.
4. Specialty Vehicles With Low Mileage (SVLM)
In response to the NPRM, Volkswagen submitted comments seeking to
adopt a new flexibility for specialty vehicles with low mileage
(SVLM).\3433\ The flexibility would apply to specialty vehicles
produced at low volumes and produced for infrequent use. They argued
these specialty vehicles do not approach the vehicle miles traveled of
typical vehicles. They requested that NHTSA and EPA allow the SVLM
flexibility for vehicles that demonstrate limited predicted driving
use. The flexibility would allot each manufacturer a limited annual
production of 5,000 SVLM vehicles. It was also proposed that, within
this limited product volume, each SVLM would retain its footprint
derived performance target (per model type), but would utilize a
modified VMT for determining any credits or debits associated with the
performance of these vehicles within the manufacturer's fleet.
---------------------------------------------------------------------------
\3433\ Volkswagen, Detailed Comments, NHTSA-2017-0069-0583.
---------------------------------------------------------------------------
The agencies have considered the request from Volkswagen for
credits or debits and fuel economy adjustments for SVLM vehicles and
are denying the request. NHTSA notes that Congress prescribed
alternative (reduced) CAFE standards for low-volume manufacturers,
codified in 49 CFR part 525. Low-volume manufacturers' vehicles are
often high-end sports cars and are not typically driven by their owners
for long distances. Congress limited this exemption under the CAFE
program to manufacturers of fewer than 10,000 passenger
automobiles.\3434\ EPA has a similar program for smallvolume
manufacturers which are defined as manufacturers with average sales for
the three most recent consecutive model years of less than 5,000
vehicles.\3435\ The flexibility proposed by Volkswagen would presumably
be in addition to these existing provisions, but Volkswagen does not
identify a source of authority for it. The agencies also have a number
of questions about how specifically a SVLM concept might be
implemented, such as whether every manufacturer would simply identify
the 5,000 vehicles with the lowest projected VMT or lowest fuel economy
and therefore qualify for credits for 5,000 vehicles every model year,
or whether there should be additional criteria for vehicles to be
included. The NPRM did not seek comment on a SVLM concept and the
agencies did not receive other comments on the requested program.
Therefore, the agencies are not adopting the SVLM concept suggested by
Volkswagen.
---------------------------------------------------------------------------
\3434\ 49 U.S.C. 32902(d)(1).
\3435\ 40 CFR 86.1818-12(g).
---------------------------------------------------------------------------
E. CO2 and CAFE Compliance Issues Not Addressed in the NPRM
1. CO2 and CAFE Adjustments for 5-Cycle Testing
EPA and NHTSA received several comments requesting that the
agencies revise current CAFE test procedures to use EPA's 5-cycle test
procedures in place of the 2-cycle test procedures that have been
largely unchanged since the inception of the CAFE program, or offset
measured 2-cycle test fuel economy and CO2 emissions for
CO2 and CAFE compliance. Walter Kreucher commented ``some
technologies (Hybrid Electric) have penalties on the road that are not
reflected on the tests used to determine CAFE compliance. . . . If the
Agencies want to provide adjustment factors for A/C and other `Off-
Cycle' conditions it must do so in both the positive and negative
direction'' (sic).\3436\ AVE commented that the agencies should use 5-
cycle procedures rather than 2-cycle procedures, arguing that the 5-
cycle model better demonstrates real-world driving conditions and would
lead to a more simplified credit allocation system.\3437\ BorgWarner
echoed those comments, stating that the 5-cycle test is more accurate
than the 2-cycle test and would reduce the need for credit
adjustments.\3438\ Jeremy Michalek commented that the fuel economy
values the public sees reflected on vehicles for purchase (e.g., on the
Monroney label or in new car advertising) is calculated from the 5-
cycle test; updating the 2-cycle test to capture more of the vehicle's
fuel efficiency factors would allow for better consistency and a more
accurate fuel efficiency measure.\3439\ The Auto Alliance proposed that
the EPA revise its methodology for calculating off-cycle improvements
when using the 5-cycle methodology by subtracting the 2-cycle benefit
from the 5-cycle benefit to ensure credits are calculated
properly.\3440\
---------------------------------------------------------------------------
\3436\ Walter Kreucher, Detailed Comments, NHTSA-2018-0067-0444.
\3437\ AVE, Detailed Comments, NHTSA-2018-0067-11696.
\3438\ BorgWarner, Detailed Comments, NHTSA-2018-0067-11895.
\3439\ Jeremy Michalek, et al., Detailed Comments, NHTSA-2018-
0067-11903.
\3440\ Auto Alliance, Detailed Comments, NHTSA-2018-0067-12073.
---------------------------------------------------------------------------
The NPRM did not seek comment on revising compliance test
procedures to use 5-cycle test procedures in place of 2-cycle test
procedures, either entirely or broadly. Such a change would require
extensive assessment and analysis to consider how changes could be
implemented and what standards might be maximum feasible for CAFE and
appropriate and reasonable for CO2 for new test procedures.
There has been no analysis conducted to estimate the impacts of such a
change on the levels of the standards. Therefore, making these
requested changes is outside the scope of this rulemaking.
2. National Zero Emissions Vehicle Concept
Although the agencies did not discuss or request comment on a
National Zero Emissions Vehicle (NZEV) program concept, several
organizations commented on that topic. Some discussed ideas from a task
force that was formed by the governors of nine States who signed a
memorandum of understanding (MOU) committing to undertake joint
cooperative actions to build a robust market for ZEVs under their
individual state programs. Collectively, these States have committed to
having at least 3.3 million ZEVs operating on their roadways by 2025.
ZEVs include battery-electric vehicles (BEVs), plug-in hybrid electric
vehicles (PHEVs), and hydrogen fuel-cell electric vehicles (FCEVs).
Comments on an NZEV concept were received from General Motors, CARB,
Edison Electric Institute, Honda, NCAT, Workhorse Group, and Volvo.
[[Page 25246]]
General Motors offered comments supporting an NZEV program, stating
that it continues to expect California to be the leader of the EV
market but hopes a national effort will be put forth, making the U.S. a
global leader in EV technology development and deployment. \3441\
General Motors stated it believes an NZEV program would further U.S.
national security interests, make the U.S. more competitive with China,
which already has an NZEV program, and reduce U.S. dependence on
foreign petroleum. General Motors requested that EPA incentivize EV
deployment, including providing credits for autonomous EVs and EVs that
are used in rideshare programs.\3442\ General Motors outlined their
proposed NZEV program which would include increasing ZEV requirements
annually, establishing credit banks for manufacturers based on national
ZEV sales, and ZEV multipliers for vehicles over 5,250 lbs., autonomous
vehicles using EV, and EVs in rideshare programs. General Motors also
proposed that requirements would be revisited if EV battery cell were
not available at the costs Argonne National Lab forecasts by 2025.
General Motors also suggested implementing a Zero Emissions Task Force
that would promote complementary policies. General Motors acknowledged
that the NZEV program would have to be subject to acceleration or delay
depending on how quickly technologies are incentivized like battery
cost.
---------------------------------------------------------------------------
\3441\ General Motors, Detailed Comments, NHTSA-2018-0067-11858.
\3442\ General Motors, Detailed Comments, NHTSA-2018-0067-11858.
---------------------------------------------------------------------------
CARB recommended a national ZEV multiplier, stating that a national
incentive would help ensure ZEVs and PHEVs were being produced for sale
beyond the ten States that have ZEV programs.\3443\ The Edison Electric
Institute supported increasing stringency of fuel economy and
CO2 standards and incorporating policies from ZEV States to
create a ``One National Program.'' \3444\ Workhorse Group commented
that a national ZEV mandate, where agencies progressively increase the
mandated percentage of electric vehicles in every fleet, merits serious
consideration by the agencies. They contended that an NZEV would have
to work with the current State ZEV mandates and not preempt the
progress already made.\3445\ Volvo, and Honda were proponents of
incorporating ZEV standards into a national program. Volvo requested
nationwide credits for ZEVs since there are 40 States without ZEV
mandates.\3446\ Honda mentioned that incorporating California's ZEV
credits into the national program would reduce compliance costs for
manufacturers while incentivizing technological development.\3447\ NCAT
recommended in their comment that EPA provide enhanced credits for EVs,
PHEVs, and FCVs that are more stringent than California (and other
States) ZEV mandates, making the national program credits
``additional'' to state ZEV compliance credits.\3448\
---------------------------------------------------------------------------
\3443\ CARB, Detailed Comments, NHTSA-2018-0067-11873.
\3444\ Edison Electric Institute, Detailed Comments, NHTSA-2018-
0067-11918.
\3445\ Workhorse Group, Detailed Comments, NHTSA-2018-0067-
12215.
\3446\ Volvo, Detailed Comments, NHTSA-2018-0067-12036.
\3447\ Honda, Detailed Comments, NHTSA-2018-0067-11818.
\3448\ NCAT, Detailed Comments, NHTSA-2018-0067-11969.
---------------------------------------------------------------------------
Northeast States for Coordinated Air Use Management (NESCAUM)
commented that an aggressive reduction in emissions will not occur
without national ZEV standards which will drive development of advanced
clean vehicle technologies.\3449\
---------------------------------------------------------------------------
\3449\ NESCAUM, Detailed Comments, NHTSA-2018-0067-11691.
---------------------------------------------------------------------------
The NPRM did not propose or request comment on an NZEV concept or
program, as such, and establishing such a program would be outside the
scope of this rulemaking. Such a concept would require thorough
assessment and full rulemaking notice and comment. There are also
policy questions about what the appropriate level of potential
incentives should be and whether certain technologies should receive
greater incentives than other technologies, and if so, on what basis
and by what amounts. Also, for the CAFE program, incentives for
technologies are almost entirely prescribed by statute, and there are
questions about how the CAFE program could implement an NZEV program in
alignment with EPCA and EISA. Therefore, the agencies have decided not
to implement an NZEV program as part of this rulemaking.
3. CO2 In-Use Requirements
Current in-use regulations outlined in 86.1845-04 provide
flexibility in determining the applicable number of test vehicles per
test group. Each large volume manufacturer is provided the flexibility
to employ small volume sampling allowances for a limited number of
total annual production units. In response to the NPRM, Volkswagen is
proposing to modify 86.1845-04 to provide a separate, additional small
volume sampling allowance allocation of annual production volume for a
manufacturer's plug-in hybrid vehicles. This additional allowance would
only be applicable through the 2025 model year and would only be
applicable to CO2 testing requirements under the in use
regulations.
The basis for this flexibility is rooted in the continuing
evolution and development of traction drive battery cell chemistries
and battery management systems. This ongoing development is aimed at
continuously improving such features as energy density, power, cost,
and durability. As such, the engineering processes for understanding
and quantifying long-term performance are still developing and subject
to reevaluation as new chemistries are examined. Manufacturers such as
Volkswagen have allocated significant capital in battery testing to
ensure that performance is maintained for consumers and are also
providing longer term battery warranty provisions.
Volkswagen believes that the targeted flexibility will provide
additional time to continue evaluating chemistries and reduce
administrative testing burdens for a very limited production allocation
per manufacturer. This provision will further support plug-in hybrid
technology development and deployment. Volkswagen proposed modifying
86.1845-04 table SO4-07 footnote 2, to read as follows:
\2\ Total annual production of groups eligible for testing under
small volume sampling plan is capped at a maximum of 14,999 vehicle 49
or 50 state annual sales, or a maximum of 4,500 vehicle California only
sales per model year, per large volume manufacturer. Through model year
2025, a separate total annual production of plug-in hybrid electric
vehicle groups shall be eligible for testing under small volume
sampling plan as described above. This allocation shall only be
applicable to exhaust CO2 emission standards under this
subpart.\3450\
---------------------------------------------------------------------------
\3450\ See EPA-HQ-OAR-2018-0283-5689-A1, p.32.
---------------------------------------------------------------------------
Regarding comments from VW on CO2 in-use requirements,
EPA did not consider the change recommended by VW in the proposal and
is not finalizing such a change. EPA believes the current program
provides enough flexibility. EPA's general approach for this final rule
is also to avoid providing incentives or other unique flexibilities to
specific technologies.
[[Page 25247]]
F. Medium and Heavy-Duty Fuel Efficiency Technical Amendments
NHTSA proposed in the NPRM to make minor technical revisions to
correct typographical mistakes and improper references adopted in the
agency's 2016 Phase 2 medium- and heavy-duty fuel efficiency
rule.\3451\ The proposed changes were as follows:
---------------------------------------------------------------------------
\3451\ 81 FR 73478 (Oct. 25, 2016).
---------------------------------------------------------------------------
NHTSA heavy-duty vehicles and engine fuel consumption
credit equations. In each credit equation in 49 CFR 535.7, the minus-
sign in each multiplication factor was omitted in the final version of
the rule sent to the Federal Register. For example, the credit equation
in Part 535.7(b)(1) should be specified as, Total MY Fleet FCC
(gallons) = (Std-Act) x (Volume) x (UL) x (10-2) instead of (102), as
currently exists. NHTSA proposed to correct these omissions.
The CO2 to gasoline conversion factor: In 49
CFR 535.6(a)(4)(ii) and (d)(5)(ii), NHTSA provides the methodology and
equations for converting the CO2 FELs/FCLs for heavy-duty
pickups and vans (gram per mile) and for engines (grams per hp-hr) to
their gallon-of-gasoline equivalence. In each equation, NHTSA proposed
to correct the conversion factor to 8,887 grams per gallon of gasoline
fuel instead of a factor of 8,877 as currently specified.
Curb weight definition: In 49 CFR 523.2, the reference in
the definition for curb weight is incorrect. NHTSA proposed to correct
the definition to incorporate a reference to 40 CFR 86.1803 instead of
49 CFR 571.3.
No public comments were received in response to NHTSA's proposed
technical corrections. Therefore, NHTSA is finalizing these amendments
and incorporating them into its heavy-duty regulations.
X. Regulatory Notices and Analyses
A. Executive Order 12866, Executive Order 13563
Executive Order 12866, ``Regulatory Planning and Review'' (58 FR
51735, Oct. 4, 1993), as amended by Executive Order 13563, ``Improving
Regulation and Regulatory Review'' (76 FR 3821, Jan. 21, 2011),
provides for making determinations whether a regulatory action is
``significant'' and therefore subject to the Office of Management and
Budget (OMB) review and to the requirements of the Executive Order. One
comment requested that the agencies provide ``a far more robust cost/
benefit analysis as required by Executive Order (E.O.) 12866 and Office
of Management and Budget Circular A-4.'' \3452\ The NPRM and this final
rule satisfy the requirements of Executive Order 12866, ``Regulatory
Planning and Review'' (58 FR 51735, Oct. 4, 1993), as amended by
Executive Order 13563, ``Improving Regulation and Regulatory Review''
(76 FR 3821, Jan. 21, 2011). Under these Executive Orders, this action
is an ``economically significant regulatory action'' because it is
likely to have an annual effect on the economy of $100 million or more.
Accordingly, EPA and NHTSA submitted this action to the OMB for review
and any changes made in response to OMB recommendations have been
documented in the docket for this action. The benefits and costs of
this proposal are described above and in the Final Regulatory Impact
Analysis (FRIA), which is located in the docket and on the agencies'
websites.
---------------------------------------------------------------------------
\3452\ See Anonymous Comment, Docket No. EPA-HQ-OAR-2018-0283-
3896, at 4-5 (footnote and citation omitted). As an example, the
comment critiqued the NPRM's discussion of the ``diminishing
returns'' of fuel economy benefits, alleging that the discussion
``is not backed by reference to data or studies regarding how this
conclusion was made.'' Id. at 5. Contrary to the comment's
allegation, the conclusion is supported by the analysis from U.S.
Energy Information Administration's (EIA's) Annual Energy Outlook
(AEO) that was cited in the discussion. Id. As noted in the NPRM,
the EIA--the statistical and analytical agency within the U.S.
Department of Energy (DOE)--is the nation's premier source of energy
information, and every fuel economy rulemaking since 2002 (and every
joint CAFE and CO2 rulemaking since 2009) has applied
fuel price projections from EIA's AEO. Id. at 42992 n.24.
---------------------------------------------------------------------------
B. DOT Regulatory Policies and Procedures
The rule is also significant within the meaning of the Department
of Transportation's Regulatory Policies and Procedures. The benefits
and costs of this proposal are described above and in the FRIA, which
is located in the docket and on NHTSA's website.
C. Executive Order 13771 (Reducing Regulation and Controlling
Regulatory Costs)
This rule is an E.O. 13771 deregulatory action. Per OMB Memorandum
M-17-21, because this rule is deregulatory, it is not required to be
offset by two deregulatory actions, as one comment suggested.\3453\
---------------------------------------------------------------------------
\3453\ Anonymous Comment, Docket No. EPA-HQ-OAR-2018-0283-3896,
at 8.
---------------------------------------------------------------------------
D. Executive Order 13211 (Energy Effects)
Executive Order 13211 applies to any rule that: (1) is determined
to be economically significant as defined under E.O. 12866, and is
likely to have a significant adverse effect on the supply,
distribution, or use of energy; or (2) that is designated by the
Administrator of the Office of Information and Regulatory Affairs as a
significant energy action. If the regulatory action meets either
criterion, the agencies must evaluate the adverse energy effects of the
rule and explain why the regulation is preferable to other potentially
effective and reasonably feasible alternatives considered.
The rule establishes passenger car and light truck fuel economy
standards and tailpipe carbon dioxide and related emissions standards.
An evaluation of energy effects of the action and reasonably feasible
alternatives considered is provided in NHTSA's EIS and in the FRIA. To
the extent that EPA's CO2 standards are substantially
related to fuel economy and, accordingly, petroleum consumption, the
EIS and FRIA analyses also provide an estimate of impacts of EPA's
rule.
E. Environmental Considerations
1. National Environmental Policy Act (NEPA)
Concurrently with this final rule, NHTSA is releasing a Final
Environmental Impact Statement (FEIS), pursuant to the National
Environmental Policy Act, 42 U.S.C. 4321-4347, and implementing
regulations issued by the Council on Environmental Quality (CEQ), 40
CFR part 1500, and NHTSA, 49 CFR part 520. NHTSA prepared the FEIS to
analyze and disclose the potential environmental impacts of the
proposed CAFE standards and a range of alternatives. The FEIS analyzes
direct, indirect, and cumulative impacts and analyzes impacts in
proportion to their significance. It describes potential environmental
impacts to a variety of resources, including fuel and energy use, air
quality, climate, land use and development, hazardous materials and
regulated wastes, historical and cultural resources, noise, and
environmental justice. The FEIS also describes how climate change
resulting from global carbon emissions (including CO2
emissions attributable to the U.S. light duty transportation sector
under the alternatives considered) could affect certain key natural and
human resources. Resource areas are assessed qualitatively and
quantitatively, as appropriate, in the FEIS.
Some commenters provided feedback on the ``flaws'' they identified
in the CAFE model, concluding that because it played a significant role
in modeling for the DEIS, the DEIS itself was flawed and
[[Page 25248]]
should be withdrawn and reissued.\3454\ The agencies address the
comments regarding the CAFE model above in this preamble and in the
FRIA. Ultimately, the findings on potential environmental impacts
presented in the FEIS are of the same level of intensity and
significance as those presented in the DEIS. While in some cases, the
directionality of potential air quality emissions changed, the overall
impact was generally small. NHTSA concludes that the CAFE model
results, as used in the FEIS, do not result in the FEIS providing
significant new information for the decisionmaker or the public
compared to the DEIS.\3455\ NHTSA therefore concludes that a
supplemental DEIS is not required.
---------------------------------------------------------------------------
\3454\ States of California, Connecticut, Delaware, Hawaii,
Iowa, Illinois, Maine, Maryland, Minnesota, North Carolina, New
Jersey, New Mexico, New York, Oregon, Rhode Island, Vermont, and
Washington; the Commonwealths of Massachusetts, Pennsylvania, and
Virginia; the District of Columbia; and the Cities of Los Angeles,
New York, Oakland, San Francisco, and San Jose (``California et.
al.--Detailed NEPA Comments''), Docket No. NHTSA-2017-0069-0625, at
6-11; Environmental Defense Fund, Docket No. NHTSA-2018-0067-11996,
at 3-4; and Center for Biological Diversity, et al., Docket No.
NHTSA-2018-0067-12123, at 19.
\3455\ 40 CFR 1502.9(c)(1)(ii).
---------------------------------------------------------------------------
NHTSA also performed a national-scale photochemical air quality
modeling and health benefit assessment for the FEIS; it is included as
Appendix E. The purpose of this assessment was to use air quality
modeling and health-related benefits analysis tools to examine the
potential air quality-related consequences of the alternatives
considered in its Draft Environmental Impact Statement (DEIS). In a
comment on the DEIS, the South Coast Air Quality Management District
stated that performing the photochemical modeling for the FEIS ``comes
too late for the public to be able to comment on that analysis,'' and
that the EIS must be recirculated to allow such public comment.\3456\
However, NHTSA publicly stated its intent to conduct the analysis as
part of the FEIS in its scoping notice published on July 26,
2017.\3457\ The agency noted that this approach was consistent with
past practice and resulted from the substantial time required to
complete such an analysis. NHTSA also announced that, due to the
substantial lead time required, the analysis would be based on the
modeling of the alternatives presented in the DEIS, not of the
alternatives as presented in the FEIS. NHTSA received no public
comments in response to the scoping notice addressing this analytical
approach, and the agency proceeded accordingly. Furthermore, while
photochemical modeling provides spatial and temporal detail for
estimating changes in ambient levels of air pollutants and their
associated impacts on human health and welfare, the analysis affirms
the estimates that appear in the EIS and does not provide significant
new information for the decisionmaker or the public. For these reasons,
NHTSA concludes that inclusion of the photochemical modeling and health
benefit assessment in the FEIS is appropriate, and recirculation of the
EIS is not required.
---------------------------------------------------------------------------
\3456\ South Coast Air Quality Management District, Docket No.
NHTSA-2018-0067-5666, at 10. See also North Carolina Department of
Environmental Quality, Docket No. NHTSA-2018-0067-12025, at 35-37.
\3457\ NHTSA, ``Notice of Intent to Prepare an Environmental
Impact Statement for Model Year 2022-2025 Corporate Average Fuel
Economy Standards,'' 82 FR 34740, 34743 fn. 15 (Jul. 26, 2017).
---------------------------------------------------------------------------
NHTSA has considered the information contained in the FEIS in
making the final decision described in this final rule.\3458\ This
preamble and final rule constitute NHTSA's Record of Decision (ROD)
under 40 CFR 1505.2 for its promulgation of CAFE standards for MYs
2021-2026. NHTSA has authority to issue its FEIS and ROD simultaneously
pursuant to 49 U.S.C. 304a(b) and U.S. Department of Transportation,
Office of Transportation Policy, Guidance on the Use of Combined Final
Environmental Impact Statements/Records of Decision and Errata Sheets
in National Environmental Policy Act Reviews (April 25, 2019).\3459\
NHTSA has determined that neither the statutory criteria nor
practicability considerations preclude simultaneous issuance.
---------------------------------------------------------------------------
\3458\ The FEIS is available for review in the public docket for
this action and in Docket No. NHTSA-2017-0069.
\3459\ The guidance is available at https://www.transportation.gov/sites/dot.gov/files/docs/mission/transportation-policy/permittingcenter/337371/feis-rod-guidance-final-04302019.pdf.
---------------------------------------------------------------------------
As required by the CEQ regulations,\3460\ this final rule (as the
ROD) sets forth the following: (1) The agency's decision (Sections V
and VIII above); (2) alternatives considered by NHTSA in reaching its
decision, including the environmentally preferable alternative
(Sections V, VII, and VIII above); (3) the factors balanced by NHTSA in
making its decision, including essential considerations of national
policy (Section VIII.B above); (4) how these factors and considerations
entered into its decision (Section VIII.B above); and (5) the agency's
preferences among alternatives based on relevant factors, including
economic and technical considerations and agency statutory missions
(Section VIII.B.4 above). This section also briefly addresses
mitigation\3461\ and whether all practicable means to avoid or minimize
environmental harm from the alternative selected have been adopted.
---------------------------------------------------------------------------
\3460\ 40 CFR 1505.2.
\3461\ See 40 CFR 1508.20(b) (``Mitigation includes . . . (b)
Minimizing impacts by limiting the degree or magnitude of the action
and its implementation. . .'')
---------------------------------------------------------------------------
In the DEIS and in the FEIS, the agency identified a Preferred
Alternative. In the DEIS, the Preferred Alternative was identified as
Alternative 1 (0.0 Percent Annual Increase in Fuel Economy, MYs 2021-
2026), which were the standards the agency proposed in the NPRM. In the
FEIS, the Preferred Alternative was identified as Alternative 3 (1.5
Percent Annual Increase in Fuel Economy, MYs 2021-2026). As the FEIS
notes, under the Preferred Alternative, on an mpg basis, the estimated
annual increases in the average required fuel economy levels between
MYs 2021 and 2026 is 1.5 percent for both passenger cars and light
trucks.\3462\ After carefully reviewing and analyzing all of the
information in the public record, the FEIS, and comments submitted on
the DEIS and the NPRM, NHTSA has decided to finalize the Preferred
Alternative described in the FEIS for the reasons described in this
ROD.
---------------------------------------------------------------------------
\3462\ Because the standards are attribute-based, average
required fuel economy levels, and therefore rates of increase in
those average mpg values, depend on the future composition of the
fleet, which is uncertain and subject to change. When NHTSA
describes a percent increase in stringency, we mean in terms of
shifts in the footprint functions that form the basis for the actual
CAFE standards (as in, on a gallon per mile basis, the CAFE
standards change by a given percentage from one model year to the
next).
---------------------------------------------------------------------------
NHTSA has considered environmental considerations as part of its
balancing of the statutory factors to set maximum feasible fuel economy
standards. As a result, the agency has limited the degree or magnitude
of the action as appropriate in light of its statutory
responsibilities. NHTSA's authority to promulgate fuel economy
standards does not allow it to regulate criteria polluants from
vehicles or refineries, nor can NHTSA regulate other factors affecting
those emissions, such as driving habits. Consequently, NHTSA must set
CAFE standards but is unable to take further steps to mitigate the
impacts of these standards. Chapter 9 of the FEIS provides a further
discussion of mitigation measures in the context of NEPA.
One commenter states that NHTSA, at a minimum, ``must include a
thorough discussion of all reasonable mitigation measures and detail
the appropriate agencies that could implement such
[[Page 25249]]
measures.'' \3463\ As examples, the commenter listed: ``creating tax
breaks for transit and biking, expanding transportation demand
management programs for federal employees, implementing a social
marketing campaign regarding VMT reduction, increasing dedicated
funding for transit and active modes, requiring VMT as a performance
measure for federal funding, and providing NEPA guidance on evaluating
VMT impacts of federal projects.'' Each of the examples listed is
beyond NHTSA's statutory authority. Furthermore, documenting the myriad
measures that could reduce VMT or address criteria pollutant or carbon
dioxide emissions would provide no added benefit to the decisionmaker
or the public. Each of these actions requires their own extensive cost-
benefit anlaysis, are beyond the purview of this action, and are beyond
the legal responsibility of NHTSA. NHTSA concludes that the commenter's
request is beyond the bounds of NEPA's ``rule of reason.'' \3464\
---------------------------------------------------------------------------
\3463\ California et. al.--Detailed NEPA Comments, Docket No.
NHTSA-2017-0069-0625, at 31.
\3464\ Dep't of Transp. v. Pub. Citizen, 541 U.S. 752, 772
(2004).
---------------------------------------------------------------------------
Another commenter disputes NHTSA's conclusion that it lacks
statutory authority to mitigate the impacts of its CAFE standards.
Specifically, the commenter cites to its very authority to set fuel
economy standards: ``It is axiomatic that fuel efficiency standards set
at levels of the No Action Alternative or at more stringent levels
would eliminate the additional pollution created by the proposed
freeze.'' \3465\ This, however, mischaracterizes mitigation as nothing
more than a choice among alternatives. NHTSA is already considering a
range of reasonable alternatives and has concluded that alternatives
more stringent than the No Action Alternative are beyond reasonable.
Furthermore, NHTSA disputes that more stringent fuel economy standards
will axiomatically lead to lower levels of criteria pollutant
emissions. In fact, because of the rebound effect, higher levels of
stringency may result in higher VMT, which may result in criteria
pollutant emission increases.
---------------------------------------------------------------------------
\3465\ Center for Biological Diversity, et al., Docket No.
NHTSA-2018-0067-12123, at 55-56.
---------------------------------------------------------------------------
The North Carolina Department of Environmental Quality commented
that the proposed changes to the CAFE standards could undermine the
integrity of many of the assumptions in various NEPA documents across
the United States, in part because EPA required the use of the
MOVES2014 model (or a subsequent revision) for transportation
conformity determinations.\3466\ That version of MOVES incorporates
CAFE and CO2 standards based on the agencies' actions in
2012 and does not reflect the actions being finalized in this rule. The
implication of the commenter's assertion, however, is that neither
NHTSA nor EPA could take any regulatory action regarding CAFE or
CO2 standards, regardless of whether such action was to
increase or decrease such standards. Clearly neither agency can be
paralyzed from undertaking its statutory obligations because of the
independent NEPA obligations related to other ongoing Federal actions.
For those actions, responsible officials may need to assess whether
this final rule triggers the need for a supplemental NEPA document.
However, it is not unique for Federal agencies to take actions or for
new information to become available that affects the underlying inputs
in models, such as EPA's MOVES model, on which NEPA and conformity
analyses rely. Over time, those models will be updated to reflect these
actions and information. EPA is responsible for approving the
availability of models for the use in State implementation plans and
transportation conformity analyses. EPA will evaluate and address, as
appropriate, the impact of this action on future SIP approval actions.
Currently approved emission factor models remain approved for SIPs and
transportation conformity analyses, and EPA will work with DOT on the
appropriate implementation of Federal requirements based on current and
available information.
---------------------------------------------------------------------------
\3466\ North Carolina Department of Environmental Quality,
Docket No. NHTSA-2018-0067-12025, at 37. See also Southern
Environmental Law Center, EPA-HQ-OAR-2018-0283-0887, at 2-4.
---------------------------------------------------------------------------
2. Clean Air Act (CAA) as Applied to NHTSA's Action
The CAA (42 U.S.C.[thinsp]7401 et seq.) is the primary Federal
legislation that addresses air quality. Under the authority of the CAA
and subsequent amendments, EPA has established National Ambient Air
Quality Standards (NAAQS) for six criteria pollutants, which are
specifically identified pollutants that have recognized adverse effects
on ambient air quality and that can accumulate in the atmosphere as a
result of human activity. EPA is required to review each NAAQS every
five years and to revise those standards as may be appropriate
considering new scientific information.
The air quality of a geographic region is usually assessed by
comparing the levels of criteria air pollutants found in the ambient
air to the levels established by the NAAQS (taking into account, as
well, the other elements of a NAAQS: averaging time, form, and
indicator). Concentrations of criteria pollutants within the air mass
of a region are measured in parts of a pollutant per million parts
(ppm) of air or in micrograms of a pollutant per cubic meter ([mu]g/
m\3\) of air present in repeated air samples taken at designated
monitoring locations using specified types of monitors. These ambient
concentrations of each criteria pollutant are compared to the levels,
averaging time, and form specified by the NAAQS in order to assess
whether the region's air quality is in attainment with the NAAQS.
When the measured concentrations of a criteria pollutant within a
geographic region are below those permitted by the NAAQS, EPA
designates the region as an attainment area for that pollutant, while
regions where concentrations of criteria pollutants exceed Federal
standards are called nonattainment areas. Former nonattainment areas
that are now in compliance with the NAAQS are designated as maintenance
areas. Each State with a nonattainment area is required to develop and
implement a State Implementation Plan (SIP) documenting how the region
will reach attainment levels within time periods specified in the CAA.
For maintenance areas, the SIP must document how the State intends to
maintain compliance with the NAAQS. When EPA revises a NAAQS, each
State must revise its SIP to address how it plans to attain the new
standard.
No Federal agency may ``engage in, support in any way or provide
financial assistance for, license or permit, or approve'' any activity
that does not ``conform'' to a SIP or Federal Implementation Plan after
EPA has approved or promulgated it.\3467\ Further, no Federal agency
may ``approve, accept, or fund'' any transportation plan, program, or
project developed pursuant to title 23 or chapter 53 of title 49,
U.S.C., unless the plan, program, or project has been found to
``conform'' to any applicable implementation plan in effect.\3468\ The
purpose of these conformity requirements is to ensure that Federally
sponsored or conducted activities do not interfere with meeting the
emissions targets in SIPs, do not cause or contribute to new violations
of the NAAQS, and do not impede the ability of a State to attain or
maintain the NAAQS or delay any interim milestones. EPA has issued two
sets of
[[Page 25250]]
regulations to implement the conformity requirements:
---------------------------------------------------------------------------
\3467\ 42 U.S.C. 7506(c)(1).
\3468\ 42 U.S.C. 7506(c)(2).
---------------------------------------------------------------------------
(1) The Transportation Conformity Rule\3469\ applies to
transportation plans, programs, and projects that are developed,
funded, or approved under title 23 or chapter 53 of title 49, U.S.C.
---------------------------------------------------------------------------
\3469\ 40 CFR part 51, subpart T, and part 93, subpart A.
---------------------------------------------------------------------------
(2) The General Conformity Rule\3470\ applies to all other federal
actions not covered under transportation conformity. The General
Conformity Rule establishes emissions thresholds, or de minimis levels,
for use in evaluating the conformity of an action that results in
emissions increases.\3471\ If the net increases of direct and indirect
emissions are lower than these thresholds, then the project is presumed
to conform and no further conformity evaluation is required. If the net
increases of direct and indirect emissions exceed any of these
thresholds, and the action is not otherwise exempt, then a conformity
determination is required. The conformity determination can entail air
quality modeling studies, consultation with EPA and state air quality
agencies, and commitments to revise the SIP or to implement measures to
mitigate air quality impacts.
---------------------------------------------------------------------------
\3470\ 40 CFR part 51, subpart W, and part 93, subpart B.
\3471\ 40 CFR 93.153(b).
---------------------------------------------------------------------------
The CAFE standards and associated program activities are not
developed, funded, or approved under title 23 or chapter 53 of title
49, United States Code. Accordingly, this action and associated program
activities are not subject to the Transportation Conformity Rule. Under
the General Conformity Rule, a conformity determination is required
where a Federal action would result in total direct and indirect
emissions of a criteria pollutant or precursor originating in
nonattainment or maintenance areas equaling or exceeding the rates
specified in 40 CFR 93.153(b)(1) and (2). As explained below, NHTSA's
action results in neither direct nor indirect emissions as defined in
40 CFR 93.152.
The General Conformity Rule defines direct emissions as ``those
emissions of a criteria pollutant or its precursors that are caused or
initiated by the Federal action and originate in a nonattainment or
maintenance area and occur at the same time and place as the action and
are reasonably foreseeable.'' \3472\ Because NHTSA's action would set
fuel economy standards for light duty vehicles, it would cause no
direct emissions consistent with the meaning of the General Conformity
Rule.\3473\
---------------------------------------------------------------------------
\3472\ 40 CFR 93.152.
\3473\ Dep't of Transp. v. Pub. Citizen, 541 U.S. at 772
(``[T]he emissions from the Mexican trucks are not `direct' because
they will not occur at the same time or at the same place as the
promulgation of the regulations.''). NHTSA's action is to establish
fuel economy standards for MY 2021-2026 passenger car and light
trucks; any emissions increases would occur in a different place and
well after promulgation of the final rule.
---------------------------------------------------------------------------
Indirect emissions under the General Conformity Rule are ``those
emissions of a criteria pollutant or its precursors (1) That are caused
or initiated by the federal action and originate in the same
nonattainment or maintenance area but occur at a different time or
place as the action; (2) that are reasonably foreseeable; (3) that the
agency can practically control; and (4) for which the agency has
continuing program responsibility.'' \3474\ Each element of the
definition must be met to qualify as indirect emissions. NHTSA has
determined that, for purposes of general conformity, emissions that may
result from its final fuel economy standards would not be caused by
NHTSA's action, but rather would occur because of subsequent activities
the agency cannot practically control. ``[E]ven if a Federal licensing,
rulemaking, or other approving action is a required initial step for a
subsequent activity that causes emissions, such initial steps do not
mean that a Federal agency can practically control any resulting
emissions.'' \3475\
---------------------------------------------------------------------------
\3474\ 40 CFR 93.152.
\3475\ 40 CFR 93.152.
---------------------------------------------------------------------------
As the CAFE program uses performance-based standards, NHTSA cannot
control the technologies vehicle manufacturers use to improve the fuel
economy of passenger cars and light trucks. Furthermore, NHTSA cannot
control consumer purchasing (which affects average achieved fleetwide
fuel economy) and driving behavior (i.e., operation of motor vehicles,
as measured by VMT). It is the combination of fuel economy
technologies, consumer purchasing, and driving behavior that results in
criteria pollutant or precursor emissions. For purposes of analyzing
the environmental impacts of the alternatives considered here and under
NEPA, NHTSA has made assumptions regarding all of these factors. The
agency's FEIS predicts that increases in air toxic and criteria
pollutants would occur in some nonattainment areas under certain
alternatives. However, the standards and alternatives do not mandate
specific manufacturer decisions, consumer purchasing, or driver
behavior, and NHTSA cannot practically control any of them.\3476\
---------------------------------------------------------------------------
\3476\ See, e.g., Dep't of Transp. v. Pub. Citizen, 541 U.S.
752, 772-73 (2004); S. Coast Air Quality Mgmt. Dist. v. Fed. Energy
Regulatory Comm'n, 621 F.3d 1085, 1101 (9th Cir. 2010).
---------------------------------------------------------------------------
In addition, NHTSA does not have the statutory authority to control
the actual VMT by drivers. As the extent of emissions is directly
dependent on the operation of motor vehicles, changes in any emissions
that result from NHTSA's CAFE standards are not changes the agency can
practically control or for which the agency has continuing program
responsibility. Therefore, the final CAFE standards and alternative
standards considered by NHTSA would not cause indirect emissions under
the General Conformity Rule, and a general conformity determination is
not required.
As this analysis was presented in the NPRM, some commenters
disagreed with NHTSA's conclusion. One commenter cited two reasons for
concluding that the General Conformity Rule applies to NHTSA's
action.\3477\ First, the commenter argues that NHTSA used
``inappropriate modeling'' in its analysis. However, this is irrelevant
to the agency's analysis, which is based on the Federal regulations and
the applicable case law. Second, the commenter asserts that NHTSA
``cannot have it both ways'' by alleging that it cannot control the
technologies that automobile manufacturers would use or consumer
purchasing behavior, yet justifies its rulemakings based on consumer
purchasing and emissions implications.3478 3479 The
rulemaking analysis presents a feasible pathway for manufacturers to
comply with the rules, based on a series of assumptions about consumer
behavior; it is not sufficiently foreseeable to trigger application of
the General Conformity Rule. Furthermore, NHTSA cannot directly control
these behaviors, and the chain of causation is too attenuated to be
responsible for the resulting emissions. Another commenter stated that
NHTSA has continuing
[[Page 25251]]
program responsibility for motor vehicle criteria pollutant emissions
because it ``retain[s] authority to revise [its] standards in a way
that affects future emission levels.'' \3480\ However, NHTSA disagrees
with this assertion. First, the agency does not have statutory
authority to regulate criteria pollutant emissions from motor vehicles.
Second, the fact that NHTSA could establish CAFE standards for
separate, future motor vehicles does not establish continuing program
responsibility over emissions that could result from the vehicles
regulated by this action.
---------------------------------------------------------------------------
\3477\ California et. al.--Detailed NEPA Comments, Docket No.
NHTSA-2017-0069-0625, at 21-22.
\3478\ The commenter also quotes CBD v. NHTSA, 538 F.3d at 1217,
for the proposition that NHTSA's regulations are the proximate cause
of the emissions because they allow particular fuel economy levels
that ``translate directly into particular tailpipe emissions.''
However, that quote was referencing carbon dioxide emissions, which
are predictable based on fuel used. NHTSA can directly regulate fuel
economy for passenger cars and light trucks. On the other hand,
criteria pollutant emissions are more significantly impacted by VMT,
technology choices, and other factors that are not directly within
the control of NHTSA.
\3479\ See also Joint Submission from the States of California
et al. and the Cities of Oakland et al., Docket No. NHTSA-2018-0067-
11735, at 35.
\3480\ Id.
---------------------------------------------------------------------------
NHTSA and EPA further discuss their obligations under the General
Conformity Rule, and further address comments received, in Section
VI.D.3 above.
3. National Historic Preservation Act (NHPA)
The NHPA (54 U.S.C. 300101 et seq.) sets forth government policy
and procedures regarding ``historic properties''--that is, districts,
sites, buildings, structures, and objects included on or eligible for
the National Register of Historic Places. Section 106 of the NHPA
requires Federal agencies to ``take into account'' the effects of their
actions on historic properties.\3481\ In the NPRM, the agencies
concluded that the NHPA is not applicable to this rulemaking because
the promulgation of CAFE and CO2 emissions standards for
light duty vehicles is not the type of activity that has the potential
to cause effects on historic properties.
---------------------------------------------------------------------------
\3481\ Section 106 is now codified at 54 U.S.C. 306108.
Implementing regulations for the Section 106 process are located at
36 CFR part 800.
---------------------------------------------------------------------------
Two commenters wrote that ``[c]limate change and air pollution
imperil historic properties throughout the country via direct
degradation, sea level rise, fire, flood, and other forms of harm.''
Therefore, the commenters concluded that NHTSA and EPA must consult
with the relevant Federal and State authorities and fully disclose any
impacts to historic properties.\3482\ However, as this final rule
establishes CAFE and CO2 standards that increase each year
for MYs 2021-2026, this action will result in reductions in climate
change-related impacts and most air pollutants compared to the absence
of regulation. Furthermore, any impacts to particular historic
properties that could be related to emissions changes associated with
this rulemaking are not reasonably certain to occur, would be de
minimis in their level of impact if they did occur, and are too
attenuated to be attributed directly to this action. (See also Section
X.E.6 below.) There is no evidence that the changes in air pollution or
CO2 emissions associated with this rulemaking, in and of
themselves, would alter the characteristics of a historic property
qualifying it for inclusion in or eligibility for the National
Register.\3483\ Nevertheless, NHTSA includes a brief, qualitative
discussion of the impacts of the alternatives on historical and
cultural resources in Section 7.3 of the FEIS. For the foregoing
reasons, the agencies continue to conclude that any potential impacts
have been accounted for in the associated analyses of this rulemaking
and that no consultation is required under the NHPA.
---------------------------------------------------------------------------
\3482\ CARB, Docket No. NHTSA-2018-0067-11873, at 411;
California et. al.--Detailed NEPA Comments, Docket No. NHTSA-2017-
0069-0625, at 30.
\3483\ 36 CFR 800.16(i).
---------------------------------------------------------------------------
4. Fish and Wildlife Conservation Act (FWCA)
The FWCA (16 U.S.C. 2901 et seq.) provides financial and technical
assistance to States for the development, revision, and implementation
of conservation plans and programs for nongame fish and wildlife. In
addition, the Act encourages all Federal departments and agencies to
utilize their statutory and administrative authorities to conserve and
to promote conservation of nongame fish and wildlife and their
habitats. The agencies conclude that the FWCA is not applicable to this
final rule because this rulemaking does not involve the conservation of
nongame fish and wildlife and their habitats. NHTSA has, however,
conducted a qualitative review in its FEIS of the related direct,
indirect, and cumulative impacts, positive or negative, of the
alternatives on potentially affected resources, including nongame fish
and wildlife and their habitats.
5. Coastal Zone Management Act (CZMA)
The Coastal Zone Management Act (16 U.S.C. 1451 et seq.) provides
for the preservation, protection, development, and (where possible)
restoration and enhancement of the Nation's coastal zone resources.
Under the statute, States are provided with funds and technical
assistance in developing coastal zone management programs. Each
participating State must submit its program to the Secretary of
Commerce for approval. Once the program has been approved, any activity
of a Federal agency, either within or outside of the coastal zone, that
affects any land or water use or natural resource of the coastal zone
must be carried out in a manner that is consistent, to the maximum
extent practicable, with the enforceable policies of the State's
program.\3484\
---------------------------------------------------------------------------
\3484\ 16 U.S.C. 1456(c)(1)(A).
---------------------------------------------------------------------------
In the NPRM, the agencies concluded that the CZMA is not applicable
to this rulemaking because this rulemaking does not involve an activity
within, or outside of, the Nation's coastal zones that affects any land
or water use or natural resource of the coastal zone. CARB commented
that California's coast is vulnerable to sea level rise from climate
change and that the proposal would exacerbate that threat. Therefore,
the commenter claimed that the proposal violated California's policies
and obligations in its management program to preserve, protect, and
enhance its coastline.\3485\ However, in its FEIS, NHTSA estimates that
the sea-level rise in 2100 associated with Alternative 1 (0 percent
annual average increase for both passenger cars and light trucks for
MYs 2021-2026), the least stringent alternative considered, would be
0.7 mm. Such a level is too small to have any meaningful impact on land
or water use or a natural resource of the coastal zone. Furthermore, as
this final rule establishes CAFE and CO2 standards that
increase each year for MYs 2021-2026, this action will result in
reductions in sea level rise resulting from climate change compared to
the absence of regulation. Therefore, the agencies continue to conclude
that the CZMA is not applicable to this rulemaking. NHTSA has, however,
conducted a qualitative review in its FEIS of the related direct,
indirect, and cumulative impacts, positive or negative, of the
alternatives on potentially affected resources, including coastal
zones.
---------------------------------------------------------------------------
\3485\ CARB, Docket No. NHTSA-2018-0067-11873, at 411.
---------------------------------------------------------------------------
6. Endangered Species Act (ESA)
Under Section 7(a)(2) of the Endangered Species Act (ESA), Federal
agencies must ensure that actions they authorize, fund, or carry out
are ``not likely to jeopardize the continued existence'' of any
Federally listed threatened or endangered species (collectively,
``listed species'') or result in the destruction or adverse
modification of the designated critical habitat of these species.\3486\
In general, if a Federal agency determines that an agency action may
affect a listed species or designated critical habitat, it must
initiate consultation with the
[[Page 25252]]
appropriate Service--the U.S. Fish and Wildlife Service (FWS) of the
Department of the Interior (DOI) and/or the National Oceanic and
Atmospheric Administration's National Marine Fisheries Service (NMFS)
of the Department of Commerce (together, ``the Services''), depending
on the species involved--in order to ensure that the action is not
likely to jeopardize the species or destroy or adversely modify
designated critical habitat.\3487\ Under this standard, the Federal
agency taking action evaluates the possible effects of its action and
determines whether to initiate consultation.\3488\
---------------------------------------------------------------------------
\3486\ 16 U.S.C. 1536(a)(2).
\3487\ See 50 CFR 402.14.
\3488\ See 50 CFR 402.14(a) (``Each Federal agency shall review
its actions at the earliest possible time to determine whether any
action may affect listed species or critical habitat.'').
---------------------------------------------------------------------------
In the NPRM, the agencies noted that they had considered the
effects of the proposed standards and alternatives in light of
applicable ESA regulations, case law, and guidance to determine what,
if any, impact there might be to listed species or designated critical
habitat. The agencies also considered the discussion in the DEIS, where
NHTSA incorporated by reference its response to a public comment on
page 9-101 of the MY 2017-2025 CAFE Standards Final EIS.\3489\ Based on
that assessment, the agencies determined that the actions of setting
CAFE and CO2 emissions standards did not require
consultation under Section 7(a)(2) of the ESA. Accordingly, the
agencies wrote that they had concluded their review of this action
under Section 7 of the ESA.
---------------------------------------------------------------------------
\3489\ For the final rule for MY 2017 and beyond CAFE standards,
NHTSA concluded that a Section 7(a)(2) consultation was not required
because any potential for a specific impact on particular listed
species and their habitats associated with emission changes achieved
by that rulemaking were too uncertain and remote to trigger the
threshold for such a consultation. In the Draft EIS, NHTSA wrote
that this conclusion, based on the discussion and analysis cited,
applied equally to the current rulemaking.
---------------------------------------------------------------------------
Several commenters disagreed with the agencies' assessment. In
general, commenters stated that the agencies' proposed action would
increase emissions of CO2 and criteria air pollutants (e.g.,
nitrogen oxide [NOX] and sulfur dioxide
[SO2]\3490\), that these emissions would have direct or
indirect (i.e., through climate change) impacts on listed species and
critical habitats, that the threshold for a finding of ``may affect''
is extremely low, and that the agencies therefore have a duty to
consult with the Services under the ESA.\3491\
---------------------------------------------------------------------------
\3490\ In fact, in Section 4.2.1.1 of NHTSA's FEIS, the agency
reports that any of the action alternatives would result in
decreased emissions of sulfur dioxide in 2025, 2035, and 2050
compared to the No Action Alternative.
\3491\ See Center for Biological Diversity, Earthjustice,
Natural Resources Defense Council, and Sierra Club, Docket Nos.
NHTSA-2017-0069-0605 and NHTSA-2018-0067-12127; Center for
Biological Diversity, Sierra Club, and Public Citizen, Inc., Docket
No. NHTSA-2018-0067-12378; Center for Biological Diversity,
Earthjustice, Environmental Law and Policy Center, Natural Resources
Defense Council, Public Citizen, Inc., Safe Climate Campaign, Sierra
Club, Southern Environmental Law Center, and Union of Concerned
Scientists, Docket No. NHTSA-2018-0067-12123, at 69; States of
California, Connecticut, Delaware, Hawaii, Iowa, Illinois, Maine,
Maryland, Minnesota, New Jersey, New Mexico, New York, North
Carolina, Oregon, Rhode Island, Vermont, and Washington, the
Commonwealths of Massachusetts, Pennsylvania, and Virginia, the
District of Columbia, and the Cities of Los Angeles, New York,
Oakland, San Francisco, and San Jose, Docket Nos. NHTSA-2018-0067-
11735, at 47-48; and California Air Resources Board, Docket Nos.
NHTSA-2018-0067-11873, at 411.
---------------------------------------------------------------------------
In light of these comments, the agencies re-evaluated their
obligations under the ESA and applicable regulations, case law, and
guidance. Ultimately, for the following reasons, the agencies arrive at
the same conclusion. Although there is a general association between
the actions undertaken in this final rule and environmental impacts, as
described in this preamble and the FEIS, the agencies' actions result
in no effects on listed species or designated critical habitat and
therefore do not require consultation under Section 7(a)(2) of the ESA.
Furthermore, the agencies lack sufficient discretion or control to
bring these actions under the consultation requirement of the ESA. The
agencies' review under the ESA is concluded.
a) The Agencies' Actions Have No Effects on Listed Species or Critical
Habitat and Do Not Trigger ESA Consultation
Commenters have stated that CO2 and criteria air
pollutant emissions are relevant to Section 7(a)(2) consultation
because of the potential impacts of climate change or the pollutants
themselves on listed species or critical habitat. The agencies have
considered the potential impacts of this action to listed species or
designated critical habitat of these species and conclude that any such
impacts cannot be attributed to the agencies' actions (e.g., they are
too uncertain and attenuated). Because the agencies conclude there are
``no effects,'' Section 7(a)(2) consultation is not required. The
agencies base this conclusion both on the language of the Section
7(a)(2) implementing regulations and on the long history of actions and
guidance provided by DOI.
The Section 7(a)(2) implementing regulations require consultation
if a Federal agency determines its action ``may affect'' listed species
or critical habitat.\3492\ The recently revised regulations define
``effects of the action'' as ``all consequences to listed species or
critical habitat that are caused by the proposed action, including the
consequences of other activities that are caused by the proposed
action. A consequence is caused by the proposed action if it would not
occur but for the proposed action and it is reasonably certain to
occur.'' \3493\ The revised definition made explicit a ``but for'' test
and the concept of ``reasonably certain to occur'' for all
effects.\3494\ However, in the preamble to the final rule, the Services
emphasized that the ``but for'' test and ``reasonably certain to
occur'' are not new or heightened standards.\3495\ In this context,
```but for' causation means that the consequence in question would not
occur if the proposed action did not go forward . . . . In other words,
if the agency fails to take the proposed action and the activity would
still occur, there is no `but for' causation. In that event, the
activity would not be considered an effect of the action under
consultation.'' \3496\
---------------------------------------------------------------------------
\3492\ 50 CFR 402.14(a). The Services recently issued a final
rule revising the regulations governing the ESA Section 7
consultation process. 84 FR 44976 (Aug. 27, 2019). The effective
date of the new regulations was subsequently delayed to October 28,
2019. 84 FR 50333 (Sep. 25, 2019). As discussed in the text that
follows, the agencies believe that their conclusion would be the
same under both the current and prior regulations.
\3493\ 50 CFR 402.02 (emphasis added), as amended by 84 FR
44976, 45016 (Aug. 27, 2019).
\3494\ The Services' prior regulations defined ``effects of the
action'' in relevant part as ``the direct and indirect effects of an
action on the species or critical habitat, together with the effects
of other activities that are interrelated or interdependent with
that action, that will be added to the environmental baseline.'' 50
CFR 402.02 (as in effect prior to Oct. 28, 2019). Indirect effects
were defined as ``those that are caused by the proposed action and
are later in time, but still are reasonably certain to occur.'' Id.
\3495\ 84 FR at 44977 (``As discussed in the proposed rule, the
Services have applied the `but for' test to determine causation for
decades. That is, we have looked at the consequences of an action
and used the causation standard of `but for' plus an element of
foreseeability (i.e., reasonably certain to occur) to determine
whether the consequence was caused by the action under
consultation.'').
\3496\ Id. We note that as the Services do not consider this to
be a change in their longstanding application of the ESA, this
interpretation applies equally under the prior regulations (which
were effective through October 28, 2019, and the current
regulations.
---------------------------------------------------------------------------
The revised ESA regulations also provide a framework for
determining whether consequences are caused by a proposed action and
are therefore ``effects'' that may trigger consultation. The
regulations provide in part:
[[Page 25253]]
To be considered an effect of a proposed action, a consequence
must be caused by the proposed action (i.e., the consequence would
not occur but for the proposed action and is reasonably certain to
occur). A conclusion of reasonably certain to occur must be based on
clear and substantial information, using the best scientific and
commercial data available. Considerations for determining that a
consequence to the species or critical habitat is not caused by the
proposed action include, but are not limited to:
(1) The consequence is so remote in time from the action under
consultation that it is not reasonably certain to occur; or
(2) The consequence is so geographically remote from the
immediate area involved in the action that it is not reasonably
certain to occur; or
(3) The consequence is only reached through a lengthy causal
chain that involves so many steps as to make the consequence not
reasonably certain to occur.\3497\
---------------------------------------------------------------------------
\3497\ 50 CFR 402.17(b).
The regulations go on to make clear that the action agency must factor
these considerations into its assessments of potential effects.\3498\
---------------------------------------------------------------------------
\3498\ 50 CFR 402.17(c) (``Required consideration. The
provisions in paragraphs (a) and (b) of this section must be
considered by the action agency and the Services.'').
DOI, the agency charged with co-administering the ESA, previously
evaluated whether CO2 emissions associated with a specific
proposed Federal action triggered ESA Section 7(a)(2) consultation. The
agencies have reviewed the long history of actions and guidance
provided by DOI. To that point, the agencies incorporate by reference
Appendix G of the MY 2012-2016 CAFE standards EIS.\3499\ That analysis
relied on the significant legal and technical analysis undertaken by
FWS and DOI. Specifically, NHTSA looked at the history of the Polar
Bear Special Rule and several guidance memoranda provided by FWS and
the U.S. Geological Survey. Ultimately, DOI concluded that a causal
link could not be made between CO2 emissions associated with
a proposed Federal action and specific effects on listed species;
therefore, no Section 7(a)(2) consultation would be required.
---------------------------------------------------------------------------
\3499\ Available on NHTSA's Corporate Average Fuel Economy
website at https://one.nhtsa.gov/Laws-&-Regulations/CAFE-%E2%80%93-Fuel-Economy/Final-EIS-for-CAFE-Passenger-Cars-and-Light-Trucks,-Model-Years-2012%E2%80%932016.
---------------------------------------------------------------------------
Subsequent to the publication of that Appendix, a court vacated the
Polar Bear Special Rule on NEPA grounds, though it upheld the ESA
analysis as having a rational basis.\3500\ FWS then issued a revised
Final Special Rule for the Polar Bear.\3501\ In that final rule, FWS
provided that for ESA Section 7, the determination of whether
consultation is triggered is narrow and focused on the discrete effect
of the proposed agency action. FWS wrote, ``[T]he consultation
requirement is triggered only if there is a causal connection between
the proposed action and a discernible effect to the species or critical
habitat that is reasonably certain to occur. One must be able to
`connect the dots' between an effect of a proposed action and an impact
to the species and there must be a reasonable certainty that the effect
will occur.'' \3502\ The statement in the revised Final Special Rule is
consistent with the prior guidance published by FWS and remains valid
today.\3503\ Likewise, the current regulations identify remoteness in
time, geography, and the causal chain as factors to be considered in
assessing whether a consequence is ``reasonably certain to occur.'' If
the consequence is not reasonably certain to occur, it is not an
``effect of a proposed action'' and does not trigger the consultation
requirement.
---------------------------------------------------------------------------
\3500\ In re: Polar Bear Endangered Species Act Listing and
Section 4(D) Rule Litigation, 818 F.Supp.2d 214 (D.D.C. Oct. 17,
2011).
\3501\ 78 FR 11766 (Feb. 20, 2013).
\3502\ 78 FR at 11784-11785.
\3503\ See DOI Solicitor's Opinion No. M-37017, ``Guidance on
the Applicability of the Endangered Species Act Consultation
Requirements to Proposed Actions Involving the Emissions of
Greenhouse Gases'' (Oct. 3, 2008).
---------------------------------------------------------------------------
The agencies' actions establishing CAFE and CO2
standards for passenger cars and light trucks do not directly affect
listed species or critical habitat. The regulations promulgated by the
agencies are used to calculate average standards for manufacturers
based on the vehicles they produce for sale in the United States. Any
potential effects of this action on listed species or designated
critical habitat would be a result of changes to CO2 or air
pollutant emissions that are caused by the individual choices of
manufacturers in producing these vehicles and of consumers in
purchasing and operating those vehicles. The agencies are not
requiring, authorizing, funding, or carrying out the operation of motor
vehicles (i.e., the proximate cause of downstream emissions), the
production or refining of fuel (i.e., a proximate cause of upstream
emissions),\3504\ the use of any land that is critical habitat for any
purpose, or the taking of any listed species or other activity that may
affect any listed species. Ultimately, the relevant decisions that
result in emissions are taken by third parties, and any on-the-ground
activities to implement and carry out those decisions are undertaken by
such third parties. These decisions are influenced by a complex series
of market factors that, though influenced by the agencies' actions,
independently could result in the same series of decisions by consumers
that commenters attribute to the agencies' actions (such as increased
VMT and therefore increased emissions). This complex and lengthy chain
of causality, which is highly dependent on market factors and therefore
uncertain, leads the agencies to conclude that the resulting impacts of
their actions to listed species or critical habitat do not satisfy the
``but for'' test or are ``reasonably certain to occur.''
---------------------------------------------------------------------------
\3504\ The agencies note that upstream emissions sources, such
as oil extraction sites and fuel refineries, remain subject to the
ESA. As future non-federal activities become reasonably certain,
Section 7 and/or other sections of the ESA may provide protection
for listed species and designated critical habitats. For example,
new oil exploration or extraction activity may result in permitting
or construction activities that would trigger consultation or other
activities for the protection of listed species or designated
critical habitat, as impacts may be more direct and more certain to
occur.
---------------------------------------------------------------------------
With regard to climate change, EPA and NHTSA are not able to make a
causal link for purposes of Section 7(a)(2) that would ``connect the
dots'' between their actions, vehicle emissions from motor vehicles
affected by their actions, climate change, and particular impacts to
listed species or critical habitats. The agencies' actions are to set
standards that are effectively footprint curves, which are used as part
of a complex calculation based on the vehicles produced by
manufacturers for sale in the United States to determine a corporate
average standard for each manufacturer. This approach, dictated by the
Federal statute, gives manufacturers significant discretion to design,
produce, and sell motor vehicles to meet consumer demand. Because
manufacturers could choose to produce more vehicles with larger
footprints (and therefore less stringent standards), fleet-average
CO2 emissions could increase to some extent year-over-year
independently of where the agencies set standards. Or the opposite may
be true, and a shift in consumer preferences could lead to increased
production of vehicles with smaller footprints (and therefore more
stringent standards), resulting in overall declines in CO2
emissions in the future compared to what the agencies are forecasting.
Importantly, consumers not only choose which vehicles to purchase
across a range of available fuel economies, they also choose how much
to operate those vehicles (and therefore the quantity of fuel used and
CO2 emitted)
[[Page 25254]]
independently of any action undertaken by the
agencies.3505 3506
---------------------------------------------------------------------------
\3505\ While VMT is affected by the cost of driving associated
with fuel economy (i.e., the rebound effect), it is also affected by
several market factors, such as economic conditions, that are far
beyond the agencies' control and arguably have a greater influence
than this rulemaking.
\3506\ The fact that overall CO2 emissions are
influenced so heavily by consumer preferences and behavior further
supports the agencies' conclusion that impacts are not ``reasonably
certain to occur.''
---------------------------------------------------------------------------
Even with so many third parties in the causal chain making
independent choices influenced by independent factors, the mechanics of
climate change further break the chain of causality between the
agencies' actions and specific effects on listed species or designated
critical habitat. Climate change is a global phenomenon, impacted by
greenhouse gas emissions that could occur anywhere throughout the
world. As these gases accumulate in the atmosphere, radiative forcing
increases, resulting in various potential impacts to the global climate
system (e.g., warming temperatures, droughts, and changes in ocean pH)
over long time scales. These changes could directly or indirectly
impact listed species and/or designated critical habitat over time.
Although this is a simplified explanation of a complex phenomenon
subject to a significant degree of scientific study, it illustrates
that the potential climate change-related consequences of this
rulemaking on listed species and designated critical habitat are not
``reasonably certain to occur'' under any of the three tests in the ESA
regulations and listed above. Not only are the consequences to listed
species or designated critical habitat geographically and temporally
remote from the emissions that result from regulated vehicles, the
chain of causality is simply too lengthy and complex. Because impacts
to listed species and designated critical habitat result from climate
shifts that, in and of themselves, result from the accumulation over
time of greenhouse gas emissions from anywhere in the world, there is
simply no way to ``connect the dots'' between the emissions from a
regulated vehicle and those impacts. While the potential impacts of
climate change have been well-documented, there is no degree of
certainty that this action (as distinct from any other source of
CO2 emissions) would be the cause of any particular impact
to listed species or critical habitats. Because greenhouse gas
emissions continue to occur from other sectors within the U.S. and from
other sources globally, there is simply no scientific way to apportion
any impact to a listed species or designated critical habitat to the
agencies' actions.\3507\
---------------------------------------------------------------------------
\3507\ See 50 CFR 402.17(b) (``A conclusion of reasonably
certain to occur must be based on clear and substantial information,
using the best scientific and commercial data available.'')
---------------------------------------------------------------------------
One comment to the NPRM documented the potential impacts of climate
change on Federally protected species and included a five-page table of
species listed during 2006 to 2015 for which the commenters claim
climate change was a listing factor.\3508\ This conflates the
requirements of ESA Section 4 (governing ESA listing) and ESA Section 7
(addressing the obligations of Federal agencies). Section 4 requires
FWS or NMFS to assess all threats to species regardless of the origin
of those threats. 16 U.S.C. 1533(a)(1). In contrast, the focus of
Section 7(a)(2) is narrower and requires agencies to assess only
effects on species that are attributable to the specific agency action.
16 U.S.C. 1536(a)(2). That climate change was considered as a factor in
a determination to list a species does not speak to the separate
inquiry of whether the specific agency action is impacting a listed
species. Here, the agencies believe this comment inappropriately
attributes the entire issue of climate change, including all
CO2 emissions no matter which sector generated them, to
NHTSA and EPA's actions. In fact, NHTSA and EPA's actions would have
only very small impacts on climate attributes, such as average
temperatures, precipitation, and sea-level rise. The likelihood that
these very small impacts, which are described above and in NHTSA's
FEIS, would jeopardize listed species or adversely modify designated
critical habitat is simply too remote to be cognizable under the ESA
consultation requirements.\3509\ The fact that the agencies would
exacerbate the impacts of climate change to a very small degree is not
enough to determine that impacts on listed species or designated
critical habitat are reasonably certain to occur.3510 3511
---------------------------------------------------------------------------
\3508\ Center for Biological Diversity, Sierra Club, and Public
Citizen, Inc., Docket No. NHTSA-2018-0067-12378, at 25-30.
\3509\ Ground Zero Center for Non-Violent Action v. U.S. Dept.
of Navy, 383 F.3d 1082 (2004).
\3510\ Such a broad interpretation of the ESA would ensnare
every Federal action that resulted in even an additional ounce of
additional carbon dioxide emissions into the Section 7(a)(2)
consultation process. See, e.g., 78 FR 11766, 11785 (Feb. 20, 2013)
(``Without the requirement of a causal connection between the action
under consultation and effects to species, literally every agency
action that contributes CO2 emissions to the atmosphere
would arguably result in consultation with respect to every listed
species that may be affected by climate change.'').
\3511\ The agencies also disagree that, for purposes of
compliance with the ESA, this action would exacerbate climate change
impacts on listed species or critical habitat. This final rule
establishes CAFE and CO2 standards that increase in
stringency on a year-by-year basis. While these standards are less
stringent than the standards considered and set forth in the 2012
rulemaking, the ESA does not serve as a one-way ratchet when
agencies use their inherent authority to reconsider decisions that
have not yet taken effect.
---------------------------------------------------------------------------
As noted above, for consultation to be required, there must exist a
sufficient nexus between the agency activity and the impact on listed
species that the ESA intends to avoid. The Services have defined that
nexus as ``but for'' causation. However, there is no ``but for''
causation associated with this final rule as the impacts of climate
change will occur regardless of this action. In fact, even if the
agencies were to set CAFE and CO2 standards at levels that
would eliminate all CO2 emissions from motor vehicles made
available for sale in the United States, the impacts of climate change
are still projected to occur due to emissions from other sectors in the
United States and other sources globally. Changes to tailpipe
greenhouse gas emissions or associated upstream emissions related to
this rulemaking and the alternatives considered would be very small
compared to global CO2 emissions, which would continue. The
agencies also note that because third parties (as described above)
undertake most of the decisions that result in emissions, increased
greenhouse gas emissions could occur regardless of the agencies'
actions in this final rule. This further demonstrates the lack of ``but
for'' causality in this case.
Criteria air pollutant emissions from passenger cars and light
trucks differ from greenhouse gas emissions in many ways. Most
significantly, because passenger cars and light trucks are subject to
gram-per-mile emissions standards for criteria pollutants, more fuel-
efficient (and, correspondingly, less CO2-intensive)
vehicles are not necessarily, from the standpoint of air quality,
``cleaner'' vehicles. Therefore, to the extent that CAFE and
CO2 standards lead to changes in overall quantities of
vehicular emissions that impact air quality, these are dominated by
induced changes in highway travel. Changes in overall fuel consumption
do lead to changes in emissions from ``upstream'' processes involved in
supplying fuel to vehicles. Depending on how total vehicular emissions
and total upstream emissions change in response to less stringent
standards, overall emissions could increase or decrease.
While small in magnitude, net impacts could also vary considerably
[[Page 25255]]
among different geographic areas depending on the locations of upstream
emission sources and where changes in highway travel occur. This is
important because of another significant difference between criteria
air pollutant emissions and greenhouse gas emissions: Criteria air
pollutant emissions are localized \3512\ whereas CO2
emissions contribute to global atmospheric concentrations and climate
change no matter where they occur. As reported in Section 4.1.1 of the
FEIS, concentrations of many air pollutants emitted from motor vehicles
are elevated in ambient air within approximately 1,000 to 2,000 feet of
major roadways. With meteorological conditions that tend to inhibit the
dispersion of emissions, concentrations of traffic-generated air
pollutants can be elevated for as much as about 8,500 feet downwind of
roads.3513 3514 But this means that impacts of criteria
pollutant emissions are dependent on where they occur, to a degree much
more significant than greenhouse gas emissions. Although the agencies
anticipate increased fuel use as a result of this final rule (compared
to the standards described in the 2012 final rule),\3515\ NHTSA and EPA
have no way to know with reasonable certainty where additional fuel
extraction and refining will occur. The agencies also cannot calculate
with reasonable certainty where changes in highway travel will occur,
as those impacts may not be uniform across the country. In fact,
changes in land use patterns could exacerbate or reduce criteria
pollutant emissions in any particular area, and such local changes are
more uncertain. Therefore, even with the best scientific and commercial
data available, the agencies cannot draw conclusions on impacts on
particular listed species or designated critical habitat.
---------------------------------------------------------------------------
\3512\ Criteria pollutant emissions contribute to local,
regional, cross-state, and cross-national air pollution. Ultimately,
however, the physical distance impacted by the pollutants is much
smaller than for CO2 emissions, which affect the global
atmosphere.
\3513\ Hu, S., S. Fruin, K. Kozawa, S. Mara, S.E. Paulson, and
A.M. Winer. A Wide Area of Air Pollutant Impact Downwind of a
Freeway during Pre-sunrise Hours. Atmospheric Environment.
43(16):2541-49 (2009). doi:10.1016/j.atmosenv.2009.02.033.
\3514\ Hu, S., S.E. Paulson, S. Fruin, K. Kozawa, S. Mara, and
A.M. Winer. Observation of Elevated Air Pollutant Concentrations in
a Residential Neighborhood of Los Angeles California Using a Mobile
Platform. Atmospheric Environment. 51:311-319 (2012). doi:10.1016/
j.atmosenv.2011.12.055. Available at: http://europepmc.org/backend/ptpmcrender.fcgi?accid=PMC3755476&blobtype=pdf.
\3515\ Although, again, the agencies note that average fleet-
wide fuel economy is projected to improve under any of the
alternatives considered in this action.
---------------------------------------------------------------------------
In short, the impacts of CAFE and CO2 standards on
criteria pollutant emissions is indirect, and the impacts on air
quality at any particular location (such as where a listed species or
designated critical habitat is located) are more ambiguous than for
global atmospheric concentrations of CO2 over the long term.
Therefore, the agencies reach the same conclusion for criteria
pollutant emissions as for CO2 emissions and climate change.
For example, the causal chain between the agencies' actions and any
impacts to listed species or designated critical habitat is attenuated
by the fact that independent third parties must choose not only how
much to operate their motor vehicles, but where to operate those motor
vehicles as well. And the agencies cannot meaningfully conclude that
any impact to a listed species and designated critical habitat would be
caused by criteria pollutant emissions from the vehicles regulated by
this rule rather than by another source. Finally, the impacts on
criteria pollutant emissions as a result of this rule, especially in
light of other emissions sources besides the regulated vehicles, are
small\3516\ and the likelihood of jeopardy or the adverse modification
of designated critical habitat is too remote. Current modeling tools
available are not designed to trace fluctuations in ambient
concentration levels of criteria and toxic air pollutants to potential
impacts on particular endangered species. The agencies therefore cannot
conclude that impacts are ``reasonably certain to occur.'' \3517\
---------------------------------------------------------------------------
\3516\ For more information, see Chapter 4 of the FEIS.
\3517\ See 50 CFR 402.17 (``A conclusion of reasonably certain
to occur must be based on clear and substantial information, using
the best scientific and commercial data available'').
---------------------------------------------------------------------------
Finally, the agencies also note the potential uncertainty related
to changes in total air pollutant and CO2 emissions as a
result of the flexibilities in the CAFE and CO2 programs.
Both programs allow manufacturers to trade and apply credits that have
been earned from over-compliance in lieu of meeting the applicable
standards for a particular model year, and manufacturers may have
planned to rely on credits to comply with the standards for the model
years regulated by this action. This could offset any changes in
emissions that would result from the agencies' final decision.
Furthermore, NHTSA's CAFE program allows manufacturers to pay civil
penalties to cover any shortfall in compliance, further offsetting
potential improvements in fuel economy (and, therefore, changes in air
pollutant and CO2 emissions) that might have occurred under
the augural standards. The existence of these flexibilities further
supports the agencies' conclusion that they can establish neither ``but
for'' causation nor a reasonable certainty that impacts will occur on
listed species or designated critical habitat.
The agencies have considered this analysis and conclude that any
consequence to specific listed species or designated critical habitats
from climate change or other air pollutant emissions is too remote and
uncertain to be attributable to the agencies' actions here. These
consequences are not ``effects'' for purposes of consultation under
Section 7(a)(2). NHTSA and EPA therefore conclude that this final rule
has no effect on listed species or their critical habitats.
(b) The Agencies Lack Sufficient Discretion or Control To Bring These
Actions Under the Consultation Requirement of the ESA
The primary purpose of EPCA, as amended by EISA, and codified at 49
U.S.C. chapter 329, is energy conservation, and NHTSA is statutorily
obligated to set attribute-based CAFE standards for each model year at
the levels it determines are ``maximum feasible.'' \3518\ But ``maximum
feasible'' is a balancing of several factors, and Congress clearly did
not envision that the CAFE program would ``solve'' energy conservation
in a single rulemaking action.\3519\ Fuel economy standards have the
related benefit of reducing CO2 emissions, and may also
result in reduced emissions of many criteria air pollutants. Similarly,
EPA has found that the elevated concentrations of greenhouse gases in
the atmosphere may reasonably be anticipated to endanger public health
and welfare. As a result of these findings, CAA section 202(a) requires
the agency to issue standards applicable to emissions of such gases
from motor vehicles. Although not a statutory requirement, EPA has
given weight to the policy goal of establishing CO2
[[Page 25256]]
standards that are coordinated with NHTSA's CAFE standards.\3520\
---------------------------------------------------------------------------
\3518\ See 49 U.S.C. 32902(a) (``At least 18 months before the
beginning of each model year, the Secretary of Transportation shall
prescribe by regulation average fuel economy standards for
automobiles manufactured by a manufacturer in that model year. Each
standard shall be the maximum feasible average fuel economy level
that the Secretary decides the manufacturers can achieve in that
model year.'').
\3519\ See, e.g., 49 U.S.C. 32902(b)(2) (setting separate
requirements for CAFE standards for MYs 2011 through 2020 and MYs
2021 through 2030).
\3520\ See Mass. v. EPA, 549 U.S. 497, 532 (2007) (``. . .there
is no reason to think the two agencies cannot both administer their
obligations and yet avoid inconsistency.'')
---------------------------------------------------------------------------
As previously indicated, commenters assert that CO2 and
criteria air pollutant emissions are relevant to Section 7(a)(2)
consultation because of the potential impacts of climate change or the
pollutants themselves on listed species or designated critical habitat.
However, it is not clear whether their comments are based on the fact
that the agencies predict increases in CO2 emissions and
most criteria pollutant emissions under all action alternatives
compared to the MY 2022-2025 CO2 and augural CAFE standards,
or the fact that any emissions from passenger cars or light trucks will
continue under any of the alternatives considered.
With regard to the latter, NHTSA does not interpret EPCA/EISA to
mean that Congress expected the CAFE program to take the U.S. auto
fleet off of oil entirely--indeed, EISA renders doing so impossible
because it amended EPCA to prohibit NHTSA from considering the fuel
economy of dedicated alternative fuel vehicles, including electric
vehicles, when setting maximum feasible standards. This means that
standards cannot be set that assume increased usage of full
electrification for compliance. As a result, no matter the level at
which NHTSA sets CAFE standards in accordance with EPCA, CO2
and criteria pollutant emissions will continue. So long as NHTSA's
obligation to set CAFE standards remains in place, it is reasonable to
assume that Congress's expectation for EPA, in coordinating with NHTSA,
is similar.
The purpose of Section 7(a)(2) consultation is to ensure that
Federal agencies are not undertaking, funding, permitting, or
authorizing actions that are likely to jeopardize the continued
existence of listed species or destroy or adversely modify designated
critical habitat. However, no matter what standards the agencies set
under the CAFE and CO2 programs, Americans will continue to
drive. Neither NHTSA nor EPA has authority to control vehicle miles
traveled. As long as there is driving, there will be emissions--whether
from vehicle tailpipes or from the stationary sources that create the
energy that the vehicles consume. Moreover, both agencies have
concluded that significant further electrification of the fleet is not
practicable at this time due to concerns about consumer acceptance in a
time of foreseeably low fuel prices. The fact that CO2 and
criteria pollutant emissions will continue after NHTSA and EPA actions
on standards cannot, alone, trigger Section 7(a)(2) consultation as the
agencies lack the discretion or control over these emissions to simply
regulate them away entirely in this action.\3521\ Consultation is not
required where an agency lacks discretion to take action that will
inure to the benefit of listed species.\3522\ Since elimination of oil
from the fleet is inconsistent with the agencies' statutory authorities
and the clear intent of Congress, consultation is not triggered under
this scenario.
---------------------------------------------------------------------------
\3521\ National Ass'n of Home Builders v. Defenders of Wildlife,
551 U.S. 644, 673 (2007) (``Applying Chevron, we defer to the
Agency's reasonable interpretation of ESA [section] 7(a)(2) as
applying only to `actions in which there is discretionary Federal
involvement or control.''' (quoting 50 CFR 402.03)).
\3522\ Id.; Sierra Club v. Babbitt, 65 F.3d 1502, 1509 (9th Cir.
1995) (ESA Section 7(a)(2) consultation is not required where an
agency lacks discretion to influence private conduct in a manner
that will inure to the benefit of listed species).
---------------------------------------------------------------------------
Commenters may instead be referring to the trend in CO2
and criteria air pollutant emissions under the action alternatives
considered in this rulemaking (e.g., whether and by how much emissions
increase or decrease). To that point, all of the action alternatives
considered result in increases in CO2 and most criteria air
pollutant emissions compared to the standards considered and set forth
in the 2012 rulemaking. However, the agencies do not believe this is
the relevant comparison for purposes of determining the applicability
of Section 7 of the ESA to this action. Model years 2021 through 2026,
for the most part, have not yet arrived. So it is not appropriate to
compare the current action to a prior action that has not been
implemented and which the agencies are reconsidering. When compared to
standards through MY 2020, under any of the alternatives considered,
fuel economy will improve and CO2 and most criteria
pollutant emissions will decrease over time, either as stringency
increases or from the turnover in the fleet to newer, cleaner vehicles.
As detailed above, however, there is no way to meaningfully
differentiate between the alternatives in terms of outcomes for listed
species and designated critical habitat. The agencies cannot reasonably
calculate how incrementally less emissions resulting from more
stringent standards would benefit those species or habitats; rather, at
most, the agencies can only posit that more stringent standards
hypothetically could lead to better outcomes. But where to draw any
line in terms of impacts to species and habitats is an impossible
exercise. Yet, as noted above, NHTSA is mandated by Congress to set
``maximum feasible'' standards and EPA's mission is to protect public
health and welfare. Under these circumstances, where the agencies must
issue standards pursuant to statutory mandate that under any scenario
will involve emissions, yet they lack the commensurate ability to take
action that will inure to the benefit of species in any meaningful way,
Section 7(a)(2) consultation is not required.
Finally, regardless of the level of stringency at which the
agencies set CAFE and CO2 standards, criteria pollutant and
CO2 emissions from motor vehicles will change to a greater
or lesser degree because of several independent factors. Because of the
complex relationships between fuel economy, vehicle sales, driver
behavior (e.g., VMT and driving location), and technology choices by
manufacturers, emissions will never uniformly increase or decrease for
all future model years, across all regulated pollutants, and in all
locations throughout the country. For example, increased stringency may
result in greater VMT, resulting in larger downstream emissions of some
criteria pollutants. On the other hand, decreased stringency may result
in greater fuel refining, result in larger upstream emissions of some
pollutants. Because vehicle operation and refinery activity depends
upon independent market forces, impacts to particular listed species or
designated critical habitat are dependent upon where vehicle operation
or increased fuel refining occur, but neither agency can control such
decisions. Regardless of whether NHTSA and EPA engage in Section
7(a)(2) consultation, the agencies lack the control necessary to negate
all emissions increases in whatever years and locations they occur
(e.g., ensure ideal technology choices by manufacturers, control
consumer purchasing behavior, or regulate driving locations or VMT), or
otherwise mitigate impacts associated with these particular emissions.
But setting stringency is, in fact, what the agencies are statutorily
obligated to do.
For the foregoing reasons, NHTSA and EPA conclude that they lack
sufficient discretion or control to bring these actions under the
consultation requirement of the ESA.
7. Floodplain Management (Executive Order 11988 and DOT Order 5650.2)
These Orders require Federal agencies to avoid the long- and short-
term adverse impacts associated with the
[[Page 25257]]
occupancy and modification of floodplains, and to restore and preserve
the natural and beneficial values served by floodplains. Executive
Order 11988 also directs agencies to minimize the impact of floods on
human safety, health, and welfare, and to restore and preserve the
natural and beneficial values served by floodplains through evaluating
the potential effects of any actions the agency may take in a
floodplain and ensuring that its program planning and budget requests
reflect consideration of flood hazards and floodplain management. DOT
Order 5650.2 sets forth DOT policies and procedures for implementing
Executive Order 11988. The DOT Order requires that the agency determine
if a proposed action is within the limits of a base floodplain, meaning
it is encroaching on the floodplain, and whether this encroachment is
significant. If significant, the agency is required to conduct further
analysis of the proposed action and any practicable alternatives. If a
practicable alternative avoids floodplain encroachment, then the agency
is required to implement it.
In this rulemaking, the agencies are not occupying, modifying and/
or encroaching on floodplains. The agencies, therefore, conclude that
the Orders are not applicable to this action. NHTSA has, however,
conducted a review of the alternatives on potentially affected
resources, including floodplains, in its FEIS.
8. Preservation of the Nation's Wetlands (Executive Order 11990 and DOT
Order 5660.1a)
These Orders require Federal agencies to avoid, to the extent
possible, undertaking or providing assistance for new construction
located in wetlands unless the agency head finds that there is no
practicable alternative to such construction and that the proposed
action includes all practicable measures to minimize harm to wetlands
that may result from such use. Executive Order 11990 also directs
agencies to take action to minimize the destruction, loss, or
degradation of wetlands in ``conducting Federal activities and programs
affecting land use, including but not limited to water and related land
resources planning, regulating, and licensing activities.'' DOT Order
5660.1a sets forth DOT policy for interpreting Executive Order 11990
and requires that transportation projects ``located in or having an
impact on wetlands'' should be conducted to assure protection of the
Nation's wetlands. If a project does have a significant impact on
wetlands, an EIS must be prepared.
In the NPRM, the agencies noted that they are not undertaking or
providing assistance for new construction located in wetlands. The
agencies, therefore, concluded that these Orders do not apply to this
rulemaking. One commenter disagreed with this conclusion, noting the
potential land use impacts of the rule and the agencies' obligation to
consider all factors relevant to the proposal's effect on the survival
and quality of wetlands.\3523\ The agencies do not believe that it is
feasible to establish the requisite causal chain between the impacts of
this action and impacts on wetlands, nor would such impacts be
reasonably foreseeable as a direct or indirect result of this
rulemaking. The agencies therefore continue to conclude that these
Orders do not apply to this rulemaking. Regardless, NHTSA addresses the
potential effects of the alternatives on resources, including wetlands,
in its FEIS.
---------------------------------------------------------------------------
\3523\ Joint Submission from the States of California et al. and
the Cities of Oakland et al., Docket No. NHTSA-2018-0067-11735, at
46-47.
---------------------------------------------------------------------------
9. Migratory Bird Treaty Act (MBTA), Bald and Golden Eagle Protection
Act (BGEPA), Executive Order 13186
The MBTA (16 U.S.C. 703-712) provides for the protection of certain
migratory birds by making it illegal for anyone to ``pursue, hunt,
take, capture, kill, attempt to take, capture, or kill, possess, offer
for sale, sell, offer to barter, barter, offer to purchase, purchase,
deliver for shipment, ship, export, import, cause to be shipped,
exported, or imported, deliver for transportation, transport or cause
to be transported, carry or cause to be carried, or receive for
shipment, transportation, carriage, or export'' any migratory bird
covered under the statute.\3524\
---------------------------------------------------------------------------
\3524\ 16 U.S.C. 703(a).
---------------------------------------------------------------------------
The BGEPA (16 U.S.C. 668-668d) makes it illegal to ``take, possess,
sell, purchase, barter, offer to sell, purchase or barter, transport,
export or import'' any bald or golden eagles.\3525\ Executive Order
13186, ``Responsibilities of Federal Agencies to Protect Migratory
Birds,'' helps to further the purposes of the MBTA by requiring a
Federal agency to develop a Memorandum of Understanding (MOU) with the
Fish and Wildlife Service when it is taking an action that has (or is
likely to have) a measurable negative impact on migratory bird
populations.
---------------------------------------------------------------------------
\3525\ 16 U.S.C. 668(a).
---------------------------------------------------------------------------
The agencies conclude that the MBTA, BGEPA, and Executive Order
13186 do not apply to this action because there is no disturbance,
take, measurable negative impact, or other covered activity involving
migratory birds or bald or golden eagles involved in this rulemaking.
10. Department of Transportation Act (Section 4(f))
Section 4(f) of the Department of Transportation Act of 1966 (49
U.S.C. 303), as amended, is designed to preserve publicly owned park
and recreation lands, waterfowl and wildlife refuges, and historic
sites. Specifically, Section 4(f) provides that DOT agencies cannot
approve a transportation program or project that requires the use of
any publicly owned land from a public park, recreation area, or
wildlife or waterfowl refuge of national, State, or local significance,
or any land from a historic site of national, State, or local
significance, unless a determination is made that:
(1) There is no feasible and prudent alternative to the use of
land, and
(2) The program or project includes all possible planning to
minimize harm to the property resulting from the use.
These requirements may be satisfied if the transportation use of a
Section 4(f) property results in a de minimis impact on the area.
NHTSA concludes that Section 4(f) is not applicable to this action
because this rulemaking is not an approval of a transportation program
or project that requires the use of any publicly owned land.
11. Executive Order 12898: ``Federal Actions To Address Environmental
Justice in Minority Populations and Low-Income Populations''
Executive Order 12898 (59 FR 7629 (Feb. 16, 1994)) establishes
Federal executive policy on environmental justice. It directs Federal
agencies, to the greatest extent practicable and permitted by law, to
make environmental justice part of their mission by identifying and
addressing, as appropriate, disproportionately high and adverse human
health or environmental effects of their programs, policies, and
activities on minority and low-income populations in the United States.
DOT Order 5610.2(a) \3526\ sets forth the Department of
Transportation's policy to consider environmental justice principles in
all its programs, policies, and activities.
---------------------------------------------------------------------------
\3526\ Department of Transportation Updated Environmental
Justice Order 5610.2(a), 77 FR 27534 (May 10, 2012).
---------------------------------------------------------------------------
Environmental justice is a principle asserting that all people
deserve fair treatment and meaningful involvement with respect to
environmental laws,
[[Page 25258]]
regulations, and policies. EPA seeks to provide the same degree of
protection from environmental health hazards for all people. DOT shares
this goal and is informed about the potential environmental impacts of
its rulemakings through the NEPA process. One comment on the NPRM
claimed that the agencies ``failed to recognize the benefits of the
existing standards'' for disadvantaged communities. Specifically, the
commenter claimed that the agencies did not provide an underlying
analysis of environmental justice issues and thereby failed to meet the
requirements of E.O. 12898.\3527\ However, the agencies addressed their
obligations under E.O. 12898 in the preamble to the NPRM and in Section
7.5 of the DEIS. The agencies received a number of comments regarding
the analysis it presented. NHTSA responds to those comments in Section
10.7 of the FEIS, and the agencies have revised their environmental
justice analysis based on the information contained in those comments.
The revised analysis is presented here and in the FEIS.
---------------------------------------------------------------------------
\3527\ CARB, Docket No. NHTSA-2018-0067-11873, at 411-12.
---------------------------------------------------------------------------
There is evidence that proximity to oil refineries could be
correlated with incidences of cancer and
leukemia.3528 3529 3530 Proximity to high-traffic roadways
could result in adverse cardiovascular and respiratory impacts, among
other possible impacts.3531 3532 3533 3534 3535 3536 3537
Climate change affects overall global temperatures, which could, in
turn, affect the number and severity of outbreaks of vector-borne
illnesses.3538 3539 In the context of this rulemaking, the
environmental justice concern is the extent to which minority and low-
income populations could be more exposed or vulnerable to such
environmental and health impacts.
---------------------------------------------------------------------------
\3528\ Pukkala, E. Cancer incidence among Finnish oil refinery
workers, 1971-1994. Journal of Occupational and Environmental
Medicine. 40(8):675-79 (1998). doi:10.1023/A:1018474919807.
\3529\ Chan, C.-C.; Shie, R.H.; Chang, T.Y.; Tsai, D.H. Workers'
exposures and potential health risks to air toxics in a
petrochemical complex assessed by improved methodology.
International Archives of Occupational and Environmental Health.
79(2):135-142 (2006). doi:10.1007/s00420-005-0028-9. Online at:
https://www.researchgate.net/publication/7605242_Workers'_exposures_and_potential_health_risks_to_air_toxics_i
n_a_petrochemical_complex_assessed_by_improved_methodology.
\3530\ Bulka, C.; Nastoupil, L.J.; McClellan, W.; Ambinder, A.;
Phillips, A.; Ward, K.; Bayakly, A.R.; Switchenko, J.M.; Waller, L.;
Flowers, C.R. Residence proximity to benzene release sites is
associated with increased incidence of non-Hodgkin lymphoma. Cancer.
119(18):3309-17 (2013). doi:10.1002/cncr.28083. Online at: http://onlinelibrary.wiley.com/doi/10.1002/cncr.28083/pdf;jsessionid=1520A90A764A95985316057D7D76A362.f02t02.
\3531\ HEI (Health Effects Institute). 2010. Traffic-Related Air
Pollution: A Critical Review of the Literature on Emissions,
Exposure and Health Effects. Special Report 17. Health Effects
Institute: Boston, MA:. HEI Panel on the Health Effects of Traffic-
Related Air Pollution, 386 pp. Available at: https://www.healtheffects.org/system/files/SR17Traffic%20Review.pdf.
(Accessed: March 3, 2018).
\3532\ Heinrich, J. and H.-E. Wichmann. 2004. Traffic Related
Pollutants in Europe and their Effect on Allergic Disease. Current
Opinion in Allergy and Clinical Immunology 4(5):341-348.
\3533\ Salam, M.T., T. Islam, and F.D. Gilliland. 2008. Recent
Evidence for Adverse Effects of Residential Proximity to Traffic
Sources on Asthma. Current Opinion in Pulmonary Medicine 14(1):3-8.
doi:10.1097/MCP.0b013e3282f1987a.
\3534\ Samet, J.M. 2007. Traffic, Air Pollution, and Health.
Inhalation Toxicology 19(12):1021-27. doi:10.1080/08958370701533541.
\3535\ Adar, S. and J. Kaufman. 2007. Cardiovascular Disease and
Air Pollutants: Evaluating and Improving Epidemiological Data
Implicating Traffic Exposure. Inhalation Toxicology 19(S1):135-49.
doi:10.1080/08958370701496012.
\3536\ Wilker, E.H., E. Mostofsky, S.H. Lue, D. Gold, J.
Schwartz, G.A. Wellenius, and M.A. Mittleman. 2013. Residential
Proximity to High-Traffic Roadways and Poststroke Mortality. Journal
of Stroke and Cerebrovascular Diseases 22(8): e366-e372.
doi:10.1016/j.jstrokecerebrovasdis.2013.03.034. Available at:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4066388/. (Accessed:
March 6, 2018).
\3537\ Hart, J.E., E.B. Rimm, K.M. Rexrode, and F. Laden. 2013.
Changes in Traffic Exposure and the Risk of Incident Myocardial
Infarction and All-cause Mortality. Epidemiology 24(5):734-42.
\3538\ U.S. Global Change Research Program (GCRP). Global
Climate Change Impacts in the United States: The Third National
Climate Assessment. U.S. Global Change Research Program. Melillo,
J.M, T.C. Richmond, and G.W. Yohe (Eds.). U.S. Government Printing
Office: Washington, DC 841 pp (2014). doi:10.7930/J0Z31WJ2.
Available at: http://nca2014.globalchange.gov/report. (Accessed:
February 27, 2018).
\3539\ GCRP. The Impacts of Climate Change on Human Health in
the United States, A Scientific Assessment (2016). April 2016.
Available at: https://health2016.globalchange.gov. (Accessed:
February 28, 2018).
---------------------------------------------------------------------------
Numerous studies have found that some environmental hazards are
more prevalent in areas where racial/ethnic minorities and people with
low socioeconomic status represent a higher proportion of the
population compared with the general population. In addition, compared
to non-Hispanic whites, some subpopulations defined by race and
ethnicity have been shown to have a greater incidence of some health
conditions during certain life stages. For example, in 2014, about 13
percent of Black, non-Hispanic and 24 percent of Puerto Rican children
were estimated to have asthma, compared with 8 percent of white, non-
Hispanic children.\3540\ The agencies have therefore considered areas
nationwide that could contain minority and low-income communities who
would most likely be exposed to the environmental and health impacts of
oil production, distribution, and consumption or the potential impacts
of climate change. These include areas where oil production and
refining occur, areas near roadways, coastal flood-prone areas, and
urban areas that are subject to the heat island effect.\3541\
---------------------------------------------------------------------------
\3540\ http://www.cdc.gov/asthma/most_recent_data.htm.
\3541\ The heat island effect refers to developed areas having
higher temperatures than surrounding rural areas.
---------------------------------------------------------------------------
The following discussion addresses environmental justice
implications related to air quality and to climate change and carbon
emissions in the context of this final rulemaking. Emissions of air
pollutants may be affected by this rulemaking due to changes in fuel
use and VMT, which are described above. To the degree to which minority
and low-income populations may be present in proximity to the locations
described in this section, they may be exposed disproportionately to
these emissions changes. In addition, the following analysis also
discusses other potential reasons why minority and low-income
populations may be susceptible to the health impacts of air pollutants.
NHTSA also discusses environmental justice in Chapter 7.5 of its FEIS.
a) Proximity to Oil Production and Refining
As stated above, numerous studies have found that some
environmental hazards are more prevaluent in areas where minority and
low-income populations represent a higher proportion of the population
compared with the general population. For example, one study found that
survey respondents who were black and, to a lesser degree, had lower
income levels, were significantly more likely to live within 1 mile of
an industrial facility listed in the EPA's 1987 Toxic Release Inventory
(TRI) national database.\3542\
---------------------------------------------------------------------------
\3542\ Mohai, P., P.M. Lantz, J. Morenoff, J.S. House, and R.P.
Mero. Racial and Socioeconomic Disparities in Residential Proximity
to Polluting Industrial Facilities: Evidence from the Americans'
Changing Lives Study. American Journal of Public Health 99(S3):
S649-S656 (2009). doi:10.2105/AJPH.2007.131383. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2774179/pdf/S649.pdf.
(Accessed: March 2, 2018).
---------------------------------------------------------------------------
A meta-analysis of 49 environmental equity studies concluded that
evidence of race-based environmental inequities is statistically
significant (although the average magnitude of these inequities is
small), while evidence supporting the existence of income-based
environmental inequities is substantially weaker.\3543\ Considering
poverty-based class effects, that meta-
[[Page 25259]]
analysis found an inverse relationship between environmental risk and
poverty, concluding that environmental risks are less likely to be
located in areas of extreme poverty.\3544\ However, individual studies
may reach contradictory conclusions in relation to race- and income-
based inequities across a range of environmental risks. Therefore, the
meta-analysis also sought to examine the reasons why conclusions vary
across studies of environmental inequity. Possible explanations for why
studies reach contrary conclusions include variability in the source of
potential environmental risk that the study considers (e.g., the type
of facility or the associated level of pollution or risk); variability
in the methodology applied to aggregate demographic data and to define
the comparison population; and the degree to which statistical models
control for other variables that may explain the distribution of
potential environmental risk.
---------------------------------------------------------------------------
\3543\ Ringquist, E.J. Evidence of Environmental Inequities: A
Meta-Analysis. Journal of Policy Analysis and Management 24(2):223-
47 (2005).
\3544\ Ringuist (2005).
---------------------------------------------------------------------------
To test whether there are disparate impacts from hazardous
industrial facilities on racial/ethnic minorities, the disadvantaged,
the working class, and manufacturing workers, one study tested the
relationship between hazard scores of Philadelphia-area facilities in
EPA's Risk-Screening Environmental Indicators (RSEI) database and the
demographics of populations near those facilities using multivariate
regression.\3545\ This study concluded that racial/ethnic minorities,
the most socioeconomically disadvantaged, and those employed in
manufacturing suffer a disparate impact from the highest-hazard
facilities (primarily manufacturing plants).
---------------------------------------------------------------------------
\3545\ Sicotte, D. and S. Swanson. Whose Risk in Philadelphia?
Proximity to Unequally Hazardous Industrial Facilities. Social
Science Quarterly 88(2):516-534 (2007).
---------------------------------------------------------------------------
Other commissioned reports and case studies provide additional
evidence of the presence of low-income and minority populations near
industrial facilities and of racial or socioeconomic disparities in
exposure to environmental risk, although these sources were not
published in peer-reviewed scientific
journals.3546 3547 3548 3549
---------------------------------------------------------------------------
\3546\ UCC (United Church of Christ). Toxic Wastes and Race at
Twenty: 1987--2007. A Report Prepared for the United Church of
Christ Justice and Witness Ministries. Available at: https://www.nrdc.org/sites/default/files/toxic-wastes-and-race-at-twenty-1987-2007.pdf (2007). (Accessed: April 9, 2018).
\3547\ National Association for the Advancement of Colored
People and Clean Air Task Force. Fumes Across the Fence-line: The
Health Impacts of Air Pollution from Oil & Gas Facilities on African
American Communities (2017). Available at: http://www.catf.us/wp-content/uploads/2017/11/CATF_Pub_FumesAcrossTheFenceLine.pdf.
(Accessed: February 24, 2019).
\3548\ Ash, M., J.K. Boyce, G. Chang, M. Pastor, J. Scoggins,
and J. Tran. Justice in the Air: Tracking Toxic Pollution from
America's Industries and Companies to our States, Cities, and
Neighborhoods. Political Economy Research Institute at the
University of Massachusetts, Amherst and the Program for
Environmental and Regional Equity at the University of Southern
California (2009). Available at: https://dornsife.usc.edu/assets/sites/242/docs/justice_in_the_air_web.pdf. (Accessed: February 24,
2019).
\3549\ Kay, J. and C. Katz. Pollution, Poverty and People of
Color: Living With Industry. Scientific American. Available at:
https://www.scientificamerican.com/article/pollution-poverty-people-color-living-industry/ (2012). (Accessed: March 4, 2018).
---------------------------------------------------------------------------
Few studies address disproportionate exposure to environmental risk
associated with oil refineries specifically. One study found that the
populations surrounding oil refineries are more often minorities,
concluding that ``56 percent of people living within three miles of
[oil] refineries in the United States are minorities--almost double the
national average.'' \3550\ Another examined whether findings of
environmental inequity varied between coke production plants and oil
refineries, both of which are significant sources of air
pollution.\3551\ This study concluded that census tracts near coke
plants had a disproportionate share of poor and nonwhite residents, and
that existing inequities were primarily economic in nature. However,
the findings for oil refineries did not strongly support an
environmental inequity hypothesis. A more recent study of environmental
justice in the oil refinery industry found evidence of environmental
injustice as a result of unemployment levels in areas around refineries
and, to a slightly lesser extent, as a result of income
inequality.\3552\ This study did not test for race-based environmental
inequities.
---------------------------------------------------------------------------
\3550\ O'Rourke, D. and S. Connolly. Just Oil? The Distribution
of Environmental and Social Impacts of Oil Production and
Consumption. Annual Review of Environment and Resources 28(1):587-
617 (2003). doi:10.1146/annurev.energy.28.050302.105617.
\3551\ Graham, J.D., N.D. Beaulieu, D. Sussman, M. Sadowitz, and
Y.C. Li. Who Lives Near Coke Plants and Oil Refineries? An
Exploration of the Environmental Inequity Hypothesis. Risk Analysis
19(2):171-86 (1999). doi:10.1023/A:1006965325489. Green, R.S., S.
Smorodinsky, J.J. Kim, R. McLaughlin, and B. Ostro. Proximity of
California public schools to busy roads. Environmental Health
Perspectives 112 (1):61-66 (2004). Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1241798/. (Accessed: May 31,
2018).
\3552\ Carpenter, A. and M. Wagner. Environmental Justice in the
Oil Refinery Industry: A Panel Analysis Across United States
Counties. Ecological Economics 159:101-109 (2019).
---------------------------------------------------------------------------
Overall, the body of scientific literature points to
disproportionate representation of minority and low-income populations
in proximity to a range of industrial, manufacturing, and hazardous
waste facilities that are stationary sources of air pollution; although
results of individual studies may vary. While the scientific literature
specific to oil refineries is limited, disproportionate exposure of
minority and low-income populations to air pollution from oil
refineries is suggested by other broader studies of racial and
socioeconomic disparities in proximity to industrial facilities
generally.
The potential increase in fuel production and consumption projected
as a result of this rulemaking (compared to the No Action Alternative)
could lead to an increase in upstream emissions of criteria and toxic
air pollutants due to increased extraction, refining, and
transportation of fuel. As described in Section VII.A.4.c.3.b.i, total
upstream emissions of criteria and toxic air pollutants in 2035 are
projected to increase under all action alternatives compared to the No
Action Alternative, with the exception that total upstream emissions of
SO2 are projected to decrease under all action alternatives
under the CAFE program (but not under the CO2 program). As
noted, a correlation between proximity to oil refineries and the
prevalence of minority and low-income populations is suggested in the
scientific literature. To the extent that minority and low-income
populations live closer to oil refining facilities, these populations
may be more likely to be adversely affected by these emissions.
However, the magnitude of the change in emissions relative to the
baseline is minor and would not be characterized as high and adverse.
Proximity to High-Traffic Roadways
Studies have more consistently demonstrated a disproportionate
prevalence of minority and low-income populations living near mobile
sources of pollutants. In certain locations in the United States, for
example, there is consistent evidence that populations or schools near
roadways typically include a greater percentage of minority or low-
income residents.3553 3554 3555 3556 3557 3558 3559 In
[[Page 25260]]
California, studies demonstrate that minorities and low-income
populations are disproportionately likely to live near a major roadway
or in areas of high traffic density compared to the general
population.3560 3561 A study of traffic, air pollution, and
socio-economic status inside and outside the Minneapolis-St. Paul
metropolitan area similarly found that populations on the lower end of
the socioeconomic spectrum and minorities are disproportionately
exposed to traffic and air pollution and at higher risk for adverse
health outcomes.\3562\ Near-road exposure to vehicle emissions can
cause or exacerbate health conditions such as
asthma.3563 3564 3565 3566 One study demonstrated that
students at schools in Michigan closer to major highways had a higher
risk of respiratory and neurological disease and were more likely to
fail to meet state educational standards, after controlling for other
variables.\3567\ In general, studies such as these demonstrate trends
in specific locations in the United States that may be indicative of
broader national trends.
---------------------------------------------------------------------------
\3553\ Green, R.S., S. Smorodinsky, J.J. Kim, R. McLaughlin, and
B. Ostro. Proximity of California public schools to busy roads.
Environmental Health Perspectives 112 (1):61-66 (2004). Available
at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1241798/. Last
accessed: May 31, 2018.
\3554\ Wu, Y-C.; Batterman, S.A. Proximity of schools in
Detroit, Michigan to automobile and truck traffic. Journal of
Exposure Science and Environmental Epidemiology 16(5): 457-470
(2006). doi:10.1038/sj.jes.7500484. Available at: http://www.nature.com/articles/7500484. Last accessed: May 31, 2018.
\3555\ Chakraborty, J., and P.A. Zandbergen. Children at risk:
measuring racial/ethnic disparities in potential exposure to air
pollution at school and home. Journal of Epidemiology & Community
Health 61:1074-1079 (2007). doi: 10.1136/jech.2006.054130.
\3556\ Depro, B., and C. Timmins. Mobility and Environmental
Equity: Do Housing Choices Determine Exposure to Air Pollution?
North Carolina State University and RTI International, Duke
University and NBER (2008). Available at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.586.7164&rep=rep1&type=pdf. (Accessed: May 31,
2018).
\3557\ Marshall, J.D. Environmental inequality: air pollution
exposures in California's South Coast Air Basin. Atmospheric
Environment 42(21):5499-5503 (2008).
\3558\ Su, J. G., T. Larson, T. Gould, M. Cohen, and M.
Buzzelli. Transboundary air pollution and environmental justice:
Vancouver and Seattle compared. GeoJournal 75(6):595-608 (2010).
doi: 10.1007/s10708-009-9269-6.
\3559\ Su, J. G., M. Jarrett, A. de Nazelle, and J. Wolch. Does
exposure to air pollution in urban parks have socioeconomic, racial
or ethnic gradients? Environmental Research 111 (3):319-328 (2011).
doi: 10.1016/j.envres.2011.01.002.
\3560\ Carlson, A.E. The Clean Air Act's Blind Spot:
Microclimates and Hotspot Pollution. 65 UCLA Law Review 1036 (2018).
\3561\ Gunier, R.B., A. Hertz, J. Von Behren, and P. Reynolds.
Traffic density in California: socioeconomic and ethnic differences
among potentially exposed children. Journal of Exposure Analysis and
Environmental Epidemiology 13(3):240-46 (2003). doi:10.1038/
sj.jea.7500276.
\3562\ Pratt, G.C., M.L. Vadali, D.L. Kvale, and K.M. Ellickson,
Traffic, air pollution, minority, and socio-economic status:
addressing inequities in exposure and risk. International Journal of
Environmental research and Public Health 12(5):53555372 (2015).
doi:10.3390/ijerph120505355.
\3563\ Carlson (2018).
\3564\ Gunier et al. (2003).
\3565\ Meng, Y-Y., M. Wilhelm, R.P. Rull, P. English, S. Nathan,
and B. Ritz. Are frequent asthma symptoms among low-income
individuals related to heavy traffic near homes, vulnerabilities, or
both? Annals of Epidemiology 18:343-350 (2008). doi:10.1016/
j.annepidem.2008.01.006.
\3566\ Khreis, H., C. Kelly, J. Tate, R. Parslow, K. Lucas, and
M. Nieuwenhuijsen. Exposure to traffic-related air pollution and
risk of development of childhood asthma: A systematic review and
meta-analysis. Environment International 100:1-31 (2017). https://doi.org/10.1016/j.envint.2016.11.012.
\3567\ Kweon, B-S., P. Mohai, S. Lee, and A.M. Sametshaw. 2016.
Proximity of Public Schools to Major Highways and Industrial
Facilities, and Students' School Performance and Health Hazards.
Environment and Planning B: Urban Analytics and City Science
45(2):312-329. doi.org/10.1177/0265813516673060.
---------------------------------------------------------------------------
Fewer studies have been conducted at the national level, yet those
that do exist also demonstrate a correlation between minority and low-
income status and proximity to roadways.3568 3569 For
example, one study found that greater traffic volumes and densities at
the national level are associated with larger shares of minority and
low-income populations living in the vicinity.\3570\ Another study
found that schools with minority and underprivileged \3571\ children
were disproportionately located within 250 meters of a major
roadway.\3572\
---------------------------------------------------------------------------
\3568\ Tian, N., J. Xue, and T. M. Barzyk. Evaluating
socioeconomic and racial differences in traffic-related metrics in
the United States using a GIS approach. Journal of Exposure Science
and Environmental Epidemiology 23 (2):215 (2013). doi: 10.1038/
jes.2012.83. Available at: http://www.nature.com/articles/jes201283.
(Accessed: May 31, 2018).
\3569\ Boehmer, T.K., S.L. Foster, J.R. Henry, E.L. Woghiren-
Akinnifesi, and F.Y. Yip. Residential Proximity to Major Highways--
United States, 2010. Morbidity and Mortality Weekly Report 62(3):46-
50 (2013). Available at: http://www.cdc.gov/mmwr/preview/mmwrhtml/su6203a8.htm. (Accessed: February 26, 2018).
\3570\ Rowangould, G.M. A Census of the US Near-roadway
Population: Public Health and Environmental Justice Considerations.
Transportation Research Part D: Transport and Environment 25:59-67
(2013). doi:10.1016/j.trd.2013.08.003.
\3571\ Public schools were determined to serve predominantly
underprivileged students if they were eligible for Title I programs
(federal programs that provide funds to school districts and schools
with high numbers or high percentages of children who are
disadvantaged) or had a majority of students who were eligible for
free/reduced-price meals under the National School Lunch and
Breakfast Programs.
\3572\ Kingsley, S.L., M.N. Eliot, L. Carlson, J. Finn, D.L.
MacIntosh, H.H. Suh, and G.A. Wellenius. Proximity of US Schools to
Major Roadways: A Nationwide Assessment. Journal of Exposure Science
and Environmental Epidemiology 24(3):253-59 (2014). doi:10.1038/
jes.2014.5.
---------------------------------------------------------------------------
As detailed in Section 10.3.8 of the PRIA and Section X.E.11.a.2 of
the FRIA, NHTSA and EPA analyzed two national databases that allowed
evaluation of whether homes and schools were located near a major road
and whether disparities in exposure may be occurring in these
environments. The American Housing Survey (AHS) includes descriptive
statistics of over 70,000 housing units across the nation. The study
survey is conducted every two years by the U.S. Census Bureau. The
second database the agencies analyzed was the U.S. Department of
Education's Common Core of Data, which includes enrollment and location
information for schools across the U.S.
In analyzing the 2009 AHS, the focus was on whether or not a
housing unit was located within 300 feet of a ``4-or-more lane highway,
railroad, or airport.'' \3573\ Whether there were differences between
households in such locations compared with those in locations farther
from these transportation facilities was analyzed.\3574\ Other
variables, such as land use category, region of country, and housing
type, were included. Homes with a nonwhite householder were found to be
22 to 34 percent more likely to be located within 300 feet of these
large transportation facilities than homes with white householders.
Homes with a Hispanic householder were 17 to 33 percent more likely to
be located within 300 feet of these large transportation facilities
than homes with non-Hispanic householders. Households near large
transportation facilities were, on average, lower in income and
educational attainment, more likely to be a rental property, and more
likely to be located in an urban area compared with households more
distant from transportation facilities.
---------------------------------------------------------------------------
\3573\ This variable primarily represents roadway proximity.
According to the Central Intelligence Agency's World Factbook, in
2010, the United States had 6,506,204 km of roadways, 224,792 km of
railways, and 15,079 airports. Highways thus represent the
overwhelming majority of transportation facilities described by this
factor in the AHS.
\3574\ Bailey, C. (2011) Demographic and Social Patterns in
Housing Units Near Large Highways and other Transportation Sources.
Memorandum to docket.
---------------------------------------------------------------------------
In examining schools near major roadways, the Common Core of Data
(CCD) from the U.S. Department of Education, which includes information
on all public elementary and secondary schools and school districts
nationwide, was examined.\3575\ To determine school proximities to
major roadways, a geographic information system (GIS) to map each
school and roadways based on the U.S. Census's TIGER roadway file was
used.\3576\ Minority students were found to be overrepresented at
schools within 200 meters of the largest roadways, and schools within
200 meters of the largest roadways also had higher than expected
numbers of
[[Page 25261]]
students eligible for free or reduced-price lunches. For example, Black
students represent 22 percent of students at schools located within 200
meters of a primary road, whereas Black students represent 17 percent
of students in all U.S. schools. Hispanic students represent 30 percent
of students at schools located within 200 meters of a primary road,
whereas Hispanic students represent 22 percent of students in all U.S.
schools.
---------------------------------------------------------------------------
\3575\ http://nces.ed.gov/ccd/.
\3576\ Pedde, M.; Bailey, C. Identification of Schools within
200 Meters of U.S. Primary and Secondary Roads. Memorandum to the
docket (2011).
---------------------------------------------------------------------------
Overall, there is substantial evidence that the population who
lives or attends school near major roadways are more likely to be
minority or low income. As described in Section VII.A.4.c.3.b.i, total
downstream (tailpipe) emissions of criteria and toxic air pollutants
for cars and light trucks in 2035 are projected to remain relatively
unchanged or decrease under all action alternatives compared to the No
Action Alternative, with the following exceptions: total downstream
emissions of SO2 would increase under all action
alternatives under both the CAFE and CO2 programs; total
downstream emissions of acrolein would increase under Alternatives 5,
6, and 7 under the CAFE program (but not under the CO2
program); and total downstream emissions of acetaldehyde and butadiene
would increase under Alternatives 6 and 7 under the CAFE program (but
not under the CO2 program). To the extent minority and low-
income populations disproportionately live or attend schools near major
roadways, these populations may be more likely to be affected by these
emissions. However, because some pollutant emissions are expected to
decrease and others are expected to increase, health impacts are mixed.
Overall, as the magnitude of the emissions changes is anticipated to be
minor compared to total tailpipe emissions for these vehicles, the
impacts to minority or low-income populations are not considered high
and adverse.
The agencies used the standards that were discussed in the 2012
rulemaking as the baseline for this rulemaking. Therefore, the agencies
project increases in certain air pollutants for purposes of this
analysis. However, as discussed above, one impact of the standards
finalized in this rulemaking is to reduce the up-front cost of new and
used vehicles. Low income populations may benefit most from the
reduction in cost of acquiring newer vehicles, which generally are more
fuel efficient and have lower air pollutant emissions than older
vehicles. This cost reduction may have the effect of encouraging the
quicker adoption of cleaner vehicles in low income communities, which
could result in air quality and health benefits for those who live or
attend school in proximity to the roadways where they are operated. To
the degree to which minority populations may also live in proximity to
these roadways, they would also experience benefits, thereby mitigating
the disparity in racial, ethnic, and economically based exposures.
c) Other Vulnerabilities to Climate Change and Health Impacts of Air
Pollutants
Some areas most vulnerable to climate change tend to have a higher
concentration of minority and low-income populations, potentially
putting these communities at higher risk from climate variability and
climate-related extreme weather events.\3577\ For example, urban areas
tend to have pronounced social inequities that could result in
disproportionately larger minority and low-income populations than
those in the surrounding nonurban areas.\3578\ Urban areas are also
subject to the most substantial temperature increases from climate
change because of the urban heat island
effect.3579 3580 3581 Taken together, these tendencies
demonstrate a potential for disproportionate impacts on minority and
low-income populations in urban areas. Low-income populations in
coastal urban areas, which are vulnerable to increases in flooding as a
result of projected sea-level rise, larger storm surges, and human
settlement in floodplains, could also be disproportionately affected by
climate change because they are less likely to have the means to
evacuate quickly in the event of a natural disaster and, therefore, are
at greater risk of injury and loss of life.3582 3583
---------------------------------------------------------------------------
\3577\ U.S. Global Change Research Program (GCRP). Global
Climate Change Impacts in the United States: The Third National
Climate Assessment. U.S. Global Change Research Program. Melillo,
J.M, T.C. Richmond, and G.W. Yohe (Eds.)]. U.S. Government Printing
Office: Washington, DC 841 pp (2014). doi:10.7930/J0Z31WJ2.
Available at: http://nca2014.globalchange.gov/report. (Accessed:
February 27, 2018).
\3578\ GCRP (2014).
\3579\ GCRP (2014).
\3580\ Knowlton, K., B. Lynn, R.A. Goldberg, C. Rosenzweig, C.
Hogrefe, J.K. Rosenthal, and P.L. Kinney. Projecting Heat-related
Mortality Impacts under a Changing Climate in the New York City
Region. American Journal of Public Health 97(11):2028-34 (2007).
doi:10.2105/AJPH.2006.102947. Available in: http://ajph.aphapublications.org/cgi/content/full/97/11/2028. Last
accessed: March 4, 2018.
\3581\ EPA. Heat Island Effect. U.S. Environmental Protection
Agency (2017). Last revised: February 20, 2018. Available at:
https://www.epa.gov/heat-islands. (Accessed: February 28, 2018.).
\3582\ GCRP. Global Climate Impacts in the United States (2009).
Cambridge, United Kingdom and New York, NY, USA. Karl, T.R., J.M.
Melillo, and T.C. Peterson (Eds.). Cambridge University Press:
Cambridge, UK. pp. 196.
\3583\ GCRP (2014).
---------------------------------------------------------------------------
Independent of their proximity to pollution sources or climate
change, locations of potentially high impact, minority and low-income
populations could be more vulnerable to the health impacts of
pollutants and climate change. Reports from the U.S. Department of
Health and Human Services have stated that minority and low-income
populations tend to have less access to health care services, and the
services received are more likely to suffer with respect to
quality.3584 3585 3586 Other studies show that low
socioeconomic position can modify the health effects of air pollution,
with higher effects observed in groups with lower socioeconomic
position.3587 3588 Possible explanations for this
observation include that low socioeconomic position groups may be
differentially exposed to air pollution or may be differentially
vulnerable to effects of exposure.\3589\
---------------------------------------------------------------------------
\3584\ U.S. Department of Health and Human Services (HHS).
National Healthcare Disparities Report. U.S. Department of Health
and Human Service. Rockville, MD, Agency for Healthcare Research and
Quality (2003). Available at: http://archive.ahrq.gov/qual/nhdr03/nhdr03.htm. (Accessed: March 3, 2018).
\3585\ HHS. Minority Health: Recent Findings. Agency for
Healthcare Research Quality (2013). Last revised: February 2013.
Available at: https://www.ahrq.gov/research/findings/factsheets/minority/minorfind/index.html. (Accessed: March 3, 2018).
\3586\ HHS. 2016 National Healthcare Disparities Report. U.S.
Department of Health and Human Service (2017). Rockville, MD. Agency
for Healthcare Research and Quality. Available at: https://www.ahrq.gov/research/findings/nhqrdr/nhqdr16/summary.html.
(Accessed: September 20, 2017).
\3587\ O'Neill, M.S., M. Jerrett, I. Kawachi, J.I. Levy, A.J.
Cohen, N. Gouveia, P. Wilkinson, T. Fletcher, L. Cifuentes, and J.
Schwartz. Health, Wealth, and Air Pollution: Advancing Theory and
Methods. Environmental Health Perspectives 111(16):1861-70 (2003).
doi: 10.1289/ehp.6334. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1241758/pdf/ehp0111-001861.pdf. (Accessed: February
24, 2019).
\3588\ Finkelstein, M.M.; Jerrett, M.; DeLuca, P.; Finkelstein,
N.; Verma, D.K.; Chapman, K.; Sears, M.R. Relation between income,
air pollution and mortality: a cohort study. Canadian Med Assn J
169: 397-402 (2003).
\3589\ O'Neill et al. (2003).
---------------------------------------------------------------------------
In terms of climate change, increases in heat-related morbidity and
mortality because of higher overall and extreme temperatures are likely
to affect minority and low-income populations disproportionately,
partially because of limited access to air conditioning and high energy
costs.3590 3591 3592 3593 Native
[[Page 25262]]
American tribes and Alaskan Native villages are also more susceptible
to the impacts of climate change, as these groups often
disproportionately rely on natural resources for livelihoods,
medicines, and cultural and spiritual purposes.\3594\ Moreover, coastal
tribal communities may have to relocate because of sea-level rise,
erosion, and permafrost thaw.\3595\ NHTSA's FEIS provides additional
discussion of health and societal impacts of climate change on
indigenous communities in Section 8.6.5.2, Sectoral Impacts of Climate
Change, under Human Health and Human Security.
---------------------------------------------------------------------------
\3590\ EPA. 2009. Technical Support Document for Endangerment
and Cause or Contribute Findings for Greenhouse Gases under Section
202(a) of the Clean Air Act. December 7, 2009. U.S. Environmental
Protection Agency, Office of Atmospheric Programs, Climate Change
Division: Washington, DC Available at: https://www.epa.gov/sites/production/files/2016-08/document/endangerment_tsd.pdf. (Accessed:
February 28, 2018).
\3591\ O'Neill, M.S., A. Zanobetti, and J. Schwartz. Disparities
by Race in Heat-Related Mortality in Four US Cities: The Role of Air
Conditioning Prevalence. Journal of Urban Health 82(2):191-97
(2005). doi:10.1093/jurban/jti043. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3456567/pdf/11524_2006_Article_375.pdf. (Accessed: March 4, 2018).
\3592\ GCRP (2014).
\3593\ Harlan, S.L. and D.M. Ruddell. Climate Change and Health
in Cities: Impacts of Heat and Air Pollution and Potential Co-
Benefits from Mitigation and Adaptation. Current Opinion in
Environmental Sustainability 3(3):126-34 (2011). doi: 10.1016/
j.cosust.2011.01.001.
\3594\ National Tribal Air Association. 2009. Impacts of climate
change on Tribes in the United States. Submitted December 11, 2009
to Assistant Administrator Gina McCarthy, USEPA, Office of Air and
Radiation. Available at: http://www.epa.gov/air/tribal/pdfs/Impacts%20of%20Climate%20Change%20on%20Tribes%20in%20the%20United%20States.pdf. Last accessed: February 24, 2019.
\3595\ Maldonado, J., C. Shearer, R. Bronen, K. Peterson, and H.
Lazrus. The Impact of Climate Change on Tribal Communities in the
US: Displacement, Relocation, and Human Rights. Climatic Change
120(3):601-14 (2013).
---------------------------------------------------------------------------
Together, this information indicates that the same set of potential
environmental effects (e.g., air pollutants, heat increases, and sea-
level rise) may disproportionately affect minority and low-income
populations because of socioeconomic circumstances or histories of
discrimination and inequity.
As described in Chapter 5 of NHTSA's FEIS, the action alternatives
are projected to increase CO2 emissions from passenger cars
and light trucks by 4 to 10 percent by 2100 compared to the No Action
Alternative. Impacts of climate change could disproportionately affect
minority and low-income populations in urban areas that are subject to
the most substantial temperature increases from climate change. These
impacts are largely because of the urban heat island effect.
Additionally, minority and low-income populations that live in flood-
prone coastal areas could be disproportionately affected. However, the
contribution of the action alternatives to climate change impacts would
be very minor rather than high and adverse. Compared to the annual U.S.
CO2 emissions of 7,193 MMTCO2e from all sources
by the end of the century projected by the GCAM Reference scenario, the
action alternatives are projected to increase annual U.S.
CO2 emissions by 0.4 to 1.2 percent in 2100. Compared to
annual global CO2 emissions, the action alternatives would
represent an even smaller percentage increase and ultimately, by 2100,
are projected to result in percentage increases in global mean surface
temperature, atmospheric CO2 concentrations, and sea level,
and decreases in ocean pH, ranging from 0.09 percent to less than 0.01
percent. Any impacts of this rulemaking on low-income and minority
communities would be attenuated by a lengthy causal chain; but if one
could attempt to draw those links, the changes to climate values would
be very small and incremental compared to the expected changes
associated with the emissions trajectories in the GCAM Reference
scenario.
As reported in Section VII.A.4.c.3.c above, adverse health impacts
over the lifetimes of vehicles through MY 2029 are projected to
increase nationwide under each of the action alternatives (except
Alternative 6 and Alternative 7 under the CAFE program, which show
decreases) compared to the No Action Alternative. Increases in these
pollutant emissions, however, would be primarily the result of
increases in upstream emissions (emissions near refineries, power
plants, and extraction sites), while downstream emissions (tailpipe
emissions near roadways) are anticipated to decrease or increase by
smaller amounts. The health impacts reported in that section occur over
a long period of time, would be incremental in magnitude, and would not
be characterized as high. Those impacts would also be borne nationwide,
so impacts to minority and low-income populations would be smaller.
d) Conclusion
Based on the foregoing, the agencies have determined that this
rulemaking (and alternatives considered) would not result in
disproportionately high and adverse human health or environmental
effects on minority or low-income populations. This rulemaking would
set standards nationwide, and although minority and low-income
populations may experience some disproportionate effects, in particular
locations, the overall impacts on human health and the environment
would not be ``high and adverse'' under E.O. 12898.
Furthermore, the agencies note that there are no mitigation
measures or alternatives available as part of this action that could
fulfill the respective statutory missions of the agencies and that
would address the considerations discussed in Section VIII (e.g.,
economic practicability) or avoid or reduce any disproportionate
effects in particular locations experienced by minority and low-income
populations. The impacts described in this analysis would result from
air pollutant and CO2 emissions that may occur from the
levels of stringency selected by the agencies. However, for the reasons
described in Section VIII, the agencies cannot select a higher level of
stringency. While the agencies have considered the potential impacts
described in this analysis, there is a substantial need, based on the
overall public interest, to address the costs associated with the
standards discussed in the 2012 rulemaking. More stringent alternatives
would have severe adverse social and economic costs, as described in
Section VIII, and necessitate the level of standards finalized in this
rulemaking.
12. Executive Order 13045: ``Protection of Children From Environmental
Health Risks and Safety Risks''
This action is subject to E.O. 13045 (62 FR 19885, April 23, 1997)
because it is an economically significant regulatory action as defined
by E.O. 12866, and the agencies have reason to believe that the
environmental health or safety risks related to this action may have a
disproportionate effect on children. Specifically, children are more
vulnerable to adverse health effects related to mobile source
emissions, as well as to the potential long-term impacts of climate
change. Pursuant to E.O. 13045, NHTSA and EPA must prepare an
evaluation of the environmental health or safety effects of the planned
regulation on children and an explanation of why the planned regulation
is preferable to other potentially effective and reasonably feasible
alternatives considered by the agencies. Further, this analysis may be
included as part of any other required analysis.
This preamble and NHTSA's Final EIS discuss air quality, climate
change, and their related environmental and health effects, noting
where these would disproportionately affect children. The EPA
Administrator has also discussed the impact of climate-related health
effects on children in the Endangerment and Cause or Contribute
Findings for
[[Page 25263]]
Greenhouse Gases Under Section 202(a) of the Clean Air Act (74 FR
66496, December 15, 2009). In addition, this preamble explains why the
agencies' final standards are preferable to other alternatives
considered. Together, this preamble and NHTSA's Final EIS satisfy the
agencies' responsibilities under E.O. 13045.
F. Regulatory Flexibility Act
Pursuant to the Regulatory Flexibility Act (5 U.S.C. 601 et seq.,
as amended by the Small Business Regulatory Enforcement Fairness Act
(SBREFA) of 1996), whenever an agency is required to publish a notice
of proposed rulemaking or final rule, it must prepare and make
available for public comment a regulatory flexibility analysis that
describes the effect of the rule on small entities (i.e., small
businesses, small organizations, and small governmental jurisdictions).
No regulatory flexibility analysis is required if the head of an agency
certifies the rule will not have a significant economic impact on a
substantial number of small entities. SBREFA amended the Regulatory
Flexibility Act to require Federal agencies to provide a statement of
the factual basis for certifying that a rule will not have a
significant economic impact on a substantial number of small entities.
Two comments argued that the agencies should prepare a regulatory
flexibility analysis and convene a small business review panel to
assess the impacts in accordance with the Regulatory Flexibility Act, 5
U.S.C. 601 et seq., as amended by SBREFA.\3596\ The agencies considered
these comments and the impacts of this rule under the Regulatory
Flexibility Act and certify that this rule will not have a significant
economic impact on a substantial number of small entities. The
following is the agencies' statement providing the factual basis for
this certification pursuant to 5 U.S.C. 605(b).
---------------------------------------------------------------------------
\3596\ See National Coalition for Advanced Transportation (NCAT)
Comment, Docket No. NHTSA-2018-0067-11969, at 64-65; Workhorse
Group, Inc. Comment, Docket No. NHTSA-2018-0067-12215, at 1-2.
---------------------------------------------------------------------------
Small businesses are defined based on the North American Industry
Classification System (NAICS) code.\3597\ One of the criteria for
determining size is the number of employees in the firm. For
establishments primarily engaged in manufacturing or assembling
automobiles, as well as light duty trucks, the firm must have less than
1,500 employees to be classified as a small business. This rule would
affect motor vehicle manufacturers. As shown in Table X-1, the agencies
have identified 15 small manufacturers of passenger cars, light trucks,
and SUVs of electric, hybrid, and internal combustion engines.\3598\
The agencies acknowledge that some newer manufacturers may not be
listed. However, those new manufacturers tend to have transportation
products that are not part of the light-duty vehicle fleet and have yet
to start production of light-duty vehicles. Moreover, NHTSA does not
believe that there are a ``substantial number'' of these newer
companies.\3599\
---------------------------------------------------------------------------
\3597\ Classified in NAICS under Subsector 336--Transportation
Equipment Manufacturing for Automobile Manufacturing (336111), Light
Truck (336112), and Heavy Duty Truck Manufacturing (336120). https://www.sba.gov/document/support-table-size-standards.
\3598\ Two comments pointed out that Workhorse Group Inc. was
not listed as a small domestic vehicle manufacturer in Table XII-1
of the proposal. See National Coalition for Advanced Transportation
(NCAT) Comment, Docket No. NHTSA-2018-0067-11969, at 64-65;
Workhorse Group, Inc. Comment, Docket No. NHTSA-2018-0067-12215, at
1-2. Workhorse Group has been added to the table here, but neither
its addition nor the existence of a small number of other new small
manufacturers does not alter the conclusion that this rule will not
have a significant economic impact on a substantial number of small
entities.
\3599\ 5 U.S.C. 605(b).
[GRAPHIC] [TIFF OMITTED] TR30AP20.758
[[Page 25264]]
NHTSA believes that the rulemaking would not have a significant
economic impact on the small vehicle manufacturers because under 49 CFR
part 525, passenger car manufacturers making less than 10,000 vehicles
per year can petition NHTSA to have alternative standards set for those
manufacturers. These manufacturers do not currently meet the 27.5 mpg
standard and must already petition the agency for relief. If the
standard is raised, it has no meaningful impact on these
manufacturers--they still must go through the same process and petition
for relief. Given there already is a mechanism for relieving burden on
small businesses, which is the purpose of the Regulatory Flexibility
Act, a regulatory flexibility analysis was not prepared.
---------------------------------------------------------------------------
\3600\ Estimated number of employees as of 2018, source:
Linkedin.com.
\3601\ Rough estimate of light duty vehicle production for model
year 2017.
---------------------------------------------------------------------------
Two comments argued that small manufacturers of electric vehicles
would face a significant economic impact because their ability to earn
credits would be ``substantially diminished.'' \3602\ The method for
earning credits applies equally across manufacturers and does not place
small entities at a significant competitive disadvantage. In any event,
even if the rule had a ``significant economic impact'' on these small
EV manufacturers, the amount of these companies is not ``a substantial
number.'' \3603\ For these reasons, their existence does not alter the
agencies' analysis of the applicability of the Regulatory Flexibility
Act. EPA believes this rulemaking would not have a significant economic
impact on a substantial number of small entities under the Regulatory
Flexibility Act, as amended by the Small Business Regulatory
Enforcement Fairness Act. EPA is exempting from the CO2
standards any manufacturer, domestic or foreign, meeting SBA's size
definitions of small business as described in 13 CFR 121.201. EPA
adopted the same type of exemption for small businesses in the 2017 and
later rulemaking. EPA estimates that small entities comprise less than
0.1 percent of total annual vehicle sales and exempting them will have
a negligible impact on the CO2 emissions reductions from the
standards. Because EPA is exempting small businesses from the
CO2 standards, the agency certifies that the rule will not
have a significant economic impact on a substantial number of small
entities. Therefore, EPA has not conducted a Regulatory Flexibility
Analysis or a SBREFA SBAR Panel for the rule.
---------------------------------------------------------------------------
\3602\ National Coalition for Advanced Transportation (NCAT)
Comment, Docket No. NHTSA-2018-0067-11969, at 65; Workhorse Group,
Inc. Comment, Docket No. NHTSA-2018-0067-12215, at 2.
\3603\ 5 U.S.C. 605.
---------------------------------------------------------------------------
EPA regulations allow small businesses voluntarily to waive their
small business exemption and optionally to certify to the
CO2 standards. This option allows small entity manufacturers
to earn CO2 credits under the CO2 program, if
their actual fleetwide CO2 performance is better than their
fleetwide CO2 target standard. However, the exemption waiver
is optional for small entities and thus the agency believes that
manufacturers opt into the CO2 program if it is economically
advantageous for them to do so, for example in order to generate and
sell CO2 credits. Therefore, EPA believes this voluntary
option does not affect EPA's determination that the standards will
impose no significant adverse impact on small entities.
G. Executive Order 13132 (Federalism)
Executive Order 13132 requires Federal agencies to develop an
accountable process to ensure ``meaningful and timely input by State
and local officials in the development of regulatory policies that have
federalism implications.'' The Order defines the term ``[p]olicies that
have federalism implications'' to include regulations that have
``substantial direct effects on the States, on the relationship between
the national government and the States, or on the distribution of power
and responsibilities among the various levels of government.'' Under
the Order, agencies may not issue a regulation that has federalism
implications, that imposes substantial direct compliance costs, unless
the Federal government provides the funds necessary to pay the direct
compliance costs incurred by State and local governments, or the
agencies consult with State and local officials early in the process of
developing the proposed regulation. The agencies complied with the
Order's requirements.
NHTSA also addressed the federalism implications of its proposal in
The Safer Affordable Fuel-Efficient Vehicles Rule Part One: One
National Program final rulemaking.\3604\
---------------------------------------------------------------------------
\3604\ 84 FR 51310 (Sep. 27, 2019).
---------------------------------------------------------------------------
H. Executive Order 12988 (Civil Justice Reform)
Pursuant to Executive Order 12988, ``Civil Justice Reform,'' \3605\
NHTSA has considered whether this rulemaking would have any retroactive
effect. This proposed rule does not have any retroactive effect.
---------------------------------------------------------------------------
\3605\ 61 FR 4729 (Feb. 7, 1996).
---------------------------------------------------------------------------
I. Executive Order 13175 (Consultation and Coordination With Indian
Tribal Governments)
This final rule does not have tribal implications, as specified in
Executive Order 13175 (65 FR 67249, November 9, 2000). This rule will
be implemented at the Federal level and impose compliance costs only on
vehicle manufacturers. Thus, Executive Order 13175 does not apply to
this rule. Some comments complained that the agencies have not
consulted or coordinated with Native American communities and Indian
Tribes in promulgating this rule.\3606\ Executive Order 13175 requires
consultation with Tribal officials when agencies are developing
policies that have ``substantial direct effects'' on Tribes and Tribal
interests.\3607\ Even accepting the comments' description of the
effects of the rule, they have identified only indirect effects of the
standards on Tribal interests.\3608\
---------------------------------------------------------------------------
\3606\ See, e.g., CARB Comment, Docket No. NHTSA-2018-0067-
11873, at 412; National Tribal Air Association Comment, Docket No.
NHTSA-2018-0067-11948, at 4; Keweenaw Bay Indian Community Comment,
Docket No. EPA-HQ-OAR-2018-0283-3325, at 1-2; Fond du Lac Band of
Lake Superior Chippewa Comment, Docket No. EPA-HQ-OAR-2018-0283-
4030, at 3; Sac and Fox Nation, Docket No. EPA-HQ-OAR-2018-0283-
4159, at 4-5; The Leech Lake Band of Ojibwe Comment, Docket No. EPA-
HQ-OAR-2018-0283-5931, at 4-5.
\3607\ 65 FR 67249, 67249 (Nov. 6, 2000).
\3608\ See, e.g., National Tribal Air Association Comment,
Docket No. NHTSA-2018-0067-11948, at 4.
---------------------------------------------------------------------------
J. Unfunded Mandates Reform Act
Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA)
requires Federal agencies to prepare a written assessment of the costs,
benefits, and other effects of a proposed or final rule that includes a
Federal mandate likely to result in the expenditure by State, local, or
Tribal governments, in the aggregate, or by the private sector, of more
than $100 million in any one year (adjusted for inflation with base
year of 1995). Adjusting this amount by the implicit gross domestic
product price deflator for 2016 results in $148 million (111.416/75.324
= 1.48).\3609\ Before promulgating a rule for which a written statement
is needed, section 205 of UMRA generally requires NHTSA and EPA to
identify and consider a reasonable number of regulatory
[[Page 25265]]
alternatives and adopt the least costly, most cost-effective, or least
burdensome alternative that achieves the objective of the rule. The
provisions of section 205 do not apply when they are inconsistent with
applicable law. Moreover, section 205 allows NHTSA and EPA to adopt an
alternative other than the least costly, most cost-effective, or least
burdensome alternative if the agency publishes with the rule an
explanation of why that alternative was not adopted.
---------------------------------------------------------------------------
\3609\ Bureau of Economic Analysis, National Income and Product
Accounts (NIPA), Table 1.1.9 Implicit Price Deflators for Gross
Domestic Product. https://bea.gov/iTable/index_nipa.cfm.
---------------------------------------------------------------------------
This rule will not result in the expenditure by State, local, or
Tribal governments, in the aggregate, of more than $148 million
annually, but it will result in the expenditure of that magnitude by
vehicle manufacturers and/or their suppliers. In developing this rule,
NHTSA and EPA considered a variety of alternative average fuel economy
standards lower and higher than those previously proposed. The fuel
economy standards for MYs 2021-2026 are the least costly, most cost-
effective, and least burdensome alternative that achieve the objectives
of the rule.
K. Regulation Identifier Number
The Department of Transportation assigns a regulation identifier
number (RIN) to each regulatory action listed in the Unified Agenda of
Federal Regulations. The Regulatory Information Service Center
publishes the Unified Agenda in April and October of each year. The RIN
contained in the heading at the beginning of this document may be used
to find this action in the Unified Agenda.
L. National Technology Transfer and Advancement Act
Section 12(d) of the National Technology Transfer and Advancement
Act (NTTAA) requires NHTSA and EPA to evaluate and use existing
voluntary consensus standards in its regulatory activities unless doing
so would be inconsistent with applicable law (e.g., the statutory
provisions regarding NHTSA's vehicle safety authority, or EPA's testing
authority) or otherwise impractical.\3610\
---------------------------------------------------------------------------
\3610\ 15 U.S.C. 272.
---------------------------------------------------------------------------
Voluntary consensus standards are technical standards developed or
adopted by voluntary consensus standards bodies. Technical standards
are defined by the NTTAA as ``performance-based or design-specific
technical specification and related management systems practices.''
They pertain to ``products and processes, such as size, strength, or
technical performance of a product, process or material.''
Examples of organizations generally regarded as voluntary consensus
standards bodies include the American Society for Testing and Materials
(ASTM), the Society of Automotive Engineers (SAE), and the American
National Standards Institute (ANSI). If the agencies do not use
available and potentially applicable voluntary consensus standards,
they are required by the Act to provide Congress, through OMB, an
explanation of the reasons for not using such standards.
For CO2 emissions, EPA will collect data over the same
tests that are used for the MY 2012-2016 CO2 standards and
for the CAFE program. This unified data collection will minimize the
amount of testing done by manufacturers because manufacturers are
already required to run these tests. For A/C credits, EPA will use a
consensus methodology developed by the Society of Automotive Engineers
(SAE) and also a new A/C test. EPA knows of no consensus standard
available for the A/C test.
There are currently no voluntary consensus standards that NHTSA
administers relevant to today's CAFE standards.
M. Department of Energy Review
In accordance with 49 U.S.C. 32902(j)(2), NHTSA submitted this rule
to the Department of Energy for review.
N. Paperwork Reduction Act
The Paperwork Reduction Act (PRA) of 1995, Public Law 104-13,\3611\
gives OMB authority to regulate matters regarding the collection,
management, storage, and dissemination of certain information by and
for the Federal government. It seeks to reduce the total amount of
paperwork handled by the government and the public. NHTSA strives to
reduce the public's information collection burden hours each fiscal
year by streamlining external and internal processes.
---------------------------------------------------------------------------
\3611\ Codified at 44 U.S.C. 3501 et seq.
---------------------------------------------------------------------------
To this end, NHTSA will continue to collect information to ensure
compliance with its CAFE program. NHTSA will reinstate its previously-
approved collection of information for Corporate Average Fuel Economy
(CAFE) reports specified in 49 CFR part 537 (OMB control number 2127-
0019), add the additional burden for reporting changes adopted in the
October 15, 2012 final rule that recently came into effect (see 77 FR
62623), and account for the change in burden in this rule as well as
for other CAFE reporting provisions required by Congress and NHTSA.
NHTSA is also changing the name of this collection to represent more
accurately the breadth of all CAFE regulatory reporting. Although NHTSA
is adding additional burden hours to its CAFE report requirement in 49
CFR 537, the agency believes there will be a reduction in the overall
paperwork burden due to the standardization of data and the streamlined
process.
In compliance with the PRA, the information collection request
(ICR) abstracted below was forwarded to OMB for review and comment. The
ICR describes the nature of the information collection and its expected
burden.
Title: Corporate Average Fuel Economy.
Type of Request: Reinstatement and amendment of a previously
approved collection.
OMB Control Number: 2127-0019.
Form Numbers: NHTSA Form 1474 (CAFE Projections Reporting Template)
and NHTSA Form 1475 (CAFE Credit Template).
Requested Expiration Date of Approval: Three years from date of
approval.
Summary of the collection of information: As part of this
rulemaking, NHTSA is reinstating and modifying its previously-approved
collection for CAFE-related collections of information. NHTSA and EPA
have coordinated their compliance and reporting requirements in an
effort not to impose duplicative burdens on regulated entities. This
information collection contains three different components: Burden
related to NHTSA's CAFE reporting requirements; burden related to CAFE
compliance, but not via reporting requirements; and information
gathered by NHTSA to help inform CAFE analyses. All templates
referenced in this section will be available in the rulemaking docket
and the NHTSA public information center.\3612\
---------------------------------------------------------------------------
\3612\ https://one.nhtsa.gov/cafe_pic/CAFE_PIC_Home.htm.
---------------------------------------------------------------------------
CAFE Compliance Reports
NHTSA is reinstating \3613\ its collection related to the reporting
requirements in 49 U.S.C. 32907, ``Reports and tests of
manufacturers.'' In that section, manufacturers are statutorily
required to submit CAFE compliance reports to the Secretary of
Transportation.\3614\ The reports must state if a manufacturer will
comply with its applicable fuel economy standard(s), describe what
actions the manufacturer
[[Page 25266]]
intends to take to comply with the standard(s), and include other
information as required by NHTSA. Manufacturers are required to submit
two CAFE compliance reports--a pre-model year report (PMY) and a mid-
model year (MMY) report--each year. In the event a manufacturer needs
to correct previously-submitted information, a manufacturer may need to
file additional reports.\3615\
---------------------------------------------------------------------------
\3613\ This collection expired on April 30, 2016.
\3614\ 49 U.S.C. 32907 (delegated to the NHTSA Administrator at
49 CFR 1.95). Because of this delegation, for purposes of
discussion, statutory references to the Secretary of Transportation
in this section will be discussed in terms of NHTSA or the NHTSA
Administrator.
\3615\ Specifically, a manufacturer shall submit a report
containing the information during the 30 days before the beginning
of each model year, and during the 30 days beginning the 180th day
of the model year. When a manufacturer decides that actions reported
are not sufficient to ensure compliance with that standard, the
manufacturer shall report additional actions it intends to take to
comply with the standard and include a statement about whether those
actions are sufficient to ensure compliance.
---------------------------------------------------------------------------
To implement this statute, NHTSA issued 49 CFR part 537,
``Automotive Fuel Economy Reports,'' which adds additional definition
to the terms of section 32907. The first report, the PMY report must be
submitted to NHTSA before December 31 of the calendar year prior to the
corresponding model year and contain manufacturers' projected
information for that upcoming model year. The second report, the MMY
report must be submitted by July 31 of the given model year and contain
updated information from manufacturers based on actual and projected
information known midway through the model year. Finally, the last
report, a supplementary report, is required to be submitted anytime a
manufacturer needs to correct information previously submitted to
NHTSA.
Compliance reports must include information on passenger and non-
passenger automobiles (trucks) describing the projected and actual fuel
economy standards, fuel economy performance values, production sales
volumes and information on vehicle design features (e.g., engine
displacement and transmission class) and other vehicle attribute
characteristics (e.g., track width, wheel base, and other light truck
off-road features). Manufacturers submit confidential and non-
confidential versions of these reports to NHTSA. Confidential reports
differ by including estimated or actual production sales information,
which is withheld from public disclosure to protect each manufacturer's
competitive sales strategies. NHTSA uses the reports as the basis for
vehicle auditing and testing, which helps manufacturers correct
reporting errors prior to the end of the model year and facilitate
acceptance of their final CAFE report by the Environmental Protection
Agency (EPA). The reports also help the agency, as well as the
manufacturers who prepare them, anticipate potential compliance issues
as early as possible, and help manufacturers plan their compliance
strategies.
Further, NHTSA is modifying this collection to account for
additional information manufacturers are required to include in their
reports. In the CAFE standards previously promulgated for MY 2017 and
beyond,\3616\ NHTSA allowed for manufacturers to gain additional fuel
economy benefits by installing certain technologies on their vehicles
beginning with MY 2017.\3617\ These technologies include air-
conditioning systems with increased efficiency, off-cycle technologies
whose benefits are not adequately captured on the Federal Test
Procedure and/or the Highway Fuel Economy Test,\3618\ and hybrid
electric technologies installed on full-size pickup trucks. Prior to MY
2017, manufacturers were unable to earn a fuel economy benefit for
these technologies, so NHTSA's reporting requirements did not include
an opportunity to report them. Now, manufacturers must provide
information on these technologies in their CAFE reports. NHTSA requires
manufacturers to provide detailed information on the model types using
these technologies to gain fuel economy benefits. These details are
necessary to facilitate NHTSA's technical analyses and to ensure the
agency can perform random enforcement audits when necessary.
---------------------------------------------------------------------------
\3616\ 77 FR 62623 (Oct. 15, 2012).
\3617\ These technologies were not included in the burden for
part 537 at the time as the additional reporting requirements would
not take effect until years later.
\3618\ E.g., engine idle stop-start systems, active transmission
warmup systems, etc.
---------------------------------------------------------------------------
In addition to a list of all fuel consumption improvement
technologies utilized in their fleet, 49 CFR 537 requires manufacturers
to report the make, model type, compliance category, and production
volume of each vehicle equipped with each technology and the associated
fuel consumption improvement value (FCIV). NHTSA is adding the
reporting and enforcement burden hours and cost for these new
incentives to this collection. Manufacturers can also petition the EPA
and NHTSA, in accordance with 40 CFR 86.1868-12 or 40 CFR 86.1869-12,
to gain additional credits based upon the improved performance of any
of the new incentivized technologies allowed starting in model year
2017. EPA approves these petitions in collaboration with NHTSA and any
adjustments are taken into account for both programs. As a part the
agencies' coordination, NHTSA provides EPA with an evaluation of each
new technology to ensure its direct impact on fuel economy and an
assessment on the suitability of each technology for use in increasing
a manufacturer's fuel economy performance. Furthermore, at times, NHTSA
may independently request additional information from a manufacturer to
support its evaluations. This information along with any research
conclusions shared with EPA and NHTSA in the petitions is required to
be submitted in manufacturer's CAFE reports.
NHTSA is also changing the burden hours for its CAFE reporting
requirements in 49 CFR part 537 by adjusting the total amount of time
spent collecting the required reporting information through the use of
a standardized reporting template to streamline the collection process.
The standardized template will be used by manufacturers to collect all
the required CAFE information under 49 CFR 537.7(b) and (c) and
provides a format which ensures accuracy, completeness, and better
alignment with the final data provided to EPA.
2. Other CAFE Compliance Collections
NHTSA is adopting a new standardized template for manufacturers
buying CAFE credits and for manufacturers submitting credit
transactions in accordance with 49 CFR part 536. In 49 CFR part
536.5(d), NHTSA is required to assess compliance with fuel economy
standards each year, utilizing the certified and reported CAFE data
provided by the EPA for enforcement of the CAFE program pursuant to 49
U.S.C. 32904(e). Credit values are calculated based on the CAFE data
from the EPA. If a manufacturer's vehicles in a particular compliance
category performs better than its required fuel economy standard, NHTSA
adds credits to the manufacturer's account for that compliance
category. If a manufacturer's vehicles in a particular compliance
category perform worse than the required fuel economy standard, NHTSA
will add a credit deficit to the manufacturer's account and will
provide written notification to the manufacturer concerning its failure
to comply. The manufacturer will be required to confirm the shortfall
and must either: Submit a plan indicating how it will allocate existing
credits or earn, transfer, and/or acquire credits or pay the equivalent
civil penalty. The manufacturer must submit a plan or
[[Page 25267]]
payment within 60 days of receiving notification from NHTSA.
Manufacturers should use the credit transaction template any time a
credit transaction request is sent to NHTSA. For example, manufacturers
that purchase credits and want to apply them to their credit accounts
will use the credit transaction template. The template NHTSA is
adopting is a simple spreadsheet that credit entities fill out. When
completed, credit entities will have an organized list of credit
transactions and will be able to click a button on the spreadsheet to
generate a joint transaction letter for trading parties to sign and
submit to NHTSA, along with the spreadsheet. Entities trading credits
are also required to provide to NHTSA all the confidential information
associated with the monetary and non-monetary price of credit trades.
NHTSA believes these changes will significantly reduce the burden on
manufacturers in managing their CAFE credit accounts and provide better
oversight of the CAFE credit program for NHTSA.
Finally, NHTSA is accounting for the additional burden due to
existing CAFE program elements. In 49 CFR part 525, small volume
manufacturers submit petitions to NHTSA for exemption from an
applicable average fuel economy standard and to request to comply with
a less stringent alternative average fuel economy standard. In 49 CFR
part 534, manufacturers are required to submit information to NHTSA
when establishing a corporate controlled relationship with another
manufacturer. A controlled relationship exists between manufacturers
that control, are controlled by, or are under common control with, one
or more other manufacturers. Accordingly, manufacturers that have
entered into written contracts transferring rights and responsibilities
to other manufacturers in controlled relationships for CAFE purposes
are required to provide reports to NHTSA. There are additional
reporting requirements for manufacturers submitting carry back plans
and when manufacturers split apart from controlled relationships and
must designate how credits are to be allocated between the
parties.\3619\ Manufacturers with credit deficits at the end of the
model year, can carry back future earned credits up to three model
years in advance of the deficit to resolve a current shortfall. The
carryback plan proving the existence of a manufacturer's future earned
credits must be submitted and approved by NHTSA, pursuant to 49 U.S.C.
32903(b).
---------------------------------------------------------------------------
\3619\ See 49 CFR part 536.
---------------------------------------------------------------------------
3. Analysis Fleet Composition
As discussed in Section VI.B, in setting CAFE standards, NHTSA
creates an analysis fleet from which to model potential future economy
improvements. To compose this fleet, the agency uses a mixture of
compliance data and information from other sources to replicate more
closely the fleet from a recent model year. While refining the analysis
fleet, NHTSA occasionally asks manufacturers for information that is
similar to information submitted as part of EPA's final model year
report (e.g., final model year vehicle volumes). Periodically, NHTSA
may ask manufacturers for more detailed information than what is
required for compliance (e.g., what engines are shared across vehicle
models). Often, NHTSA requests this information from manufacturers
after manufacturers have submitted their final model year reports to
EPA, but before EPA processes and releases final model year reports.
Information like this, which is used to verify and supplement the
data used to create the analysis fleet, is tremendously valuable to
generating an accurate analysis fleet, and setting maximum feasible
standards. The more accurate the analysis fleet is, the more accurate
the modeling of what technologies could be applied will be. Therefore,
NHTSA is accounting for the burden on manufacturers to provide the
agency with this additional information. In almost all instances,
manufacturers already have the information NHTSA seeks, but it might
need to be reformatted or recompiled. Because of this, NHTSA believes
the burden to provide this information will often be minimal.
Affected Public: Respondents are manufacturers of engines and
vehicles within the North American Industry Classification System
(NAICS) and use the coding structure as defined by NAICS including
codes 33611, 336111, 336112, 33631, 33631, 33632, 336320, 33635, and
336350 for motor vehicle and parts manufacturing.
Respondent's obligation to respond: Regulated entities are required
to respond to inquiries covered by this collection. 49 U.S.C. 32907. 49
CFR part 525, 534, 536, and 537.
Frequency of response: Variable, based on compliance obligation.
Please see PRA supporting documentation in the docket for more detailed
information.
Average burden time per response: Variable, based on compliance
obligation. Please see PRA supporting documentation in the docket for
more detailed information.
Number of respondents: 23.
4. Estimated Total Annual Burden Hours and Costs:
[GRAPHIC] [TIFF OMITTED] TR30AP20.759
[GRAPHIC] [TIFF OMITTED] TR30AP20.760
[[Page 25268]]
O. Privacy Act
In accordance with 5 U.S.C. 553(c), the agencies solicited comments
from the public to inform the rulemaking process better. These comments
are posted, without edit, to www.regulations.gov, as described in DOT's
system of records notice, DOT/ALL-14 FDMS, accessible through
www.transportation.gov/privacy. In order to facilitate comment tracking
and response, the agencies encouraged commenters to provide their
names, or the names of their organizations; however, submission of
names is completely optional.
List of Subjects
40 CFR Part 86
Administrative practice and procedure, Confidential business
information, Incorporation by reference, Labeling, Motor vehicle
pollution, Reporting and recordkeeping requirements.
40 CFR Part 600
Administrative practice and procedure, Electric power, Fuel
economy, Labeling, Reporting and recordkeeping requirements.
49 CFR Parts 523, 531, and 533
Fuel economy.
49 CFR Parts 536 and 537
Fuel economy, Reporting and recordkeeping requirements.
Environmental Protection Agency
40 CFR Chapter I
For the reasons set forth in the preamble, the Environmental
Protection Agency is amending part 86 of title 40, Chapter I of the
Code of Federal Regulations as follows:
PART 86--CONTROL OF EMISSIONS FROM NEW AND IN-USE HIGHWAY VEHICLES
AND ENGINES
0
1. The authority citation for part 86 continues to read as follows:
Authority: 42 U.S.C. 7401-7671q.
0
2. Section 86.1818-12 is amended by revising paragraphs (c)(2)(i)(A)
through (C) and (c)(3)(i)(A), (B), and (D), to read as follows:
Sec. 86.1818-12 Greenhouse gas emission standards for light-duty
vehicles, light-duty trucks, and medium-duty passenger vehicles.
* * * * *
(c) * * *
(2) * * *
(i) * * *
(A) For passenger automobiles with a footprint of less than or
equal to 41 square feet, the gram/mile CO2 target value
shall be selected for the appropriate model year from the following
table:
------------------------------------------------------------------------
CO2 target
Model year value (grams/
mile)
------------------------------------------------------------------------
2012.................................................... 244.0
2013.................................................... 237.0
2014.................................................... 228.0
2015.................................................... 217.0
2016.................................................... 206.0
2017.................................................... 195.0
2018.................................................... 185.0
2019.................................................... 175.0
2020.................................................... 166.0
2021.................................................... 161.8
2022.................................................... 159.0
2023.................................................... 156.4
2024.................................................... 153.7
2025.................................................... 151.2
2026 and later.......................................... 148.6
------------------------------------------------------------------------
(B) For passenger automobiles with a footprint of greater than 56
square feet, the gram/mile CO2 target value shall be
selected for the appropriate model year from the following table:
------------------------------------------------------------------------
CO2 target
Model year value (grams/
mile)
------------------------------------------------------------------------
2012.................................................... 315.0
2013.................................................... 307.0
2014.................................................... 299.0
2015.................................................... 288.0
2016.................................................... 277.0
2017.................................................... 263.0
2018.................................................... 250.0
2019.................................................... 238.0
2020.................................................... 226.0
2021.................................................... 220.9
2022.................................................... 217.3
2023.................................................... 213.7
2024.................................................... 210.2
2025.................................................... 206.8
2026 and later.......................................... 203.4
------------------------------------------------------------------------
(C) For passenger automobiles with a footprint that is greater than
41 square feet and less than or equal to 56 square feet, the gram/mile
CO2 target value shall be calculated using the following
equation and rounded to the nearest 0.1 grams/mile, except that for any
vehicle footprint the maximum CO2 target value shall be the
value specified for the same model year in paragraph (c)(2)(i)(B) of
this section:
Target CO2 = [a x f] + b
Where: f is the vehicle footprint, as defined in Sec. 86.1803; and
a and b are selected from the following table for the appropriate
model year:
------------------------------------------------------------------------
Model year a b
------------------------------------------------------------------------
2012.................................... 4.72 50.5
2013.................................... 4.72 43.3
2014.................................... 4.72 34.8
2015.................................... 4.72 23.4
2016.................................... 4.72 12.7
2017.................................... 4.53 8.9
2018.................................... 4.35 6.5
2019.................................... 4.17 4.2
2020.................................... 4.01 1.9
2021.................................... 3.94 0.2
2022.................................... 3.88 -0.1
2023.................................... 3.82 -0.4
2024.................................... 3.77 -0.6
2025.................................... 3.71 -0.9
2026 and later.......................... 3.65 -1.2
------------------------------------------------------------------------
* * * * *
(3) * * *
(i) * * *
(A) For light trucks with a footprint of less than or equal to 41
square feet, the gram/mile CO2 target value shall be
selected for the appropriate model year from the following table:
[[Page 25269]]
------------------------------------------------------------------------
CO2 target
Model year value (grams/
mile)
------------------------------------------------------------------------
2012.................................................... 294.0
2013.................................................... 284.0
2014.................................................... 275.0
2015.................................................... 261.0
2016.................................................... 247.0
2017.................................................... 238.0
2018.................................................... 227.0
2019.................................................... 220.0
2020.................................................... 212.0
2021.................................................... 206.5
2022.................................................... 203.0
2023.................................................... 199.6
2024.................................................... 196.2
2025.................................................... 193.2
2026 and later.......................................... 189.9
------------------------------------------------------------------------
(B) For light trucks with a footprint that is greater than 41
square feet and less than or equal to the maximum footprint value
specified in the table below for each model year, the gram/mile
CO2 target value shall be calculated using the following
equation and rounded to the nearest 0.1 grams/mile, except that for any
vehicle footprint the maximum CO2 target value shall be the
value specified for the same model year in paragraph (c)(3)(i)(D) of
this section:
Target CO2 = (a x f) + b
Where:
f is the footprint, as defined in Sec. 86.1803; and a and b are
selected from the following table for the appropriate model year:
----------------------------------------------------------------------------------------------------------------
Maximum
Model year footprint a b
----------------------------------------------------------------------------------------------------------------
2012............................................................ 66.0 4.04 128.6
2013............................................................ 66.0 4.04 118.7
2014............................................................ 66.0 4.04 109.4
2015............................................................ 66.0 4.04 95.1
2016............................................................ 66.0 4.04 81.1
2017............................................................ 50.7 4.87 38.3
2018............................................................ 60.2 4.76 31.6
2019............................................................ 66.4 4.68 27.7
2020............................................................ 68.3 4.57 24.6
2021............................................................ 68.3 4.51 21.5
2022............................................................ 68.3 4.44 20.6
2023............................................................ 68.3 4.37 20.2
2024............................................................ 68.3 4.31 19.6
2025............................................................ 68.3 4.23 19.6
2026 and later.................................................. 68.3 4.17 19.0
----------------------------------------------------------------------------------------------------------------
* * * * *
(D) For light trucks with a footprint greater than the minimum
value specified in the table below for each model year, the gram/mile
CO2 target value shall be selected for the appropriate model
year from the following table:
------------------------------------------------------------------------
CO2 target
Model year Minimum value (grams/
footprint mile)
------------------------------------------------------------------------
2012.................................... 66.0 395.0
2013.................................... 66.0 385.0
2014.................................... 66.0 376.0
2015.................................... 66.0 362.0
2016.................................... 66.0 348.0
2017.................................... 66.0 347.0
2018.................................... 66.0 342.0
2019.................................... 66.4 339.0
2020.................................... 68.3 337.0
2021.................................... 68.3 329.4
2022.................................... 68.3 324.1
2023.................................... 68.3 318.9
2024.................................... 68.3 313.7
2025.................................... 68.3 308.7
2026 and later.......................... 68.3 303.7
------------------------------------------------------------------------
* * * * *
0
3. Section 86.1866-12 is amended by revising paragraph (a)(2), removing
paragraph (a)(3), and revising (b) introductory text, (b)(1), and
(b)(2)(i) to read as follows:
Sec. 86.1866-12 CO2 credits for advanced technology vehicles.
* * * * *
(a) * * *
(2) Model years 2017 through 2026: For electric vehicles, plug-in
hybrid electric vehicles, and fuel cell vehicles produced for U.S.
sale, where ``U.S.'' means the states and territories of the United
States, in the 2017 through 2026 model years, such use of zero (0)
grams/mile CO2 is unrestricted.
(b) For electric vehicles, plug-in hybrid electric vehicles, fuel
cell vehicles, dedicated natural gas vehicles, and dual-fuel natural
gas vehicles as those terms are defined in Sec. 86.1803-01, that are
certified and produced for U.S. sale in the specified model years and
that meet the additional specifications in this section, the
manufacturer may use the production multipliers in this paragraph (b)
when determining additional credits for advanced technology vehicles.
Full size pickup trucks eligible for and using a production multiplier
are not eligible for the performance-based credits described in Sec.
86.1870-12(b).
[[Page 25270]]
(1) The production multipliers, by model year, for model year 2017
through 2021 electric vehicles and fuel cell vehicles are as follows:
------------------------------------------------------------------------
Production
Model year multiplier
------------------------------------------------------------------------
2017.................................................... 2.0
2018.................................................... 2.0
2019.................................................... 2.0
2020.................................................... 1.75
2021.................................................... 1.5
------------------------------------------------------------------------
(2)(i) The production multipliers, by model year, for model year
2017 through 2021 plug-in hybrid electric vehicles and model year 2017
through 2026 dedicated natural gas vehicles and dual-fuel natural gas
vehicles are as follows:
------------------------------------------------------------------------
Production
Model year multiplier
------------------------------------------------------------------------
2017.................................................... 1.6
2018.................................................... 1.6
2019.................................................... 1.6
2020.................................................... 1.45
2021.................................................... 1.3
2022-2026 (dedicated and dual fuel natural gas vehicles 2.0
only)..................................................
------------------------------------------------------------------------
* * * * *
0
4. Section 86.1868-12 is amended by adding an entry to the end of the
table in paragraph (a)(2) and by adding paragraph (h)(7) to read as
follows:
Sec. 86.1868-12 CO2 credits for improving the efficiency of air
conditioning systems.
* * * * *
(a) * * *
(2) * * *
------------------------------------------------------------------------
Passenger
Air conditioning technology automobiles Light trucks
(g/mi) (g/mi)
------------------------------------------------------------------------
* * * * * * *
Advanced technology air conditioning 1.1 1.1
compressor with improved efficiency
relative to fixed-displacement
compressors achieved through the
addition of a variable crankcase
suction valve..........................
------------------------------------------------------------------------
* * * * *
(h) * * *
(7) Advanced technology air conditioning compressor means an air
conditioning compressor with improved efficiency relative to fixed-
displacement compressors. Efficiency gains are derived from improved
internal valve systems that optimize the internal refrigerant flow
across the range of compressor operator conditions through the addition
of a variable crankcase suction valve.
0
5. Section 86.1869-12 is amended by revising paragraph (a), by adding
paragraphs (b)(1)(ix), (b)(1)(x), (b)(4)(xiii) and (b)(4)(xiv), and by
revising paragraph (d)(2) to read as follows:
Sec. 86.1869-12 CO2 credits for off-cycle CO2 reducing technologies.
* * * * *
(a) Manufacturers may generate credits for CO2-reducing
technologies where the CO2 reduction benefit of the
technology is not adequately captured on the Federal Test Procedure
and/or the Highway Fuel Economy Test such that the technology would not
be otherwise installed for purposes of reducing emissions (directly or
indirectly) over those test cycles for compliance with the GHG
standards. These technologies must have a measurable, demonstrable, and
verifiable real-world CO2 reduction that occurs outside the
conditions of the Federal Test Procedure and the Highway Fuel Economy
Test. These optional credits are referred to as ``off-cycle'' credits.
The technologies must not be integral or inherent to the basic vehicle
design, such as engine, transmission, mass reduction, passive
aerodynamic design, and tire technologies. Technologies installed for
non-off-cycle emissions related reasons are also not eligible as they
would be considered part of the baseline vehicle design. The technology
must not be inherent to the design of occupant comfort and
entertainment features except for technologies related to reducing
passenger air conditioning demand and improving air conditioning system
efficiency. Notwithstanding the provisions of this paragraph (a), off-
cycle menu technologies included in paragraph (b) of this section
remain eligible for credits. Off-cycle technologies used to generate
emission credits are considered emission-related components subject to
applicable requirements and must be demonstrated to be effective for
the full useful life of the vehicle. Unless the manufacturer
demonstrates that the technology is not subject to in-use
deterioration, the manufacturer must account for the deterioration in
their analysis. Durability evaluations of off-cycle technologies may
occur at any time throughout a model year, provided that the results
can be factored into the data provided in the model year report. Off-
cycle credits may not be approved for crash-avoidance technologies,
safety critical systems or systems affecting safety-critical functions,
or technologies designed for the purpose of reducing the frequency of
vehicle crashes. Off-cycle credits may not be earned for technologies
installed on a motor vehicle to attain compliance with any vehicle
safety standard or any regulation set forth in Title 49 of the Code of
Federal Regulations. The manufacturer must use one of the three options
specified in this section to determine the CO2 gram per mile
credit applicable to an off-cycle technology. Note that the option
provided in paragraph (b) of this section applies only to the 2014 and
later model years. The manufacturer should notify EPA in their pre-
model year report of their intention to generate any credits under this
section.
(b) * * *
(1) * * *
(ix) High efficiency alternator. The credit for a high efficiency
alternator for passenger automobiles and light trucks shall be
calculated using the following equation, and rounded to the nearest 0.1
grams/mile:
[GRAPHIC] [TIFF OMITTED] TR30AP20.761
[[Page 25271]]
Where:
VDAHEA is the ratio of the alternator output power to the
power supplied to the alternator, as measured using the Verband der
Automobilindustrie (VDA) efficiency measurement methodology and
expressed as a whole number percent from 68 to 100.
* * * * *
(4) * * *
(xiii) High efficiency alternator means an alternator where the
ratio of the alternator output power to the power supplied to the
alternator is greater than 67 percent, as measured using the Verband
der Automobilindustrie (VDA) efficiency measurement methodology.
* * * * *
(d) * * *
(2) Notice and opportunity for public comment. (i) The
Administrator will publish a notice of availability in the Federal
Register notifying the public of a manufacturer's proposed alternative
off-cycle credit calculation methodology. The notice will include
details regarding the proposed methodology but will not include any
Confidential Business Information. The notice will include instructions
on how to comment on the methodology. The Administrator will take
public comments into consideration in the final determination and will
notify the public of the final determination. Credits may not be
accrued using an approved methodology until the first model year for
which the Administrator has issued a final approval.
(ii) The Administrator may waive these notice and comment
requirements for technologies for which EPA has previously approved a
methodology for determining credits. To qualify for this waiver, the
new application must be substantially identical in form, content, and
methodology to the application for a previously approved methodology,
and must include the following:
(A) A cite to the appropriate previously approved methodology,
including the appropriate Federal Register Notice and any subsequent
EPA documentation of the Administrator's decision;
(B) All necessary manufacturer- and vehicle-specific test data,
modeling, and credit calculations; and,
(C) Any other vehicle- or technology-specific details required
pursuant to the previously approved methodology to assess and support
an appropriate credit value.
(iii) A waiver of the notice and comment requirements does not
imply a determination that a specific credit value for a given
technology is appropriate, and nor does it imply a waiver from the
requirements in paragraphs (d)(1) and (e) of this section.
(iv) The Administrator retains the option to require a notice and
opportunity for public comment in cases where a new application
deviates in significant respects from a previously approved methodology
or raises novel substantive issues.
* * * * *
0
6. Section 86.1870-12 is amended by revising paragraphs (a)(2) and
(b)(2) to read as follows:
Sec. 86.1870-12 CO2 credits for qualifying full-size light pickup
trucks.
* * * * *
(a) * * *
(2) Full size pickup trucks that are strong hybrid electric
vehicles and that are produced in the 2017 through 2021 model years are
eligible for a credit of 20 grams/mile. To receive this credit in a
model year, the manufacturer must produce a quantity of strong hybrid
electric full size pickup trucks such that the proportion of production
of such vehicles, when compared to the manufacturer's total production
of full size pickup trucks, is not less than 10 percent in that model
year.
* * * * *
(b) * * *
(2) Full size pickup trucks that are produced in the 2017 through
2021 model years and that achieve carbon-related exhaust emissions less
than or equal to the applicable target value determined in Sec.
86.1818-12(c)(3) multiplied by 0.80 (rounded to the nearest gram/mile)
in a model year are eligible for a credit of 20 grams/mile. A pickup
truck that qualifies for this credit in a model year may claim this
credit for a maximum of four subsequent model years (a total of five
consecutive model years) if the carbon-related exhaust emissions of
that pickup truck do not increase relative to the emissions in the
model year in which the pickup truck first qualified for the credit.
This credit may not be claimed in any model year after 2021. To qualify
for this credit in a model year, the manufacturer must produce a
quantity of full size pickup trucks that meet the emission requirements
of this paragraph (b)(2) such that the proportion of production of such
vehicles, when compared to the manufacturer's total production of full
size pickup trucks, is not less than 10 percent in that model year.
* * * * *
PART 600--FUEL ECONOMY AND GREENHOUSE GAS EXHAUST EMISSIONS OF
MOTOR VEHICLES
0
7. The authority citation for part 600 continues to read as follows:
Authority: 49 U.S.C. 32901--23919q, Pub. L. 109-58.
0
8. Section 600.113-12 is amended by revising paragraphs (n)
introductory text, (n)(1), and (n)(3) to read as follows:
Sec. 600.113-12 Fuel economy, CO2 emissions, and carbon-related
exhaust emission calculations for FTP, HFET, US06, SC03 and cold
temperature FTP tests.
* * * * *
(n) Manufacturers shall determine CO2 emissions and
carbon-related exhaust emissions for electric vehicles, fuel cell
vehicles, and plug-in hybrid electric vehicles according to the
provisions of this paragraph (n). Subject to the limitations on the
number of vehicles produced and delivered for sale as described in
Sec. 86.1866 of this chapter, the manufacturer may be allowed to use a
value of 0 grams/mile to represent the emissions of fuel cell vehicles
and the proportion of electric operation of a electric vehicles and
plug-in hybrid electric vehicles that is derived from electricity that
is generated from sources that are not onboard the vehicle, as
described in paragraphs (n)(1) through (3) of this section. For
purposes of labeling under this part, the CO2 emissions for
electric vehicles shall be 0 grams per mile. Similarly, for purposes of
labeling under this part, the CO2 emissions for plug-in
hybrid electric vehicles shall be 0 grams per mile for the proportion
of electric operation that is derived from electricity that is
generated from sources that are not onboard the vehicle. For all 2027
and later model year electric vehicles, fuel cell vehicles, and plug-in
hybrid electric vehicles, the provisions of this paragraph (n) shall be
used to determine the non-zero value for CREE for purposes of meeting
the greenhouse gas emission standards described in Sec. 86.1818 of
this chapter.
(1) For electric vehicles, but not including fuel cell vehicles,
the carbon-related exhaust emissions in grams per mile is to be
calculated using the following equation and rounded to the nearest one
gram per mile:
CREE = CREEUP - CREEGAS
Where:
CREE means the carbon-related exhaust emission value as defined in
Sec. 600.002, which may be set equal to zero for eligible 2012
through 2026 model year electric vehicles as described in Sec.
86.1866-12(a) of this chapter.
[[Page 25272]]
[GRAPHIC] [TIFF OMITTED] TR30AP20.762
[GRAPHIC] [TIFF OMITTED] TR30AP20.763
Where:
EC = The vehicle energy consumption in watt-hours per mile, for
combined FTP/HFET operation, determined according to procedures
established by the Administrator under Sec. 600.116-12.
GRIDLOSS = 0.935 (to account for grid transmission losses).
AVGUSUP = 0.534 (the nationwide average electricity greenhouse gas
emission rate at the powerplant, in grams per watt-hour).
2478 is the estimated grams of upstream greenhouse gas emissions per
gallon of gasoline.
8887 is the estimated grams of CO2 per gallon of
gasoline.
TargetCO2 = The CO2 Target Value for the fuel
cell or electric vehicle determined according to Sec. 86.1818 of
this chapter for the appropriate model year.
* * * * *
(3) For 2012 and later model year fuel cell vehicles, the carbon-
related exhaust emissions in grams per mile shall be calculated using
the method specified in paragraph (n)(1) of this section, except that
CREEUP shall be determined according to procedures established by the
Administrator under Sec. 600.111-08(f). As described in Sec. 86.1866
of this chapter, the value of CREE may be set equal to zero for 2012
through 2026 model year fuel cell vehicles.
* * * * *
0
9. Section 600.510-12 is amended by revising paragraphs (c)(2)(vi)
introductory text, adding paragraph (c)(2)(vii) introductory text,
revising the introductory text of paragraphs (c)(2)(vii)(B), (j)(2)(v),
(vii)(A) and (vii)(B) to read as follows:
Sec. 600.510-12 Calculation of average fuel economy and average
carbon-related exhaust emissions.
* * * * *
(c) * * *
(2) * * *
(vi) For natural gas dual fuel model types, for model years 1993
through 2016, and optionally for 2021 and later model years, the
harmonic average of the following two terms; the result rounded to the
nearest 0.1 mpg:
* * * * *
(vii) This paragraph (c)(2)(vii) applies to model year 2017 through
2020 natural gas dual fuel model types. Model year 2021 and later
natural gas dual fuel model types may use the provisions of paragraph
(c)(2)(vi) of this section or this paragraph (c)(2)(vii).
* * * * *
(B) Model year 2017 through 2020 natural gas dual fuel model types
must meet the following criteria to qualify for use of a Utility Factor
greater than 0.5:
* * * * *
(j) * * *
(2) * * *
(v) For natural gas dual fuel model types, for model years 2012
through 2015, and optionally for 2021 and later model years, the
arithmetic average of the following two terms; the result rounded to
the nearest gram per mile:
* * * * *
(vii)(A) This paragraph (j)(2)(vii) applies to model year 2016
through 2020 natural gas dual fuel model types. Model year 2021 and
later natural gas dual fuel model types may use the provisions of
paragraph (j)(2)(v) of this section or this paragraph (j)(2)(vii).
* * * * *
(B) Model year 2016 through 2020 natural gas dual fuel model types
must meet the following criteria to qualify for use of a Utility Factor
greater than 0.5:
* * * * *
National Highway Transportation Administration
Chapter V
For the reasons discussed in the preamble, the National Highway
Traffic Safety Administration amends 49 CFR chapter V as follows:
PART 523--VEHICLE CLASSIFICATION
0
10. The authority citation for part 523 continues to read as follows:
Authority: 49 U.S.C 32901; delegation of authority at 49 CFR
1.95.
0
11. Amend Sec. 523.2 by revising the definitions of ``Curb weight''
and ``Full-size pickup truck'' to read as follows:
Sec. 523.2 Definitions.
* * * * *
Curb weight has the meaning given in 40 CFR 86.1803-01.
* * * * *
Full-size pickup truck means a light truck or medium duty passenger
vehicle that meets the specifications in 40 CFR 86.1803-01.
* * * * *
PART 531--PASSENGER AUTOMOBILE AVERAGE FUEL ECONOMY STANDARDS
0
12. The authority citation for part 531 is revised to read as follows:
Authority: 49 U.S.C. 32902; delegation of authority at 49 CFR
1.95.
0
13. Amend Sec. 531.5 by revising the introductory text of paragraph
(c), Table III to paragraph (c), and paragraph (d), removing paragraph
(e), and redesignating paragraph (f) as paragraph (e) to read as
follows:
Sec. 531.5 Fuel economy standards.
* * * * *
(c) For model years 2012-2026, a manufacturer's passenger
automobile fleet shall comply with the fleet average fuel economy level
calculated for that model year according to this Figure 2 and the
appropriate values in this Table III.
* * * * *
Table III--Parameters for the Passenger Automobile Fuel Economy Targets, MYs 2012-2026
----------------------------------------------------------------------------------------------------------------
Parameters
---------------------------------------------------------------
Model year c (gal/mi/
a (mpg) b (mpg) ft\2\) d (gal/mi)
----------------------------------------------------------------------------------------------------------------
2012............................................ 35.95 27.95 0.0005308 0.006057
2013............................................ 36.80 28.46 0.0005308 0.005410
2014............................................ 37.75 29.03 0.0005308 0.004725
2015............................................ 39.24 29.90 0.0005308 0.003719
2016............................................ 41.09 30.96 0.0005308 0.002573
[[Page 25273]]
2017............................................ 43.61 32.65 0.0005131 0.001896
2018............................................ 45.21 33.84 0.0004954 0.001811
2019............................................ 46.87 35.07 0.0004783 0.001729
2020............................................ 48.74 36.47 0.0004603 0.001643
2021............................................ 49.48 37.02 0.000453 0.00162
2022............................................ 50.24 37.59 0.000447 0.00159
2023............................................ 51.00 38.16 0.000440 0.00157
2024............................................ 51.78 38.74 0.000433 0.00155
2025............................................ 52.57 39.33 0.000427 0.00152
2026............................................ 53.37 39.93 0.000420 0.00150
----------------------------------------------------------------------------------------------------------------
(d) In addition to the requirements of paragraphs (b) and (c) of
this section, each manufacturer shall also meet the minimum fleet
standard for domestically manufactured passenger automobiles expressed
in Table IV:
Table IV--Minimum Fuel Economy Standards for Domestically Manufactured
Passenger Automobiles, MYs 2011-2026
------------------------------------------------------------------------
Minimum
Model year standard
------------------------------------------------------------------------
2011.................................................... 27.8
2012.................................................... 30.7
2013.................................................... 31.4
2014.................................................... 32.1
2015.................................................... 33.3
2016.................................................... 34.7
2017.................................................... 36.7
2018.................................................... 38.0
2019.................................................... 39.4
2020.................................................... 40.9
2021.................................................... 39.9
2022.................................................... 40.6
2023.................................................... 41.1
2024.................................................... 41.8
2025.................................................... 42.4
2026.................................................... 43.1
------------------------------------------------------------------------
* * * * *
0
14. Amend Sec. 531.6 by revising paragraphs (a) and (b) to read as
follows:
Sec. 531.6 Measurement and calculation procedures.
(a) The fleet average fuel economy performance of all passenger
automobiles that are manufactured by a manufacturer in a model year
shall be determined in accordance with procedures established by the
Administrator of the Environmental Protection Agency under 49 U.S.C.
32904 and set forth in 40 CFR part 600. For model years 2017 to 2026, a
manufacturer is eligible to increase the fuel economy performance of
passenger cars in accordance with procedures established by the EPA set
forth in 40 CFR part 600, subpart F, including any adjustments to fuel
economy the EPA allows, such as for fuel consumption improvements
related to air conditioning efficiency and off-cycle technologies.
(1) A manufacturer that seeks to increase its fleet average fuel
economy performance through the use of technologies that improve the
efficiency of air conditioning systems must follow the requirements in
40 CFR 86.1868-12. Fuel consumption improvement values resulting from
the use of those air conditioning systems must be determined in
accordance with 40 CFR 600.510-12(c)(3)(i).
(2) A manufacturer that seeks to increase its fleet average fuel
economy performance through the use of off-cycle technologies must
follow the requirements in 40 CFR 86.1869-12. A manufacturer is
eligible to gain fuel consumption improvements for predefined off-cycle
technologies in accordance with 40 CFR 86.1869-12(b) or for
technologies tested using the EPA's 5-cycle methodology in accordance
with 40 CFR 86.1869-12(c). The fuel consumption improvement is
determined in accordance with 40 CFR 600.510-12(c)(3)(ii).
(b) A manufacturer is eligible to increase its fuel economy
performance through use of an off-cycle technology requiring an
application request made to the EPA in accordance with 40 CFR 86.1869-
12(d). The request must be approved by the EPA in consultation with
NHTSA. To expedite NHTSA's consultation with the EPA, a manufacturer
shall concurrently submit its application to NHTSA if the manufacturer
is seeking off-cycle fuel economy improvement values under the CAFE
program for those technologies. For off-cycle technologies that are
covered under 40 CFR 86.1869-12(d), NHTSA will consult with the EPA
regarding NHTSA's evaluation of the specific off-cycle technology to
ensure its impact on fuel economy and the suitability of using the off-
cycle technology to adjust the fuel economy performance. NHTSA will
provide its views on the suitability of the technology for that purpose
to the EPA. NHTSA's evaluation and review will consider:
(1) Whether the technology has a direct impact upon improving fuel
economy performance;
(2) Whether the technology is related to crash-avoidance
technologies, safety critical systems or systems affecting safety-
critical functions, or technologies designed for the purpose of
reducing the frequency of vehicle crashes;
(3) Information from any assessments conducted by the EPA related
to the application, the technology and/or related technologies; and
(4) Any other relevant factors.
PART 533--LIGHT TRUCK FUEL ECONOMY STANDARDS
0
15. The authority citation for part 533 is revised to read as follows:
Authority: 49 U.S.C. 32902; delegation of authority at 49 CFR
1.95.
0
16. In Sec. 533.5, amend paragraph (a) by revising Table VII and
removing paragraph (k) to read as follows:
Sec. 533.5 Requirements.
(a) * * *
[[Page 25274]]
Table VII--Parameters for the Light Truck Fuel Economy Targets for MYs 2017-2026
--------------------------------------------------------------------------------------------------------------------------------------------------------
Parameters
-------------------------------------------------------------------------------------------------------
Model year c (gal/mi/ g (gal/mi/
a (mpg) b (mpg) ft\2\) d (gal/mi) e (mpg) f (mpg) ft\2\) h (gal/mi)
--------------------------------------------------------------------------------------------------------------------------------------------------------
2017............................................ 36.26 25.09 0.0005484 0.005097 35.10 25.09 0.0004546 0.009851
2018............................................ 37.36 25.20 0.0005358 0.004797 35.31 25.20 0.0004546 0.009682
2019............................................ 38.16 25.25 0.0005265 0.004623 35.41 25.25 0.0004546 0.009603
2020............................................ 39.11 25.25 0.0005140 0.004494 35.41 25.25 0.0004546 0.009603
2021............................................ 39.71 25.63 0.000506 0.00443 NA NA NA NA
2022............................................ 40.31 26.02 0.000499 0.00436 NA NA NA NA
2023............................................ 40.93 26.42 0.000491 0.00429 NA NA NA NA
2024............................................ 41.55 26.82 0.000484 0.00423 NA NA NA NA
2025............................................ 42.18 27.23 0.000477 0.00417 NA NA NA NA
2026............................................ 42.82 27.64 0.000469 0.00410 NA NA NA NA
--------------------------------------------------------------------------------------------------------------------------------------------------------
* * * * *
0
17. Amend Sec. 533.6 by revising paragraphs (b) and (c) to read as
follows:
Sec. 533.6 Measurement and calculation procedures.
* * * * *
(b) The fleet average fuel economy performance of all light trucks
that are manufactured by a manufacturer in a model year shall be
determined in accordance with procedures established by the
Administrator of the Environmental Protection Agency under 49 U.S.C.
32904 and set forth in 40 CFR part 600. For model years 2017 to 2026, a
manufacturer is eligible to increase the fuel economy performance of
light trucks in accordance with procedures established by the EPA set
forth in 40 CFR part 600, subpart F, including any adjustments to fuel
economy the EPA allows, such as for fuel consumption improvements
related to air conditioning efficiency, off-cycle technologies, and
hybridization and other performance-based technologies for full-size
pickup trucks that meet the requirements specified in 40 CFR 86.1803.
(1) A manufacturer that seeks to increase its fleet average fuel
economy performance through the use of technologies that improve the
efficiency of air conditioning systems must follow the requirements in
40 CFR 86.1868-12. Fuel consumption improvement values resulting from
the use of those air conditioning systems must be determined in
accordance with 40 CFR 600.510-12(c)(3)(i).
(2) A manufacturer that seeks to increase its fleet average fuel
economy performance through the use of off-cycle technologies must
follow the requirements in 40 CFR 86.1869-12. A manufacturer is
eligible to gain fuel consumption improvements for predefined off-cycle
technologies in accordance with 40 CFR 86.1869-12(b) or for
technologies tested using the EPA's 5-cycle methodology in accordance
with 40 CFR 86.1869-12(c). The fuel consumption improvement is
determined in accordance with 40 CFR 600.510-12(c)(3)(ii).
(3) The eligibility of a manufacturer to increase its fuel economy
using hybridized and other performance-based technologies for full-size
pickup trucks must follow 40 CFR 86.1870-12 and the fuel consumption
improvement of these full-size pickup truck technologies must be
determined in accordance with 40 CFR 600.510-12(c)(3)(iii).
(c) A manufacturer is eligible to increase its fuel economy
performance through use of an off-cycle technology requiring an
application request made to the EPA in accordance with 40 CFR 86.1869-
12(d). The request must be approved by the EPA in consultation with
NHTSA. To expedite NHTSA's consultation with the EPA, a manufacturer
shall concurrently submit its application to NHTSA if the manufacturer
is seeking off-cycle fuel economy improvement values under the CAFE
program for those technologies. For off-cycle technologies that are
covered under 40 CFR 86.1869-12(d), NHTSA will consult with the EPA
regarding NHTSA's evaluation of the specific off-cycle technology to
ensure its impact on fuel economy and the suitability of using the off-
cycle technology to adjust the fuel economy performance. NHTSA will
provide its views on the suitability of the technology for that purpose
to the EPA. NHTSA's evaluation and review will consider:
(1) Whether the technology has a direct impact upon improving fuel
economy performance;
(2) Whether the technology is related to crash-avoidance
technologies, safety critical systems or systems affecting safety-
critical functions, or technologies designed for the purpose of
reducing the frequency of vehicle crashes;
(3) Information from any assessments conducted by the EPA related
to the application, the technology and/or related technologies; and
(4) Any other relevant factors.
PART 535--MEDIUM- AND HEAVY-DUTY VEHICLE FUEL EFFICIENCY PROGRAM
0
18. The authority citation for part 535 continues to read as follows:
Authority: 49 U.S.C. 32902 and 30101; delegation of authority
at 49 CFR 1.95.
0
19. Amend Sec. 535.6 by revising paragraphs (a)(4)(ii) and (d)(5)(ii)
to read as follows:
Sec. 535.6 Measurement and calculation procedures.
* * * * *
(a) * * *
(4) * * *
(ii) Calculate the equivalent fuel consumption test group results
as follows for spark-ignition vehicles and alternative fuel spark-
ignition vehicles. CO2 emissions test group result (grams
per mile)/((8,887 grams per gallon of gasoline fuel) x
(10-2)) = Fuel consumption test group result (gallons per
100 mile).
* * * * *
(d) * * *
(5) * * *
(ii) Calculate equivalent fuel consumption FCL values for spark-
ignition engines and alternative fuel spark-ignition engines.
CO2 FCL value (grams per hp-hr)/((8,887 grams per gallon of
gasoline fuel) x (10-2)) = Fuel consumption FCL value
(gallons per 100 hp-hr).
* * * * *
0
20. Amend Sec. 535.7 by revising the equations in paragraphs (b)(1),
(c)(1), (d)(1), (e)(2), and (f)(2)(iii)(E) to read as follows:
[[Page 25275]]
Sec. 535.7 Averaging, banking, and trading (ABT) credit program.
* * * * *
(b) * * *
(1) * * *
Total MY Fleet FCC (gallons) = (Std - Act) x (Volume) x (UL) x
(10-2)
Where:
Std = Fleet average fuel consumption standard (gal/100 mile).
Act = Fleet average actual fuel consumption value (gal/100 mile).
Volume = the total U.S.-directed production of vehicles in the
regulatory subcategory.
UL = the useful life for the regulatory subcategory. The useful life
value for heavy-pickup trucks and vans manufactured for model years
2013 through 2020 is equal to the 120,000 miles. The useful life for
model years 2021 and later is equal to 150,000 miles.
* * * * *
(c) * * *
(1) * * *
Vehicle Family FCC (gallons) = (Std - FEL) x (Payload) x (Volume) x
(UL) x (10-\3\)
Where:
Std = the standard for the respective vehicle family regulatory
subcategory (gal/1000 ton-mile).
FEL = family emissions limit for the vehicle family (gal/1000 ton-
mile).
Payload = the prescribed payload in tons for each regulatory
subcategory as shown in the following table:
------------------------------------------------------------------------
Payload
Regulatory subcategory (tons)
------------------------------------------------------------------------
Vocational LHD Vehicles................................. 2.85
Vocational MHD Vehicles................................. 5.60
Vocational HHD Vehicles................................. 7.5
MDH Tractors............................................ 12.50
HHD Tractors, other than heavy-haul Tractors............ 19.00
Heavy-haul Tractors..................................... 43.00
------------------------------------------------------------------------
Volume = the number of U.S.-directed production volume of vehicles
in the corresponding vehicle family.
UL = the useful life for the regulatory subcategory (miles) as shown
in the following table:
------------------------------------------------------------------------
Regulatory subcategory UL (miles)
------------------------------------------------------------------------
LHD Vehicles.............................. 110,000 (Phase 1).
150,000 (Phase 2).
Vocational MHD Vehicles and tractors at or 185,000.
below 33,000 pounds GVWR.
Vocation HHD Vehicles and tractors at or 435,000.
above 33,000 pounds GVWR.
------------------------------------------------------------------------
* * * * *
(d) * * *
(1) * * *
Engine Family FCC (gallons) = (Std - FCL) x (CF) x (Volume) x (UL) x
(10-2)
Where:
Std = the standard for the respective engine regulatory subcategory
(gal/100 hp-hr).
FCL = family certification level for the engine family (gal/100 hp-
hr).
CF= a transient cycle conversion factor in hp-hr/mile which is the
integrated total cycle horsepower-hour divided by the equivalent
mileage of the applicable test cycle. For engines subject to spark-
ignition heavy-duty standards, the equivalent mileage is 6.3 miles.
For engines subject to compression-ignition heavy-duty standards,
the equivalent mileage is 6.5 miles.
Volume = the number of engines in the corresponding engine family.
UL = the useful life of the given engine family (miles) as shown in
the following table:
------------------------------------------------------------------------
Regulatory subcategory UL (miles)
------------------------------------------------------------------------
SI and CI LHD Engines..................... 120,000 (Phase 1).
150,000 (Phase 2).
CI MHD Engines............................ 185,000.
CI HHD Engines............................ 435,000.
------------------------------------------------------------------------
* * * * *
(e) * * *
(2) * * *
Vehicle Family FCC (gallons) = (Std - FEL) x (Payload) x (Volume) x
(UL) x (10-3)
Where:
Std = the standard for the respective vehicle family regulatory
subcategory (gal/1000 ton-mile).
FEL = family emissions limit for the vehicle family (gal/1000 ton-
mile).
Payload = 10 tons for short box vans and 19 tons for other trailers.
Volume = the number of U.S.-directed production volume of vehicles
in the corresponding vehicle family.
UL = the useful life for the regulatory subcategory. The useful life
value for heavy-duty trailers is equal to 250,000 miles.
* * * * *
(f) * * *
(2) * * *
(iii) * * *
(E) * * *
Off-cycle FC credits = (CO2 Credit/CF) x Production x VLM
Where:
CO2 Credits = the credit value in grams per mile
determined in 40 CFR 86.1869-12(c)(3), (d)(1), (d)(2) or (d)(3).
CF = conversion factor, which for spark-ignition engines is 8,887
and for compression-ignition engines is 10,180.
Production = the total production volume for the applicable category
of vehicles
VLM = vehicle lifetime miles, which for 2b-3 vehicles shall be
150,000 for the Phase 2 program.
The term (CO2 Credit/CF) should be rounded to the nearest
0.0001
* * * * *
PART 536--TRANSFER AND TRADING OF FUEL ECONOMY CREDITS
0
21. The authority citation for part 536 is revised to read as follows:
Authority: 49 U.S.C. 32903; delegation of authority at 49 CFR
1.95.
0
22. Amend Sec. 536.4 by revising paragraph (c) to read as follows:
Sec. 536.4 Credits.
* * * * *
(c) Adjustment factor. When traded or transferred and used, fuel
economy credits are adjusted to ensure fuel oil savings is preserved.
For traded credits, the user (or buyer) must multiply the calculated
adjustment factor by the number of shortfall credits it plans to offset
in order to determine the number of equivalent credits to acquire from
the earner (or seller). For transferred credits, the user of credits
must multiply the calculated adjustment factor by the number of
shortfall credits it plans to offset in order to determine the number
of equivalent credits to transfer from the compliance category holding
the available credits. The adjustment factor is calculated according to
the following formula:
[GRAPHIC] [TIFF OMITTED] TR30AP20.764
Where:
A = Adjustment factor applied to traded and transferred credits. The
quotient shall be rounded to 4 decimal places;
* * * * *
0
23. Amend Sec. 536.5 by revising paragraphs (c) and (d)(6) to read as
follows:
[[Page 25276]]
Sec. 536.5 Trading infrastructure.
* * * * *
(c) Automatic debits and credits of accounts.
(1) To carry credits forward, backward, transfer credits, or trade
credits into other credit accounts, a manufacturer or credit holder
must submit a credit instruction to NHTSA. A credit instruction must
detail and include:
(i) The credit holder(s) involved in the transaction.
(ii) The originating credits described by the amount of the
credits, compliance category and the vintage of the credits.
(iii) The recipient credit account(s) for banking or applying the
originating credits described by the compliance category(ies), model
year(s), and if applicable the adjusted credit amount(s) and adjustment
factor(s).
(iv) For trades, a contract authorizing the trade signed by the
manufacturers or credit holders or by managers legally authorized to
obligate the sale and purchase of the traded credits.
(2) Upon receipt of a credit instruction from an existing credit
holder, NHTSA verifies the presence of sufficient credits in the
account(s) of the credit holder(s) involved as applicable and notifies
the credit holder(s) that the credits will be debited from and/or
credited to the accounts involved, as specified in the credit
instruction. NHTSA determines if the credits can be debited or credited
based upon the amount of available credits, accurate application of any
adjustment factors and the credit requirements prescribed by this part
that are applicable at the time the transaction is requested.
(3) After notifying the credit holder(s), all accounts involved are
either credited or debited, as appropriate, in line with the credit
instruction. Traded credits identified by a specific compliance
category are deposited into the recipient's account in that same
compliance category and model year. If a recipient of credits as
identified in a credit instruction is not a current account holder,
NHTSA establishes the credit recipient's account, subject to the
conditions described in Sec. 536.5(b), and adds the credits to the
newly-opened account.
(4) NHTSA will automatically delete unused credits from holders'
accounts when those credits reach their expiry date.
(5) Starting in model year 2021, manufacturers or credit holders
issuing credit instructions or providing credit allocation plans as
specified in Sec. 536.5(d), must use the NHTSA Credit Template
fillable form (OMB Control No. 2127-0019, NHTSA Form 1475). The NHTSA
Credit Template is available for download on NHTSA's website. If a
credit instruction includes a trade, the NHTSA Credit Template must be
signed by managers legally authorized to obligate the sale and/or
purchase of the traded credits from both parties to the trade. The
NHTSA Credit Template signed by both parties to the trade serves as an
acknowledgement that the parties have agreed to trade credits, and does
not dictate terms, conditions, or other business obligations of the
parties. All parties trading credits must also provide NHTSA the price
paid for the credits including a description of any other monetary or
non-monetary terms affecting the price of the traded credits, such as
any technology exchanged or shared for the credits, any other non-
monetary payment for the credits, or any other agreements related to
the trade. Manufacturers must submit this information to NHTSA in a PDF
document along with the Credit Template through the CAFE email,
[email protected]. NHTSA reserves the right to request additional
information from the parties regarding the terms of the trade.
(6) NHTSA will consider claims that information submitted to the
agency under this section is entitled to confidential treatment under 5
U.S.C. 552(b) and under the provisions of part 512 of this chapter if
the information is submitted in accordance with the procedures of that
part.
* * * * *
(d) * * *
(6) Credit allocation plans received from a manufacturer will be
reviewed and approved by NHTSA. Starting in model year 2021, use the
NHTSA Credit Template (OMB Control No. 2127-0019, NHTSA Form 1475) to
record the credit transactions requested in the credit allocation plan.
The template is a fillable form that has an option for recording and
calculating credit transactions for credit allocation plans. The
template calculates the required adjustments to the credits. The credit
allocation plan and the completed transaction template must be
submitted to NHTSA. NHTSA will approve the credit allocation plan
unless it finds that the proposed credits are unavailable or that it is
unlikely that the plan will result in the manufacturer earning
sufficient credits to offset the subject credit shortfall. If the plan
is approved, NHTSA will revise the respective manufacturer's credit
account accordingly. If the plan is rejected, NHTSA will notify the
respective manufacturer and request a revised plan or payment of the
appropriate fine.
PART 537--AUTOMOTIVE FUEL ECONOMY REPORTS
0
24. The authority citation for part 537 is revised to read as follows:
Authority: 49 U.S.C. 32907; delegation of authority at 49 CFR
1.95.
0
25. Amend Sec. 537.5 by redesignating paragraph (d) as paragraph (e)
and adding a new paragraph (d) to read as follows:
Sec. 537.5 General requirements for reports.
* * * * *
(d) Beginning with model year 2023, each manufacturer shall
generate reports required by this part using the NHTSA CAFE Projections
Reporting Template (OMB Control No. 2127-0019, NHTSA Form 1474). The
template is a fillable form.
(1) Select the option to identify the report as a pre-model year
report, mid-model year report, or supplementary report as appropriate;
(2) Complete all required information for the manufacturer and for
all vehicles produced for the current model year required to comply
with CAFE standards. Identify the manufacturer submitting the report,
including the full name, title, and address of the official responsible
for preparing the report and a point of contact to answer questions
concerning the report.
(3) Use the template to generate confidential and non-confidential
reports for all the domestic and import passenger cars and light truck
fleet produced by the manufacturer for the current model year.
Manufacturers must submit a request for confidentiality in accordance
with part 512 of this chapter to withhold projected production sales
volume estimates from public disclosure. If the request is granted,
NHTSA will withhold the projected production sales volume estimates
from public disclose until all the vehicles produced by the
manufacturer have been made available for sale (usually one year after
the current model year).
(4) Submit confidential reports and requests for confidentiality to
NHTSA on CD-ROM in accordance with Part 537.12. Email copies of non-
confidential (i.e., redacted) reports to NHTSA's secure email address:
[email protected]. Requests for confidentiality must be submitted in a PDF
or MS Word format. Submit 2 copies of the CD-ROM to: Administrator,
National Highway Traffic Administration, 1200 New Jersey Avenue SE,
Washington, DC 20590, and submit emailed reports electronically to
[[Page 25277]]
the following secure email address: [email protected];
(5) Confidentiality Requests. Manufacturers can withhold
information on projected production sales volumes under 5 U.S.C.
552(b)(4) and 15 U.S.C. 2005(d)(1). In accordance, the manufacturer
must:
(i) Show that the item is within the scope of sections 552(b)(4)
and 2005(d)(1);
(ii) Show that disclosure of the item would result in significant
competitive damage;
(iii) Specify the period during which the item must be withheld to
avoid that damage; and
(iv) Show that earlier disclosure would result in that damage.
* * * * *
0
26. Amend Sec. 537.6 by revising paragraphs (b) and (c) to read as
follows:
Sec. 537.6 General content of reports.
* * * * *
(b) Supplementary report. Except as provided in paragraph (c) of
this section, each supplementary report for each model year must
contain the information required by Sec. 537.7(a)(1) and (a)(2), as
appropriate for the vehicle fleets produced by the manufacturer, in
accordance with Sec. 537.8(b)(1), (2), (3), and (4) as appropriate.
(c) Exceptions. The pre-model year report, mid-model year report,
and supplementary report(s) submitted by an incomplete automobile
manufacturer for any model year are not required to contain the
information specified in Sec. 537.7 (c)(4) (xv) through (xviii) and
(c)(5). The information provided by the incomplete automobile
manufacturer under Sec. 537.7(c) shall be according to base level
instead of model type or carline.
0
27. Amend Sec. 537.7 by revising paragraph (a) to read as follows:
Sec. 537.7 Pre-model year and mid-model year reports.
(a)(1) Provide a report with the information required by paragraphs
(b) and (c) of this section for each domestic and import passenger
automobile fleet, as specified in part 531 of this chapter, for the
current model year.
(2) Provide a report with the information required by paragraphs
(b) and (c) of this section for each light truck fleet, as specified in
part 533 of this chapter, for the current model year.
(3) For model year 2023 and later, provide the information required
by paragraphs (a)(1) and (2) of this section for pre-model and mid-
model year reports in accordance with the NHTSA CAFE Projections
Reporting Template (OMB Control No. 2127-0019, NHTSA Form 1474). The
required reporting template can be downloaded from NHTSA's website.
* * * * *
0
28. Amend Sec. 537.7 by revising paragraphs (b)(3), (c)(1), (c)(3),
(c)(7)(i), (c)(7)(ii), and (c)(7)(iii) to read as follows:
Sec. 537.7 Pre-model year and mid-model year reports.
* * * * *
(b) * * *
(3) State the projected required fuel economy for the
manufacturer's passenger automobiles and light trucks determined in
accordance with Sec. Sec. 531.5(c) and 533.5 of this chapter and based
upon the projected sales figures provided under paragraph (c)(2) of
this section. For each unique model type and footprint combination of
the manufacturer's automobiles, provide the information specified in
paragraph (b)(3)(i) and (ii) of this section in tabular form. List the
model types in order of increasing average inertia weight from top to
bottom down the left side of the table and list the information
categories in the order specified in paragraphs (b)(3)(i) and (ii) of
this section from left to right across the top of the table. Other
formats, such as those accepted by the EPA, which contain all the
information in a readily identifiable format are also acceptable. For
model year 2023 and later, for each unique model type and footprint
combination of the manufacturer's automobiles, provide the information
specified in paragraph (b)(3)(i) and (ii) of this section in accordance
with the CAFE Projections Reporting Template (OMB Control No. 2127-
0019, NHTSA Form 1474).
(i) In the case of passenger automobiles:
(A) Beginning model year 2013, base tire as defined in Sec. 523.2
of this chapter,
(B) Beginning model year 2013, front axle, rear axle, and average
track width as defined in Sec. CFR 523.2 of this chapter,
(C) Beginning model year 2013, wheelbase as defined in Sec. 523.2
of this chapter, and
(D) Beginning model year 2013, footprint as defined in Sec. 523.2
of this chapter.
(E) The fuel economy target value for each unique model type and
footprint entry listed in accordance with the equation provided in part
531 of this chapter.
(ii) In the case of light trucks:
(A) Beginning model year 2013, base tire as defined in Sec. 523.2
of this chapter,
(B) Beginning model year 2013, front axle, rear axle, and average
track width as defined in Sec. 523.2 of this chapter,
(C) Beginning model year 2013, wheelbase as defined in Sec. 523.2
of this chapter, and
(D) Beginning model year 2013, footprint as defined in Sec. 523.2
of this chapter.
(E) The fuel economy target value for each unique model type and
footprint entry listed in accordance with the equation provided in part
533 of this chapter.
* * * * *
(c) * * *
(1) For each model type of the manufacturer's automobiles, provide
the information specified in paragraph (c)(2) of this section in
tabular form. List the model types in order of increasing average
inertia weight from top to bottom down the left side of the table and
list the information categories in the order specified in paragraph
(c)(2) of this section from left to right across the top of the table.
For model year 2023 and later, CAFE reports required by part 537 of
this chapter, shall for each model type of the manufacturer's
automobiles, provide the information in specified in paragraph (c)(2)
of this section in accordance with the NHTSA CAFE Projections Reporting
Template (OMB Control No. 2127-0019, NHTSA Form 1474) and list the
model types in order of increasing average inertia weight from top to
bottom.
* * * * *
(3) (Pre-model year reports only through model year 2022.) For each
vehicle configuration whose fuel economy was used to calculate the fuel
economy values for a model type under paragraph (c)(2) of this section,
provide the information specified in paragraph (c)(4) of this section
in accordance with the NHTSA CAFE Projections Reporting Template (OMB
Control No. 2127-0019, NHTSA Form 1474).
* * * * *
(7) * * *
(i) Provide a list of each air conditioning efficiency improvement
technology utilized in your fleet(s) of vehicles for each model year.
For each technology identify vehicles by make and model types that have
the technology, which compliance category those vehicles belong to and
the number of vehicles for each model equipped with the technology. For
each compliance category (domestic passenger car, import passenger car,
and light truck), report the air conditioning fuel consumption
improvement value in gallons/mile in accordance with the equation
specified in 40 CFR 600.510-12(c)(3)(i).
(ii) Provide a list of off-cycle efficiency improvement
technologies
[[Page 25278]]
utilized in your fleet(s) of vehicles for each model year that is
pending or approved by the EPA. For each technology identify vehicles
by make and model types that have the technology, which compliance
category those vehicles belong to, the number of vehicles for each
model equipped with the technology, and the associated off-cycle
credits (grams/mile) available for each technology. For each compliance
category (domestic passenger car, import passenger car, and light
truck), calculate the fleet off-cycle fuel consumption improvement
value in gallons/mile in accordance with the equation specified in 40
CFR 600.510-12(c)(3)(ii).
(iii) Provide a list of full-size pickup trucks in your fleet that
meet the mild and strong hybrid vehicle definitions. For each mild and
strong hybrid type, identify vehicles by make and model types that have
the technology, the number of vehicles produced for each model equipped
with the technology, the total number of full-size pickup trucks
produced with and without the technology, the calculated percentage of
hybrid vehicles relative to the total number of vehicles produced, and
the associated full-size pickup truck credits (grams/mile) available
for each technology. For the light truck compliance category, calculate
the fleet pickup truck fuel consumption improvement value in gallons/
mile in accordance with the equation specified in 40 CFR 600.510-
12(c)(3)(iii).
* * * * *
0
29. Amend Sec. 537.8 by revising paragraph (a)(3), adding paragraphs
(a)(4) and (b)(4), and revising paragraph (c)(1) to read as follows:
Sec. 537.8 Supplementary reports.
(a) * * *
(3) For model years through 2022, each manufacturer whose pre-model
or mid-model year report omits any of the information specified in
Sec. 537.7(b) or (c) shall file a supplementary report containing the
information specified in paragraph (b)(3) of this section. Starting
model year 2023, each manufacturer whose pre-model or mid-model year
report omits any of the information shall resubmit the information with
other information required in accordance with the NHTSA CAFE
Projections Reporting Template (OMB Control No. 2127-0019, NHTSA Form
1474).
(b) * * *
(4) The supplementary report required by paragraph (a)(4) of this
section must contain:
(i) All information omitted from the pre-model or mid-model year
reports under Sec. 537.6(c)(2); and
(ii) Such revisions of and additions to the information submitted
by the manufacturer in its pre-model or mid-model year reports
regarding the automobiles produced during the current model year as are
necessary to reflect the information provided under paragraph (b)(4)(i)
of this section.
(c)(1) Each report required by paragraphs (a)(1), (2), (3), or (4)
of this section must be submitted in accordance with Sec. 537.5(c) not
more than 45 days after the date on which the manufacturer determined,
or could have determined with reasonable diligence, that the report was
required.
* * * * *
Dated: March 30, 2020.
Andrew Wheeler,
Administrator, Environmental Protection Agency.
Issued on March 30, 2020 in Washington, DC, under authority
delegated in 49 CFR 1.95 and 501.5
James Clayton Owens,
Acting Administrator, National Highway Traffic Safety Administration.
[FR Doc. 2020-06967 Filed 4-20-20; 4:15 pm]
BILLING CODE 4910-59-P