Highway Funding: Alternatives for Distributing Federal Funds (Chapter
Report, 11/28/95, GAO/RCED-96-6).

Pursuant to a legislative requirement, GAO reviewed the formula for
distributing federal highway funds, focusing on the: (1) relevancy of
the data used for the formula; (2) major funding objectives implicit in
the formula; and (3) implications of alternative formula factors for
achieving these objectives.

GAO found that: (1) the federal highway funding formula is complex and
cumbersome; (2) the underlying data and factors used in the formula are
to a large extent irrelevant, since funding outcomes are essentially
predetermined; (3) annual combined funding for the four largest highway
programs is fixed throughout the 6-year life of the Intermodal Surface
Transportation Efficiency Act (ISTEA); (4) some of the factors used in
formula calculations are based on outdated information, are unresponsive
to changing conditions, and often do not reflect the highway system's
utilization; (5) equity adjustments increase many states' final funding
levels; (6) funding for demonstration projects is not determined by
formula; (7) ISTEA objectives include maintaining and improving the
highway infrastructure, returning Highway Trust funds to the states
where the revenue was generated, advancing selected goals, and
safeguarding the states' historical funding shares; and (8) a
combination of objectives based on states' needs and resources could
form the basis for a new formula, but any new formula is likely to
change the states' highway funding levels.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  RCED-96-6
     TITLE:  Highway Funding: Alternatives for Distributing Federal Funds
      DATE:  11/28/95
   SUBJECT:  Trust funds
             Road construction
             Interstate highway system
             Federal aid for highways
             Highway planning
             Maintenance (upkeep)
             Formula grants
             Intergovernmental fiscal relations
IDENTIFIER:  National Highway System
             DOT Surface Transportation Program
             Highway Trust Fund
             FHwA Interstate Maintenance Program
             FHwA Highway Bridge Replacement and Rehabilitation Program
             FHwA Congestion Mitigation and Air Quality Improvement 
             Program
             Highway Trust Fund Highway Account
             Highway Trust Fund Mass Transit Account
             Interstate Highway System
             FHwA Interstate Construction Program
             DOT Interstate Substitution Program
             FHwA Scenic Byways Program
             FHwA Safety Belt and Motorcycle Helmet Program
             FHwA Federal-Aid Highway Program
             FHwA Classification System
             
**************************************************************************
* This file contains an ASCII representation of the text of a GAO        *
* report.  Delineations within the text indicating chapter titles,       *
* headings, and bullets are preserved.  Major divisions and subdivisions *
* of the text, such as Chapters, Sections, and Appendixes, are           *
* identified by double and single lines.  The numbers on the right end   *
* of these lines indicate the position of each of the subsections in the *
* document outline.  These numbers do NOT correspond with the page       *
* numbers of the printed product.                                        *
*                                                                        *
* No attempt has been made to display graphic images, although figure    *
* captions are reproduced. Tables are included, but may not resemble     *
* those in the printed version.                                          *
*                                                                        *
* A printed copy of this report may be obtained from the GAO Document    *
* Distribution Facility by calling (202) 512-6000, by faxing your        *
* request to (301) 258-4066, or by writing to P.O. Box 6015,             *
* Gaithersburg, MD 20884-6015. We are unable to accept electronic orders *
* for printed documents at this time.                                    *
**************************************************************************


Cover
================================================================ COVER


Report to Congressional Committees

November 1995

HIGHWAY FUNDING - ALTERNATIVES FOR
DISTRIBUTING FEDERAL FUNDS

GAO/RCED-96-6

Alternatives for Distributing Federal Highway Funds

(342883)


Abbreviations
=============================================================== ABBREV

  AASHTO - American Association of State Highway and Transportation
  Officials
  CMAQ - Congestion Mitigation and Air Quality program
  DOT - Department of Transportation
  FHWA - Federal Highway Administration
  ISTEA - Intermodal Surface Transportation Efficiency
  Act of 1991
  LOE - level of effort
  NHS - National Highway System
  STP - Surface Transportation Program

Letter
=============================================================== LETTER


B-261718

November 28, 1995

The Honorable John H.  Chafee
Chairman
The Honorable Max S.  Baucus
Ranking Minority Member
Committee on Environment
 and Public Works
United States Senate

The Honorable Bud Shuster
Chairman
The Honorable James L.  Oberstar
Ranking Democratic Member
Committee on Transportation
 and Infrastructure
House of Representatives

In response to section 1098 of the Intermodal Surface Transportation
Efficiency Act of 1991 and following discussions with your offices,
this report discusses the way the formula for distributing federal
highway funds works and the relevancy of the data used for the
formula.  It also discusses the major funding objectives implicit in
the formula and the implications of alternative formula factors for
achieving these objectives. 

We are sending copies of this report to the Secretary of
Transportation; the Administrator, Federal Highway Administration;
the heads of the state departments of transportation; and other
interested parties.  We will also make copies available to others on
request. 

If you or your staff have any questions about this report, please
call me at (202) 512-2834.  Major contributors to this report are
listed in appendix XII.

John H.  Anderson, Jr.
Director, Transportation and
 Telecommunications Issues


EXECUTIVE SUMMARY
============================================================ Chapter 0


   PURPOSE
---------------------------------------------------------- Chapter 0:1

Under the federal-aid highway program, billions of dollars are
distributed to the states annually for the construction and repair of
highways and related activities.  The Intermodal Surface
Transportation Efficiency Act of 1991 (ISTEA) authorized
approximately $120 billion for this program for fiscal years 1992
through 1997.\1 ISTEA charged GAO with reviewing the formula by which
these highway funds are distributed to the states.  As agreed with
the Senate Committee on Environment and Public Works and the House
Committee on Transportation and Infrastructure, this report discusses
(1) the way the formula works and the relevancy of the data used for
the formula and (2) the major funding objectives implicit in the
formula and the implications of alternative formula factors for
achieving these objectives. 


--------------------
\1 The full ISTEA authorization for all surface transportation
programs, including mass transit, totals $155 billion for fiscal
years 1992-97.  In addition, ISTEA offers states and localities
unprecedented opportunities to use federal highway and mass transit
capital funds across different modes of transportation. 


   BACKGROUND
---------------------------------------------------------- Chapter 0:2

The federal-aid highway formula is a series of mathematical
calculations that determines how federal highway funds are
distributed among the states each year.  The formula is established
by law and has been periodically revised on the basis of additions
and modifications to the program.  As such, the formula has evolved
over many decades as new programs and apportionment factors have been
layered on top of existing rules.  The result is a multistep process
that encompasses several objectives, including preserving the highway
infrastructure and attaining certain social goals, such as improved
air quality. 

The current formula, established by ISTEA, determines the
distribution of funds for 13 funding categories.  These categories
include eight individual programs, the two largest of which are the
National Highway System and the Surface Transportation Program, and
five separate mechanisms for increasing individual states' funding in
order to achieve certain goals for equity among the states.  The
calculations that determine the level of funding that each state
receives for the various categories occur in a strict sequence.  Each
calculation can incorporate one or more factors.  For example, during
one step in the calculation, states gain funding to preserve their
Interstate highways in accordance with their number of lane miles and
vehicle miles traveled; in later steps, additional funding is
provided to certain states under categories referred to as equity
adjustments.  The Congress created such adjustments primarily to (1)
address the concerns of the states that contribute a greater share of
highway user taxes than they receive in federal-aid highway funds and
(2) provide each state with the same relative share of overall
funding that it received in the past, recognizing the legislative
funding compromises embedded in ISTEA.  Taken together, all 13 steps
of the formula process, corresponding to the 13 funding categories,
result in an apportionment for each state. 

Federal highway funding is supported through federal highway user
taxes on, among other things, motor fuels, tires, and trucks. 
Revenues from these taxes are credited to the Highway Trust Fund's
highway account.  The Department of Transportation has estimated that
in fiscal year 1996, these taxes will generate about $20.5 billion. 


   RESULTS IN BRIEF
---------------------------------------------------------- Chapter 0:3

The federal highway funding formula is a complex, iterative process
that is based on an array of data and factors.  To a significant
extent, however, the underlying data and factors are not meaningful
because the funding outcome is largely predetermined.  This outcome
occurs because the annual combined funding for the four largest
highway programs (accounting for 70 percent of all the funding
apportioned in fiscal year 1995) is fixed throughout the 6-year life
of ISTEA, even though the funding for each individual program
ostensibly derives from a separate calculation.  Furthermore, some of
the factors used in the formula's calculations for major programs are
based, in part, on outdated information, are unresponsive to changing
conditions, and often do not reflect the current extent or use of the
nation's highway system.  For example, the mileage of postal roads
has been included as either a direct or underlying factor in the
calculation since 1917, although this factor is not relevant to
today's federal-aid highway network.  Finally, equity adjustments
ultimately increase many states' final level of funding. 

GAO's review of the existing formula and its legislative
underpinnings, as well as discussions with federal and state
transportation officials, indicated that four overarching objectives
are entwined in the current process for distributing highway funds:\2
(1) maintaining and improving the highway infrastructure; (2)
returning the majority of the funds contributed to the Highway Trust
Fund to the state where the revenue was generated; (3) advancing
selected goals, such as improving air quality and conserving energy;
and (4) safeguarding the states' historical funding shares.  Since
needs vary among the states, the extent to which these objectives are
met also varies.  Furthermore, while these four overarching
objectives can to some extent be mutually supporting, they also
conflict in some cases.  For example, returning funds to the states
where the revenue was generated may not be in harmony with an
approach that seeks to safeguard the states' historical funding
shares. 

One or more of the overarching objectives could be the foundation for
a new formula.  For instance, the Congress could choose to emphasize
just one objective, such as preserving the highway infrastructure--an
objective aligned with those formula factors that reflect the use and
extent of each state's highway network.  Alternatively, two or more
objectives and their associated factors could be blended so as to
balance multiple goals.  Regardless of which objective or combination
of objectives is chosen, some states may receive more funds than they
do under the existing formula, others less.  The Congress could
temper these effects by also incorporating the objective of
safeguarding historical funding shares into the formula.  This result
could be accomplished through a component designed to place a cap on
the maximum percentage of loss that any individual state would be
expected to bear as a result of the changes. 


--------------------
\2 While the majority of funds are distributed to the states by
formula, a few exceptions arise because of minor deductions, such as
those for federal administrative expenses, and because of
congressionally designated projects. 


   PRINCIPAL FINDINGS
---------------------------------------------------------- Chapter 0:4


      FORMULA PROCESS IS
      CUMBERSOME, YIELDS A LARGELY
      PREDETERMINED OUTCOME, AND
      PARTIALLY RELIES ON OUTDATED
      AND IRRELEVANT FACTORS
-------------------------------------------------------- Chapter 0:4.1

The formula for apportioning federal highway funds among the states
derives from a complicated set of calculations involving
consideration of 13 specific funding categories.  In some cases,
these complex calculations can prove to be an essentially meaningless
exercise.  One prime example is the treatment of the four major
highway programs (the Interstate Maintenance, Bridge Replacement and
Rehabilitation, National Highway System, and Surface Transportation
Program), which together accounted for 70 percent of all the funds
apportioned in fiscal year 1995.  Separate calculations are used to
determine each state's share of funding for each of these four
programs.  However, the outcome of each separate calculation is
obscured because an adjustment is made for the Surface Transportation
Program in each state's apportionment.  The result of this adjustment
is that each state's total share of funding for these four programs
must equal the adjusted share of funding that the state received for
the programs' predecessors in fiscal years 1987 through 1991. 

A further concern with the existing formula is that irrelevant or
outdated factors underlie the funding calculations for certain
programs.  GAO reported in 1986 that two of the factors that underlie
certain key decisions about apportionment--postal road mileage and
land area--were irrelevant to either the extent or use of the modern
highway system.\3

ISTEA restructured the major highway programs, but the states'
funding for the two largest programs--the National Highway System and
Surface Transportation Program, together accounting for 40 percent of
all the apportioned funding--remains linked to these irrelevant
factors. 

Near the end of the apportionment process, most states' total funding
is increased through various funding categories known as equity
adjustments.  In fiscal year 1995, 41 states and the District of
Columbia received a total of $2.8 billion in funding for equity
adjustments.  This funding represented 16 percent of the
approximately $18 billion apportioned to the states that year. 


--------------------
\3 Highway Funding:  Federal Distribution Formulas Should Be Changed
(GAO/RCED-86-114, Mar.  31, 1986). 


      ALTERNATIVES FOR
      DISTRIBUTING FEDERAL HIGHWAY
      FUNDS
-------------------------------------------------------- Chapter 0:4.2

Reauthorization of the federal-aid highway program provides an
opportunity to reassess the overarching objectives for the federal
highway program's funds and the formula governing the distribution of
the funds.  Depending on which objectives and formula factors are
selected and how they are weighted, significant amounts of funding
could shift among the states.  GAO analyzed the funding distributions
that would result from certain alternative configurations of the
formula.  This analysis consisted of a hypothetical redistribution of
the actual apportionments in fiscal year 1995 according to a series
of formula options.\4 Such examples represent but a small sample of
the myriad alternative formulas available to the Congress, but these
hypothetical redistributions illustrate both the pervasiveness of
funding shifts under a variety of formula options and the magnitude
of gains and losses that each state would experience, depending on
the selection of formula factors and weighting schemes.  Under these
redistribution alternatives,\5 in some cases, a state would lose 50
percent or more of its funds.  While the losses would not always be
so sizable and a number of states would gain funds under the
redistributions, this result would be of little comfort to the states
whose relative position would worsen. 


--------------------
\4 Distributions in fiscal year 1995--the most recent year for which
data were available at the time of our analysis--were used.  However,
different funding patterns may emerge on the basis of (1) the total
distributions over the life of ISTEA or (2) the choice of a different
year. 

\5 The formula alternatives are keyed to the existing funding
objectives, but new components--such as disbursements for highways as
a percentage of the states' and localities' total
disbursements--could be added to recognize differences in the states'
fiscal capacity and costs. 


         MAINTAINING AND IMPROVING
         THE HIGHWAY
         INFRASTRUCTURE
------------------------------------------------------ Chapter 0:4.2.1

Highway funds could be returned to the states on the basis of
indicators of highway needs using actual needs or proxies for such
needs.  Direct measures of a system's needs, such as the miles of
poor pavement or number of deficient bridges in a state, reflect the
physical condition of the highway and bridge network and its
performance.  However, basing a formula on actual needs could foster
a perverse incentive, since the more needs a state has, the more
money it would receive.  This situation could be remedied by using
proxies for needs, such as ones reflecting how extensive a state's
highway system is and how heavily it is used.  Transportation
officials, however, do not agree on the appropriate proxies for
distributing highway funds.  A previous GAO report and a study
sponsored by the Federal Highway Administration both indicated that
proxies, such as lane miles and vehicle miles traveled, are closely
aligned with highway needs.  Some transportation officials argue,
however, that such proxies promote the use of highways and are at
odds with energy conservation and clean-air goals.  Alternative
proxies less directly tied to highway use include population and
population density.  Population data, however, also have limitations. 
For example, while funds would be focused on congested urban areas,
the use of population data to apportion highway funds would do little
to accommodate the needs of rural areas. 


         RETURNING FUNDS TO THE
         SOURCE
------------------------------------------------------ Chapter 0:4.2.2

States' contributions to the Highway Trust Fund are not currently
returned to the states in proportion to the amount collected,
although this approach to distributing funds would be a relatively
simple and direct method of fund distribution.  Some state
transportation officials support this approach because it would
guarantee that all or a substantial amount of the revenues collected
in their state would be returned to them.  However, the
return-to-origin approach would not be universally attractive, as a
number of states would lose funds.  In 1993, distributions of federal
highway funds as a percentage of states' contributions to the Highway
Trust Fund's highway account ranged from 83 percent for South
Carolina to 707 percent for Hawaii.  Some transportation officials
observe that this redistribution of funds is to be expected, since
federal highway taxes are collected to address national objectives,
such as preserving the National Highway System, not merely to return
the funds to their source.  Furthermore, these officials question the
need for a federal program if the states' Highway Trust Fund
contributions are simply returned to them. 


         USING SET-ASIDES TO
         ADVANCE SELECTED GOALS
------------------------------------------------------ Chapter 0:4.2.3

A portion of highway funds could be set aside to advance specific
goals before the remaining funds were distributed to the states.  For
example, incentive payments drawn from this set-aside could be used
to provide bonuses to advance quality-of-life objectives, reward
improvements in the condition of the highway infrastructure above a
certain defined floor, and advance highway safety.  The payments
could be simply added on to the states' apportionments or could be
channeled to the states through a separate program category, such as
the current authorization of approximately $1 billion annually for
the Congestion Mitigation and Air Quality Improvement program. 


         CREATING A SAFEGUARD TO
         MITIGATE AGAINST THE
         SUDDEN LOSS OF HISTORICAL
         FUNDING SHARES
------------------------------------------------------ Chapter 0:4.2.4

Altering the existing formula could eliminate its current emphasis on
historical funding shares and thus cause shifts in the amount of
funds distributed to the states.  In some cases, the funding shifts
could be dramatic, warranting consideration of ways to reduce the
magnitude of the losses.  For instance, any new formula might include
a component designed to place a cap on the maximum percentage of loss
that any individual state would be expected to bear as a result of
changes in the formula.  This cap could be either permanent or
established for a set period during the transition to a new funding
amount. 

The funds needed to make such an adjustment could derive from a
variety of sources.  As an example, the funds devoted to existing
equity adjustments in fiscal year 1995--$2.8 billion--would more than
offset the states' cumulative losses under all of the sample formula
scenarios that GAO analyzed.  As another possibility, the funding
authorized by the Congress for specific demonstration projects, which
is not distributed by formula, could in the future be used to offset
the states' losses resulting from a formula change instead of being
used for additional authorizations for specific projects.  In fiscal
year 1995, funds for demonstration projects distributed to the states
under ISTEA totaled approximately $1 billion. 


   RECOMMENDATIONS
---------------------------------------------------------- Chapter 0:5

Because the selection of a highway apportionment formula is a
judgment for the Congress, GAO is making no specific recommendations. 


   AGENCY COMMENTS
---------------------------------------------------------- Chapter 0:6

GAO provided copies of a draft of this report to the Department of
Transportation for its review and comment.  GAO met with Department
officials--including the chiefs of the Program Analysis Division, the
Policy Evaluation Branch, and the Highway Funding and Motor Fuels
Division of the Federal Highway Administration--who provided
comments.  The Department agreed with the information presented and
observations made throughout the report and considered it a
well-prepared, balanced report.  Technical comments provided by the
Department have been incorporated where appropriate. 


INTRODUCTION
============================================================ Chapter 1

The existing federal-aid highway formula is the vehicle for
distributing billions of dollars annually for highway construction
and repair and related activities to the 50 states, the District of
Columbia, and Puerto Rico (hereafter called the states, unless
otherwise noted).  Since the mid-1980s, a number of organizations
(including GAO) have suggested fundamental changes in the formula for
apportioning these federal-aid funds because of perceived problems
with the formula, such as its reliance, at least in part, on outdated
data.  Section 1098 of the Intermodal Surface Transportation
Efficiency Act of 1991 (ISTEA) tasked GAO with reviewing the process
for distributing highway funds to the states.  Chapter 2 of this
report evaluates the current apportionment formula.  Chapter 3
discusses a process by which the Congress may reconsider the formula
during the next reauthorization of the federal-aid highway program
and comments on the advantages and disadvantages of several
alternative formula options. 


   ISTEA AUTHORIZED UNPRECEDENTED
   FUNDING
---------------------------------------------------------- Chapter 1:1

ISTEA authorized funding to sustain and enhance the nation's surface
transportation infrastructure.  The act provided an unprecedented
authorization of $122 billion for highways, bridges, and related
activities for fiscal years 1992-97.\1 Figure 1.1 shows the annual
authorization for federal highway funding since 1987 and demonstrates
the dramatic increases effected under ISTEA. 

   Figure 1.1:  Highway
   Authorization Levels, Fiscal
   Years 1987-97

   (See figure in printed
   edition.)

Note:  Authorization levels refer to authorizations originally
provided under title I of the Surface Transportation and Uniform
Relocation Authorization Act of 1987 and the Intermodal Surface
Transportation Efficiency Act of 1991.  Section 1003(c) of ISTEA,
however, placed a $98.6 billion cap on highway authorizations from
fiscal years 1992 through 1996, to comply with the 1990 budget
resolution.  This cap, according to officials of the Federal Highway
Administration, will result in a 13-percent reduction in the original
authorization level for fiscal year 1996. 

Source:  Federal Highway Administration. 

Except for a few minor deductions, such as those for federal
administrative expenses, federal highway funds are provided to the
states through the Federal Highway Administration (FHWA), which is
part of the U.S.  Department of Transportation (DOT).  The money is
distributed to the states through various formula calculations and,
to a lesser extent, through congressionally designated projects. 

ISTEA's authorization is funded primarily through federal highway
user taxes such as those on motor fuels (gasoline, gasohol, and
diesel), tires, and trucks.  Funds from these sources are collected
from users and credited to the Highway Trust Fund for highway and
mass transit projects or related activities.  The fund is divided
into a highway account and a mass transit account.\2 DOT forecasts
that the income to the highway account will total $20.5 billion in
fiscal year 1996. 


--------------------
\1 The full ISTEA authorization for all surface transportation
programs, including mass transit, totals $155 billion for fiscal
years 1992-97.  In addition,ISTEA offers states and localities
unprecedented opportunities to use federal highway and mass transit
capital funds across different modes of transportation. 

\2 Throughout this report, references to the Highway Trust Fund refer
only to the highway account, unless otherwise noted. 


   ISTEA REVAMPED THE FEDERAL-AID
   HIGHWAY SYSTEMS
---------------------------------------------------------- Chapter 1:2

Before ISTEA, the federal-aid systems--designated routes on which
federal funds may be used--were at the core of the federal-aid
highway program.  Designation of a road as part of a federal-aid
system does not mean the road is owned, operated, or maintained by
the federal government.  The designation is simply the first step in
establishing the eligibility of selected state and local roads for
federal assistance.  Previously, federal aid was apportioned to
Interstate, primary, secondary, and urban highways. 

ISTEA, however, discarded this approach by creating only two systems: 
the National Highway System (NHS) and the Interstate System, which is
a component of the NHS.  The NHS is the centerpiece of ISTEA, and the
system is expected to be the major focus for the federal-aid highway
program into the 21st century.  In a speech on December 9, 1993, the
Administrator of FHWA noted that since the Interstate was begun in
1956, the nation's population has grown and shifted, the economy has
changed, and needs are different.  To serve these needs--to extend
the benefits of the Interstate system to areas not served directly by
it--the NHS was conceived as a way of focusing federal resources on
the nation's most important highways.  DOT, working cooperatively
with state and local officials as well as the private sector,
proposed to the Congress in December 1993 an NHS network of about
159,000 miles.  This network is about 17 percent of the approximately
950,000-mile federal-aid network and includes only 4 percent of the
approximately 4 million miles of public roads.  However, this system
would handle about 40 percent of all vehicle miles traveled\3

and accommodate over 70 percent of all commercial truck traffic. 

For other roads eligible for federal assistance, a program with the
characteristics of a block grant, the Surface Transportation Program
(STP), provides financial assistance.  In addition, ISTEA continued
authorizations for a separate Bridge Replacement and Rehabilitation
Program and Interstate Maintenance Program and an array of other
separate highway program initiatives as well as funding categories
addressing various equity issues, such as each state's share of
funding as compared with what it received in past years.\4


--------------------
\3 Vehicle miles traveled measures traffic by the number of miles
traveled by automobiles or other classes of vehicles during a
specific period of time. 

\4 The use of the label "equity" is based on its traditional usage in
the debate on the highway formula.  It does not reflect any judgment
on our part that these provisions increase the equity or fairness of
the formula's allocations.  Since this report does not employ any
specific criteria, we make no attempt here to gauge the performance
of the current formula and the alternatives on equity. 


   ISTEA EXPANDED THE GOALS OF
   SURFACE TRANSPORTATION
---------------------------------------------------------- Chapter 1:3

ISTEA broadened the overall goals of surface transportation. 
Previously, the federal-aid highway program had focused on completing
and preserving the Interstate Highway system and on maintaining other
federal-aid highways as well as bridges eligible for federal funds. 
While these goals remain a part of the overall surface transportation
program, ISTEA broadened the goals and included new programs,
planning processes, and management systems that are intended to help
ensure that the states' transportation plans are intermodal (that is,
coordinate various modes of transportation), environmentally sound,
and energy efficient.  For example, the Congestion Mitigation and Air
Quality Improvement Program (CMAQ) directs funds to transportation
projects in clean air nonattainment areas--areas that have not
achieved federal standards for air quality.  ISTEA also provides for
increased emphasis on mobility for the elderly, disabled, and
economically disadvantaged. 

ISTEA expanded the use of equity adjustments for the apportionment of
federal-aid funds among the states.  For example, it modified minimum
allocation funding.\5 It also created "hold harmless" funding, which
establishes the state's share of overall federal highway
apportionments.  These adjustments, which are more fully explained in
chapter 2, are generally used to increase the states' return on their
contributions to the Highway Trust Fund. 

ISTEA also embodied quality-of-life objectives, stating that the
nation's transportation system should be economically efficient and
environmentally sound, provide the foundation for the nation to
compete in the global economy, and move people and goods in an
energy-efficient manner.  Additionally, ISTEA's emphasis is
intermodal--providing links in a seamless intermodal network that
will enhance economic growth, international competitiveness, and
national security.  The NHS is expected to reflect this emphasis. 


--------------------
\5 Minimum allocation funding guarantees each state an amount so that
its apportionments and allocations for selected programs in the prior
year equal 90-percent of the percentage of the state's estimated
contributions to the Highway Account of the Highway Trust Fund. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
---------------------------------------------------------- Chapter 1:4

Section 1098 of ISTEA tasked us with reviewing the process for
distributing highway funds to the states.  In discussion with the
congressional committees identified in section 1098, we agreed to
address (1) the way the formula works and the relevancy of the data
used for the formula and (2) the major funding objectives implicit in
the formula and the implications of alternative formula factors for
achieving these objectives. 

To understand the evolution of the current formula and assist in
clarifying the process by which alternative formula options might be
crafted, we reviewed the history of the federal-aid highway programs. 
Key documents included Development and Evaluation of Alternative
Factors and Formulas, published by Jack Faucett Associates in
December 1986; Review and Analysis of Federal-Aid Apportionment
Factors, a 1969 paper prepared in FHWA's Policy Planning Division;
Alternative Financial Formulas for Allocating Federal Highway Funds,
a 1990 report by the American Association of State Highway and
Transportation Officials (AASHTO); Moving Ahead--1991 Surface
Transportation Legislation, a 1991 report by the congressional Office
of Technology Assessment; and a report we published in March 1986.\6

To understand how the current formula works and the ramifications of
possible changes, we reviewed an FHWA publication, Financing
Federal-Aid Highways, published in 1992, and discussed the formula
with FHWA's Office of Fiscal Services, which is responsible for
making formula apportionments to the states.  We interviewed
officials from FHWA's Legislation and Strategic Planning Division,
Office of Highway Information Management, and Bridge Division.  We
also solicited the states' views on the current formula and on future
apportionment issues in meetings with state transportation officials
from 34 states and the District of Columbia at regional or national
transportation meetings in Atlanta, Chicago, and Detroit.  We also
held meetings with state transportation officials in our Washington,
D.C., offices.  In addition, we met with representatives from various
transportation organizations, including the American Association of
State Highway and Transportation Officials and the Surface
Transportation Policy Project. 

While we focused our review on existing, overarching highway
objectives, new components could be added to recognize the states'
capacity to fund highway needs from state resources, the states'
level of effort in meeting their own needs, and geographic
differences in the cost of maintaining existing highway networks. 
Although similar factors have been applied in other programs, they
have not been applied to highway programs in the past.  But FHWA did
provide a report to the Congress in 1994 that addresses measures for
assessing how much of their available resources the states or local
areas devote to surface transportation.\7

Finally, working with FHWA's Office of Policy Development, we
analyzed the effect of a series of hypothetical changes to the
current formula.  This analysis was based on comparing the actual
fiscal year 1995 funding that the states received with what they
would have received under the various alternatives.  In this
analysis, the states' contributions to the Highway Trust Fund were
based on estimates for fiscal year 1993--the most recent year for
which data were available at the time of our analysis. 

We performed this review in accordance with generally accepted
government auditing standards.  We conducted our review from January
1994 through October 1995. 


--------------------
\6 Highway Funding:  Federal Distribution Formulas Should Be Changed
(GAO/RCED-86-114, Mar.  31, 1986). 

\7 Report on State Level of Effort, required by section 6013 of the
Intermodal Surface Transportation Efficiency Act of 1991
(P.L.102-240), Mar.  1994. 


THE CURRENT APPORTIONMENT FORMULA
IS A COMPLEX PROCESS WITH A
LARGELY PREDETERMINED OUTCOME
============================================================ Chapter 2

Although federal-aid highway funds are apportioned among the states
in 13 funding categories, four programs--Interstate Maintenance,
Bridge Replacement and Rehabilitation, the NHS, and the
STP--accounted for 70 percent of the funds apportioned in fiscal year
1995.  While each state's share of funds is calculated annually for
each of these separate programs, these separate calculations are
essentially meaningless since the total funding for the four programs
is fixed over the 6-year authorization period for ISTEA. 
Consequently, the total funding for the four programs does not
respond to changing conditions in a state, such as increased highway
use.  Furthermore, the factors underlying the distribution of highway
funds to the states, such as land area and postal mileage, are
generally outdated and often do not reflect the extent or use of the
nation's highway system.  Our March 1986 report\1 and a study
commissioned by FHWA from a contractor\2 noted that alternative
factors, such as lane miles, are more closely aligned with highway
needs. 

The Congress has used funding adjustments to improve equity among the
states.  These equity adjustments, which occur towards the end of the
13-step apportionment process, increase the total amount of funds for
eligible states.  In fiscal year 1995, the equity funding categories
increased the amount of federal highway funds apportioned to 41
states and the District of Columbia.  The amount of funding that the
majority of states received through the highway formula process was
therefore ultimately increased by these equity adjustments. 

The Congress can further adjust the federal highway funds a state
receives by authorizing specific projects, commonly referred to as
demonstration projects.  Funding for these demonstration projects is
not distributed by formula.  Rather, the Congress requires that
particular projects receive a specified amount of funding.  In ISTEA,
for instance, the Congress provided $6.2 billion in funds for over
500 demonstration projects over the 6-year authorization period. 


--------------------
\1 GAO/RCED-86-114. 

\2 Development and Evaluation of Alternative Factors and Formulas,
Jack Faucett Associates, Dec.  1986. 


   THE DISTRIBUTION OF FEDERAL-AID
   FUNDS AMONG THE STATES IS A
   COMPLEX PROCESS
---------------------------------------------------------- Chapter 2:1

The formula for apportioning federal-aid highway funds established in
ISTEA is a complex arithmetic tool used by FHWA to determine each
state's share of the funds.  On the basis of the formula, funding is
provided for eight programs, including the NHS and STP, and for five
separate mechanisms to raise individual states' funding levels to
achieve certain goals for equity among the states.  The calculations
that determine the level of funding that each state receives for
these various categories occur in a strict sequence, as illustrated
in figure 2.1.  During the first step of the calculation, for
example, funding is provided to complete the construction of the
Interstate Highway System.  Funding for the other program categories
is also based on separate calculations.  However, as depicted in
figure 2.1 and discussed later in this report, the funding for four
programs--Interstate Maintenance, Bridge Replacement and
Rehabilitation, the NHS, and the STP--is interdependent since a
state's total share of funding for all four programs is fixed. 

Later steps in the formula's calculation provide additional funding
to certain states; these funding categories are legislatively
designated as equity adjustments.  Equity adjustments generally
address the concerns of states that contribute a greater share of
highway user taxes than they receive in federal-aid highway funds. 
Equity adjustments also provide each state with the same relative
share of overall funding that it received in the past.\3 In fiscal
year 1995, these equity adjustments represented 16 percent of the
total funds apportioned.  Figure 2.1 outlines the sequence of the
equity adjustments and program funding categories (app.  I provides
additional details).  However, DOT has proposed changes to the
existing equity adjustments and program categories.  The changes,
proposed in DOT's fiscal year 1996 budget justification, were
preceded by the statement that if less federal money will be invested
in transportation, state and local governments need to have greater
authority and flexibility to decide which projects are most
important.  DOT has stated that it will provide an authorization
proposal for such changes at an appropriate time. 

   Figure 2.1:  Sequence of
   Calculations to Determine
   States' Highway Apportionments

   (See figure in printed
   edition.)

While apportionments for highway programs are based on individual
calculations, for some programs the dollar amount apportioned by
formula has little practical meaning because the states have
substantial flexibility to transfer funds from one program category
to another.  For example, with the Secretary of Transportation's
approval, up to 100 percent of a state's apportionment for the
Interstate Maintenance and NHS programs can be transferred to the
state's surface transportation program.  In addition, ISTEA's
flexible funding provisions have allowed decisionmakers at the state
and regional level to decide for themselves whether to allocate
transportation funds to highway or transit projects.  ISTEA provided
for a potential $70 billion in such flexible funding for transit or
highway projects over 6 years.  According to DOT's preliminary data
through the end of fiscal year 1995, $2,160.6 million in highway
funds had been transferred to transit projects and $2.2 million in
transit funds had gone to highway projects. 


--------------------
\3 One funding category designated as an equity
adjustment--Reimbursement for Interstate Segments--reimburses the
states for the cost of constructing segments of the Interstate System
completed in the early days of the Interstate Construction Program. 
Funding in this category is scheduled to begin in fiscal year 1996. 


   THE TOTAL FUNDING SHARE FOR
   MAJOR PROGRAMS IS FIXED
---------------------------------------------------------- Chapter 2:2

In fiscal year 1995, 70 percent of the funding under the formula went
to the four largest programs--the Interstate Maintenance Program,
Bridge Program, NHS, and STP.  Separate calculations determine each
state's share of funds for the Interstate Maintenance Program, Bridge
Program, and NHS.  Nonetheless, these program-specific calculations
are essentially meaningless because each state's ultimate share of
funding for all four programs is fixed.  With a few minor
adjustments, these fixed shares derive from the shares of funds that
the states received, on average, during fiscal years 1987-91 for the
predecessor programs that ISTEA consolidated into these four new
programs.  As a result, the states' funding shares for the four major
programs are divorced from current conditions, as the states' current
and future shares of total funding for these programs must equal the
adjusted historical shares. 

In practice, the states' funding shares for these programs remain
fixed over time because the final program included in the four-part
calculation--the STP--behaves as an adjuster.  The states' funding
levels for the other three programs--the Interstate Maintenance
Program, Bridge Program, and NHS--are independently calculated on the
basis of factors specific to each program.  After those calculations
are completed, however, each state's STP funding is determined by
simply taking the difference between (1) the state's predetermined
share of the total funding available for the four programs and (2)
the amount the state is actually scheduled to receive for the three
independently calculated programs.  This means that any annual
increase or decrease in a state's funding for the Interstate
Maintenance Program, Bridge Program, or NHS must be offset by a
corresponding, reciprocal change in the STP funds the state receives
for that same year. 

Figure 2.2 illustrates this zero-sum game through a hypothetical
example involving 2 years and two states.  In the example, both State
A and State B experience shifts in their apportioned funding between
fiscal years 1993 and 1994.  State A loses funds for the Interstate
Maintenance and Bridge programs, while State B gains funds in both of
these categories.  However, for both states these shifts are rendered
irrelevant because they are offset by a corresponding change in the
states' STP funding levels.  As a result, State A has 1.9 percent of
the total funding available for the four programs in both fiscal
years, despite its losses in funding for the Interstate Maintenance
and Bridge programs.  State B is locked into a 1.75-percent share in
both years, despite its gains in funding for the Interstate
Maintenance and Bridge programs. 

   Figure 2.2:  Hypothetical
   Example Illustrating That the
   States' Total Shares of Funding
   for the Four Largest Programs
   Is Fixed

   (See figure in printed
   edition.)


   FUNDING IS DETERMINED IN PART
   BY IRRELEVANT OR OUTDATED
   FACTORS
---------------------------------------------------------- Chapter 2:3

Not only is the total funding for the four major programs fixed over
the life of ISTEA, but the funding for the two largest programs--the
NHS and STP, together accounting for 40 percent of all the funding
apportioned in fiscal year 1995--is based, in part, on underlying
factors that are largely irrelevant to the highway system's needs. 
As we reported in March 1986,\4 the factors that influenced the
historical targets for funding in the federal-aid highway
program--land area, postal mileage, and population--are not closely
related to the highway system's needs.  Furthermore, our March 1986
report and an FHWA-sponsored study indicated that alternative
factors, such as lane miles and annual contributions to the Highway
Trust Fund, are more closely aligned with highway needs. 


--------------------
\4 GAO/RCED-86-114. 


      CERTAIN FACTORS USED TO
      APPORTION HIGHWAY FUNDS ARE
      IRRELEVANT OR OUTDATED
-------------------------------------------------------- Chapter 2:3.1

In our March 1986 report, we found that the factors used to apportion
certain highway funds--land area, postal mileage, and
population--were not closely related to the highway system.  At the
time of our report, the data on which these factors were based were
already between 40 and 70 years old.  Specifically, the report
detailed the following problems with the factors: 

  A state's land area was originally included as a factor in the
     distribution formula in 1916.  Land area was thought to provide
     a balance for the factor based on population and to reflect a
     state's future highway needs.  However, this approach resulted
     in large but sparsely populated states receiving larger
     apportionments than they otherwise would have.  In addition,
     land area no longer bears a close relationship to future highway
     needs, namely the need for new construction, since the highway
     system is no longer growing rapidly throughout the country. 

  Postal mileage was included as a formula factor in 1916 to provide
     a constitutional justification for federal involvement in
     highways (the power to establish post offices and post roads). 
     By 1919, changes to the highway legislation had ended the need
     for this justification.  In addition, since postal mileage is
     computed on the basis of the distance traveled both on and off
     the federal-aid highway system, it is unrelated to either the
     extent of the federal-aid highway network or its use. 

  Population figures for formula use were derived every 10 years from
     the census.  As a result, changes in the states' populations
     were accounted for only at 10-year intervals.  This problem has
     been exacerbated under ISTEA, since the population data
     underlying the states' historical shares for ISTEA's major
     funding calculations are, in part, based on 1980 population
     data, not the more current 1990 data. 

In our March 1986 report, we also identified those factors previously
suggested to the Congress as consistent with basic federal highway
programs and for which data were available.  Our results supported
lane miles as a direct measure of the size of the road network and
thus as a reflection of the extent of the system to be preserved.  In
addition, we found that vehicle miles traveled and motor fuel
consumption reflected the extent of highway use.  We recognized that
each of these factors has its own advantages and disadvantages in
establishing a formula.  (The advantages and disadvantages of certain
formula factors are addressed in ch.  3.)

Finally, we recognized that changing the factors used in certain
highway apportionment formulas would result in some states' receiving
more or less funds than they did under the then-current formulas.  We
suggested that to lessen these impacts, a transition period could be
provided during which the full effect of the formulas would be
gradually introduced.  However, the Congress elected not to change
the basic formula structure. 


      LINKAGES EXIST BETWEEN
      APPORTIONMENT FACTORS AND
      HIGHWAY NEEDS
-------------------------------------------------------- Chapter 2:3.2

In December 1986, Jack Faucett Associates, a consultant for FHWA,
issued a report evaluating alternative apportionment formulas for
highway funds that included correlation analysis.\5 Using this tool,
the report showed, for example, that a state with a large number of
vehicle miles traveled on the Interstate would also have a high
requirement for repairs to the Interstate.  Similarly, the states
that contributed large amounts of revenue to the Highway Trust Fund,
reflecting substantial use of motor fuels, were also shown to require
more repairs of the Interstate. 

The correlation analysis was reported in terms of values between zero
and one.\6 The closer the value is to 1, the closer the correlation
between the factor and the need for repairs to the Interstate.  Table
2.1 shows the correlation between selected apportionment factors and
the states' need for Interstate repairs, as reported. 



                               Table 2.1
                
                      Correlation Between Selected
                 Apportionment Factors and Repair Needs
                         on Interstate Highways

Factor                                                     Correlation
--------------------------------------------------  ------------------
Interstate vehicle miles traveled                                0.913
Highway Trust Fund contributions (annual)                        0.900
Interstate lane mileage                                          0.883
Highway Trust Fund contributions (historical)                    0.870
Total motor vehicle registrations                                0.861
Population's weighted income                                     0.780
Total population                                                 0.778
Interstate mileage                                               0.776
Urban population                                                 0.766
Rural population                                                 0.427
Daily mean temperature                                           0.116
Annual snowfall                                                  0.102
Per capita personal income                                       0.038
Annual precipitation                                             0.015
----------------------------------------------------------------------
Source:  Jack Faucett Associates. 

As table 2.1 indicates, the highest correlations--at least
0.900--existed for vehicle miles traveled and annual contributions to
the Highway Trust Fund.  Interstate lane mileage, contributions to
the Highway Trust Fund over time, and total motor vehicle
registrations also showed fairly strong correlations with the need to
repair the Interstate.  This was not the case, however, for
weather-related variables or per capita income.  Furthermore, the
strong correlations between certain of these factors and major needs
for repair diminished for federal-aid highways other than the
Interstate.\7


--------------------
\5 Correlation analysis produces a statistic, called the correlation
coefficient, that measures the extent to which the values of two
variables are associated with each other. 

\6 None of the factors showed a negative correlation with the need
for repairs to the Interstate. 

\7 Highway repair needs for other major highways showed correlations
of 0.741 for contributions to the Highway Trust Fund in 1984,
compared with 0.9 for the Interstate.  The correlations became weaker
for highways that were not considered major and were located in rural
areas. 


   EQUITY ADJUSTMENTS BENEFIT MOST
   STATES
---------------------------------------------------------- Chapter 2:4

Equity adjustments were designed to address the concerns of the
states that contribute a greater share of highway user taxes than
they receive in federal-aid highway funds.  In addition, another
adjustment provides each state with the same relative share of
overall funding that it received in the past. 

The three equity adjustment categories described below--Minimum
Allocation, 90 Percent of Payments Adjustment, and Donor State
Bonus--address the concerns of those states that contribute more in
highway user taxes than they receive in federal-aid highway funds: 

  The Minimum Allocation guarantees a state an amount such that its
     percentage of the total apportionments and prior-year
     allocations from certain highway funding categories is not less
     than 90 percent of the state's estimated percentage of
     contributions to the Highway Trust Fund's Highway Account.\8

  The 90 Percent of Payments Adjustment ensures a state that selected
     apportionments for the fiscal year and allocations in the
     previous fiscal year will equal at least 90 percent of its
     contributions to the Highway Trust Fund's Highway Account. 

  The Donor State Bonus, as implemented by FHWA, compares each
     state's projected contributions to the Highway Trust Fund in the
     fiscal year with the apportionments that the state will receive
     in that fiscal year.  Starting with the state having the lowest
     return (apportionments compared with contributions), each state
     is brought up to the level of return for those states with the
     next highest level of return.  This process is repeated
     successively for each state until the funds authorized for this
     funding category in that fiscal year are exhausted. 

Finally, a fourth adjustment category, referred to as Hold Harmless,
addresses a different objective--preserving the states' historical
funding share, recognizing the legislative compromises embedded in
ISTEA.  ISTEA established a percentage for selected apportionments
and prior-year allocations that each state must receive annually. 
For example, this legislatively prescribed funding percentage is 1.74
for Alabama, 0.41 for Delaware, 0.69 for Idaho, 3.72 for Illinois,
and 4.36 for Massachusetts.  These funding percentage shares can
result in a state's receiving an addition to the regular
apportionments, so that the state's total apportionment will equal
the established percentage. 

As figure 2.1 showed, the calculations that determine the level of
funding that each state receives for the various funding categories
occur in a strict sequence.  All of the equity adjustments come into
play late in the sequential calculation.\9

Therefore, these adjustments essentially increase the funding
calculated for a state up to that point.  For example, if a state is
hypothetically entitled to a total apportionment of $500 million on
the basis of the Hold Harmless provision, it will receive that amount
regardless of whether all the calculations up to that point yielded a
total of $200 million, $300 million, or $400 million. 

In fiscal year 1995, equity adjustments accounted for $2.8 billion
(16 percent) of the approximately $18 billion distributed to the
states.  Only nine states--Colorado, Connecticut, Hawaii, Maryland,
Pennsylvania, Rhode Island, South Carolina, Virginia, and
Washington--and Puerto Rico did not receive funding through equity
adjustments in fiscal year 1995, as highway apportionments for each
of these jurisdictions met all of ISTEA's stated equity criteria on
the basis of the funding for the programs alone.  For the other 41
states and the District of Columbia, the total amount of federal
highway funding apportioned in fiscal year 1995 was ultimately
increased by equity adjustments. 


--------------------
\8 In determining whether a state qualifies for a Minimum Allocation
adjustment, the funds considered in the calculation are those for
grants for the Interstate Construction Program, Interstate
Maintenance Program, Interstate Substitution Program, NHS, STP,
Bridge Program, Scenic Byways Program, and Safety Belt and Motorcycle
Helmet Program and the allocations from any of these programs. 

\9 A fifth equity funding category--Reimbursement for Interstate
Segments--is not described here, since reimbursement for the costs
that the states incurred in constructing the Interstate System
without federal assistance will not begin until fiscal year 1996. 


   FUNDING FOR DEMONSTRATION
   PROJECTS IS NOT GOVERNED BY A
   FORMULA
---------------------------------------------------------- Chapter 2:5

Funding for demonstration projects is distinct from apportionments to
the states in that the authorized funding for such projects is not
distributed by formula.  Rather, the Congress directs how certain
funds are to be distributed by requiring that particular projects
receive a specified amount of funding.  Funding for such projects is
authorized by the congressional committees with jurisdiction over
highway appropriations and authorizations. 

The amount of federal funds authorized for demonstration projects has
grown since 1982.  ISTEA alone authorized over $6.2 billion over 6
years for 539 demonstration projects.  While some demonstration
projects address critical transportation problems and can be
considered nationally significant, authorizing a large number of such
projects could prove troublesome.  As we noted in a 1991 report\10
and testimony in 1993 and 1995\11 before the Subcommittee on
Transportation, House Committee on Appropriations, demonstration
projects often cost more than expected.  In our 1991 report, we found
that for 66 projects reviewed, the federal funding and state matching
funds together accounted for only 37 percent of the projects' total
anticipated costs.  Future finances could be drained if extra federal
funds are needed to cover the cost of completing the projects. 
Demonstration projects can also yield a low payoff for a variety of
reasons, including the fact that they frequently are not aligned with
the states' transportation priorities, can languish in the early
stages of project development, or may never get started at all.  For
instance, in our 1991 report, we found that for 22 of the 66 projects
reviewed, none of the authorized funds ($92 million) had been
obligated, even though the projects had been authorized 4 years
earlier. 

Figure 2.3 depicts the funding to each state for highway programs,
and, if applicable, any modifications to that funding realized
through either equity adjustments or funding for demonstration
projects provided under ISTEA in fiscal year 1995. 


   Figure 2.3:  Highway Funds
   Provided to States for
   Programs, Equity Adjustments,
   and Demonstration Projects in
   Fiscal Year 1995

   (See figure in printed
   edition.)



   (See figure in printed
   edition.)


--------------------
\10 Highway Demonstration Projects:  Improved Selection and Funding
Controls Are Needed (GAO/RCED-91-146, May 28, 1991). 

\11 Surface Transportation:  Funding Limitations and Barriers to
Cross-Modal Decision Making (GAO/T-RCED-93-25, Mar.  31, 1993) and
Surface Transportation:  Reorganization, Program Restructuring, and
Budget Issues (GAO/T-RCED-95-103, Feb.  13, 1995). 


   CONCLUSIONS
---------------------------------------------------------- Chapter 2:6

ISTEA authorized approximately $120 billion for highway construction
and repair and related activities over 6 years, emphasized
quality-of-life and intermodal objectives, revamped major highway
programs, and offered states and localities unprecedented
opportunities to use federal highway and mass transit capital funds
across modal lines.  But the factors underlying the distribution of
funds for two of the largest highway programs--the NHS and
STP--essentially remained the same, since each state's funding was to
be based on the historical share of funds the state received from
major programs before ISTEA was enacted.  Locking in the status quo
on the basis of historical funding averages has also been supported
through two other funding avenues.  First, a state's total funding
share for the four largest programs is fixed over the life of ISTEA. 
Second, the Hold Harmless equity adjustment category serves to raise
the states' ultimate level of annual funding to a predetermined
percentage share of the total funding available.  These percentage
figures, which are spelled out in ISTEA and remain fixed for the
act's duration, were derived primarily from historical averages
rather than current circumstances. 

For major highway programs, the data underlying the distribution of
highway funds to the states are generally outdated, unresponsive to
changing conditions, and often not reflective of the nation's highway
system or its usage.  Furthermore, as mentioned above, because the
percentage share is fixed for the four largest programs, any updated
data that are factored into the calculation for two of these programs
are negated. 


ALTERNATIVES FOR DISTRIBUTING
FEDERAL HIGHWAY FUNDS
============================================================ Chapter 3

On the basis of our analysis and discussions with federal and state
transportation officials, ISTEA's myriad objectives for highways can
be placed into four overarching categories:  (1) maintaining and
improving the highway infrastructure, (2) returning the majority of
funds to the state where the revenue was generated, (3) fostering
social benefits, and (4) safeguarding the states' historical funding
shares.  The first two objectives translate into formula components
that are at the core of the distribution process.  Addressing the
states' highway needs, such as the miles of highway in need of repair
and the deterioration of the highway associated with traffic loads,
is a primary objective in the distribution of highway funds.  The
second component, calling for a return of funds to the state in which
they were generated, supports a congressional objective of having the
states receive a substantial return on the federal fuel and other tax
receipts that they generate and contribute to the Highway Trust
Fund.\1


The third and fourth objectives discussed in this chapter could be
met through formula components that would distribute funds set aside
from the regular apportionment process.  A portion of formula funding
could be devoted to social goals by, for example, directing a portion
of funding to selected purposes such as improving air quality and
conserving energy.  Finally, a share of funding could be set aside
and used to protect the states' historical funding shares.  The
formula objectives may be used singly or in combination and may
further be targeted to specific program categories--such as the
NHS--that are deemed to merit special attention. 

While we focus in this report on the existing, overarching highway
objectives, new components could be added to recognize the states'
capacity to fund highway needs from state resources, the states'
level of effort (LOE) in meeting their own needs, and geographic
differences in the cost of maintaining existing highway networks. 
Although similar factors have been applied in other programs, they
have not been applied to highway programs in the past.  But FHWA did
provide a report to the Congress in 1994 that addresses measures for
assessing how much of their available resources the states or local
areas devote to surface transportation.\2 (App.  III provides
additional details on FHWA's study.)

The task of revising the formula for distributing highway funds will
be difficult because needs vary across the country and objectives
conflict among themselves.  For example, relief of congestion is more
pressing in urban areas of the country, whereas connecting rural
areas is more pressing in sparsely populated areas.  The analysis
presented in this report is intended to provide the Congress with
formula alternatives that reflect the key objectives governing the
current federal-aid highway formula. 


--------------------
\1 As discussed later in this chapter, there is some overlap between
these first two objectives, as states' fuel tax receipts also tend to
be a strong measure of highway needs. 

\2 Report on State Level of Effort, required by section 6013 of the
Intermodal Surface Transportation Efficiency Act of 1991
(P.L.102-240) Mar.  1994. 


   COMPONENT 1:  DISTRIBUTION OF
   FUNDS BASED ON INDICATORS OF
   NEED
---------------------------------------------------------- Chapter 3:1

Many individual factors making up a formula are capable of supporting
the principle of distributing funds on the basis of the states'
relative needs.  One possibility would be to use factors that relate
to the states' actual needs, such as the states' miles of poor
pavement or number of deficient bridges.  In this approach, the
states with the poorest highway conditions would be granted a larger
share of the funds than the states with better highway and bridge
conditions. 

However, a formula based on direct measures of need could prove
problematic.  The use of actual needs can foster a perverse incentive
by potentially encouraging the states to permit their highway
infrastructure to worsen in order to capture a greater share of
federal highway funds.\3 Moreover, this approach would reward the
states with the poorest highway and bridge conditions while
penalizing the states that have maintained these structures.  In
addition, the condition of highways and bridges varies considerably
among the states.  For instance, as of December 1993 the percentage
of deficient bridges on the NHS ranged from a low of 8 percent in
North Dakota to a high of 64 percent in Massachusetts. 

The disadvantages of basing a formula on actual needs can be remedied
through the use of proxies of need, such as those reflecting the
extent or usage of a highway system, or more highway-neutral measures
such as population.  Such proxies have the advantage of being
relatively objective and neutral.  However, there is debate among the
states and other transportation experts on what factors can
appropriately serve as proxies for distributing highway funds.  Some
insight can be gained from the Faucett study performed for FHWA in
1986 and discussed in chapter 2.  This study indicated a strong
correlation, particularly for repairs of the Interstate, between
highway needs and lane miles and vehicle miles traveled.  The
following sections discuss the advantages and disadvantages of
certain proxies in more detail. 


--------------------
\3 To some extent, the problem of the perverse incentive can be dealt
with by limiting the time during which a state can get more money
because a particular highway or bridge is in poor condition.  For
instance, if after 5 years a particular highway or bridge would not
count towards a state's needs regardless of its condition, then a
state would have no incentive to keep that highway or bridge
unrepaired beyond that time to get more federal money. 


      EXTENT OF A SYSTEM IS
      REFLECTED IN THE NUMBER OF
      MILES COVERED
-------------------------------------------------------- Chapter 3:1.1

The primary measures of the extent of the federal-aid highway network
are center-line miles and lane miles.  Center-line miles reflect the
length of the system, whereas lane miles represent the number of
lanes per section multiplied by the actual length of the section. 
For example, a four-lane section that is 2 miles long would equal 2
center-line miles or 8 lane miles. 

Some states believe that center-line miles, not lane miles, are a
more appropriate factor for distributing highway funds.  For
instance, transportation officials from Idaho, Montana, North Dakota,
South Dakota, and Wyoming told us that to the extent that road
mileage is considered in a formula, center-line miles more accurately
reflect interconnectivity on a national and regional basis.  However,
while center-line miles accurately depict overall connections though
a linear measurement of highways, this measure does not capture any
information on the various widths of highways, because a two-lane
highway and an eight-lane highway are considered equal under this
measure. 

The width and length of highways is reflected in lane miles, and as
we noted in our 1986 report, lane miles are a good measure of the
extent of the highway system (capital stock) to be preserved.  In
addition, using lane miles as a factor for apportioning highway funds
was endorsed by a Policy Review Committee of the American Association
of State Highway and Transportation Officials (AASHTO) in fiscal year
1991.  As the committee noted, lane miles are a direct measure of the
extent of public roads in both rural and urban areas.  The committee
further noted that a measure of lane miles is probably the simplest
and most efficient potential apportionment factor on which to obtain
accurate information and that annual data are generally available
within 6 to 9 months of the close of the calendar year. 

Regardless of whether center-line miles or lane miles are used to
indicate the extent of a system, some observers criticize the use of
mileage for apportioning future highway funds because such usage
could reward expansion of the system.  Thus, this type of
apportionment factor would tend to encourage more highway
construction, to the possible detriment of adequately preserving the
existing network and of considering air quality.  Several actions
could be taken to counterbalance such tendencies.  First, as part of
the third component of the formula framework discussed later in this
chapter, set-asides could be established to reward those states that
meet certain preservation or maintenance goals.  Second, greater use
of performance measures geared to preserving the existing
infrastructure would help FHWA ensure that the states do not neglect
needed preservation and maintenance.  As we noted in our July 1994
testimony before the Senate Committee on Environment and Public
Works,\4 performance expectations need to be established for
preservation and maintenance and other important goals for the NHS. 
A well-maintained system is the necessary foundation for pursuing the
myriad goals for the system, which include economic development,
enhanced mobility, and improved air quality.  Without such a
foundation, system enhancements such as alleviating congestion and
improving the efficient movement of goods may not be fully realized. 


--------------------
\4 National Highway System:  Refinements Would Strengthen the System
(GAO/T-RCED-94-266, July 15, 1994). 


      USE OF THE SYSTEM IS ALSO AN
      INDICATOR OF HIGHWAY NEEDS
-------------------------------------------------------- Chapter 3:1.2

While measures of a system's extent provide part of the story on
highway needs, the condition of the road is also an important
element.  Condition can be captured by measures of the use of a
system, as distinct from the extent of the system.  A system's usage
is typically gauged using factors such as vehicle miles traveled or
consumption of motor fuel. 

One advantage of using data on the vehicle miles traveled as a
formula factor is that they tend to be quite reliable.  The AASHTO
Policy Review Committee observed that data on vehicle miles traveled
have been statistically designed for a high level of measurable
accuracy and are relevant as an indicator of both capital and system
preservation needs.  Also, in the Faucett study, vehicle miles
traveled garnered one of the highest correlation values, 0.913, of
all the factors related to Interstate repair needs.  That is, a state
with a high number of vehicle miles traveled would also likely have
high needs for repair of the Interstate. 

Another proxy of system use is motor fuel consumption.  Motor fuel
consumption reflects travel on all roads, not just on the federal-aid
system or on roads under a state's jurisdiction.  Therefore, it would
not be a precise measure for apportioning funds to specific groups of
roads.  These data are reported by states monthly and adjusted at
year's end.  Annual data are generally available within 6 to 9 months
of the close of the calendar year.  Fuel consumption patterns may
differ across states because of the urban-rural population mix, the
amount of travel done under congested conditions, differences in
physical terrain, and fuel purchases by transients in those states
with lower fuel taxes, among other things. 

While vehicle miles traveled and motor fuel consumption correlate
well with system usage, they do have some drawbacks.  For example,
vehicle miles traveled measure the vehicles moved rather than the
people and do not account for different vehicle classifications. 
Moreover, both factors are largely at odds with air quality
objectives, and the principle of rewarding motor fuel consumption
with more highway funding also conflicts with the goal of encouraging
energy conservation.  New Jersey transportation officials, for
instance, noted that such factors reward energy consumption and air
pollution and penalize those who successfully enact measures to
reduce the use of single-occupant vehicles.  Similarly,
transportation officials from several other states noted that the
Congress has previously rejected the notion of giving vehicle miles
traveled greater weight in apportioning funds, in part because of the
strong environmental objections raised.  As in the case of the
factors related to the system's extent, the disadvantages associated
with measures of the system's usage could be at least partially
counteracted by building incentives into the formula or by creating
appropriate performance standards. 


      OTHER PROXIES ARE LINKED
      LESS DIRECTLY TO EXTENT AND
      USAGE OF HIGHWAY SYSTEM
-------------------------------------------------------- Chapter 3:1.3

A host of other factors--such as population, climatic conditions
(daily mean temperature, annual snowfall, and annual precipitation),
and per capita income--could also be used to determine how highway
funds are distributed.  Yet, as the Faucett study demonstrated, a low
correlation exists between highway needs as reported by FHWA and
climatic variables and per capita income.\5

The Executive Director of the Surface Transportation Policy Project
supports the use of population levels for distributing highway
funds.\6 The Executive Director stated that to the extent that the
formula uses factors such as vehicle miles traveled, lane miles, and
fuel consumption, it encourages behavior that runs counter to the
objectives of reducing congestion and improving air quality.  In his
view, population and population density would be preferable
alternatives as proxies.  These proxies were recommended because they
were perceived as avoiding the perverse effects tied to a system's
extent and usage, and because the data are sound. 

Population data, however, also have limitations.  As noted by the
Executive Director of AASHTO, the link between population and the
states' highway needs is questionable.  First, while funds would be
targeted to congested urban areas, the approach would do little to
accommodate the needs of rural areas.  Second, the approach does not
recognize that goods produced in sparsely populated areas ultimately
must be transported to dense areas.  And some state transportation
officials from sparsely populated states believe that much of the
traffic that occurs in densely populated areas is local.  Officials
from these states maintain that the promotion of interstate commerce
should be a principal objective of the federal-aid highway program
and that federal funds should target the highways that tend to carry
national, not local, traffic. 


--------------------
\5 Although income was not shown to be highly correlated with highway
needs, an income factor could also be included in an apportionment
formula to reflect the ability of a state to fund its highway needs
from state resources. 

\6 The Surface Transportation Policy Project is a coalition of over
100 groups seeking to ensure that transportation policy and
investments help meet a variety of social goals, such as energy
conservation. 


      A COMBINATION OF PROXIES FOR
      NEED COULD BE EMPLOYED
-------------------------------------------------------- Chapter 3:1.4

As we reported in March 1986,\7 factors reflecting a system's extent
and use in isolation do not provide a complete picture on the states'
needs.  Combining such factors helps to round out the formula's
capacity to reflect the states' total needs.  Introducing neutral
factors, such as population, into the formula further diversifies the
mix of factors and alters the amounts the states receive.  The
analysis that follows focuses on two possible blends of proxies for
need.  Table 3.1 provides an outline of the factors to be considered
in the two alternatives. 



                               Table 3.1
                
                 Factors Considered in Two Hypothetical
                 Formula Alternatives Based on Proxies
                                for Need

----------------------------------  ----------------------------------
Alternative 1                       Distribution based equally on
                                    total lane miles
                                    total vehicle miles traveled

Alternative 2                       Distribution based equally on
                                    total lane miles
                                    Interstate vehicle miles
                                    traveled
                                    state population
----------------------------------------------------------------------
The first alternative assumes that 100 percent of total highway funds
are distributed to the states based equally on total lane miles and
total vehicle miles traveled.  Under this alternative, 13 percent of
the overall highway funds would be redistributed.  Twenty-three
states and Puerto Rico would receive more funds than they were
apportioned in fiscal year 1995.  The average dollar gain would be
$102 million; $643 million would be the high end of the range
(California), and $7 million would be the low end (North Carolina). 
Twenty-seven states and the District of Columbia would receive less
funding.  For these recipients, the average loss would be $87
million; the greatest loss would be $417 million (Pennsylvania), and
the smallest loss would be $5 million (Oklahoma).  (State-by-state
details are provided in app.  IV.)

The second alternative assumes that 100 percent of the total highway
funds are returned to the states based equally on total lane miles,
vehicle miles traveled on the Interstate, and state population. 
Under this approach, 10 percent of overall highway funds would be
redistributed.  Twenty-six states and Puerto Rico would receive more
funds than they were apportioned in fiscal year 1995.  The average
dollar gain would be $73 million; the high end of the range would be
$366 million (California), and the low end would be $2 million
(Nevada).  Twenty-four states and the District of Columbia would
receive less funding.  For these states, the average loss would be
$79 million; the greatest loss would be $359 million (Pennsylvania),
and the smallest loss would be $1.6 million (Alabama). 
(State-by-state details are provided in app.  V.)


--------------------
\7 GAO/RCED-86-114. 


   COMPONENT 2:  RETURN OF FUNDS
   TO THE SOURCE
---------------------------------------------------------- Chapter 3:2

One indicator of need that we intentionally omitted from the above
discussion is that of the states' contributions to the federal
Highway Trust Fund.  As a formula factor, these contributions have a
special status because they align with two key objectives of the
highway program.  Not only do contributions to the Trust Fund
correlate strongly with highway needs, particularly for major
highways, but the states' returns on these contributions have also
been considered a key measure of equity.  For years, the highway
apportionment formula has endorsed, through one or more equity
adjustments, the principle that the states ought to receive back a
substantial portion of what they deposit into the Trust Fund.\8 If
the formula were restructured to encompass a pure return-to-origin
approach, each state's contribution to the Trust Fund would simply be
returned to that state.  This does not currently occur.  FHWA's data
indicate that in 1993, federal highway apportionments as a percentage
of the states' contributions to the Highway Trust Fund's highway
account ranged from 83 percent for South Carolina to 707 percent for
Hawaii. 

FHWA estimates the states' contributions to the Trust Fund, which
derive from various federal excise taxes such as the gasoline and
diesel tax.  Because the majority of revenues credited to the Trust
Fund derive from the federal fuel tax, the states' contributions to
the Trust Fund tend to be quite closely linked with fuel consumption. 
As a potential formula factor, these contributions therefore offer
the same kinds of advantages and disadvantages as fuel consumption
does. 


--------------------
\8 The principal funding categories in ISTEA that support a
return-to-origin definition of equity are Minimum Allocation, 90
Percent of Payments Adjustment, and Donor State Bonus, as implemented
by FHWA. 


      RETURN-TO-ORIGIN APPROACH
      OFFERS ADVANTAGES AND
      DISADVANTAGES
-------------------------------------------------------- Chapter 3:2.1

Returning the states' contributions to the Highway Trust Fund to
their source is a relatively simple and direct way of distributing
these funds.  Some state transportation officials could be expected
to support this approach because it would guarantee that all or a
substantial amount of the revenues collected in their states would be
returned to them.  An advantage of returning funds to their source is
that, as the 1986 Faucett study shows, contributions to the Highway
Trust Fund tend to correlate highly with highway needs, particularly
for major highways. 

However, the return-to-origin approach would not be universally
attractive, as a number of states would lose funds.  For instance,
those states whose fuel usage is low relative to their land area and
extent of highway network would be financially hurt.  A prime
argument made by officials from these states is that the national
interest requires highways to span the wide expanses of large,
sparsely populated states that are the source of goods for citizens
in the population centers, but the financial resources of those
states are often insufficient to construct, maintain, and operate
such networks. 

Two additional arguments are made against the return-to-origin
approach.  First, as New York transportation officials noted,
formulas based on returning contributions to the Trust Fund to the
state where they are raised meet neither federal or state
transportation goals nor national policy as set forth in ISTEA.  If
the primary goal of federal apportionment formulas is to return
revenues from motor fuel taxes to the place they were earned, these
officials questioned whether there was a need for a federal program. 
Second, state officials have questioned the wisdom of selecting a
formula factor that is geared predominantly to fuel use.  They argue
that such an approach rewards greater use of motor fuel and as such
contradicts federal goals of improving air quality and conserving
energy. 

Finally, this approach would not necessarily preclude congressional
direction of the use of those funds.  Legislation could still specify
that the returned funds be used in certain proportions for certain
programs, such as the NHS.  Moreover, the return could function as
(1) a simple return of funds, in which states would be exempt from
any or most federal oversight, or (2) a distribution of funds, in
which FHWA would oversee the programs for which the funds were
returned. 


      RETURN-TO-ORIGIN APPROACH
      ALSO YIELDS WINNERS AND
      LOSERS
-------------------------------------------------------- Chapter 3:2.2

Under a return-to-origin approach, we considered three different
alternatives, which are summarized in table 3.2.\9




                               Table 3.2
                
                   Outline of Three Return-to-Origin
                              Alternatives

----------------------------------  ----------------------------------
Alternative 1                       Excluding Interstate Construction
                                    funds, all funds are returned to
                                    the source.

Alternative 2                       Excluding demonstration project
                                    funds, all funds are returned to
                                    the source.

Alternative 3                       With no exclusions, all funds are
                                    returned to the source.
----------------------------------------------------------------------
Under the first alternative, $17.8 billion of the total $19.1 billion
would be returned to the source.  This amount would represent all the
funds (including ISTEA's funds for demonstration projects)
distributed to the states in fiscal year 1995, except funds for
Interstate Construction.  These funds were excluded since the
Interstate Construction program's final apportionment was made at the
beginning of fiscal year 1995, and only 14 states and the District of
Columbia received funds in the program's last year.  Under this
alternative, 24 states would receive more funds than they were
apportioned in fiscal year 1995, while the remaining states, along
with the District of Columbia and Puerto Rico, would lose funds.  The
average dollar gain would be $67 million; the average loss would be
$58 million.  (State-by-state details are presented in app.  VI.)

Under the second alternative, the total amount of funds ($18.1
billion) apportioned to the states in fiscal year 1995 would be
returned, including funds for Interstate Construction funds but
excluding those for demonstration projects.  Funds for demonstration
projects are excluded from this analysis because these funds are not
distributed by formula.  Rather, the Congress directs how certain
funds are to be distributed by requiring that particular projects
receive a specified amount of funding.  Under this alternative, 27
states would receive more funds than they were apportioned in fiscal
year 1995.  The average dollar gain would be about $68 million; 23
states, along with the District of Columbia and Puerto Rico, would
receive less funding.  For these recipients, the average loss would
be $73 million.  (State-by-state details are provided in app.  VII.)

Under the third alternative, all funds would be returned to the
states, including funds for Interstate Construction and demonstration
projects along with other program funding.  Thus, this alternative
recognizes the full $19.1 billion distributed to the states in fiscal
year 1995.  Under this alternative, 24 states would gain an average
of $86 million, while 26 states along with the District of Columbia
and Puerto Rico would lose an average of $74 million. 
(State-by-state details are provided in app.  VIII.)


--------------------
\9 The level of funding used for this analysis is based on
contributions to the Highway Trust Fund's highway account in fiscal
year 1993--the latest year for which data were available. 


      NEEDS-BASED AND RETURN-BASED
      COMPONENTS CAN BE COMBINED
-------------------------------------------------------- Chapter 3:2.3

As mentioned previously, the first two formula components discussed
above--based on needs and based on returning funds to the source--can
be combined.  A significant advantage of blending these components is
that programs of particular concern (notably, the NHS) could receive
special attention through the use of carefully targeted formula
factors.  In contrast, a return-to-origin approach might be more
appropriate for the STP, which already has characteristics that
resemble those of a block grant program and which would thus lend
itself well to an approach under which funds are returned to the
states. 

For purposes of illustration, the following two hypothetical
distributions blend needs-based and return-to-origin approaches along
the existing split between the STP and two other primary highway
programs--Interstate Maintenance and the NHS.  The current funding
level for the STP represents about 40 percent of the total funds
authorized for these programs.  The two alternatives outlined in
table 3.3 and described below maintain this distribution of funding. 



                               Table 3.3
                
                    Combination of Principles in Two
                      Redistribution Alternatives

----------------------------------  ----------------------------------
Alternative 1                       Distribution based on
                                    40% returned to source
                                    remaining 60% based equally on
                                    NHS lane miles and Interstate
                                    vehicle miles traveled

Alternative 2                       Distribution based on
                                    40% returned to source
                                    remaining 60% based equally on
                                    NHS lane miles, Interstate vehicle
                                    miles traveled, and population
----------------------------------------------------------------------
Under the first alternative, 33 states would receive an average of
$64 million more than they were apportioned in fiscal year 1995,
while 17 states, along with the District of Columbia and Puerto Rico,
would lose $111 million on average.  Overall, 11 percent of highway
funds would be redistributed.  (State-by-state details are presented
in app.  IX.)

Under the second alternative, a slightly different redistribution
pattern would emerge.  The average dollar gain for 30 states would be
$60 million; 20 states, along with the District of Columbia and
Puerto Rico, would receive less funding than they did in fiscal year
1995.  For these recipients, the average loss would be $82 million. 
In total, about 9 percent of the highway funds would be
redistributed.  (State-by-state details are presented in app.  X.)


   COMPONENT 3:  SET-ASIDES TO
   FOSTER IMPROVEMENTS
---------------------------------------------------------- Chapter 3:3

While many social objectives are probably best addressed through
means other than the highway apportionment formula, a portion of
highway funds might nonetheless be retained to advance specific
objectives and/or to counterbalance some of the potential
disadvantages of the principal formula factors.  For instance, a
certain percentage of funds--10 percent, for example--could be set
aside before the remaining funds were distributed to the states. 
Payments drawn from this set-aside could be used to provide bonuses
to advance quality-of-life objectives, to reward improvements in the
condition of highway infrastructure above a certain defined floor,
and to advance highway safety.  These and similar objectives are all
laudable; however, constraint in selecting the objectives may be
warranted to prevent the dilution of funds that could result from
attempting to meet numerous objectives. 

One approach to distributing the set-aside moneys would be to direct
set-aside funds to those states suffering from unique or concentrated
needs in certain areas.  A prime example of this approach is the
existing Congestion Mitigation and Air Quality Improvement (CMAQ)
program, which directs funds to states with particularly severe
problems in air quality.  ISTEA provided CMAQ with a $6 billion
authorization-- approximately $1 billion annually for 6 years.  CMAQ
is focused on investment in air quality improvements and provides
funds for projects that expand or initiate transportation services
that benefit air quality.  It is directed to those states that are
classified as nonattainment areas for ozone and carbon monoxide
(although every state, regardless of its air quality status, is
guaranteed an annual minimum apportionment of 0.5 percent of the
program's total funding.) The advantage of such a program is that it
focuses funding on precisely those areas with the greatest needs. 
The disadvantage is that, as occurs with the needs-based formula
factors discussed earlier in this chapter, directing funding to
states with specific needs can foster a perverse incentive.  In the
case of the CMAQ program, questions have been raised about the wisdom
of essentially rewarding states for their nonattainment status,
particularly given that a state loses CMAQ funding if it makes "too
much progress" in improving air quality. 

A second approach to directing set-aside funding towards specific
goals is to treat the funds as incentive payments.  Incentive
payments, as the name implies, do not redress shortcomings, but
instead reward desired behaviors or accomplishments.  For example,
shared set-aside funding could be used to reward states that make
notable and measurable improvements in the percentage of the state's
pavement condition rated as "good" under FHWA's classification
system.\10 To emphasize the condition of the nation's most heavily
traveled highways, such rewards could be further refined to focus on
improvements in the condition of the NHS. 

One concern with providing incentives for improvements in highway
conditions, however, is that the data from the states on the
condition and performance of their roadways are not always reliable,
making it more difficult to equitably distribute such incentive
payments.  In subcommittee hearings for the House Committee on
Appropriations in fiscal year 1994, FHWA was questioned on
significant swings in the percentage of Interstate pavement rated in
poor condition, as illustrated by table 3.4. 



                               Table 3.4
                
                 Examples of Significant Changes in the
                Percentage of Interstate Pavements Rated
                    in Poor Condition in Some States

                                     Percentage of       Percentage of
                                        Interstate          Interstate
                                  pavement in poor    pavement in poor
State                              condition, 1989     condition, 1991
------------------------------  ------------------  ------------------
Alaska                                        26.4                 5.1
Arizona                                       27.4                 1.2
Colorado                                       7.6                30.3
Georgia                                       11.7                   0
Michigan                                      10.7                19.0
Nevada                                        33.9                11.6
New Mexico                                     1.3                38.1
North Dakota                                   0.2                32.2
Rhode Island                                  31.4                 1.4
Vermont                                       18.7                 5.9
Wisconsin                                     18.6                   0
----------------------------------------------------------------------
FHWA explained that the data on the condition of the pavement were
based on the use of an index, referred to as the Present
Serviceability Index.  This index, however, represents a subjective
measure of the pavement's ride quality and can be arrived at by a
variety of procedures.  Furthermore, FHWA noted that from time to
time the states have attempted to improve their estimation of this
measure, thus invalidating comparisons with data from previous years. 
As a result, until the reliability of these data is improved, their
use as an indicator for distributing federal highway funds would be
suspect and arbitrary.  An alternative measure, the International
Roughness Index, is a more objective measure of pavement condition
(roughness), and FHWA expects this data source to play a more
prominent role in the future.  In 1993, the most recent year for
which data are available on pavement condition, 37 states used the
International Roughness Index to measure pavement condition on
Interstate highways, while the remaining 13 states continued to rely
on the Present Serviceability Index.  For other major highways, the
proportion of states using the International Roughness Index dropped
to about half the states.  An FHWA official noted that some states do
not use the International Roughness Index because they do not have
the money to purchase the necessary equipment.  Another impediment to
using the International Roughness Index is that the equipment must be
operated at a speed of 35-55 miles per hour, thus making in
infeasible for use on certain major highways in urban areas because
of the presence of other traffic, traffic signals, and other
disruptions. 


--------------------
\10 FHWA uses data from the states that classify their pavement into
broad categories--poor, mediocre, fair, good and very good--on the
basis of the roughness of the ride and surface defects.  According to
FHWA's statistics, about 65 percent of the NHS' urban mileage and 61
percent of the system's rural mileage is classified in less than good
condition. 


   COMPONENT 4:  SAFEGUARDS TO
   MITIGATE AGAINST SUDDEN LOSSES
---------------------------------------------------------- Chapter 3:4

Altering the existing formula would undoubtedly cause shifts in the
states' relative shares of annual highway funding.  Under a number of
the scenarios presented in this chapter, more states would gain funds
than would lose funds, but this overall result would be of little
comfort to the states whose relative position worsened.  Some
individual states such as Alaska and Hawaii could lose 50 percent or
more of their highway funds under any of the scenarios derived from
the approaches based on needs and return to origin.  Sudden and
significant losses would likely play havoc with the states' planning
processes and programs, and it is doubtful that the affected states
would be prepared to cope with losses of this magnitude. 

In addition, the effect that a change in the formula would have on
any state would depend on the percentage of the state's highway
revenue provided by federal funds.  Figure 3.1 depicts federal funds
as a percentage of the states' total highway revenue. 

   Figure 3.1:  Federal
   Contributions as a Percentage
   of Total Highway Funding, by
   State, Fiscal Year 1994

   (See figure in printed
   edition.)

To help temper the effects of changes in the formula, any new formula
might include a component designed to place a cap on the maximum
percentage of loss that any individual state would be expected to
bear as a result of the changes.  For example, a maximum-loss cap of
20 percent might be established.  Thus, if a new formula calculation
caused a given state's funding to fall by 50 percent from the level
it would otherwise be, the cap would come into effect and funding for
the state in question would be reinstated to 80 percent of what the
state would otherwise have received.  The cap could be either
permanent or established for a set period of time during the
transition to a new funding amount. 

Finding the funds to shield the states from severe losses might not
be as difficult as it would first appear.  If the existing, intricate
equity adjustments were replaced with a single, simple cap, the funds
devoted to these equity adjustments in fiscal year 1995--$2.8
billion--would more than offset the states' combined losses in that
year under all of the scenarios discussed in this report.  The
scenario resulting in the greatest adverse impact on the
states--alternative 1 of the needs proxy approach--produced a
combined loss of $2.4 billion.  Alternatively, other categories of
funding, such as those supporting highway demonstration projects
(currently commanding about $1 billion per year), could be redirected
to provide safeguards against sudden losses. 


   CONCLUSIONS
---------------------------------------------------------- Chapter 3:5

Reauthorization of the federal-aid highway program presents the
Congress with the opportunity to review the objectives associated
with providing federal highway funds and the accompanying formula for
distributing the funds.  A review of the program's objectives could
be structured to recognize differences among highways and the federal
role associated with important highways, such as those included in
the NHS. 

There are no perfect factors that embrace the breadth of ISTEA's
diverse objectives as well as the states' different needs. 
Regardless of the factors chosen, some states will experience
disadvantages that the construction of a formula may not be able to
compensate for.  Which states are negatively affected changes with
the factors chosen and the percentage weights assigned to various
factors.  Moreover, as noted in chapter 2, DOT has proposed changes
in the system for delivering grants. 

Whether DOT's proposed changes are adopted or other scenarios for
delivering grants are developed, the Congress will have to reach a
consensus on the national objective(s) that are critical for the
highway program to address; decide whether a formula is the
appropriate vehicle for addressing these objective(s); and for those
objectives that the formula can best address, determine the most
representative factors and corresponding weight to be assigned to
those factors.  If an alternative formula is adopted for distributing
highway funds in the future and this formula would result in dramatic
funding losses for certain states, ways could be considered to reduce
the magnitude of the losses--by, for example, providing for a cap on
the maximum percentage of loss that any one state would be expected
to bear. 


FUNDING CATEGORIES FOR FEDERAL-AID
APPORTIONMENTS
=========================================================== Appendix I

                      Authorized funding
                      level (billions of
Funding category              dollars)\a  Funding factor(s)
--------------------  ------------------  --------------------------------------
Interstate                           7.2  Relative federal share of cost to
 Construction                              complete the Interstate


Interstate                          17.0  Interstate lane miles, 55 percent;
 Maintenance                               vehicle miles traveled on the
                                           Interstate, 45 percent

National Highway                    21.0  Essentially based on each state's
 System                                    share of funds for Interstate
 (NHS)                                     Maintenance, Primary, Secondary,
                                           Urban, and Bridge programs and
                                           adjusted minimum allocations for
                                           1987-91
Bridge                              16.1  Each state's relative share of the
                                           total cost of deficient bridges

Surface                             23.9  Basically the same as for the NHS,
 Transportation                            reduced by apportionments for
 Program (STP)                             Interstate Maintenance, NHS, and
                                           Bridge programs

Interstate                           1.0  Relative federal share of costs to
 Substitution                              complete Interstate Substitution
                                           projects
Minimum Allocation                 5.2\b  Each state's share of funds is not
                                           less than 90 percent of the state's
                                           share of contributions to the Highway
                                           Trust Fund for a specified set of
                                           programs
Congestion                           6.0  Each state's share of population in
 Mitigation and Air                        air quality nonattainment areas, but
 Quality                                   each state is guaranteed at least 0.5
                                           percent of the total funding

Donor State Bonus                    3.0  Rate of return on contributions to
                                           Highway Trust Fund increased for
                                           states that have the lowest return
Interstate                           4.0  Fixed percentages based on the states'
 Reimbursement                             investments in Interstate highways
                                           made before federal funding of the
                                           Interstate
Hold Harmless                      3.6\b  Legislated minimum percentage for each
                                           state as provided in section 1015 of
                                           ISTEA

90 Percent of                      0.4\b  Ensures that each state receives at
 Payments                                  least a 90-percent rate of return on
 Adjustment                                its contributions to the Highway
                                           Trust Fund

Metropolitan                 No separate  Share of national urban population
 planning                authorization\c
 organizations
--------------------------------------------------------------------------------
\a Funding categories are presented in the order in which they occur
in the sequential funding calculation process; funding levels are the
6-year (1992-97) totals for each category under ISTEA. 

\b Estimated amount.  Actual annual amount depends on the formula
calculations for each year. 

\c Metropolitan planning organizations are funded through 1 percent
set-asides from authorizations for the Interstate Maintenance,
National Highway System, Bridge, and Congestion Mitigation and Air
Quality programs and the Surface Transportation Program. 


FISCAL YEAR 1995 FUNDING TO STATES
FOR PROGRAMS, EQUITY ADJUSTMENTS,
AND DEMONSTRATION PROJECTS
========================================================== Appendix II

                        (demonstration projects Dollars in
                                    thousands)

                                                                   Total funding
                                                     Allocations   for programs,
                                                             for          equity
                         Program          Equity   demonstration    adjustments,
State             apportionments     adjustments        projects             and
----------------  --------------  --------------  --------------  --------------
Alabama                  257,242          57,583          26,901         341,726
Alaska                   202,719          28,589               0         231,308
Arizona                  182,802          84,491           2,245         269,538
Arkansas                 149,132          67,297          51,152         267,581
California             1,385,030         372,253          58,963       1,816,246
Colorado                 202,575               0             534         203,109
Connecticut              351,562               0          14,610         366,172
Delaware                  69,248           5,155               0          74,403
Dist. of                  91,496           4,125           4,066          99,687
 Columbia
Florida                  529,411         232,980          33,039         795,430
Georgia                  401,480         135,155          19,394         556,029
Hawaii                   120,851               0           1,104         121,955
Idaho                    102,881          24,206          12,954         140,041
Illinois                 539,955         104,904          45,711         690,570
Indiana                  285,518         122,087          17,278         424,883
Iowa                     201,247          17,738           7,286         226,271
Kansas                   183,832          21,957          13,432         219,221
Kentucky                 214,308          75,278           3,974         293,560
Louisiana                213,335          50,823          12,909         277,067
Maine                     84,053           6,027          34,426         124,506
Maryland                 352,645               0          17,682         370,327
Massachusetts            763,756          23,962           1,086         788,804
Michigan                 382,487         118,959          22,827         524,273
Minnesota                223,780          64,805          36,237         324,822
Mississippi              161,242          46,198           5,106         212,546
Missouri                 316,707          85,985          20,608         423,300
Montana                  142,777          32,302           3,312         178,391
Nebraska                 138,884           1,027             957         140,868
Nevada                   101,695           9,115          13,542         124,352
New Hampshire             81,730           4,011           5,906          91,647
New Jersey               479,176          52,991          37,334         569,501
New Mexico               139,067          51,789           1,987         192,843
New York                 877,479          87,146          65,657       1,030,282
North Carolina           340,200         137,790          18,106         496,096
North Dakota              99,141          13,016          13,064         125,221
Ohio                     518,138         153,384          29,043         700,565
Oklahoma                 196,586          59,820          16,291         272,697
Oregon                   185,186          27,605           8,464         221,255
Pennsylvania             886,137               0         159,585       1,045,722
Puerto Rico               87,783               0               0          87,783
Rhode Island             103,615               0          10,563         114,178
South Carolina           190,515               0           7,121         197,636
South Dakota             109,094          17,821               0         126,915
Tennessee                296,715          79,067           7,084         382,866
Texas                    984,510         193,612          43,498       1,221,620
Utah                     124,454           9,665           2,006         136,125
Vermont                   74,241           5,420           3,680          83,341
Virginia                 378,204               0          25,668         403,872
Washington               237,523               0          16,486         254,009
West Virginia            159,118          10,107          57,371         226,596
Wisconsin                223,886         131,230          13,156         368,272
Wyoming                  109,126           6,820           3,680         119,626
================================================================================
Total                 15,234,274       2,834,295       1,027,085      19,095,654
--------------------------------------------------------------------------------

DOT'S ASSESSMENT OF POSSIBLE
MEASURES RELATED TO STATES'
TRANSPORTATION FINANCING
========================================================= Appendix III

Each state has a unique fiscal and economic framework, and various
factors determine its capability to plan and pay for public services,
such as highway construction and mass transit services.  For example,
the strength of the economic base of a state is tied to its ability
to raise both public and private funds.  Some states tax their
residents almost as heavily as the economic base will allow, while
others are wealthier than the tax burden suggests.  A bonus could be
structured to essentially reward a state whose financial
contributions to transportation services are high relative to the
state's wealth.  This approach was proposed in the highway bill
passed in the Senate in June 1991 (S.  1204), which would have
authorized $4.1 billion over a 4-year period for bonuses for states
with higher-than-average state gasoline taxes and lower-than-average
per capita disposable income.  Although ISTEA did not incorporate the
bonuses proposed in the Senate bill, the Congress did direct the
Secretary of Transportation and the Director of the Bureau of
Transportation Statistics to study and recommend the most appropriate
and accurate methods of calculating the states' level of effort (LOE)
in funding surface transportation programs. 

A Department of Transportation (DOT) report of March 1994 responded
to the legislative mandate to review the states' LOE.\1

DOT evaluated a range of measurements against a series of questions
to determine the "best" LOE measures.  The questions included the
following: 

  Does the measure consider actual spending for highways and mass
     transit as well as the ability to pay? 

  Is the measure accurate, objective, and equitable? 

  Can the measure be updated as necessary? 

  Is the measure uniform across the nation? 

DOT found that four types of state and local measurements of revenue
and spending met their criteria:  (1) equivalent motor-fuel tax
rates--the total receipts for a state's highway and transit use; (2)
highway and mass transit disbursements as a percentage of state and
local disbursements, which considers the relative importance of
spending on surface transportation to other spending; (3)
disbursements for highways and mass transit financed only by state
and local funds compared with the state's and localities' total
disbursements (since this measure eliminates the direct impact of
federal spending, it considers the relative effort that states and
local governments make with their own financial resources); and (4)
per capita state and local spending on highways and mass transit as a
percentage of per capita personal income. 

DOT observed that combining the four recommended measurements into
one provided a simple, comprehensive measurement.  For illustrative
purposes in DOT's report, each of the recommended measures was
weighted equally at one-fourth.  DOT noted that use of any one
measure alone would not be the most appropriate or accurate
measurement of the states' LOE.  For instance, the equivalent
motor-fuel tax does not address the considerable assistance that the
state and local transportation sector receives from nonuser fees such
as general fund appropriations and other taxes.  Also, equivalent
motor-fuel tax rates do not recognize differences in fiscal capacity
among the states, since a poorer state must impose a relatively
greater economic burden on its highway and mass transit users to
achieve the same motor-fuel equivalent tax rate as a wealthier state. 

DOT concluded that there is no "perfect" single factor to measure
LOE, as the individual factors have strengths and weaknesses. 
Therefore, DOT noted that considering more than one measurement gives
a more accurate picture of state and local LOE and reduces the impact
of year-to-year fluctuations in any single measurement.  As
previously mentioned, DOT found that combining four measures met
certain criteria it had established.  In its report, DOT did not
recommend either for or against measuring LOE as a factor in a
formula for providing assistance to highways or mass transit. 
Rather, DOT stated that if LOE is considered as a formula factor, it
should be considered within the total context of highway and mass
transit financing, investment requirements, goals of the
transportation program, environmental considerations, and other
public policy considerations not explicitly considered in the study. 

While we did not evaluate the reasonableness of DOT's assumptions and
selection of these measurements, we note that better LOE indicators
are possible.  For example, per capita personal income is the measure
of fiscal capacity most commonly used in federal grant formulas and
is a measure DOT recommended.  However, we have found\2 that total
taxable resources, developed by the U.S.  Department of the Treasury,
is a better measure of fiscal capacity than per capita personal
income because it measures all the income produced within a
state--whether received by residents or nonresidents or retained by
business corporations--rather than personal income alone.  The
federal block grant formula for Alcohol and Drug Abuse and Mental
Health Services currently uses total taxable resources. 


--------------------
\1 Report on State Level of Effort, required by section 6013 of
Intermodal Surface Transportation Efficiency Act of 1991 (P.L. 
102-240), Mar.  1994. 

\2 Older Americans Act:  Funding Formula Could Better Reflect State
Needs (GAO/HEHS-94-41, May 12, 1994) and Maternal and Child Health: 
Block Grant Funds Should Be Distributed More Equitably (GAO/HRD-92-5,
Apr.  2, 1992). 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A NEEDS PROXY APPROACH: 
ALTERNATIVE ONE
========================================================== Appendix IV

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions
resulting from a combination of two proxies for needs.  The
alternative distributions assume that each state receives one-half of
its funding on the basis of its total lane miles and one-half of its
funding on the basis of its total vehicle miles traveled. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A NEEDS PROXY APPROACH: 
ALTERNATIVE TWO
=========================================================== Appendix V

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions
resulting from a combination of three proxies for needs.  The
alternative distributions assume that each state receives one-third
of its funding on the basis of its total lane miles, one-third on the
basis of the vehicle miles traveled on its Interstate highways, and
one-third on the basis of its population. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A RETURN-TO-ORIGIN APPROACH: 
ALTERNATIVE ONE
========================================================== Appendix VI

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions
resulting from returning the states' tax contributions to the Highway
Trust Fund back to the state of origin.  Interstate Construction is
excluded from this analysis to adjust for the augmented funding some
states receive through this program.  Fiscal year 1995 is the final
year of funding for the Interstate Construction program. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A RETURN-TO-ORIGIN APPROACH: 
ALTERNATIVE TWO
========================================================= Appendix VII

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions
resulting from returning the states' tax contributions to the Highway
Trust Fund to the state of origin.  Funding for demonstration
projects is excluded from this analysis to adjust for the augmented
funding that the states receive outside of the apportionment process. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain
funds under the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A RETURN TO ORIGIN APPROACH: 
ALTERNATIVE THREE
======================================================== Appendix VIII

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions
resulting from returning the states' tax contributions to the Highway
Trust Fund to the state of origin.  The comparison includes all
funding categories. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A COMBINED APPROACH: 
ALTERNATIVE ONE
========================================================== Appendix IX

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions. 
The alternative formula under consideration blends a needs-proxy
approach and a return-to-origin approach.  The alternative
distributions assume that each state receives 40 percent of its
funding on the basis of its contributions to the Highway Trust Fund,
30 percent on the basis of the lane miles on its National Highway
System highways, and 30 percent on the basis of the vehicle miles
traveled on its Interstate highways. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


ALTERNATIVE FORMULA DISTRIBUTIONS
UNDER A COMBINED APPROACH: 
ALTERNATIVE TWO
=========================================================== Appendix X

The following table compares the states' actual fiscal year 1995
federal-aid highway distributions with alternative distributions. 
The alternative formula under consideration blends a needs-proxy
approach and a return-to-origin approach.  The alternative
distributions assume that each state receives 40 percent of its
funding on the basis of its contributions to the Highway Trust Fund,
20 percent on the basis of the lane miles on its National Highway
System highways, 20 percent on the basis of the vehicle miles
traveled on its Interstate highways, and 20 percent on the basis of
its population. 



   (See figure in printed
   edition.)



   (See figure in printed
   edition.)

Notes:  Shading indicates those states that would gain funds under
the alternative distribution. 

Totals may not add because of rounding. 


FEDERAL FUNDING AS A PERCENTAGE OF
TOTAL HIGHWAY RECEIPTS, BY STATE,
FISCAL YEAR 1994
========================================================== Appendix XI

                         (Dollars in thousands)

                                                               Federal
                                                          contribution
                                                                    as
                               Contribution                 percentage
                               from federal        Total      of total
State                            government     receipts      receipts
-----------------------------  ------------  -----------  ------------
Alabama                             332,310    1,007,101            33
Alaska                              214,957      436,056            49
Arizona                             181,052    1,083,145            17
Arkansas                            225,215      668,242            34
California                        1,340,780    5,111,941            26
Colorado                            274,892      890,138            31
Connecticut                         336,934    1,053,229            32
Delaware                             78,620      309,563            25
Dist. of Columbia                    48,727      289,250            17
Florida                             649,151    2,923,270            22
Georgia                             398,908    1,298,222            31
Hawaii                              226,930      365,242            62
Idaho                               104,962      323,923            32
Illinois                            556,491    2,727,428            20
Indiana                             386,082    1,359,131            28
Iowa                                272,938    1,005,267            27
Kansas                              167,030    1,062,946            16
Kentucky                            228,715    1,399,035            16
Louisiana                           228,248      745,851            31
Maine                                86,650      333,338            26
Maryland                            250,764    1,194,821            21
Massachusetts                       731,936    2,313,017            32
Michigan                            389,334    1,661,472            23
Minnesota                           334,497    1,303,184            26
Mississippi                         225,839      698,567            32
Missouri                            346,634    1,112,755            31
Montana                             160,988      313,632            51
Nebraska                            149,493      558,549            27
Nevada                               97,429      370,178            26
New Hampshire                        94,834      334,175            28
New Jersey                          460,033    2,490,037            18
New Mexico                          181,164      518,644            35
New York                            848,026    4,483,431            19
North Carolina                      383,600    1,690,327            23
North Dakota                        107,996      234,866            46
Ohio                                568,413    2,490,193            23
Oklahoma                            177,402      710,506            25
Oregon                              226,527      851,961            27
Pennsylvania                        746,074    3,085,837            24
Rhode Island                        134,177      313,557            43
South Carolina                      223,026      631,700            35
South Dakota                        124,300      275,850            45
Tennessee                           265,171    1,210,728            22
Texas                             1,011,466    3,383,666            30
Utah                                150,855      404,685            37
Vermont                              97,717      227,338            43
Virginia                            219,948    1,927,064            11
Washington                          377,940    1,454,616            26
West Virginia                       292,444      774,711            38
Wisconsin                           354,736    1,167,722            30
Wyoming                             170,236      269,185            63
Total                            16,242,591   62,849,292            26
----------------------------------------------------------------------
Source:  1993 Highway Statistics, Federal Highway Administration. 


MAJOR CONTRIBUTORS TO THIS REPORT
========================================================= Appendix XII

RESOURCES, COMMUNITY, AND ECONOMIC
DEVELOPMENT DIVISION, WASHINGTON,
D.C. 

Gary Jones, Assistant Director
Charles Barchok, Jr., Assistant Director
Yvonne Pufahl, Project Manager
Laurie S.  Zeitlin, Evaluator-in-Charge
Miriam Roskin, Staff Evaluator
Nancy A.  Boardman, Staff Evaluator
Philip G.  Farah, Senior Economist


*** End of document. ***