[Federal Register Volume 67, Number 141 (Tuesday, July 23, 2002)]
[Proposed Rules]
[Pages 48099-48110]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 02-18579]


-----------------------------------------------------------------------

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Parts 122 and 412

[FRL-7250-2]


Notice of Data Availability; National Pollutant Discharge 
Elimination System Permit Regulation and Effluent Limitations 
Guidelines and Standards for Concentrated Animal Feeding Operations

AGENCY: Environmental Protection Agency (EPA).

ACTION: Notice of data availability.

-----------------------------------------------------------------------

SUMMARY: On January 12, 2001 (66 FR 2959), EPA published a proposal to 
revise two regulations that address manure, wastewater, and other 
process waters generated by concentrated animal feeding operations 
(CAFOs). These two regulations are: The National Pollutant Discharge 
Elimination System (NPDES) provisions that define which operations are 
CAFOs and establish permit requirements; and the Effluent Limitations 
Guidelines, or effluent guidelines, for feedlots (beef, dairy, swine 
and poultry subcategories), which establish the technology-based 
effluent discharge standards for CAFOs.
    In the proposal, and in a subsequent notice of data availability 
published on November 21, 2001 (66 FR 58556), EPA solicited comment on 
various aspects of the proposed revisions and data used to analyze the 
proposed revisions. Due to additional data and comments received, EPA 
is considering changes to certain aspects of the proposed rulemaking. 
Specifically, today's notice presents information on the following: 
Establishing alternative regulatory thresholds for chicken operations 
using dry litter management practices; the potential creation of 
alternative performance standards to encourage CAFOs to implement new 
technologies; and financial data and changes EPA is considering to 
refine its economic analysis models. Today, EPA is making these data 
and potential changes available for public review and comment.

DATES: You must submit comments by August 22, 2002.

ADDRESSES: You are encouraged to submit your comments electronically to 
[email protected]. Electronic comments should specify docket 
number W-00-27 and must be submitted as an ASCII, Microsoft Word, or 
WordPerfect file avoiding the use of special characters and any form of 
encryption. Electronic comments on this action may be filed online at 
many Federal Depository Libraries. No confidential business information 
(CBI) should be sent via e-mail.
    You also may submit comments by mail to: Concentrated Animal 
Feeding Operation Proposed Rule, Office of Water, Engineering and 
Analysis Division (4303T), USEPA, 1200 Pennsylvania Avenue, NW, 
Washington, DC 20460. Hand deliveries (including overnight mail) should 
be submitted to the Concentrated Animal Feeding Operation Proposed 
Rule, USEPA, EPA West Building, Room 6231, 1301 Constitution Avenue, 
NW, Washington, DC 20004. Please submit an original and three copies of 
your written comments and enclosures, as well as any references cited 
in your comments.
    The public record for this action and the proposed rulemaking has 
been established under docket number W-00-27 and is located at 1200 
Pennsylvania Avenue NW, Washington, DC. The record is available for 
inspection from 8 a.m. to noon, Monday through Thursday, excluding 
legal holidays. For access to the docket materials, call (202) 566-1000 
for the room number and to schedule an appointment. A reasonable fee 
may be charged for copying.

FOR FURTHER INFORMATION CONTACT: Renee Selinsky Johnson at (202) 566-
1077 or at the following e-mail address: [email protected].

SUPPLEMENTARY INFORMATION:   

Contents of This Document

I. Purpose of this Notice
II. Background
    A. Proposed Rule
    B. Notice of Data Availability
III. Thresholds for Chicken Operations Using Dry Litter Management
IV. Voluntary Alternative Performance Standards for Innovative 
Technologies
V. Changes to the Economic Analysis
    A. Changes to Model Framework and Assumptions
    B. Changes to the Baseline Financial Data
    C. Preliminary Analysis Results

I. Purpose of This Notice

    There are three main components to today's notice: (1) Discussion 
of potential new regulatory thresholds for chicken operations with dry 
litter management practices; (2) the potential creation of alternative 
performance standards to encourage CAFOs to implement new wastewater 
treatment technologies and/or practices; and (3) discussion of new 
financial data and changes EPA is considering to refine the economic 
analysis models used to evaluate economic effects that potential 
regulatory options may have on CAFOs.
    For chicken operations with dry litter management, EPA is 
considering alternative approaches for determining the number of 
broilers or laying hens that would be considered equivalent to 1,000 
animal units (AU). In the proposed rule, EPA presented a scenario where 
100,000 chickens would be considered equivalent to 1,000 AU. In today's 
notice, EPA presents two possible alternative approaches for setting 
this metric for chicken operations.
    EPA's long-term environmental vision for CAFOs includes continuing 
research and progress toward environmental improvement. The Agency 
believes that individual CAFOs can be encouraged to voluntarily develop 
and install new technologies and management practices equal to or 
better than those required by baseline best available technology 
economically achievable (BAT) and new source performance standards 
(NSPS) effluent guidelines regulations. Further, EPA recognizes that 
some CAFOs, as well as land grant universities, state

[[Page 48100]]

agencies, equipment vendors, and agricultural organizations, are 
working to develop new technologies that achieve reductions in nutrient 
and pathogen losses to surface water, ammonia and other air emissions, 
and groundwater contamination. The development of new technologies 
offers the potential to match or surpass the pollutant reduction that 
would be achieved by compliance with limits and standards promulgated 
in the final CAFO rule.
    Today's notice includes EPA's preliminary approach for developing a 
voluntary program intended to facilitate the development of new 
technologies and management practices that perform as well as or better 
than BAT (and NSPS) and may also help address the multimedia 
environmental issues confronting CAFOs. A key tenet of these programs 
is that CAFOs would voluntarily choose an alternative BAT/NSPS 
performance standard as the basis for their technology-based NPDES 
permit limits (e.g., inclusion of effluent limitations in their NPDES 
permits that are different from those based on the baseline effluent 
guideline).
    Data that EPA is considering to use in the economic analysis models 
include both farm level and enterprise level financial data, as well as 
data and information pertaining to various modeling assumptions used by 
EPA. The financial data include data from the U.S. Department of 
Agriculture (USDA), the Food and Agricultural Policy Research Institute 
(FAPRI) at the University of Missouri, and the National Cattlemen's 
Beef Association (NCBA). Other enterprise level data for some sectors 
were collected by EPA from various land grant universities. In today's 
notice, EPA describes the set of financial data that EPA is considering 
using to depict baseline financial conditions at regulated CAFOs. This 
notice also discusses methodological changes EPA is considering, based 
on comments received on the proposed rule and the previous notice of 
data availability (the ``2001 Notice''), to the analytical framework 
used to evaluate economic effects that potential regulatory options may 
have on CAFOs. Among the changes being considered are the inclusion of 
enterprise level financial data, inclusion of additional measures of 
profitability to evaluate post-regulatory effects at the enterprise 
level, and a few ancillary assumptions. These data and analytical 
changes are in addition to those already presented by EPA in the 2001 
Notice.
    EPA is seeking further public comment on the specific data and 
issues identified in this notice. However, EPA is seeking public 
comment only on these specific data and issues. Nothing in today's 
notice is intended to reopen any other issues discussed in the CAFO 
proposal or the 2001 Notice, or to reopen the proposal in general for 
additional public comments. EPA is continuing to review the comments 
already submitted on the proposed rule and the 2001 Notice and will 
address those comments, along with comments submitted on the data and 
issues identified in today's notice, in the final rulemaking.

II. Background

A. Proposed Rule

    On January 12, 2001 (66 FR 2959), EPA published proposed revisions 
to the existing effluent guidelines for CAFOs (40 CFR Part 412) and to 
certain provisions of the NPDES regulations applicable to CAFOs. 
Effluent guidelines and standards for CAFOs establish the technology-
based effluent discharge and performance standards for both existing 
and new sources for each of the beef, dairy, swine and poultry 
subcategories. The NPDES permit program for CAFOs defines which animal 
feeding operations (AFOs) are CAFOs and need to obtain a NPDES permit, 
and establishes the specific requirements that must be complied with 
under a permit. These two existing interrelated regulations affecting 
CAFOs were originally promulgated in the 1970s.
1. Effluent Limitations Guidelines and Standards
    Under the current effluent guidelines regulations, CAFOs are 
prohibited from discharging process wastewater, except when rainfall 
events cause an overflow from a facility designed, constructed, and 
operated to contain all process-generated wastewater plus the runoff 
from a 25-year, 24-hour rainfall event.
    EPA proposed requiring all existing and new CAFOs spreading manure 
on cropland to limit the application rate to the nitrogen needs of the 
crops and, for those fields where additional constraints are considered 
necessary, to also ensure that the manure application rate would not 
exceed the phosphorus needs of the crops.
    EPA also proposed to require all existing beef and dairy operations 
to implement controls (e.g., retrofitting lagoons and ponds with 
impervious liners) to minimize leaching to ground water if the ground 
water beneath the production area has a direct hydrologic connection to 
surface water. EPA proposed requiring all existing swine, veal, and 
poultry CAFOs to eliminate all discharges from the animal production 
area (i.e., for these sectors, eliminating the effluent guidelines 
provision that allows for certain overflows due to chronic or 
catastrophic rainfall).
    EPA proposed that newly constructed CAFOs should meet the same 
requirements as were proposed for existing CAFOs, except that new 
swine, veal and poultry operations also would need to implement ground 
water controls where there is a direct hydrologic connection to surface 
water.
    For more information on the proposed technology options, see 
section VIII of the proposed rule (66 FR 3050-3070). Section VIII of 
the proposed rule also describes certain other technology options that 
were considered by EPA at proposal, such as prohibiting manure 
application on frozen, snow-covered, or saturated ground; mandatory use 
of anaerobic digester systems; composting; and surface water monitoring 
requirements.
2. NPDES Regulations
    Under the current NPDES regulations for CAFOs, a ``three-tier'' 
structure is used to determine which animal feeding operations (AFOs) 
also meet the criteria under which they are considered concentrated 
animal feeding operations (CAFOs). Under this current NPDES structure, 
(1) all AFOs with more than 1,000 AU are automatically defined as 
CAFOs; (2) AFOs with 301 to 1,000 AU are defined as a CAFO only if they 
meet certain conditions; and (3) AFOs with 301 to 1,000 AU that do not 
meet these conditions, and all AFOs with 300 or less AU, become CAFOs 
only if they are designated as such by the permitting authority. (See 
40 CFR 122.23 and Part 122, Appendix B).
    EPA proposed several alternatives for revising the existing CAFO 
definition. Under one scenario, the current ``three-tier'' structure 
would be retained, but there would be certain changes to the conditions 
that define an operation as a CAFO in the middle tier (i.e., 300-1,000 
AU). EPA also proposed an alternative regulatory approach that would 
replace the existing ``three-tier'' structure with a ``two-tier'' 
scenario for defining operations as CAFOs. Under the ``two-tier'' 
scenario, all animal feeding operations with more than a specified 
number of animals would be defined as a CAFO. EPA considered several 
potential thresholds that could be set under the two-tier scenario.
    EPA also proposed to revise the definition of a CAFO to expressly 
include chicken operations using dry litter management techniques, 
swine nurseries, and heifer operations. EPA proposed to explicitly 
address manure

[[Page 48101]]

application on land under the control of the CAFO, and considered 
alternatives for collecting information regarding manure transferred to 
off-site locations. The proposed rule also considered certain changes 
that affect which entities would be required to obtain NPDES permits, 
and proposed to add provisions requiring CAFOs that cease operation to 
retain their NPDES permits until all wastes that were generated by the 
operation no longer have the potential to reach waters of the United 
States.
    For more information on the proposed changes to the NPDES 
regulations, see section VII of the proposed rule (66 FR 2993-3050).

B. Notice of Data Availability

    On November 21, 2001 (66 FR 58556), EPA published a notice of data 
availability presenting a summary of new data and information submitted 
to EPA during the public comment period on the proposed CAFO 
regulations, including data received from USDA. The 2001 Notice also 
discussed new data and changes being considered to refine the cost and 
economic analysis models, and to improve estimates of pollutant 
reductions and monetized benefits that would result from changes to the 
CAFO regulations. EPA presented information on potential changes that 
would enhance flexibility for using State NPDES and non-NPDES CAFO 
programs, discussed options intended to encourage broader 
implementation of environmental management systems, and described 
certain refinements to the CAFO definition that were under 
consideration.

III. Thresholds for Chicken Operations Using Dry Litter Management

    EPA's existing effluent guidelines for CAFOs apply to chicken 
operations with 30,000 laying hens or broilers when the facility has a 
liquid manure handling system, and to chicken operations with 100,000 
laying hens or broilers when the facility has unlimited continuous flow 
watering systems. (See 40 CFR Part 412.10). Under the proposed rule, 
the CAFO regulations would be revised to remove language referring to 
the type of manure handling or watering system employed at laying hen 
and broiler operations and would, as a result, expand the scope of the 
rule to also address chicken operations with ``dry'' litter management 
systems. (The term ``dry'' does not mean that there are no wastewaters 
associated with these types of operations. For example, poultry waste 
includes manure, poultry mortalities, litter, spilled water, waste 
feed, water associated with cleaning houses, runoff from litter 
stockpiles, and runoff from land where manure has been applied.) As 
proposed, the revised CAFO regulations would establish 100,000 chickens 
as equal to 1,000 AU. (See 66 FR 3010-3012).
    At proposal, EPA presented two alternative ways to structure the 
NPDES regulations and define which animal feeding operations are CAFOs. 
Under EPA's proposed ``two-tier'' structure, all AFOs with more than a 
certain AU threshold level would be defined as CAFOs, and those with 
fewer than the threshold would become CAFOs only if they were 
designated as such by the permittting authority. Under this 
alternative, with a threshold of 500 AU, for example, all chicken 
operations with more than 50,000 chickens would be defined as a CAFO. 
Under this two-tier structure with the threshold set at 500 AU, EPA 
estimates 9,300 broiler operations and 1,000 laying hen operations 
would be automatically defined as CAFOs.
    In the second alternative discussed at proposal, EPA proposed to 
retain the current ``three-tier'' structure. Under this three-tier 
scenario, operations with more than 100,000 chickens would be 
automatically defined as a CAFO, and operations with 30,000 to 100,000 
chickens would be defined as a CAFO only if they met certain 
conditions. Under the three-tier structure, EPA estimates 2,950 broiler 
operations and 550 laying hen operations would have more than 1,000 AU 
and would automatically be defined as CAFOs. EPA also estimates an 
additional 600 broiler operations and 50 laying hen operations would be 
defined as middle-tier CAFOs (i.e., those with 301-1,000 AU) using 
EPA's current middle-tier criteria. See 66 FR 2996-3004 for additional 
discussion of the two-tier and three-tier regulatory structures.
    In developing the proposed rule, EPA evaluated several methods for 
equating poultry operations with dry litter management to the existing 
definition of an animal unit (See 66 FR 3010-3012). One factor 
considered is that the existing CAFO regulations already apply to 
chicken operations with 100,000 laying hens or broilers when the 
facility has unlimited continuous flow watering systems. Another factor 
considered relates to the manure generated by chickens in comparison to 
the manure generated by beef cattle. The average daily manure 
generation from 100,000 broilers and laying hens (EPA's proposed metric 
for the number of chickens being equal to 1,000 AU) is comparable to 
the average daily manure generation from 1,000 beef cattle (1,000 AU). 
Using manure waste characterization data from USDA's Agricultural Waste 
Management Field Handbook, EPA's analysis indicated a range of 82,000 
laying hens to 111,000 broilers--or approximately 100,000 chickens--as 
being equivalent to 1,000 AU. EPA's methodology for calculating these 
values is presented in the record. This analysis suggested a similar 
threshold for chickens whether basing the comparison of manure on the 
amount of nitrogen, phosphorus, or biochemical oxygen demand 
(BOD5) in the manure.
    EPA is considering other thresholds both higher and lower than the 
100,000-chicken threshold presented in the proposed rule.
    Several comments were received on EPA's proposed thresholds for 
chicken operations, asserting EPA should maintain a distinction between 
laying hens and broilers. Other comments asserted that EPA should 
determine the value (i.e., number of birds) equating to 1,000 AU by 
evaluating phosphorus content in the manure on an annual basis as 
opposed to a daily basis. For example, these comments further assert 
that estimates of the annual phosphorus production should reflect that 
five to six flocks of broilers are produced per year, and should not 
assume phosphorus production continues during the periods of the year 
(i.e., cleanout time between flocks when no broilers are present) when 
no manure is generated. Using an approach for setting the threshold 
that compares the phosphorus produced by chickens annually to the 
phosphorus produced by beef cattle, based on manure waste 
characterization data from USDA's Agricultural Waste Management Field 
Handbook, EPA would estimate the 1,000 AU equivalent as 125,000 
broilers (in contrast to the 111,000 value estimated using the daily 
manure generation rates). EPA's methodology for calculating these 
values (e.g., average bird live weight; typical number of flocks 
produced per year; average time between bird placements ) is presented 
in the record. Using an alternative threshold of 125,000 broilers, EPA 
estimates 1,800 broiler operations would have greater than 1,000 animal 
units. Because laying hens typically are kept at CAFOs for 
approximately 94 weeks of production, they continue to produce manure 
throughout the year and EPA's previous estimate of 82,000 laying hens 
as being equivalent to 1,000 AU remains unchanged.
    Additional information regarding the nutrient and BOD5 
content of beef and chicken manure can be found in section 17 of the 
public record for the CAFO rulemaking. The USDA data used by

[[Page 48102]]

EPA to estimate the number of broiler and laying hen operations that 
would have more than 1,000 AU under the alternative thresholds 
discussed in this notice are included in section 19.1 of the record. 
Detailed information on EPA's analyses and assumptions appears in 
section 19.5 of the record. Section 21 of the record contains public 
comments received on the proposed rule and 2001 Notice regarding the 
threshold for chicken operations. See the ADDRESSES section of this 
notice for information on how to obtain access to the public record for 
the CAFO rulemaking.
    EPA is considering whether the 1,000 AU equivalent for broilers 
should remain as proposed at 100,000 broilers, or whether it should be 
changed to either 125,000 broilers. EPA is also considering whether the 
1,000 AU equivalent for laying hens should remain as proposed at 
100,000 laying hens, or whether it should be changed to 82,000 laying 
hens. EPA notes that the thresholds codified in the current regulations 
for operations with liquid manure handling systems or unlimited 
continuous flow watering systems may remain unchanged in the final 
rule. EPA solicits comment on these alternative thresholds for broiler 
and laying hen operations with dry litter management systems, the 
assumptions and data used to derive the thresholds (e.g., average bird 
liveweight; typical number of flocks produced per year; average time 
between bird placements), and if other alternative thresholds (and 
their technical basis) exist that may be appropriate for these 
operations.

IV. Voluntary Alternative Performance Standards for Innovative 
Technologies

    EPA's long-term environmental vision for CAFOs includes continuing 
research and progress toward environmental improvement. The Agency 
believes that individual CAFOs should be encouraged to voluntarily 
develop and install technologies and management practices that achieve 
pollutant reductions equivalent to or better than those required by the 
baseline effluent guidelines regulations.
    Further, EPA recognizes that some CAFOs, as well as USDA, land 
grant universities, equipment vendors and agricultural organizations, 
are working to develop new technologies that achieve reductions in 
nutrient and pathogen losses to surface water, ammonia and other air 
emissions, and ground water contamination. The development of new 
technologies offers the potential to match or surpass the pollutant 
reductions that would be achieved by compliance with limits and 
standards in the final CAFO rule.
    EPA received suggestions from a number of stakeholders on the 
merits of creating a framework for alternative performance standards. 
Several stakeholders believe that the current and proposed effluent 
guidelines discourage the use of innovative treatment and pollution 
prevention technologies and that EPA should include incentives to 
encourage CAFOs to use improved technologies that would protect all 
environmental media (particularly surface water, but also air and 
ground water). A number of commenters expressed support for the 
inclusion of voluntary alternative technologies which are equivalent to 
or better than BAT effluent guidelines (or NSPS requirements for new 
CAFOs), and specifically requested a provision that would allow CAFOs 
to discharge treated process wastewater generated from the production 
area of the CAFO.
    A number of stakeholders commented that EPA should include controls 
for pathogens or antibiotics, as well as atmospheric emissions of 
ammonia, methane, or hydrogen sulfide. Other commenters suggested that 
adding flexibility in the rule to allow for the discharge of treated 
process wastewater could lead to better approaches for addressing 
environmental concerns in all environmental media, including air, 
ground water, and surface water.
    In view of these suggestions, today's notice presents two 
approaches, described below, to encourage the development of new 
technologies and management practices that achieve pollutant reductions 
equivalent to or better than those that would be achieved by the 
baseline BAT (and NSPS) that will be promulgated in the final rule, and 
possibly also help address multimedia issues related to air emissions 
and ground water. Under a Production Area Approach, alternative 
performance standards would focus on the manure and wastewater 
discharges from the CAFO production area and CAFOs would be allowed to 
discharge process wastes that have been treated by technologies that 
result in equivalent or better pollutant removals than would be 
achieved under the baseline BAT standard. Under the Whole Farm 
Approach, CAFOs would conduct a site-specific ``whole farm'' multimedia 
review to target optimal pollutant load reduction and pollution 
prevention opportunities for the production and land application areas. 
The Whole Farm Approach could include an allowance for wastewater 
discharge from the production area as described for the Production Area 
Approach, but most importantly, would require the CAFO to evaluate and 
implement whole farm improvements through the use of an audit process 
as a condition for obtaining alternate effluent limits.
    A key tenet of these approaches is that CAFOs would voluntarily 
choose to comply with an alternative BAT/NSPS performance standard as 
the basis for their technology-based NPDES permit limits (e.g., 
inclusion of effluent limitations in their NPDES permits that are 
different from those based on the baseline effluent guideline, to be 
established by the NPDES permitting authority on the basis of best 
professional judgement). CAFOs would not be required to enter the 
alternative standards program. A CAFO choosing not to participate in 
the alternative standards program would instead be subject to the 
baseline BAT limitations (discussed below in section IV.A). EPA 
previously used a similar approach in the effluent guidelines for the 
pulp and paper industry. See 63 FR 18504, 18593-18611 (April 15, 1998).
    EPA solicits comment on the concepts presented in this notice for 
creating alternative performance standards to encourage CAFOs to 
implement new technologies. In sections IV.B, IV.C, and IV.D below, EPA 
also solicits comment specifically on certain aspects related to the 
Production Area and Whole Farm Approaches for creating alternative 
performance standards, and on the potential incentives that may be made 
available to CAFOs participating in an alternative performance 
standards program.

A. Baseline BAT

    Under the current effluent guidelines regulations, CAFOs are 
prohibited from discharging process wastewater, except when rainfall 
events cause an overflow from a facility designed, constructed, and 
operated to contain all process-generated wastewater plus the runoff 
from a 25-year, 24-hour rainfall event. The limits included in the 
effluent guidelines are based on the use of storage ponds and lagoons 
to contain the process wastes and runoff, but they do not prevent CAFOs 
from using alternative technologies, as long as those technologies also 
meet zero discharge or the containment requirement. These limitations 
were established on the basis of factors specified in Clean Water Act 
sections 304(b) and 306(b), including the cost of achieving the 
effluent reductions and any non-water quality environmental impacts. 
EPA continues to assess the large number of comments and data received 
on the proposed rule regarding the appropriate technology basis for the 
BAT/NSPS requirements

[[Page 48103]]

that will be promulgated in December 2002 (referred to in this notice 
as the ``baseline BAT'').

B. Production Area Approach

    The Production Area Approach focuses on manure and wastewater 
discharges from the CAFO production area. Under this approach, CAFOs 
would be allowed to discharge process wastewater that has been treated 
by technologies that the CAFO demonstrates will result in equivalent or 
better pollutant removals than would be achieved by the baseline BAT 
standard. The requirements applying to wastewater discharges could also 
be coupled with either a regulatory provision or non-regulatory 
guidance for participating CAFOs to develop a plan for achieving 
improvement in multiple environmental media.
    As discussed above, the baseline BAT standard, though nominally 
zero discharge, allows for untreated overflow discharges if the system 
is designed, constructed and operated to contain process wastewater 
plus the runoff from a 25-year, 24-hour rainfall. Thus, to demonstrate 
that the alternative technology would achieve equivalent or better 
pollutant reductions than baseline BAT requirements, the CAFO would be 
required to submit a technical analysis, which would include 
calculating the mass-based pollutant reductions based on the site-
specific modeled performance of the baseline BAT system (currently, 
defined as a 25-year/24-hour storage lagoon). Under this approach, a 
computer simulation model could be used to evaluate site-specific or 
region-specific climate data, along with wastewater characterization 
data, to determine the mass-based pollutant discharge that would be 
projected for a system designed, constructed and operated to achieve 
compliance with the baseline BAT standard. The model would evaluate the 
daily inputs to the storage system, including all process wastes, 
direct precipitation, and runoff. It would also evaluate the daily 
outputs from the storage system, including losses due to evaporation, 
sludge removal, and the removal of wastewater for use on cropland at 
the CAFO or transport off site. The model would be used to predict the 
overflow from the BAT system that would occur over a 25-year period, 
and these overflow predictions would be used to determine the median 
annual predicted overflow over the 25 years evaluated by the model. 
Site-specific or other appropriate pollutant characterization data for 
the wastewater from the waste storage system (i.e., the overflow) would 
then be coupled with the overflow volume output from the model 
described above to predict the mass pollutant discharge that would 
occur from a baseline BAT system. CAFOs would be required to meet NPDES 
permit conditions that result in equivalent or improved pollutant 
reductions, as compared to the predicted mass discharge from overflow 
of the baseline BAT system, for example, on an annual basis or over the 
lifetime of the permit. If a CAFO elected to use this approach it would 
be meeting the same limitations as a CAFO under the baseline BAT, but 
expressed in a different fashion (e.g., numeric limits on a continuous 
discharge versus a limit of zero discharge with an allowance for 
discontinuous overflows). To illustrate this type of analysis, EPA has 
prepared an example evaluation using model farm characteristics. This 
example is available in section 19.6.2 of the rulemaking record. Land 
application activities would be required to correspond to an approved 
nutrient management plan.
    A variation of this approach could be based on a more holistic 
approach that considers other environmental media besides discharges to 
surface water. Under this approach, CAFOs would be authorized to comply 
with alternative BAT/NSPS performance standards if they implement 
technologies and management practices that result in equivalent or 
improved pollutant reductions, including all media. CAFOs that achieve 
significant reductions in air emissions or ground water discharges for 
a pollutant would qualify for less stringent limits on discharges to 
surface water to be established on the basis of best professional 
judgement. In essence, EPA would be using the authority of Clean Water 
Act section 304(b) to establish alternative BAT requirements that 
address the non-water quality environmental impacts from controls of 
discharges to other media, as well as the costs of those controls. One 
challenge with this approach is how to determine ``equivalence'' across 
environmental media.
    This approach would essentially divide up CAFOs within a 
subcategory into different segments. Those CAFOs which have undertaken 
or will voluntarily undertake actions to control air emissions or 
ground water discharges can be distinguished from facilities which have 
not under Clean Water Act sections 304(b) and 306(b), because they face 
different economic achievability concerns related to the costs of 
compliance with the effluent guidelines, or because their activities 
will have fewer non-water quality environmental impacts. EPA adopted a 
similar set of alternate requirements for the pesticide chemicals 
formulating, packaging and repackaging industry when EPA found that 
facilities using an alternative pollution prevention approach would 
reduce air emissions. See 61 FR 57518, 57525-26 (November 6, 1996).
    EPA solicits comment on the following: (1) The criteria and process 
that would be used under the Production Area Approach to demonstrate 
performance equivalent to or better than the baseline BAT technology; 
(2) the appropriate methodology for translating annual mass discharge 
estimates into an NPDES permit limitation; (3) approaches for comparing 
the intermittent overflow discharge that would occur under the baseline 
BAT requirement to the continuous treated discharge that may be allowed 
under an alternative performance standard; and (4) whether a holistic 
approach that considers pollutant reductions across all environmental 
media would be appropriate, how equivalence across media could be 
determined operationally and embodied in NPDES permit limits, on what 
statutory basis could EPA distinguish CAFOs that employ the holistic 
approach, and whether the NPDES permit could and should mandate 
compliance with the pollutant reductions achieved across media.

C. Whole Farm Approach

    The Whole Farm Approach is based on conducting a site-specific 
multimedia review to target optimal pollutant load reduction and 
pollution prevention opportunities for both the production and land 
application areas. This approach could include an allowance for 
wastewater discharge from the production area as described for the 
Production Area Approach, but most importantly, would require the CAFO 
to evaluate and implement whole-farm environmental improvements through 
the use of a site-specific audit process as a condition for qualifying 
for alternative BAT limits. At a minimum, as part of the audit, the 
CAFO would be required to use a mass-balance approach to address site-
specific concerns (e.g., karst geology, flood plains) and to quantify 
their existing releases; identify the potential to reduce losses from 
the production area, land application area, and transfer of manure off 
site; and identify specific opportunities to reduce the largest 
releases (to surface water, ground water, air, or land). In general, 
EPA would expect the CAFO to evaluate releases that occur at the point 
of generation first to minimize or eliminate waste

[[Page 48104]]

production and air emissions, followed by an evaluation of the waste 
handling and management systems, and ending with an evaluation of land 
application and off-site transfer operations.
    CAFOs would need to develop and implement a plan for the operation 
that generates improvement across multiple environmental media. The 
plan would identify the specific technologies or practices that will be 
installed or implemented to achieve the estimated pollutant reductions, 
and provide criteria that demonstrate effective performance of these 
technologies or practices that could be used to determine compliance. 
The specific approaches used would be expected to vary somewhat among 
operations and would be selected by the CAFO as being effective for the 
particular operation. Potential approaches could include:
     Implementation of feeding strategies (to reduce or 
eliminate nutrients, hormones, and/or antibiotics);
     Installation of new and innovative waste management 
technologies;
     Changes to animal housing;
     Changes to the type and frequency of cleaning operations;
     Controls for the existing waste management system (e.g., 
storage liners, covers);
     Energy recovery systems;
     Centralized waste treatment or processing;
     Stabilization and production of value-added products;
     Changes to land application methods (e.g., erosion control 
measures, incorporation/injection);
     Controls for air emissions (e.g., ammonia, particulate 
matter, methane, hydrogen sulfide);
     Implementation of methods to ensure off-site land 
application follows nutrient management plan approach; and
     Implementation of a mortality disposal plan.
    The implementation plan would need to present data to demonstrate 
that the plan results in whole-farm reductions in pollutant releases to 
surface waters equivalent to or better than would be achieved by the 
baseline BAT requirements. As discussed under the Production Area 
Approach, this would result in equivalent BAT effluent limitations, but 
expressed in a different fashion.
    Alternatively, the Whole Farm Approach could also be based on a 
more holistic measure of pollutant reduction and allow trade-offs among 
reductions in discharges to different media, as long as the plan 
resulted in equivalent or improved pollutant reduction overall. As 
discussed above in section IV.B, such an approach would need to 
determine how to compare reductions across environmental media. As 
discussed under the Production Are Approach, EPA would utilize its 
statutory authority to distinguish between facilities that voluntarily 
achieve reductions to other media and those that do not, on the basis 
of cost, non-water quality environmental impacts, or other factors.
    To illustrate the Whole Farm Approach, EPA has prepared a 
hypothetical example process evaluation using model farm 
characteristics. This example is available in section 19.6.2 of the 
record.
    The whole farm approach offers many benefits to the CAFO and to the 
environment. By targeting reductions of pollutant releases to all 
media, the CAFO may find ways tailored to the individual site to more 
cost-effectively minimize environmental impacts. The approach offers 
flexibility in choosing an environmental system that is most effective 
and affordable for the specific site, and encourages CAFOs to go beyond 
the minimum regulatory requirements. This type of program also offers 
opportunities for state and local partnerships to evaluate location-
specific issues and develop targeted approaches.
    A potential obstacle to implementing new technologies is the 
tension between a requirement that CAFOs comply immediately with BAT at 
the time of permit issuance, and the time that may be required to 
develop and implement a new technology. While immediate compliance may 
promote, in the short term, prompt implementation of BAT technologies, 
EPA is concerned that such a requirement can also discourage CAFOs from 
fully investigating and implementing alternative technologies that may 
be better than the baseline BAT technology. EPA is considering, as part 
of the Whole Farm Approach only, providing CAFOs who choose to 
implement whole-farm multimedia approaches under the alternative 
standards program additional time to implement and meet the alternative 
performance standards. In this way, EPA hopes to provide an incentive 
for CAFOs to implement whole-farm reductions in pollutant releases. EPA 
used a similar approach in the effluent guidelines for the pulp, paper, 
and paperboard industry. Facilities were required to meet BAT 
reflecting existing practice in the short-term in order to implement a 
more aggressive BAT (not economically achievable in the short-term) at 
a later date.
    While EPA public recognition programs already exist, the Agency 
believes that it may also be appropriate to develop and implement a 
program unique to this industry as an incentive to invest in new 
technologies and whole-farm approaches to reducing pollutant releases. 
As part of a public recognition program, EPA could establish criteria 
under which CAFOs would qualify to receive public recognition on an 
annual basis. In addition to commitments leading to and achievement of 
the limits specified or additional reductions associated with a whole 
farm approach, such criteria would likely include some form of periodic 
compliance audit and could be structured to give CAFOs flexibility to 
implement an environmental management system approach. EPA would then 
recognize the qualifying CAFOs each year through a public event.
    EPA solicits comment on the following: (1) The criteria and process 
that would be used under the Whole Farm Approach to demonstrate 
pollutant reductions equivalent to or better than the baseline BAT 
technology; (2) the appropriate methodology for translating the actions 
identified in the plan for the Whole Farm Approach into an NPDES permit 
limitation; (3) approaches for comparing the intermittent overflow 
discharge that would occur under the baseline BAT requirement to the 
whole-farm actions CAFOs propose to undertake; (4) the length of time 
CAFOs should be afforded to implement whole-farm pollutant reduction 
actions; and (5) the possible incentives described in this section for 
CAFOs implementing the Whole Farm Approach, the applicable criteria 
used to qualify for the incentives, the type of public recognition that 
would be afforded, and the frequency for recurring public recognition.

D. Process and Incentives for Participating in Alternative Standards

    CAFOs interested in pursuing either alternative standards approach 
should have a good compliance history. The facility would also be 
expected to conduct an analysis of their operation (as described above 
in sections V.B. and V.C.) and prepare a proposed alternative program 
plan including the results of the analysis, the proposed method for 
implementing new technologies and practices, and the results 
demonstrating that these technologies and practices perform equivalent 
to or better than baseline BAT. This plan would be included with their 
permit application or renewal, and would be incorporated into the 
permit. EPA solicits comment on: (1) The process and criteria that

[[Page 48105]]

should be used by CAFOs to apply and qualify for participation in the 
alternative performance standards program; (2) whether CAFOs that have 
a deadline for ``future BAT'' under an alternative performance 
standards program should have interim milestones incorporated in their 
permits towards meeting the ultimate BAT standard; (3) how the program 
should address CAFOs that volunteer to participate in the alternative 
standards program, yet later back out of the alternative standards 
program without implementing the changes outlined in their plan; (4) on 
the length of time that CAFOs should be afforded for development and 
implementation of the plan; and (5) what should the BAT basis be for 
requirements during the period of development of the alternative 
standards program (e.g., ``existing effluent quality,'' as EPA used for 
the pulp and paper effluent guidelines, or some other basis).
    CAFOs potentially may derive substantial benefits from 
participation in the alternative standards approach, through greater 
flexibility in operation, increased good will of neighbors, reduced 
odor emissions, and potentially lower costs. EPA is also exploring 
opportunities for other possible incentives to encourage participation 
in this program. EPA solicits comment on the possible incentives 
discussed in this notice, and invites suggestions for other incentives 
that should be made available.

V. Changes to the Economic Analysis

    This section presents changes that EPA is considering to the 
methodology and underlying financial data that it uses to assess the 
economic effects of the final regulations to CAFOs. Many of these 
changes reflect comments and new data that EPA has obtained since 
proposal, which were broadly described in the 2001 Notice. Today, EPA 
presents additional information on the approach and data that would be 
used for an economic analysis of the final rule. Section V.A of this 
notice describes the modeling framework and changes being considered to 
assess financial effects to regulated CAFOs. Section V.B of this notice 
describes the financial data that EPA is considering using to depict 
baseline conditions at model CAFO facilities. Section V.C discusses 
preliminary results of analyses using these alternative data and 
approaches.

A. Changes to Model Framework and Assumptions

    EPA expects the economic analysis for the final rule will retain 
the general modeling framework that the Agency used to assess economic 
effects for the proposed rule (see 66 FR 3079-3103), with the 
modifications discussed in the 2001 Notice (see 66 FR 58577-58591). The 
2001 Notice describes a range of methodological changes and financial 
data EPA was considering using to improve its analysis. Today's notice 
provides further information on the specific changes being considered 
for the modeling framework and financial data EPA will use to analyze 
the regulatory options for the final rule.
1. Farm Level Analysis
    The farm level analysis that supports the final rulemaking is 
expected to retain the same general framework used for the proposed 
rule. Specifically, financial impacts are assessed using a sales test, 
discounted cash flow analysis, and a debt-asset test. This evaluation 
is conducted using farm level financial data that are described in 
Section V.B of this notice and are available in EPA's record. These 
farm level data reflect income and cost information spanning an 
operation's primary livestock production, as well as secondary 
livestock and crop production, government payments, and other farm-
related income. As conducted for the proposed rule, EPA would divide 
the resultant regulatory impacts into defined categories. Operations 
with estimated financial effects that are ``affordable'' or 
``moderate'' would not be considered to be vulnerable to closure post-
compliance and would, therefore, be considered to indicate economic 
achievability. Operations with estimated financial ``stress'' would be 
considered to be vulnerable to closure and may not be considered to 
indicate economic achievability, subject to other considerations.
    To address public comments submitted to EPA on the overall 
analysis, EPA is considering making three general changes to its 
analytical framework at the farm level.
    First, for the final analysis, EPA proposes to use a sales test 
that would use pre-tax incremental cost, as opposed to costs that take 
into account potential tax savings (post-tax), which was assumed at 
proposal. These pre-tax costs would be compared to total farm level 
revenues and that ratio would be used as an initial screener to 
determine the need for additional analysis using EPA's discounted cash 
and debt-asset tests.
    Second, EPA is considering using alternate baseline debt and asset 
information for several livestock sectors (beef, heifer, veal, dairy, 
and hog) that EPA has obtained since proposal and is considering a 
change to the debt-asset threshold values that would indicate financial 
stress for these sectors. Consideration of alternative debt and asset 
data for these sectors is consistent with recommendations by National 
Cattlemen's Beef Association (NCBA), the National Milk Producers 
Federation (NMPF), and the National Pork Producers Council (NPPC) and 
other industry commenters. Data submitted to EPA by NCBA and the Food 
and Agricultural Policy Research Institute (FAPRI) during the comment 
period indicate that larger, more intensive, or expanding cattle 
feeding operations tend to carry more debt than that assumed by EPA for 
the proposal. (Average USDA-reported data cover a broader range of farm 
types and sizes, including small farms and non-confinement operations 
that are not covered by the regulations.) Financial data submitted by 
NCBA and FAPRI indicate that confinement operations with more than 
1,000 AU have baseline debt-asset levels greater than the USDA-
recommended 40 percent, ranging from 60 percent to more than 70 percent 
in the beef, dairy and hog sectors. Since USDA's recommended 40 percent 
benchmark value may not be suitable for assessing the larger confined 
cattle, dairy, and hog operations affected by EPA's regulations, EPA is 
considering using an alternate threshold value for its debt-asset test 
for these sectors. Based on recommendations by NCBA submitted to EPA 
since the proposal, EPA is considering an 80 percent threshold value to 
indicate financial stress (see information submitted by NCBA at DCN 
375047 in the record). EPA requests comment on the use of an 
alternative benchmark, such as 80 percent, for these sectors, if 
alternative data are used. EPA also requests comment on the use of 
alternative debt and asset data for the cattle, dairy, and hog sectors. 
These data are available in the rulemaking record (see: DCN 175044 and 
DCN 175038). (Due to limited data, EPA will continue to use USDA-
reported average debt and asset information for the poultry sectors 
(broiler, egg, and turkey) and will continue to assess changes in the 
debt-asset test for these sectors assuming a 40 percent benchmark, as 
was done in the analysis for the proposed rule.)
    A third change being considered for the final analysis involves the 
use of time series data to project available financial data onto a 10-
year time horizon for EPA's discounted cash flow analysis. For the 
proposed rule, EPA used data projections by USDA. As discussed in the 
2001 Notice, many

[[Page 48106]]

commenters disagree with the use of this data series as the basis for 
EPA's projections. To address these comments for the final analysis, 
EPA is considering using alternative timeline data from FAPRI (hog and 
poultry sectors), USDA (dairy sector), and NCBA (cattle sector) to 
project future earnings from the 1997 baseline data. A summary of these 
data and EPA's projected values based on these data is available for 
review at DCN 375084. The method that EPA uses to project the baseline 
data follows the approach used for the proposal analysis, as discussed 
in the Economic Analysis for the proposed rule.
2. Enterprise Level Analysis
    For the final rule, EPA is considering expanding upon the analysis 
developed for the proposed rule by including an assessment of the 
financial effects on the enterprise level (e.g., an operation's 
livestock or poultry enterprise). This modeling change would address 
comments expressed by many commenters, including FAPRI, other land 
grant university researchers, and industry, as well as USDA, as 
discussed in the 2001 Notice (66 FR 58580-58582). These comments are 
supported by alternate enterprise level data that were submitted to EPA 
since proposal and presented in the 2001 Notice. An enterprise level 
analysis would recognize that a farm may be unwilling to cross-
subsidize a continually failing livestock operation. Also, this 
approach would recognize that a failing enterprise with continuous cash 
flow problems would have limited access to financing for capital 
replacement and/or expansion, despite the health of the overall 
business. EPA is considering addressing this concern by including, as 
part of its final analysis, an assessment of changes in enterprise 
level profitability, in addition to the results of the farm level 
analysis. This analysis would be conducted using the enterprise level 
financial data described in Section V.B of this notice. A summary of 
these data are available at DCN 375084 in the rulemaking record.
    Since the publication of the 2001 Notice, EPA has evaluated ways to 
incorporate an enterprise level analysis as part of its assessment. To 
evaluate enterprise level effects, EPA is considering using enterprise 
level net cash income to develop a discounted cash flow (DCF) estimate 
for each model enterprise over the 10-year time frame of the analysis. 
The net present value of cash flow is compared to the net present value 
of the total cost of the regulatory options. If the farm level analysis 
shows that the regulations impose ``affordable'' or ``moderate'' 
effects on the operation, the enterprise level analysis would be 
conducted to determine whether the enterprise's cash flow is able to 
cover the cost of regulations. Over the analysis period, if an 
operation's livestock or poultry enterprise maintains a cash flow 
stream that both exceeds the cash costs of the BAT option (operating 
and maintenance costs plus interest) and also covers the net present 
value of the principal payments on the capital, EPA would assume that 
the enterprise will likely not close due to the CAFO regulations. EPA 
is also considering whether to add some measure of capital replacement 
costs to both its farm and enterprise level cash flow analysis. This 
analysis would be conducted on a pass/fail basis. If the net present 
value of cash flow minus the net present value of the BAT costs is 
greater than zero, the enterprise passes the test and the enterprise is 
assumed to continue to operate. If the net present value of cash flow 
is not sufficient to cover the net present value of the cost of the 
regulations, EPA would assume that the CAFO operator would consider 
shutting down its livestock or poultry enterprise.
    The enterprise level analysis would build on the farm level 
analysis, evaluating effects at a farm's livestock or poultry 
enterprise. If the operation shows farm level impacts that are 
``affordable'' or ``moderate,'' then an enterprise level analysis is 
conducted to determine whether the operation's livestock or poultry 
enterprise remains viable. If enterprise level profitability remains 
positive over the period of the analysis, then the requirements would 
be determined to be economically achievable to the entire operation, as 
well as the livestock or poultry enterprise at the business. Enterprise 
level results would be presented in addition to estimated farm level 
effects (i.e., estimated farm impacts would comprise a subset of 
reported enterprise impacts) and both the farm and enterprise level 
results could be considered in determining economic achievability. 
Results indicating ``affordable'' or ``moderate'' farm level effects, 
but where enterprise level profitability is negative (i.e., the farm 
remains in business but the livestock or poultry enterprise at that 
business is discontinued) would be subject to further analysis before a 
final assessment is made. Operations that are determined to experience 
financial ``stress'' at the farm level would not be further evaluated 
because it is assumed that these facilities would go out of business. 
Additional information about this analysis is provided in the 
rulemaking record (DCN 375084).
3. Other Model Framework Changes
    A summary of other changes being considered for the economic models 
is as follows. First, EPA is considering expanding the range of cost 
estimates per representative farm to account for variability across 
operations based on expected capital and management improvements 
needed, using data from USDA. These data were discussed in the 2001 
Notice (see 66 FR 58572-58573). This change, along with other changes 
to expand EPA's costing approach, would effectively increase the number 
of cost models in EPA's analysis from about 200 to approximately 1,600 
representative models. Second, for reasons outlined in the 2001 Notice, 
EPA may not include a debt feasibility test as part of its analysis of 
the final rule because a down payment assumption is not necessary given 
EPA's joint analysis of debt-to-asset ratios and cash flow (see 66 FR 
58583-58584).
    EPA continues to review options to consider additional potential 
cost offsets as part of its final analysis, including available cost-
sharing and technical assistance to farmers under various Federal, 
State and local conservation programs. In particular, at the Federal 
level, new farm bill legislation passed this spring by Congress may 
significantly raise government expenditures for USDA conservation 
programs. For example, total Environmental Quality Incentives Program 
(EQIP) authorization for FY 2002-2007 is $5.8 billion, ranging from 
$400 million to $1.3 billion per year over the period. This compares to 
current authorized levels of about $200 million per year. The new 
legislation targets 60 percent of available EQIP funds to livestock and 
poultry producers, including confinement and grass-based systems (the 
latter accounting for about 70 percent of total livestock and poultry 
operations). The new legislation also removed the previous eligibility 
requirements under EQIP that restricted funding for certain structural 
practices to operations with fewer than 1,000 animal units (as measured 
by USDA), replacing this with an overall payment limitation of $450,000 
per producer over the authorized life of the 2002 Farm Bill. Under 
EQIP, cost sharing may cover up to 75 percent of the costs of certain 
conservation practices. The debate surrounding these increased funding 
levels included a focus on assisting producers to comply with 
environmental regulations.
    EPA believes that this increased funding in EQIP and other USDA 
conservation programs may benefit

[[Page 48107]]

farmers and offset compliance costs incurred by some facilities under 
the CAFO regulations by increasing farm access to government cost-share 
dollars and increased technical assistance. EPA is considering two 
approaches that would incorporate cost share assumptions as part of 
EPA's CAFO level analysis. One approach would assume that cost sharing 
would cover up to 75 percent of the estimated capital compliance costs, 
spread out over the 10 year period of the analysis. A second approach 
would be similar to that adopted for a previous USDA and EPA impact 
analysis of confined animal operations and would assume average per-
farm cost share information, as reported by USDA, as an offset to 
estimated capital costs. EPA solicits comment on these possible 
approaches and requests additional information to incorporate cost 
share assumptions as part of EPA's CAFO level analysis. Specifically, 
EPA requests information on how to account for uncertainty about actual 
program funding levels and uncertainty about which producers would 
obtain these funds and in what amount.
    EPA will also continue to evaluate expected broader market level 
changes and adjustments. EPA is considering adjustments to the approach 
used for the proposal analysis by instead utilizing predicted price and 
quantity changes from EPA's market model analysis. The market model 
output information would be used to adjust the baseline financial data 
that are assumed for EPA's CAFO level analysis. Such an approach is 
more consistent with previous regulatory analyses conducted by EPA's 
effluent guideline program (e.g.: 65 FR 49686). EPA solicits comment on 
this modification to the approach used for proposal.

B. Changes to the Baseline Financial Data

    This section provides information specific to each animal sectors 
and describes the data that EPA is considering using, given the 
availability of financial data from a variety of sources. More detailed 
citations and the actual farm and enterprise level input data that EPA 
is proposing to use for its analyses are included in the rulemaking 
record, with a summary of these data available at DCN 375084.
1. Overview
    EPA received many comments on the financial data used to estimate 
CAFO level effects for the proposed rule. For proposal, EPA 
incorporated only farm level financial data into its analysis. For the 
final regulations, EPA is considering using these farm level data for 
some animal sectors, substituting the 1997 USDA with other data 
received by EPA in conjunction with financial data specified at the 
enterprise level. This change in the approach and underlying data for 
the analysis is discussed in the 2001 Notice (see 66 FR 58585-58590). 
This section discusses the data that EPA is considering using for its 
final analysis.
    For most sectors, EPA will continue to use available 1997 data from 
USDA reflecting financial conditions at the farm level, which EPA used 
for proposal. For two sectors--the cattle feeding and hog sectors--EPA 
is replacing the 1997 USDA data used for proposal with other data 
presented in the 2001 Notice. For cattle, EPA uses financial data 
provided by NCBA and FAPRI; for hogs, EPA uses farm level data from 
USDA. For the dairy and poultry sectors, EPA will continue to employ 
the 1997 USDA financial data used for proposal.
    To address comments that criticize EPA's use of a single year of 
financial data to reflect baseline conditions, EPA is considering 
adjusting the financial data for the cattle, hog, and dairy sectors 
based on other available published data from USDA, FAPRI, and the land 
grant universities to average out conditions over multiple years. This 
approach involves incorporating other available data into the analysis 
to obtain average conditions over a multiple year time frame, as 
discussed in the 2001 Notice (see 66 FR 58590-58591). Due to lack of 
multiple years of financial data for the poultry sectors, EPA is not 
able to use this approach for those types of operations and is instead 
continuing to use a single year of data.
2. Cattle Sector
    As discussed in the 2001 Notice, EPA is considering not using the 
farm level data used for the cattle feeding sector used in the proposed 
rule analysis because of USDA concerns that these data are reflective 
of cow-calf operations and are not suitable for evaluating impacts to 
cattle feeding operations (see 66 FR 58585-58587). Instead, for its 
final analysis of impacts to the cattle feeding sector, EPA is 
considering using financial data submitted by NCBA and FAPRI (see: DCN 
175044 and DCN 175038).
    For operations with more than 1,000 AU, EPA would use data provided 
by NCBA for operations with an average of 52,000 head. For operations 
with between 300 and 1,000 AU, EPA would use data submitted by FAPRI 
for a 500-head feedlot enterprise. For the purposes of EPA's analysis, 
and because of lack of additional available data, EPA assumes these 
data reflect baseline financial conditions for operations with fed 
cattle, veal, and heifers. Both the NCBA and FAPRI data represent 
enterprise level conditions. Farm level data are not available; 
therefore, EPA's analysis will assume that farm and enterprise 
conditions are the same. Information on EPA's rationale for selecting 
these data for its analysis is provided in the rulemaking record (DCN 
375084).
    To address recommendations that EPA average out baseline conditions 
to better account for year-to-year variability and pricing cycles (see 
66 FR 58590), EPA uses the three years of survey data (1997-1999) 
provided by NCBA to calculate an average gross revenue value for its 
analysis using the sales test. Using the FAPRI data, which provides a 
2000 base year along with several years of projected data (2000-2011), 
EPA uses the first 3-years of reported revenue (2000 to 2002) to obtain 
an average total revenue value. EPA uses average values to address 
recommendations expressed during the public comment period that EPA 
consider ways to depict financial conditions over multiple years, 
despite the availability of a single year of available data only in 
some cases (see 66 FR 58590).
    Accounting for variability and changing conditions over multiple 
years is already incorporated into EPA's DCF analysis, which spans a 
10-year time frame (1997-2006) and utilizes time series projections. 
This approach is consistent with that used for the 2001 proposal. For 
this analysis, EPA obtains net cash income estimates at both the farm 
and enterprise level for the base year (1997) from the available data. 
EPA uses NCBA data for 1997 for cattle operations with more than 1,000 
AU; EPA derives a base year estimate from available FAPRI data for 
2000, back-calculated to 1997 using the NCBA time series data.
    EPA projects out the 1997 baseline data using NCBA-reported data on 
costs and returns to feedlot enterprises, expressed as dollars per 
marketed head to obtain a cash flow stream over the analysis period 
(1997 to 2006). NCBA's projection covers the 10-year analysis period, 
relying on historical data and pricing trends in the cattle cycle that 
correspond to the three years of data in their survey. EPA uses 
projected returns made by NCBA that were submitted to EPA, along with 
comments and alternate financial data on the proposed rule because both 
FAPRI and USDA baseline projections report net returns to cow-calf 
operations only, which do not

[[Page 48108]]

correspond to cattle feeding operations that are affected by the 
regulations; other cattle sector projections provided by FAPRI do not 
cover the 1997-2006 time period for EPA's analysis. The method that EPA 
uses to project the baseline data follows the approach used for the 
proposal analysis, as discussed in the Economic Analysis supporting the 
2001 proposal. From this projected cash stream, EPA estimates the net 
present value estimates for use in its DCF analysis. Additional 
information is available in the rulemaking record (DCN 375084).
    For the debt-asset test, EPA is considering using FAPRI data on 
total assets and total liabilities for similar size operations in this 
sector, replacing USDA asset and liability data (used for proposal) 
with alternative FAPRI data. Use of these alternative data address 
concerns expressed during the public comment period about EPA's 
assumptions of baseline debt and equity conditions at CAFOs and the 
data on debt and assets assumed for the proposed rulemaking, as 
discussed in the 2001 Notice (66 FR 58582-58583).
    A summary of the baseline financial data that EPA is considering 
using for its final analysis of this sector is available for review at 
DCN 375084.
3. Dairy Sector
    For dairy operations, EPA is continuing to use the 1997 USDA farm 
level data that were used for the proposal analysis. However, USDA 
recently submitted alternative farm level data for dairy operations 
from a 2000 USDA survey of this sector that EPA is considering using. 
These data include farm and enterprise level data and are available for 
review at DCN 375085. For the enterprise level analysis, EPA is 
considering using financial data obtained during the comment period 
from FAPRI (DCN 175038), as presented in the 2001 Notice (66 FR 58588-
58589). Information on EPA's rationale for selecting these data for its 
analysis is provided in the rulemaking record (DCN 375084).
    To address recommendations that EPA average out baseline conditions 
to better account for year-to-year variability and pricing cycles (see 
66 FR 58590), EPA would adjust the available 1997 gross income data 
prior to evaluating these data as part of EPA's sales test using 
published USDA cost and returns data for U.S. dairy operations, 
spanning 1993 to 2000. These national level data are used to create an 
index of 8 years of farm level financial data from which to project out 
1997 gross sales data, producing an average 8-year revenue value.
    Accounting for variability and changing conditions over multiple 
years is already incorporated into EPA's DCF analysis, which spans a 
10-year time frame (1997-2006). This approach is consistent with that 
used for the 2001 proposal. For this analysis, EPA obtains net cash 
income estimates at both at the farm and enterprise level for the base 
year (1997) from the available data. At the farm level, EPA projects 
out the 1997 baseline data using USDA-reported net returns for the 
dairy sector to obtain a cash flow stream over the analysis period 
(1997 to 2006). At the enterprise level, EPA is considering using the 
2000 net cash income for representative dairy operations submitted by 
FAPRI. The 2000 data are back calculated to 1997 and projected from 
2000 to 2006 using the same USDA-reported net returns for the dairy 
sector used for farms. EPA continues to use USDA's projections because 
other available projections do not regularly report net returns per 
milk cow or cover the 1997-2006 time period for EPA's analysis. The 
method that EPA uses to project the baseline data follows the approach 
used for the proposal analysis. From this projected cash stream, EPA 
estimates the net present value estimates for use in its DCF analysis. 
Additional information is available in the rulemaking record (DCN 
375084).
    For the debt-asset test, EPA is considering using FAPRI data on 
total assets and total liabilities for similar size operations in this 
sector, replacing USDA asset and liability data (used for proposal) 
with alternative FAPRI data. Use of these alternative data address 
concerns expressed during the public comment period about EPA's 
assumptions of baseline debt and equity conditions at CAFOs and the 
data on debt and assets assumed for the proposed rulemaking, as 
discussed in the 2001 Notice (66 FR 58582-58583).
    A summary of the baseline financial data that EPA is considering 
using for its final analysis of this sector is available for review at 
DCN 375084.
4. Hog Sector
    As discussed in the 2001 Notice, EPA is substituting the 1997 USDA 
data for hog operations used for proposal with other data obtained by 
EPA since proposal (see 66 FR 58587-58588). For the hog sector, EPA is 
not using the financial data that it used for the proposal analysis 
because of concerns expressed by USDA that 1997 data are not 
representative, because they reflect conditions where hog prices were 
unusually high. For the final analysis, EPA proposes to use alternate 
farm level and enterprise level data from USDA. These cover a broader 
range of hog production types, including both farm and enterprise level 
conditions across three types of operations: Independent owner-operator 
farrow-finish and farrowing operations, contract grow-finish 
operations, and independent grow-finish operations. Information on 
EPA's rationale for selecting these data for its analysis is provided 
in the rulemaking record (DCN 375084).
    As anticipated by EPA in its 2001 Notice, initial data obtained by 
EPA from USDA were not readily analyzable by EPA and since the 
publication of the Notice, EPA has been working with USDA to resolve 
these issues and obtain additional data. Since the publication of the 
2001 Notice, EPA has obtained data from USDA that report farm income 
excluding non-cash items that were included by USDA in the original 
submittal of these data. USDA's new submittal also includes 
corresponding farm level data. These data are available in the 
rulemaking record (DCN 375064).
    To average the available baseline financial data over multiple 
years, EPA adjusts the 1998 data using published USDA cost and returns 
data for both farrow-finish and grow-finish operations. These data 
cover 1995 to 1999. For this analysis, EPA uses national level data to 
create an index to develop 5-years of farm level financial data from 
which to extrapolate the 1998 farm data. The 1998 data are extrapolated 
over the time frame by apportioning costs and revenues on the basis of 
changes in costs, revenues, and returns reported for 1995 through 1999. 
This type of adjustment is discussed in the 2001 Notice (66 FR 58590-
58591) and addresses comments received on the proposal analysis by 
averaging out baseline conditions to better account for year-to-year 
variability and pricing cycles. Using this approach and USDA data, EPA 
obtains the average farm level revenue values that EPA uses for its 
sales test.
    EPA's DCF analysis already incorporates changes over multiple 
years, spanning a 10-year time frame (1997-2006). This approach is 
consistent with that used for the 2001 proposal. However, net cash 
income reported by USDA for hog enterprises in 1998 continues to be 
negative in some cases. When these 1998 values were extrapolated to the 
1995-1999 time period, as is done for the farm level data, cash flow on 
average over this 5-year period continues to be negative for some 
representative facilities. The primary reason for these negative income 
values is that 1998 was a year where hog prices dropped dramatically. 
At the farm level, USDA-reported net

[[Page 48109]]

cash income is positive, although likely low when compared to other 
years.
    Because of persistently negative net cash income due to 1998 market 
conditions in the hog sector, EPA is unable to readily analyze these 
data for its analysis and is considering additional modifications to 
the data obtained by USDA. The principal modification to these data by 
EPA would be the adjustment of these data to reflect expected price 
rather than actual price for 1998 and 1999. The approach that EPA 
proposes to use is based on an approach recommended by USDA Economic 
Research Service (ERS) personnel. This recommended approach uses price 
projections from USDA's World Agricultural Supply and Demand Estimates 
(WASDE) published in 1997 as an indicator of expected 1998 price level 
in the hog sector. Applying this approach provides an expected price of 
about $47 per hundredweight (cwt.) across all hog operations for that 
year, compared to the actual price of under $35 per cwt. reported in 
1998. Adjustment of the original USDA data is necessary to avoid the 
need for EPA to regard these operations as baseline closures and remove 
them from the analysis.
    EPA is considering using the resultant expected price for 1998 to 
adjust the enterprise level data provided to EPA by USDA. (EPA would 
not adjust USDA-reported farm level data since these data may be 
analyzed by EPA without adjustment.) Once the 1998 enterprise level 
data are adjusted, EPA would derive a base year estimate by back-
calculating to 1997 using a 5-year index that EPA created based on the 
same USDA national level cost and returns data for farrow-finish and 
grow-finish operations from 1995 to 1999, as is used to extrapolate 
farm level revenues. EPA is proposing to replace the USDA reported data 
for 1998 and 1999 with EPA adjusted values based on the expected market 
prices during this period. EPA solicits comment on this approach. EPA 
has presented the results of these adjustments of the original data to 
USDA ERS personnel, who are reviewing the approach and resultant 
adjustments to these data. EPA would project out the 1997 baseline data 
using FAPRI timeline data of net returns for the hog sector to obtain a 
cash flow stream over the analysis period (1997 to 2006). From these 
data, EPA would estimate the net present value of expected cash flow 
for use in its DCF analysis. Additional information on EPA's adjustment 
of these data is available in the rulemaking record (DCN 375083).
    For the debt-asset test, EPA is considering using FAPRI data on 
total assets and total liabilities for similar size operations in this 
sector, replacing USDA asset and liability data (used for proposal) 
with alternative FAPRI data. Use of these alternative data address 
concerns expressed during the public comment period about EPA's 
assumptions of baseline debt and equity conditions at CAFOs and the 
data on debt and assets assumed for the proposed rulemaking, as 
discussed in the 2001 Notice (66 FR 58582-58583).
    A summary of the baseline financial data that EPA is considering 
using for its final analysis of this sector is available for review at 
DCN 375084.
5. Poultry Sector
    For EPA's farm level analysis, EPA is continuing to use the 1997 
USDA farm level data for broiler, egg layer, and turkey operations used 
by EPA for its proposal analysis. Since proposal, additional farm level 
data for these sectors have not been made available. EPA also continues 
to use USDA data on total assets and total liabilities for the debt-
asset test, which EPA used for proposal. Despite concerns expressed 
during the public comment period about EPA's assumptions of baseline 
debt and equity conditions at CAFOs, EPA was not able to obtain 
alternate debt-asset information.
    For the enterprise level analysis, EPA is considering using 
enterprise budget data collected by EPA from Oklahoma State University 
(contract broiler operations), North Carolina State University 
(contract turkey hen and turkey tom operation), and Iowa State 
University (independent-owner egg operation). These data are available 
in the rulemaking record (see: DCN 175024, DCN 375036, DCN 375048, and 
DCN 375049). Despite an extensive search of available data, EPA is 
unable to locate financial data that capture each of the possible types 
of poultry operations, including whether an operation is independently 
owned and operated or whether the operation raises animals under 
contract. Additional information on EPA's rationale for selecting these 
data for its analysis is provided in the rulemaking record (DCN 
375084).
    Because limited data are available that characterize conditions at 
farms that raise chickens and turkeys, EPA is not able to locate 
multiple years of financial data in order to average available data 
over a multiple year time frame. Therefore, EPA's analysis of the 
financial effects on broiler, egg, and turkey operations would be based 
on a single year of input data. Using available data, EPA obtains net 
cash income estimates at both the farm and enterprise level. EPA would 
project out the 1997 baseline data using FAPRI timeline data of net 
returns for the broiler, egg, and turkey sectors to obtain a cash flow 
stream over the analysis period (1997 to 2006). From these data, EPA 
would estimate the net present value of expected cash flow for use in 
its DCF analysis. Additional information is available in the rulemaking 
record (DCN 375084).
    A summary of the baseline financial data that EPA is considering 
using for its final analysis of this sector is available for review at 
DCN 375084.

C. Preliminary Analysis Results

    EPA's rulemaking record presents a summary of estimated total 
compliance costs by sector and technology option (pre-tax, 2001 
dollars), which are relatively consistent compared to EPA's estimates 
for the proposed rule across the various technology options. EPA's 
rulemaking record also provides a comparison of the results at proposal 
with preliminary results using the data and methodological changes 
presented in today's notice. As anticipated by EPA in its 2001 Notice, 
the cumulative effect of each of the methodological changes and uses of 
alternative financial data for some sectors results in changes to EPA's 
estimate of the number of operations that may be vulnerable to closure 
post-regulation (66 FR 58580-58583).
    EPA's preliminary results show that the inclusion of an enterprise 
level financial analysis does not significantly alter the results of 
EPA's overall analysis (i.e., the enterprise level results do not 
always differ substantially from the farm level results across all 
sectors). The use of alternative financial data in the beef and hog 
sectors, however, does result in changes in EPA's analysis compared to 
that conducted for the proposed rule, with more beef operations but 
fewer hog operations expected to experience financial stress from 
estimated compliance costs. These preliminary results, however, are not 
driven solely by changes to EPA's financial models but are also driven 
by underlying changes to EPA's engineering cost models. As discussed in 
the 2001 Notice, EPA is considering expanding the range of cost 
estimates per representative farm to account for variability across 
operations based on expected capital and management improvements needed 
(see 66 FR 58572-58573).
    Overall, EPA's preliminary analysis results show that combined 
changes to EPA's cost and financial models and input data to address 
public comments

[[Page 48110]]

do not result in significant changes to EPA's regulatory analysis 
compared to that conducted for proposal. More detailed information on 
the results of this analysis is provided in the rulemaking record (DCN 
375084). These results are preliminary and subject to change, depending 
on ongoing refinements and corrections made to both EPA's cost and 
financial models and input data. In addition, these results do not yet 
consider potential longer-term market adjustment and structural 
adjustment by regulated facilities. These results also do not take into 
account potential cost offsets due to available cost share assistance, 
given increases in government expenditures and changes to program 
eligibility requirements under the new farm bill legislation.

    Dated: July 16, 2002.
G. Tracy Mehan, III,
Assistant Administrator for Water.
[FR Doc. 02-18579 Filed 7-22-02; 8:45 am]
BILLING CODE 6560-50-P