CONGRESS OF THE UNITED STATES
CONGRESSiONAl, BUDGET OFFICE
A
CBO
STUDY
innovative financing of highways:
an analysis of proposals
The Congress of the United States
Congressional Budget Office
For sale by the U.S. Government Pt inting Ollke
Suircrintentient of Dcx-uments. Mail Stop: SSOP, Washington, DC 2(M02-i).f28
ISBN 0-16-049413-3
NOTES
Numbers in the text and tables may not add up to totals because of rounding.
Cover photo, which shows toll booths on the Dulles Greenway in Virginia, by David
Galen.
Preface
The federal government provides about $20 billion a year in grants to states for high¬
ways—about one-quarter of the total amount spent on roads each year by all levels of
government. Most of the federal money is raised through taxes on motor fuels. States,
in financing their road-building programs, also rely heavily on motor fuel taxes and on fees
paid by highway users. But revenues from those user taxes and fees are insufficient to build
as many new roadways as transportation officials would like. As a result, they have been
exploring irmovative ways of financing such projects. This study, which was prepared by the
Congressional Budget Office (CBO) in response to a request by the Senate Committee on
Environment and Public Works, reviews several approaches to augment traditional sources of
funding. The analysis covers changes in rules governing federal aid, state infrastructure
banks, federal credit assistance, and private-sector financing of roads.
Elizabeth Pinkston of CBO's Natural Resources and Commerce Division wrote the study
under the supervision of Jan Paul Acton and Elliot Schwartz. Nabeel Alsalam, Clare Doherty,
Peter Fontaine, Douglas Hamilton, Arlene Holen, Pearl Richardson, Robin Seiler, and Paul
Van de Water of CBO all contributed helpful comments. Many members of the staff of the
Federal Highway Administration (FHWA) provided information and explanations of technical
matters. The author is particularly grateful to Max Inman, David Seltzer, Lucinda Eagle,
Bryan Grote, and Larry Dwyer of FHWA, who reviewed drafts and offered valuable advice.
Several state and local transportation officials also made useful suggestions; they include Jon
Bloom, Norman Foster, Adeel Lari, Christopher Mann, Frank Smith, and Constance Sorrell.
In addition, the author wishes to thank J. Richard Aronson, Stewart Butler, Randall Eberts,
David Ehrlich, and David Klinges for their càreful reviews and eomments.
Leah Mazade edited the manuscript, and Melissa Burman provided editorial assistance.
Rae Wiseman typed the final draft. Kathryn Quattrone and Martina Wojak-Piotrow prepared
the study for publication. Laurie Brown prepared the electronic version for CBO's World
Wide Web site.
June E. O'Neill
Director
January 1998
This study and other CBO publications
are available at CBO's Web site.
http://www.cbo.gov/
Contents
ONE
TWO
THREE
FOUR
FIVE
APPENDIXES
A
SUMMARY ix
INTRODUCTION 1
Traditional and Innovative Financing 1
Evaluating Innovative Financing Measures 3
RELAXING RESTRICTIONS ON FEDERAL AID 5
Overview of the Federal-Aid Highway Program 5
FHWA's Innovative Financing Initiatives 12
Evaluating FHWA's Innovative Financing Initiatives 15
Federal-Aid Rules and Innovative Financing Initiatives:
Costs and Trade-Offs 17
STATE INFRASTRUCTURE BANKS 19
The State Infrastructure Bank Program 19
Evaluating State Infrastructure Banks 22
The Outlook for State Infrastructure Banks 26
FEDERAL CREDIT ASSISTANCE 27
Federal Loans, Loan Guarantees, and Lines of Credit 27
Federal Infrastructure Financing Institutions 29
Evaluating Federal Credit Assistance 30
PRIVATE-SECTOR PARTICIPATION 35
Types of Private-Sector Participation 36
Evaluating Private-Sector Participation
in Highway Financing 46
The Outlook for Private-Sector Participation 50
Highway Financing: Sources and Trends 55
B
The Matching-Share Requirement 65
vi INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
TABLES
1. Potential Sources of Financing for Highways 2
2. Funding Authorized by Title I of the 1991 Intermodal
Surface Transportation Efficiency Act, 1992-1997 8
3. Innovative Financing Tools Used by the States Under
the Innovative Financing Initiatives of the Federal
Highway Administration, April 1994 to July 1996 16
4. Sponsors and Features of Highway Financing 37
5. Motivation, Contributions, and Risks of Partners in
Public/Private Ventures in Highway Financing 40
A-1. Types of Revenue Used for Highway Financing, 1994 5 8
A-2. Revenues Flowing to the Highway Trust Fimd, 1995 59
A-3. Disposition of Revenues from State Highway Users, 1995 60
A-4. Revenues Available for Repaying State Highway Bonds 61
A-5. Sources of State Financing for Highways, 1995 62
A-6. Sources of Local Financing for Highways, 1994 63
CONTENTS vii
BOXES
1. Highway User Taxes and the Highway Trust Fund 6
2. Proposais for Devolution 12
3. Types of Financial Assistance SIBs Can Offer 20
4. Federal Infrastructure Financing Institution 31
Summary
Motorists want better, safer, less congested
highways, but the money to build them is
scarce. Transportation officials strain to
juggle limited resources to meet the demands. Elected
representatives eould raise taxes, but voters resist.
How then can transportation plarmers secure funding
for new highway projects? To augment money raised in
traditional ways, highway officials are exploring the use
of innovative financing techniques.
In the past and in large part today, states have fi¬
nanced roads primarily through a combination of state
revenues and federal aid. States have raised their share
of the funds by taxing motor fuels and charging user
fees—for example, for motor vehicle registration and
driver's licenses and, to a lesser extent, tolls. The fed¬
eral program of aid to the states for highways is also
financed through motor fuel taxes and other levies on
highway users. Federal aid for highways is entirely on
a cash basis ifom the Highway Trust Fund. At the state
level, most highway spending is in the form of cash
raised from taxes and user fees that are accumulated in
designated highway accounts. Debt financing consti¬
tutes only about 6 percent of the revenues states use for
highways.
Transportation officials at all levels of government
have recognized in recent years that funding from tradi¬
tional sources is not keeping pace with demands for
new, expanded, or improved highways. As a result,
they have begun to explore new sources of highway
financing. Highway officials use the term "irmovative
financing" to refer broadly to any funding measures
other than the traditional pay-as-you-go approach.
Most of the innovative measures currently under con¬
sideration entail debt financing. Although most finan¬
cial experts would not consider debt fmancing innova¬
tive, the term is used to contrast that approach with tra¬
ditional methods of funding highway projects.
The 1990s have seen significant innovation in high¬
way fmancing. Encouraged by measures that the Con¬
gress adopted in the Intermodal Surface Transportation
Efficiency Act of 1991 (ISTEA), new ideas have been
proposed, discussed, tried on an experimental basis,
and in some cases enacted into law. Most such mea¬
sures are linked closely to an individual project, in
many instances, a toll facility; they do not cover a
state's overall highway program. Although they may
enable state and local governments to get important
projects built sooner than traditional fmancing would
allow, they are unlikely to replace pay-as-you-go user
taxes for the bulk of roadway needs. The reason is that
as long as most roads are toll-free, new toll projects will
be at a disadvantage in competing with them.
To some observers, "irmovative" has a positive
coimotation suggesting a new and better approach. To
others, especially when it is used to deseribe methods of
fmancing, the term raises the specters of gimmickry and
smoke and mirrors. Each side has some justifieation
for its view. Some innovative techniques can work well
to finance specific projects without imposing addi¬
tional, unanticipated, and unwanted eost burdens. But
careful scmtiny of some financing proposals may reveal
that they cost more than they appear to at first glance
and they shift the cost burden in hidden and unfavor¬
able ways.
X INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Innovative financing may enable states to increase
the incomes, wealth, and well-being of their residents
by building highway projects sooner than would be pos¬
sible with traditional financing. Like other investments,
judiciously selected highway projects have the potential
to yield benefits over many years. If the benefits ex¬
ceed the costs of a project, after factoring in the cost of
capital that could have been used for alternative
income-generating ventures, the project is a worthwhile
investment for the community. Borrowing money to
build it enables people to enjoy its benefits sooner.
Borrowing has risks, however, and those risks may
be greater for public projects than for private invest¬
ments. And even if the risks are not greater, they are,
at the very least, different and thus continue to raise
concerns for policymakers. Observers who brand inno¬
vative measures as gimmickry worry that the risks and
costs associated with debt financing may be hidden or
downplayed by enthusiasts who are eager to get a proj¬
ect built. For example, if a public project receives a
subsidy in its borrowing (perhaps by obtaining a below-
market interest rate), the subsidy should be factored
into the cost of the project. Also requiring consider¬
ation are indirect subsidies, such as making the interest
on bonds issued by state and local governments exempt
from federal income taxes. Such subsidies impose
costs on federal taxpayers that may be obscured in ana¬
lyzing the merits of any individual highway project.
In the end, money to pay for highways can come
only from general taxpayers or from users or other ben¬
eficiaries of highways. Innovative financing measures
generally shift the burden of costs from current users of
highways in general—that is, payers of taxes on motor
fuels and other taxes and fees imposed on motorists
—to future general taxpayers, motorists in general, and
users of the specific projects built with innovative fi¬
nancing. Some shifting may also occur hetween tax¬
payers at the state and federal levels.
Innovative financing can affect how efficiently re¬
sources are allocated in both making and using an in¬
vestment. Ventures that must attract financing from
lenders or equity investors face intense scrutiny, which
raises the likelihood that the investment will be a sound
one. It does not, however, ensure a profitable outcome.
Tight controls over costs are often critical to the suc¬
cess of such projects.
To the extent that they impose user fees to repay
debt, projects financed by innovative means have
greater potential for allocating resources efficiently than
do tax-financed, toll-free highways. Tolls that reflect
the marginal social cost of use of a roadway provide
incentives for efficiency. In congested corridors, tolls
that take into account the costs of traffic delays are a
way of allocating use to motorists who place the great¬
est value on avoiding those delays. But highway opera¬
tors do not necessarily set user fees at the efficient
level. If a highway is not congested, the cost of one
additional vehicle—the efficient toll rate—may be so
low that revenues from a toll set at that level will not
cover debt service, operations and maintenance, and
other costs. In those eases, the sponsors of the project
may need to charge tolls that are higher than the eco¬
nomically efficient level to cover their costs.
Although innovative financing is unlikely to replace
the current system of pay-as-you-go financing, it can
augment traditional sources of funding in specific eases
and enable state and local governments to proceed with
major projects sooner than they might otherwise. This
study analyzes several innovative measures that have
been tried or proposed in recent years. It describes the
way they would raise money to accelerate projects and
discusses their potential effects on the allocation and
distribution of resources. It also analyzes the implica¬
tions such approaches have for the federal budget.
Innovative fmaneing measures for highways fall
into several broad categories: relaxing financial restric¬
tions on the use of federal aid, establishing financing
institutions at the state level, providing federal credit
assistance, and tapping private-sector resources for in¬
vestment in highway projects.
Relaxing Restrictions on
Federal Aid
The federal-aid highway program is characterized by a
maze of requirements that states must fulfill as a condi¬
tion of receiving grants. In recent years, the federal
government has removed some long-standing restric¬
tions on highway aid and given states greater latitude in
their use of it. Those reforms, which help states stretch
SUMMARY
xi
the value of federal aid, have provided greater flexibil¬
ity in three major areas: sponsoring toll roads, provid¬
ing financial assistance to transportation projects built
by public/private partnerships, and engaging in debt
financing. The federal government has also modified
the matching-share requirement, a move that states
have found especially helpful. The changes have en¬
abled states to launch highly valued projects more
quickly than would have been possible under the stan¬
dard funding approach. In addition, the reforms have
broadened the set of projects that are eligible for federal
assistance.
The federal budget is unlikely to be much affected
by relaxing financial restrictions on federal highway
aid. Of course, the relaxation might lead to somewhat
faster spendout rates if states drew on every dollar of
aid as soon as it became available instead of amassing
enough money to pay up front for a large, new project.
But even if federal outlays accelerated initially, over
time they would probably resume a steady rate.
State Infrastructure Banks
State infrastructure banks (SIBs) are investment funds
that provide loans or other forms of financial assistance
to public or combined public/private sponsors of trans¬
portation projects. As loans are repaid, the proceeds
can be re-lent to fund additional projects. In the Na¬
tional Highway System Designation Act of 1995, the
Congress established a SIB pilot program and autho¬
rized the Secretary of Transportation to enter into
agreements with up to 10 states to form such banks.
Subsequently, the Congress opened the SIB pilot pro¬
gram to all states. The Department of Transportation
has given definite or conditional approval for SIBs in
38 states and Puerto Rico.
SIBs give states greater flexibility in financing
transportation projects than they have under the stan¬
dard financial regulations of the federal-aid highway
program. Such rules impose constraints on the timing
and use of aid and the types of projects eligible for it.
By providing flexibility, SIBs can help states get proj¬
ects under way sooner. SIBs may also aid states in an¬
other aspect of highway ventures. Besides the financiai
rules of the federal highway program, the government
imposes conditions on federal aid that may increase a
project's cost. For example, a project built with federal
aid must meet the prevailing-wage requirements of the
Davis-Bacon Act. Under current interpretation of the
law, those requirements would apply only to the first
round of projects financed by SIBs and not to projects
financed with recycled funds.
In addition to flexibility in financing, SIBs can of¬
fer credit enhancements to attract private investment.
In making loans rather than grants, SIB financing
makes federal highway aid go farther because loan re¬
payments are available for additional highway projects.
How much additional money SIBs will make available,
and when, depends in large part on how much aid the
federal government provides to help capitalize the
SIBs. It is also a function of how the states respond,
including the way in which they structure their banks.
Using SIB funding increases efficiency in invest¬
ment because it loosens federal constraints on a state's
choice of projects. With fewer restrictions on its deci¬
sions, a state is free to choose projects with the highest
overall economic returns and not just the highest re¬
turns within each category of federal aid (as traditional
financing would require). The use of SIBs may also
enhance efficiency in resource allocation through the
mechanisms chosen for repaying SIB loans—for exam¬
ple, tolls and other user fees.
SIB financing has implications for who bears the
costs of a project. Issuing debt shifts the burden of
costs from current general taxpayers and payers of user
taxes to future taxpayers, users, and other beneficiaries.
Moreover, because debt issued by the SIBs is tax-
exempt, using SIB financing transfers some of the costs
from taxpayers at the state and local levels to those at
the federal level.
Participation in the SIB program is voluntary, an
attractive feature since not all states will find this kind
of tool useful. Some states have projects ready that
could benefit immediately from access to SIB financ¬
ing. Other states are considering whether they have
viable candidates. Still others are restricted by their
own laws or constitutions in their ability to take advan¬
tage of this innovative financing mechanism.
At this point, the effect SIBs would have on the
r deral budget seems to be relatively small. The Con¬
gress provided $150 million in 1997 to capitalize the
xii INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
banks, but it appropriated no additional funds for 1998.
The effect of the potential loss of revenues associated
with the SIBs' tax-exempt debt must also be consid¬
ered. The magnitude of the loss is hard to predict but
appears minor in the near term.
Federal Credit Assistance
The Congress is considering several proposals for pro¬
viding federal loans, loan guarantees, or other credit
assistance to state and local governments for use in
transportation projects. By making money available to
lend or by promising to assist with repayment, those
measures can enhance the quality of the credit that gov¬
ernments obtain to finance a project. They may also
enable the sponsors of a project to borrow at lower in¬
terest rates.
How federal credit assistance affects efficiency, the
distribution of resources, and the federal budget de¬
pends on the specific form that assistance takes. More¬
over, analysis of such effects depends on whether fed¬
eral credit assistance is viewed as a separate program in
addition to the federal-aid highway program or as a
substitute for all or part of it. In general, however, by
leveragmg federal aid, credit enhancements may be able
to generate funding for more projects, compared with
traditional outright grants of the same amount. A draw¬
back to be kept in mind, though, is that a credit pro¬
gram might require additional federal bureaucracy.
Private-Sector Participation
In recent years, private firms have built two toll roads
in the United States and formed partnerships with state
and local governments to consider additional invest¬
ments in transportation projects. Private investors are
motivated by the prospect of profits, although such ex¬
pectations are not always met. Viewing private partici¬
pation as a source of badly needed capital, officials in
some states have welcomed that investment.
The degree of private participation in highway
building varies from project to project. For some ven¬
tures, it means complete private development and oper¬
ation of a roadway; for others, it consists of incentive
contracts m which governments have primary owner¬
ship and responsibility but private firms bear some de¬
gree of risk—and are rewarded for efficiency. Private
investors generally scrutinize highway projects closely
to make sure revenues will cover costs and provide a
reasonable return on their investment. Their care in
that regard enhances the probability that projects built
with private money will use resources efficiently. Effi¬
ciency can be further enhanced by tolls or other user
fees that allocate use—for example, of a congested
highway—to motorists who are willing to pay.
Using private investment to fund highway projects
shifts some or all of the cost and risk from governments
and taxpayers to private investors and users of those
roads. That change in turn reduces demands on govem-
ment funds and conforms to the principle of public fi¬
nance that the beneficiary pays, factors that may create
greater public acceptance of privately financed toll
roads. However, states need to be aware that private
ventures can create new problems for them. For in¬
stance, a project may not generate enough revenues to
cover operations, maintenance, and debt repayment,
and the developers may then ask the state to bail them
out. The agreements that governments enter into with
private firms should make clear the responsibilities of
all parties in the event of such contingencies.
As investors and public officials consider an in¬
creasing number of private or public/private projects,
the most successful ways of fmancing and operating
them are likely to emerge. However, the potential of
private investment to fund a wide variety of highway
activities at this time remains limited. The most prom¬
ising candidates in the near term appear to be toll lanes
that augment congested highways.
Conclusion
Which innovative financing measures are the most suc¬
cessful? The evidence suggests that no one approach is
the magic bullet that would solve all highway fmancing
problems. Rather, different measures work better in
different situations. And of course, as economists say,
there is no free lunch. The money for highway projects
must come from somewhere. To some analysts, obtain¬
ing it from debt backed by tolls or from other fees im-
SUMMARY
Xlll
posed on motorists who use specific roadways is fairer
and more efficient than using conventional tax financ¬
ing. But tolls are not always structured efficiently, and
users of toll roads also pay the same taxes on fuel that
users of toll-free roads pay. Questions remain, more¬
over, about what will happen if a project does not gen¬
erate enough revenues to repay its debt.
Until the 1990s, federal policies generally limited
the ability of states to employ nontraditional financing.
Relaxing restrictions on the use of federal aid has
cleared the way for state and local governments to pur¬
sue new ways of fmancing transportation projects.
Their experience suggests that the federal government
could further the development of innovative financing
tools by according states even greater flexibility in their
use of federal aid. With freedom to experiment, state
and local governments could test various approaches
and adopt the measures that work best in their particu¬
lar circumstances.
Chapter One
Introduction
A perennial problem for transportation officials
is how to fmance highways. Motorists clamor
for better, safer, less congested roads, but gov¬
ernment budgets are constrained and taxpayers balk at
higher taxes. As increases in personal travel and com¬
merce impose growing demands on the nation's high¬
ways, state governments are exploring new approaches
to funding. They are seeking innovative ways to add
capacity—new roads, additional lanes on existing roads,
new or expanded interchanges—sooner than is now
possible with traditional financing.
Traditional and Innovative
Financing
Traditionally, states have financed roads for the most
part through a combination of state resources and fed¬
eral aid. State funds have come from taxes on motor
fuels and user fees, such as fees from motor vehicle
registration and driver's licenses and, to a lesser extent,
tolls. Federal aid is also financed through motor fuel
and other taxes on highway users. At both the federal
and state levels, spending is essentially on a pay-as-
you-go basis. Receipts from user taxes and fees accu¬
mulate, typically in a designated highway account, and
then are spent. Debt financing—issuing bonds to pay
for highway development and construction—is not
widespread, although many states have used it on occa¬
sion. In summary, traditional financing involves:
0 user taxes and fees, often earmarked for transporta¬
tion;
0 federal aid, with matching state funds; and
0 current, pay-as-you-go financing.
Highway officials use the term "innovative financ¬
ing" to refer broadly to any funding measures other than
traditional financing. Although financial experts would
hardly view debt financing as innovative, it comes un¬
der that heading insofar as highways are concerned. In
fact, most innovative financing measures now being
considered entail debt financing.
An additional distinction can be drawn between
traditional and innovative financing. The innovative
measures under consideration tend to be tied closely to
an individual project or set of projects rather than ap¬
plied to a state's overall highway program. They may
enable state and local governments to get important
projects built sooner than traditional financing would
allow, but they are unlikely to replace pay-as-you-go
user taxes for the bulk of roadway needs. A major rea¬
son is that most innovative financing measures involve
tolls as the mechanism for recouping investments. As
long as most roads in the United States are toll-free,
new toll projects will be at a disadvantage in competing
with them.
Some proponents of innovative financing view it as
a lower-cost alternative to traditional financing. As
economists often say, however, there is no free lunch.
Money to pay for highways must come from some¬
where. The potential sources of those funds are general
taxpayers, users or other beneficiaries of highways, and
borrowing paid for by future taxpayers, users, or bene¬
ficiaries (see Table 1).
2 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Under certain circumstances, debt financing can
play a socially beneficial role. If a road project yields
benefits that exceed its costs, then society gains by pro¬
ceeding with the investment. (Of course, such calcula¬
tions must discount costs and factor in the opportunity
cost of borrowed funds and any implicit or explicit sub¬
sidies.) Several concerns have led the federal govern¬
ment and most state governments to shy away from
financing highways with debt and to opt instead to fi¬
nance them primarily on a pay-as-you-go basis.
One concern is that borrowing may reduce fiscal
discipline and open the floodgates for projects of lim¬
ited merit. Estimates of the benefits and costs of any
project are generally uncertain and often subjective.
Without the restraint inherent in a pay-as-you-go ap¬
proach, fiscal conservatives worry that proponents of a
project will overestimate benefits and underestimate
costs. Program advocates take the opposite position
and worry that benefits will be undervalued and costs
overstated.
Another concem to states is their creditworthiness.
With too much outstanding debt, they may encounter
difficulties in the credit markets. States may have trou¬
ble borrowing what they need because investors will be
uncertain about repayment and require higher interest
rates in compensation. Therefore, most states have
statutory or constitutional limits on the amount of debt
they may incur.
Given those concerns, governments have implicitly
opted for a trade-off: they will risk underspending, thus
forgoing earlier returns on investments, to avoid wast-
Tablel.
Potential Sources of Financing for Highways
Where Would the Money Come From?
Who Might Pay?^ If Paid Now if Paid Later
Generai Taxpayers
Users of Roads and Highways in
Generai
Users of Specific Highways
Property Owners
Developers and Businesses that
Benefit from Specific Highway im¬
provements
Income taxes (federal and state)
General sales taxes (state and
local)
Property and other taxes (local)
Motor fuel taxes (federal and state)
Excise taxes (federal)
Vehicle registration taxes and fees
(federal and state)
Operator's licenses (state)
Tolls (including charges for conges¬
tion)
Development impact fees (special
assessments)
Development impact fees (special
assessments)
Debt backed by general taxes paid
by future taxpayers
Debt backed by taxes on future us¬
ers of highways (with or without fur¬
ther backing by general tax reve¬
nues)
Debt backed by toll revenues (with
or without further backing by tax rev¬
enues)
Property taxes and special assess¬
ments (in the future)
Property taxes and special assess¬
ments (in the future)
SOURCE: Congressional Budget Office.
a. The initial payer might be able to pass the cost on to another person or entity.
CHAPTER ONE
INTRODUCTION 3
ing money on projects for which benefits had been in¬
correctly estimated to exceed eosts. Another force
pushing them in that direction is the increasingly fre¬
quent laek of consensus about the value of new roads.
Bitter disputes between developers and environmental¬
ists indieate that what some people view as a worthy
and needed investment is to others an unconscionably
costly plundering of the environment.
Innovative financing cannot solve all of the prob¬
lems associated with transportation projects. What it
can do is to help some projects get under way sooner
than would be possible with traditional financing.
Suppose a state wants to build a new highway, such
as a limited-access roadway segment connecting two
major, existing highways. Under the traditional ap¬
proach, a state would list the segment on its transporta¬
tion improvement plan along with other projects it
wants to build. Depending on the priority of that proj¬
ect compared with others and depending on the project's
cost, it might remain on the list for several years until
enough funding was available from federal aid and state
sources to begin work.
The state might be able to start the project some¬
what sooner if it used a funding strategy different from
the pay-as-you-go method. For example, the state
could take advantage of initiatives by the Federal High¬
way Administration (FHWA) to allow states greater
flexibility in using federal aid.' Or it could obtain
funds sooner by issuing bonds backed by taxes, tolls, or
other dedicated revenues. The state could get credit
assistance m marketing the bonds from a state infra¬
structure bank or the federal government. (Currently,
federal credit assistance is limited to a few specific pro¬
jects, but the Clinton Administration and the Senate
have proposed broader federal credit programs.)
The state could also try to tap resources of the pri¬
vate sector to build the road. Public/private ventures
can take several forms, with varying degrees and types
of private capital at risk. At one extreme, the state
could simply rely on a private firm, which would use
the debt and equity it raised by itself, to finance the en¬
tire road. At the other extreme—and not strictly inno¬
vative, since this approach has been used to some ex¬
tent in the past—the state could issue bonds that were
backed solely by revenues from the project and not by
taxes. Buyers of those so-called nonrecourse bonds
take the risk that the project will not generate enough
revenue to repay the debt, and they recognize that they
will have no reeourse to the state's taxing powers in that
event. In comparison, bonds backed by taxes also tap
private-sector resources: bond buyers purchase them
with their private funds. But the risk is different from
that of nonrecourse bonds because the bond buyers ex¬
pect the government to raise taxes, if necessary, to re¬
pay the debt. That understanding changes the nature of
the risk associated with the private capital.
Evaluating Innovative
Financing Measures
How do the various innovative financing measures
compare with one another and with traditional financ¬
ing? Overall, the innovative mechanisms are more sim¬
ilar to each other than they are to traditional financing.
The primary motivation for innovative financing is
to fill the gap between demands for new projects and
the funding available from traditional sources of reve¬
nue. As a consequence, all innovative measures make
funding available sooner. Some do so by borrowing,
some by attracting private equity investment, some by
advancing federal aid, and some by a combination of
those mechanisms. Credit assistance from the federal
government or the state helps support some of the debt-
financing measures.
Innovative financing measures can tap a variety of
funding sources to repay the debt incurred building a
highway and to provide a return on equity investment.
Some projects use dedicated taxes, some use tolls, and
some use other fees imposed on beneficiaries of the
work. The differences between those sources of funds
result in different effects on economic efficiency, the
distribution of the cost burden of the project, and the
federal budget.
1. The Federal Highway Administration is the agency within the Depart¬
ment of Transportation that oversees the federal-aid highway program.
4 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Innovative measures that entail user fees can pro¬
mote efficiency in two ways.^ First, a project that must
repay debt and provide a reasonable rate of return on
investment requires a certain level of revenues from
tolls or other user fees or from dedicated taxes. Such a
requirement implicitly subjects the project to a market
test, letting it stand or fall on its merits. Knowing that
m advance gives transportation planners an incentive to
build projects that have the greatest expected returns
and not squander funds on oversized or gold-plated
projects.
Second, projects financed with user fees have the
potential to allocate scarce resources to their most
highly valued use. Such an allocation occurs when the
price that consumers are willing and able to pay equals
the marginal cost of the resources used in making the
product or service available.^ Sponsors of a project will
not always set user fees to equal the marginal cost,
however. To collect enough to service the debt incurred
in building the highway, they might set user fees at a
level that would maximize total revenue rather than
efficiency. That choice might entail fees that were
higher or lower than the level that would induce effi¬
ciency.
Innovative financing measures may have effects on
the distribution of the cost burden that differ not only
from traditional financing but also from each other.
Financing projects by issuing debt shifts the burden to
luture users or taxpayers, which at first glance may ap¬
pear unfair. Yet it is not necessarily a bad idea. For
long-lived investments, it accords with the economic
2. Economic efficiency means allocating scarce resources to uses that are
of the highest value to society as a whole.
3. If coasumers are willing to pay a price that is greater than the marginal
cost for a good, their willingness signals that they value the good more
than the resources that go into providing it. Therefore, more of the
good should be produced. If the price consumers are willing to pay is
less than the marginal cost, their behavior indicates that they place less
value on the good than on the resources used to produce it and those
resources could produce greater value if used elsewhere.
principle that those who use and benefit from an invest¬
ment should pay for it. Still, debt financing has draw¬
backs—for example, if the revenues projected for a new
venture turn out to be unrealistic and the project de¬
faults on the debt. In that ease, someone must pay, and
the prime candidates are bondholders and taxpayers.
Innovations in the rules governing federal aid to
states would also have distributional consequences.
Relaxing requirements concerning the way states must
match federal aid or changing provisions intended to
ensure that states act accountably would reduce the fed¬
eral government's control over resources. Eliminating
requirements intended to achieve social objectives in
the areas of safety, the environment, and labor would
help highways but at a cost to those goals.
Always of major concern with any proposed
changes in federal aid is their effect on the federal bud¬
get. Some innovative measures for highway financing
make demands—or establish potential claims—on the
budget, whereas others appear to be neutral. Some
proposals would create new federal programs and pro¬
vide funds in addition to those authorized under the
federal-aid program. Others would not increase federal
spending but would give states greater flexibility in
their use of federal aid. Still others would lead to
greater use of tax-exempt financing and thus reduce
federal tax revenues.
To sum up, no single innovative financing mecha¬
nism stands out from the rest. Projects differ from one
another, as do their prospective sponsors. Some states
have been more receptive than others to debt financing,
toll roads, and user fees. Of course, states do not need
to adopt such measures if doing so is impracticable be¬
cause the innovative financing measures under consid¬
eration are all voluntary. The measures discussed in
this study illustrate the possibilities and limits of alter¬
native financing mechanisms. In certain circumstanees,
those mechanisms can augment existing funding, but
they are not likely to generate enough funds to replace
traditional sources of revenue.
Chapter Two
Relaxing Restrictions on Federal Aid
The federal government currently provides about
$20 billion a year in grants to states for high¬
ways. That amount is about one-quarter of the
annual spending for roads by all levels of government.
(Appendix A provides an overview of highway funding
at the federal, state, and local levels.) The Congress is
debating measures that would increase federal aid to
between $25 billion and $30 billion a year. Also under
consideration is whether to modify some of the rules
governing states' use of that aid.
The federal-aid highway program involves a maze
of requirements that states must navigate if they are to
receive aid. In recent years, some long-standing restric¬
tions have been relaxed, initially through administrative
action by the Federal Highway Administration and later
through legislation ratifying some of FHWA's initia¬
tives. Removing restrictions on federal aid can stretch
its value to the states by giving them greater flexibility
without the cost of new procedures, institutions, and
bureaucracies.
Overview of the Federal-Aid
Highway Program
Since 1916, the Congress has authorized—and reau¬
thorized—federal aid to states for highways in legisla¬
tion passed every few years. The legislation sets forth
rules governing how states can use that aid. During the
1990s, the Congress has broadened the set of financing
options open to the states.
The federal government owns and operates very
few highways—mainly roads on Indian reservations, in
national parks, and on other federal lands. Neverthe¬
less, it influences the ways in which state and local gov¬
ernments spend their highway funds. And that influ¬
ence applies not only to the $20 billion annually in fed¬
eral aid for highways but also, to a lesser extent, to the
$60 billion or so a year that state and local governments
raise from their own sources.
The federal government has a long history of pro¬
viding funds to build roads. An early rationale was that
roads were an important means of bringing the nation
together, linkmg interior agricultural markets with man¬
ufacturing and trade at ports. By the middle of the 20th
century, the nation's defense also loomed large m fed¬
eral policies toward highways. The Federal-Aid High¬
way Act of 1956, which created the National System of
Interstate and Defense Highways (the Interstate Sys¬
tem), authorized a 41,000-mile system of highways to
promote commerce, provide for more expeditious
movement of militaiy supplies and personnel, and ex¬
pand and improve travel opportunities for citizens.'
The 1956 act also created the federal Highway Trust
Fund, increased the federal tax on motor fuels to 3
cents a gallon, and ordered the government to deposit
revenues from the motor fuel and other federal taxes on
highway users into the trust fund (see Box 1).
1. Federal-Aid Highway Act of 1956, 70 Stat. 378. In an earlier mea¬
sure (section 7 of the Federal-Aid Highway Act of 1944, 58 Stat. 842)
passed during World War II, the Congress outlined plans for building
a system of interstate highways once the war was over.
6 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
The 1956 highway act authorized federal aid for a
period of three years. Since then, the Congress has
reauthorized the highway program several times, gener¬
ally for a period of four to six years. The most recent
multiyear authorization expired at the end of September
1997. In November 1997, the Congress passed a six-
month stop-gap measure while it considers proposals
for reauthorization.
Box 1.
Highway User Taxes and the Highway Trust Fund
In 1956, the nation launched a massive effort to build
the Interstate System. Although federal taxes on gaso¬
line and diesel fuel predated that activity, the new high¬
way initiative required substantial increases in revenues.
Thus, in that same year, the Congress established the
federal Highway Trust Fund as a way of providing a
steady, predictable source of funding for highways. It
also raised tax rates on gasoline and diesel fuel to 3
cents a gallon from the previous rate of 2 cents. Re¬
ceipts from the taxes were earmarked for the trust fund
so that the Interstate program would have a dedicated
stream of funding.
Federal taxes on gasoline subsequently increased to
4 cents a gallon in 1959, 9 cents in 1983, 9.1 cents in
1987, 14.1 cents in 1990, and 18.4 cents in 1993. On
December 31, 1995, the rate dropped to 18.3 cents a
gallon with the expiration of the 0.1 -cent-per-gallon tax
designated for the Leaking Underground Storage Tank
Trust Fund. Tax rates on diesel fuel were the same as
the gasoline rates until 1983, when they were raised to
15 cents a gallon. In the years following, they have ex¬
ceeded gasoline tax rates by 6 cents a gallon. The dif¬
ference is intended to reflect the greater wear and tear
on roads from heavy trucks that use diesel fuel.
For almost 30 years, all revenues from federal taxes
on lliel went to the Highway Trust Fund to finance high¬
ways. But in 1983, provisions in the Surface Transpor¬
tation Assistance Act of 1982 came into effect requiring
that a penny per gallon in fuel taxes be set aside for
mass transit. To accommodate that requirement, a tran¬
sit account was established within the Highway Trust
Fund. On December 1, 1990, the amount set aside for
mass transit increased to 1.5 cents a gallon.
Another wedge between taxes on fuels and highway
financing took shape with the passage of the Omnibus
Budget Reconciliation Act of 1990 (OBRA-90).
OBRA-90 not only raised the fuel tax rates by 5 cents a
gallon but also for the first time directed revenues from
those taxes into the general fund of the Treasury. Thus,
2.5 cents of the increase went into the general fund, and
2.5 cents went to the Highway Trust Fund. Of the
amount going to the trust fund, 2 cents flowed into the
highway account, and 0.5 cents went to the mass transit
account. On October 1, 1995, the 2.5-cent portion of
the OBRA-90 tax that had gone to the general fund was
redirected to the Highway Trust Fund (2 cents for the
highway account and 0.5 cents for the transit account).
The emphasis in the Congress on reducing the fed¬
eral deficit brought yet another wedge on October 1,
1993. Fuel taxes increased by 4.3 cents a gallon, and all
the revenues from that increase went into the general
fund for deficit reduction. The budget reconciliation bill
of 1997 reversed the pattern, however, and redirected
revenues from that portion of the fuel tax to the High¬
way Trust Fund. The increase was divided in such a
way that 3.45 cents now goes to the highway account
and 0.85 cents goes to the transit account.
Thus, of the 18.3-cent-per-gallon federal tax on
gasoline in place as of October 1997, 15.45 cents goes
to the highway account, 2.85 cents to the transit account,
and nothing to the general fund. Of the 24.3-cent-per-
gallon tax on diesel fuel, 21.45 cents goes to the high¬
way account, 2.85 cents to the transit account, and noth¬
ing to the general fund. '
Diverting revenues from highway users to the tran¬
sit account and the general fund has weakened the "user-
pays" principle on which the Highway Trust Fund was
established. Such diversions have also contributed to
debates over whether the Highway Trust Fund should be
subject to controls on federal spending and whether its
balances should be included in computing the federal
deficit or surplus.
1. Federal Highway Administration, Highway Statistics 1994
(October 1995), Tables FE-IOIA and FE-2I, pp. IV-20 and
rV-22. Gasohol and other fuels made with ethanol and meth¬
anol are taxed at rates that range between 11 cents and 18.3
cents a gallon; the splits between the highway and transit ac¬
counts are similar to those for gasoline and diesel fuel.
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 7
Several factors determine the amount of federal aid
apportioned to each state. ^ Commonly referred to as
formulas, such factors include the number of lane-miles
and vehicle-miles traveled on the Interstate System and
local air quality. However, no simple mathematical
expression determines the amount a state will receive.
Several additional provisions of the law attempt to
achieve what the Congress considers to be an equitable
distribution of federal funds among the states.^
Rules Governing the Use of Federal Aid
Once federal funds are apportioned to the states, the
states are subject to federal laws and regulations re¬
garding their distribution. Provisions of the federal-aid
program are codified in title 23 of the U.S. Code. It
contains about 60 sections that establish the program,
spell out the policy objectives, and specify the rules for
distributing federal aid. The rules tell the states what
kinds of highway projects are eligible for different cate¬
gories of federal aid, and they dictate requirements for
states in matching the aid. They also impose require¬
ments that are intended to advance social objectives
unrelated to transportation such as environmental and
employment goals. Of course, states do not have to
accept federal money for highways and the restrictions
that come with it. But the amount of federal aid is so
large that it would be difficult for a state to decline.
Categories of Federal Aid. The federal-aid program
specifies how much federal funding is available for var¬
ious categories of surfaee transportation. The Inter-
modal Surface Transportation Efficiency Act of 1991
(ISTEA) authorizes spending on eight major highway
programs and numerous smaller programs. In addition,
it contains several "equity adjustment categories" that
are intended by the Congress to distribute funding equi-
2. The U.S. Code makes distinctions between apportionments and alloca¬
tions. For a technical description, see Federal Highway Administra¬
tion, Financing Federal-Aid Highways, FHWA-PL-92-016 (May
1992), pp. 13-14.
3. In 1997, the distribution of funding was the focus of controversy as the
Congress considered measures to reauthorize the highway program.
The measures include distributing funds according to how much states
have paid in fiiel taxes, according to need, and according to the level of
effort put forth by each state to finance its highways.
tably among the states. The legislation speeifies how
the funding provided under those categories can be
spent.''
The eight major programs under ISTEA are Inter¬
state constmction, Interstate substitute, Interstate main¬
tenance, the National Highway System (NHS), the sur¬
face transportation program (STP), congestion mitiga¬
tion and air quality improvement, bridge replacement
and rehabilitation, and federal lands highways (see
Table 2). States are allowed to transfer some funds
between categories under certain conditions. For exam¬
ple, they may transfer up to 50 percent of NHS funds to
the STP (they can transfer even more with the approval
of the Secretary of Transportation), or they may trans¬
fer up to 40 percent of bridge funds to either the NHS
or STP (or both).^ A state may also shift any amount of
Interstate maintenance funds to the NHS or STP as
long as it certifies that it is adequately maintaining its
highways on the Interstate System.®
The federal government also sets rules for how
states may spend funds within various categories of aid.
The surface transportation program, which includes
most major state and local highways as well as mass
transit capital projects, is subject to numerous funding
constraints:
0 Fifty percent of STP funds must be divided be¬
tween urbanized areas with populations above
200,000 and other areas of the state, in proportion
to their relative share of the state's population.
o Thirty percent can be used in any area of the state.'
o Ten percent must be used for safety construction.^
4. Federal Highway Administration, Financing Federal-Aid Highways,
pp. 14-17.
5. 23 U.S.C. 104(c) and 23 U.S.C. 144(g). The transferred funds are not
subject to allocation requirements specified at 23 U.S.C. 133(d).
6. 23 U.S.C. 119(f).
7. 23 U.S.C. 321(d). There are exceptions for states in which more than
80 percent of the population lives in a metropolitan area, for states in
which the federal government owns more than 80 percent of the land,
and for areas with less than 5,000 people.
8. 23 U.S.C. 133(d)(1).
8 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Table 2.
Funding Authorized by Title I of the 1991
Intermodal Surface Transportation
Efficiency Act, 1992-1997
As a
In Billions Percentage
Category of Dollars of Total
Major Programs
Interstate construction 7.20 6.0
Interstate substitute 0.96 0.8
Interstate maintenance 17.00 14.1
National Highway System 21.00 17.4
Surface transportation 23.90 19.8
Congestion mitigation and air
quality improvement 6.00 5.0
Bridge replacement and
rehabilitation 16.10 13.3
Federal lands highways 2.60 2.2
Equity Adjustments'
Donor state bonus 3.00 2.5
Interstate reimbursement 4.00 3.3
Hold harmless 3.64 3.0
90 Percent of payments
guarantee 0.42 0.3
Minimum allocation'' 5.18 4.3
Other
Earmarked projects 6.23 5.2
Other surface transportation 3.59 3.0
Total 120.81 100.0
SOURCE: Federal Highway Administration, Financing Federai-Aid
Highways (May 1992), pp. 35-36.
a. The Congress's Intent in adding these categories was to ensure
equitable distribution of funds among the states. All funds from
these categories except those under minimum allocation flow to
the surface transportation program.
b. Funds from this category may be used for any major program
except Interstate maintenance and federal lands highways. They
may also be used for hazard elimination and rail/highway cross¬
ings.
O Ten percent may be used only for "transportation
enhancements," such as pedestrian and bicycle fa¬
cilities, beautification, scenic highways, and his¬
toric preservation.®
0 Up to 1/16 of 1 percent of SIP funds may be used
for education and training of state and local high¬
way department employees, but those funds can
pay no more than 80 percent of the cost of tuition
and direct educational expenses.'"
Other categories of aid are also constrained by
rules. At least 15 percent but not more than 35 percent
of the bridge apportionment must be used for bridge
projects that are not on a federal-aid road." The Secre¬
tary of Transportation may waive the 15 percent re¬
quirement, however, if he or she determines that the
state does not need such bridge work. For states with
Indian reservations, at least 1 percent of the bridge ap¬
portionment must go for bridges on reservation roads.
Two percent of the funds apportioned to states for
the major categories of aid can be used only for plan¬
ning and research activities, and one-quarter of that 2
percent must be used for research, development, and
technology transfer. The state may claim an exception
to the requirement if it certifies that expenditures for
transportation planning will require more than 75 per¬
cent of the earmarked amount.'^
Recognizing the complex nature of the federal-aid
program, the Federal Highway Administration has is¬
sued a 355-page handbook that describes the types of
programs and projects that are eligible for federal aid
and the terms and conditions that apply.
In addition to categorical aid, ISTEA authorized
$6.2 billion for 539 demonstration projects.'" Some of
those projects appeared on lists of priorities that the
states maintain. Some, however, were clearly of lower
priority to the states than other projects for which they
were seeking funds.
10. 23 U.S.C. 321(b).
11. 23 U.S.C. 144(g)(3).
12. Federal Highway Administration, Financing Federal-Aid Highways^
pp. 14-15.
13. Federal Highway Administration, Federal-Aid Program Branch, A
Guide to Federal-Aid Programs, Projects, and Other Uses of High¬
way Funds, FHWA-PD-92-018 (September 1992).
9. 23 U.S.C. 133(dX2).
14.
Federal Highway Mxmmstiaúon, Financing Federal-Aid Highways,
pp. 14 and 35.
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 9
Matching Shares. In general, the federal-aid highway
program requires states to mateh federal funding. The
federal share for most types of projects is 80 percent.
For Interstate maintenance, the federal share is gener¬
ally 90 percent, but for additional lanes that are neither
designated as auxiliary nor reserved for high-oecupancy
vehicles, the federal share is 80 percent.'® The federal
share can be larger if a substantial percentage of a
state's area is Indian lands or public domain. The fed¬
eral share may also be increased for certain emergency
relief or safety projects."
Many of the innovative financing mechanisms put
forward by the Department of Transportation adminis¬
tratively or enacted in ISTEA or the National Highway
System Designation Act of 1995 (the NHS Act) relax
rules governing how states can provide their share of
highway fmancing.
Federal-Aid Rules to Promote Social Objectives.
The federal-aid highway program also contains some
conditions for receiving aid that are intended to pro¬
mote social objectives. A full review is beyond the
scope of this study. Nevertheless, a description of sev¬
eral past and present conditions attached to such aid
shows how aspects of the program can affect the aid's
value, either by imposing rules on private individuals or
firms or by making highway construction more expen¬
sive than it might otherwise be.
For example, as a result of the 1974 Arab oil em¬
bargo and subsequent shortage of motor fuels, the Con¬
gress decreed that states wanting to receive federal
highway aid had to impose a maximum speed limit of
55 miles per hour. The initial objective of the condition
was to reduce fuel consumption. However, when stud¬
ies showed that a significant decline in automobile fa¬
talities was also associated with the lower speed limit,
pressure developed to keep the lower limit for safety
reasons, even after the flow of oil resumed. Later,
states were permitted to raise the speed limit to 65
miles per hour on rural highways on the Interstate Sys¬
tem. ISTEA brought additional relaxation to the 55-
mile-per-hour limit; the act allowed states to set speed
15. 23 U.S.C. 120(b).
16. 23 U.S.C. 120(a).
17. 23 U.S.C. 120.
limits of 65 miles per hour on some non-Interstate high¬
ways. In 1995, with the NHS Act, the Congress elimi¬
nated the federal policy on speed limits after much de¬
bate over the appropriate role of the federal govern¬
ment.'^ The Congress also removed a number of other
requirements, such as laws that motorcyclists wear hel¬
mets, that had been imposed on states as conditions of
federal highway aid.
The NHS Act added a provision requiring action by
the states in regard to drinking and operating motor
vehicles. Begimiing in 1998, states must have enacted
and be enforcing a law that considers an individual un¬
der the age of 21 who has a blood alcohol concentration
of 0.02 percent or greater while operating a motor vehi¬
cle to be driving while intoxicated or driving under the
influenee of alcohol.'® If the state is not fulfilling those
conditions, part of its apportionments for the National
Highway System, surface transportation program, and
Interstate maintenance will be withheld. Beginning in
1996, the Secretary of Transportation was required to
withhold 10 percent of a state's apportionments for
those programs if the state did not enact and enforce
legislation to revoke or suspend the driver's licenses of
people convicted of drug offenses.^"
ISTEA requires that at least 10 percent of federal
aid for highway and transit projects "be expended with
small business concerns owned and controlled by so¬
cially and economically disadvantaged individuals."^'
That requirement may mean that states will pay more
for some aspects of their programs than they would in
the absence of the provision. In addition, each state
must annually survey and compile a list of such small
business concerns. It must then notify the Secretary of
Transportation of the percentage of firms that are con¬
trolled by women and the percentages controlled by
women or by men who are considered socially and eco¬
nomically disadvantaged.^^
18. Section 205, 109 Stat. 577.
19. Section 320, 23 U.S.C. 161, 109 Stat. 589. Five percent of the funds
are to be withheld in 1998 and 10 percent in subsequent years.
20. 23 U.S.C. 159.
21. Section 1003(h), 105 Stat. 1919.
22. Section 1003(h)(3), 105 Stat. 1920.
10 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Contractors and subcontractors who perform work
on highway projects built with federal aid are subject to
another such requirement that may increase the cost of
their operations. Even if they are able to hire workers
for less, they must pay laborers and mechanics at least
the prevailing wage for that kind of work in that local¬
ity. The prevailing wage is determined by the Secre¬
tary of Labor according to provisions of the Davis-Ba¬
con Act.^'^
The federal government can also use the federal
highway program to achieve environmental goals. Fed¬
eral aid for highways can be withheld if states do not
meet standards set forth in the Clean Air Act, as
amended.^'
The present system causes some friction between
state and federal governments and among the state gov¬
ernments as eaeh state seeks to receive as much federal
aid as possible and to have the authority to spend those
funds according to its own priorities. The total amount
of federal aid is limited and essentially fixed after
reauthorization. As a result, the states are engaged in a
zero-sum game: if one state gets more federal money,
another gets less.^® That reality has led to the search for
other methods of highway financing.
Legislation Promoting Innovative
Financing Initiatives
By the late 1980s, transportation planners had started
to search for nontraditional means of financing new
projects. The Federal Highway Administration began
to explore ways of giving states greater flexibility in
using debt and toll financing and strategies for encour¬
aging investment by the private sector. It relaxed some
rules on an experimental basis and suggested legislation
to adopt the most promising reforms. Subsequently,
the Congress passed ISTEA and the NHS Act, which
23. 23U.S.C. 113.
24. The Davis-Bacon Act is codified at 40 U.S.C. 276a.
25. 42 U.S.C. 7509(b)(1), 104 SUt. 2420.
26. Federal flmdingfor highways has increased in the 1990s as many other
programs within the domestic discretionary part of the budget have
shrunk. However, pressures to increase federal highway funding are
counterbalanced by pressures to continue to reduce domestic discre¬
tionary spending.
ratified FHWA's initiatives and also eased other restric¬
tions on the use of federal aid. Those changes have
helped states to use their federal aid more productively.
The Intermodal Surface Transportation Efficiency
Act of 1991. ISTEA brought significant changes in
federal policies toward toll roads. It allowed federal-aid
funds to be used on any toll facility owned by a public
entity and on approved private facilities.^' Tolls no
longer had to be removed once a road was paid for (see
below), and any money left over after providing proper
maintenance could be used for other highways.
The toll provisions of ISTEA marked a major break
with tradition. Beginning in 1916, the federal govern¬
ment had maintained a policy that required roads built
with federal aid to be free of tolls.In 1956, the Con¬
gress reinforced the policy by decreeing that highways
on the Interstate System would be toll-free. Over the
years, some exceptions were carved out—for existing
toll roads that became part of the Interstate System, for
bridges and tunnels on the system, and for cases in
which states paid back the federal aid they had received
for a highway. But in general, until ISTEA, the federal
government had discouraged states from developing toll
roads.
ISTEA also broadened the set of toll projects eligi¬
ble for federal aid. Included under the new provisions
were most construction and major maintenance except
for highways on the Interstate System.^' (Major main¬
tenance comprises the so-called 4R projects: recon¬
struction, resurfacing, restoration, and rehabilitation.)
ISTEA authorized states to use federal aid to pay from
50 percent to 80 percent of the cost of the toll projects,
and it permitted states to lend the federal share of a pro¬
ject's cost to a public or private entity to build a toll
facility. The latter provision expanded the opportuni¬
ties for states to engage in debt financing and cleared
the way for greater investment by private firms in high¬
way projects.
A further change brought about by the legislation's
toll provisions was in the rules goveming the state's
27. Section 1012, 23 U.S.C. 129, 105 Stat. 1936.
28. For a bnef overview of federal policies toward toll roads, see Congres¬
sional Budget Office, Toll Roads: A Review of Recent Experience,
GBO Memorandum (February 1997).
29. Section 1012, 23 U.S.C. 129, 105 Stat. 1936.
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 11
matching share. ISTEA permitted states to count to¬
ward their matching-share requirement "toll revenues
that are generated and used by public, quasi-public, and
private agencies to build, improve, or maintain high¬
ways, bridges, or tuimels that serve the public purpose
of interstate commerce."^" With the "toll credit" provi¬
sion, the legislation made toll roads more attractive to
states and encouraged them to form partnerships with
the private sector.
The National Highway System Designation Act of
1995. The NHS Act contained several innovative fi¬
nancing provisions. It continued along the trail blazed
by ISTEA, making toll roads more attractive to states
by raising the federal share for toll facilities to 80 per¬
cent.^' It also authorized a pilot program of state infra¬
structure banks (SIBs) for making loans and providing
other financial assistance for highway and mass transit
projects. Funding from certain categories of federal aid
could be used to capitalize the banks; the act permitted
states to deposit up to 10 percent of those funds. The
NHS Act did not provide any new funding for capitali¬
zation, but the following year, the Congress appropri¬
ated an additional $150 million to be distributed to
states for the SIB program.
Several other loan-related provisions of the NHS
Act gave states an incentive to consider debt financing.
The act provided greater latitude for states in making
loans for projects, setting the terms of those loans, and
usmg the repaid funds. And it codified states' authority
to use federal aid to pay for costs related to the issuing
of bonds—^not only the bond principal but other charges
such as interest and bond insurance.
The NHS Act also contained provisions that rati¬
fied several financing measures that the Federal High¬
way Administration had instituted on an experimental
basis. Those measures relaxed rules governing the way
states met the requirement to match federal aid with
state funding.
30. Section 1044, 23 U.S.C. 120, 105 Stat. 1994.
31. Section 313(a), 23 U.S.C. 129(a)(5), 109 Stat. 585.
32. Section 311, 23 U.S.C. 122, 109 Stat. 583.
Legislative Proposals in
the 105th Congress
As the 105th Congress considers reauthorizing the
highway and mass transit programs, innovative financ¬
ing has played at best a secondary role. Indeed, the ma¬
jority of the recent debate has centered on the formulas
by which federal aid is doled out to the states. States
that have sent more money to Washington than they
have received in aid have pressed for adjustments that
would return more of their money to them. Another
contentious issue has been the growing balances in the
federal Highway Trust Fund and the restrictions on
spending those funds that are being used to meet bal¬
anced budget targets.
Yet despite its relegation to the sidelines during the
reauthorization process, innovative financing has not
been unaffected by the debate. The major issues that
the Congress has been considering relate to innovative
financing in at least two ways. First, they suggest that
traditional fundmg continues to be limited and that state
and local officials will have to be creative in making the
best use of the available funds. Second, they call into
question the appropriate role of the federal government
in providing aid to states for highways. (See Box 2 for
an overview of proposals for devolution.)
In the area of innovative financing, two legislative
proposals would create federal programs offering credit
assistance. The Senate's bill to reauthorize ISTEA
(S. 1173) would establish a pilot program of federal
credit assistance for large surface transportation proj¬
ects." The program would provide up to $530 million
in federal budget authority to support as much as $10
billion in credit assistance. Projects eligible for assis¬
tance would be those costing more than $ 100 million or
usmg more than half of a state's armual federal aid,
whichever was less. Senator Chafee has also intro¬
duced legislation (S. 275) that would make tax-exempt
bond financing available for up to 15 highway projects
nationwide to be developed by the private sector.
The Clinton Administration has prepared its own
proposal for helping states fund their highway pro¬
grams. The Administration's plan would expand the
33. Title I, Subtitle C, Chapter 2.
12 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Box 2.
Proposals for Devolution
Some policymakers have proposed eliminating most of
the federal highway program and returning the job of
funding highway construction and maintenance to the
states. Bills introduced by Representative Kasich (H.R.
1470) and Senator Mack (S. 667), which together make
up the Transportation Empowerment Act, would retain
only a limited federal role that would focus on programs
with a strong national interest. To that end, the pro¬
posed legislation would cut federal taxes on motor fuels
from their current rates of 18.3 cents a gallon for gaso¬
line and 24.3 cents for diesel fuel to 6.3 cents a gallon
by 2002. Only 2 cents would go to the Highway Trust
Fund.
Although states might welcome having greater con¬
trol over their highway programs, they may take a dim¬
mer view of their new responsibility for providing most
of the funds to pay for them. Even if states raised their
own taxes on gasoline by only as much as (or by less
than) federal taxes were reduced, elected officials would
still be vulnerable to the unwelcome charge that they
had increased taxes. Some proposals address that con¬
cern by retaining the federal tax but refunding to each
state approximately the amount of revenue estimated to
have been collected from it. Measures of that type were
proposed by Senator Warner and Representative DeLay
in the ISTEA Integrity Restoration Act (S. 335 and H.R.
674, also known as the STEP 21 proposals).
In essence, those proposals would make the federal
government the tax-collecting agent but substantially
reduce its existing redistributive function. Why does the
federal government redistribute money among the states
for highway purposes? The principal economic justifi¬
cation for returning more money to some states and less
to others is the presence of spillover benefits. If the
highways of one state are used disproportionately by
residents of other states (and if those users buy a dispro¬
portionate amount of fuel in other states and thus do not
contribute much to the first state's tax revenues), an ar¬
gument can be made that lunds should be redistributed.
Another justification relies on the contributions
highways make to economic development. If federal
policy favors helping poorer regions of the nation by
stimulating economic development, one way of advanc¬
ing toward that objective will be to build highways.
Highway programs create construction jobs and eventu¬
ally spur industry and commerce by reducing transporta¬
tion costs. However, building highways is a bltmter in¬
strument for redistributing wealth than are health or
welfare policies that target aid to individuals with low
incomes. As a result, highway projects are likely to be a
more expensive way of aiding poor people. The stron¬
ger argument for federal redistribution of highway funds
is the one of spillover benefits.
program of state infrastructure banks to all states by
providing $150 million annually to assist state and lo¬
cal governments in capitalizing the banks.It has also
proposed a new federal program of credit enhancements
for publie or private sponsors of publicly owued proj¬
ects. The program would finance projects of national
significance by awarding grants to project sponsors for
establishing revenue stabilization funds to back project
debt. The projeets would use tolls or other user charges
or dedicated sources of revenue for repayment.^^ The
Administration's plan proposes that obligations secured
by the fund not be considered federally guaranteed un-
34. Section 1022 of the National Economic Crossroads Transportation
Efficiency Act of 1997, or NEXTEA (H.R. 1268 and S. 468).
der the tax code. Therefore, they could be set up as
either taxable or tax-exempt debt.^®
FHWA's Innovative
Financing Initiatives
With the passage of the Intermodal Surface Transporta¬
tion Efficiency Act of 1991, which authorized greater
flexibility in financing, FHWA was free to consider
new initiatives in funding for the nation's highways. In
response to a 1994 executive order promoting more
35. Title V of NEXTEA
36. Section 5005 of NEXTEA
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 13
effective investment in infrastructure, FHWA estab¬
lished a program known as the Innovative Financing—
Test and Evaluation Project, or the TE-045 program^'
Its goals were to;
0 speed construction;
o create incentives for states to use federal financing
to its fullest;
0 assist states in leveraging their funds to produce
more investments; and
o obtain and disseminate information about creative
techniques used by the states.^®
The National Highway System Designation Act of
1995 subsequently ratified most of the TE-045 initia¬
tives, which can be divided into two categories: those
that enable projects to get under way sooner than they
would under traditional financing and those that pro¬
vide greater leverage of federal aid.
Getting Projects Under Way Sooner:
Cash Flow Tools
In the past, rules relating to the timing of federal aid
and the state's matching share have sometimes eon-
strained states from proceeding with highway projects
as quickly as they might have liked. For a highway pro¬
ject to receive federal aid, a state had to set aside the
full amount of federal and state funds needed before
beginning the work. That restriction often presented
problems for projects that took large shares of a state's
federal aid and its own highway funds. Typically, such
projects required that the state put funds aside for sev¬
eral years until it had accumulated enough to proceed,
an approach that could crowd out most other projects.
FHWA developed several TE-045 initiatives to over-
37. The TE-045 work was initiated under 23 U.S.C. 307(a), which gives
FHWA authority to engage in research on highway financing as well
as on many other issues related to the highway program. For a more
detailed description, see Federal Highway Administration, Rebuilding
America: Partnership for Investment, Innovative Financing Hand¬
book, Test and Evaluation 045 {TE-045) (October 1995). The execu¬
tive order is no. 12893, Principles for Federal Infirastructure Invest¬
ments, January 26, 1994.
38. Federal Highway Administration, "Innovative Finance and Statewide
Financial Planning" (unofficial briefing book, 1996), Unit 2, p. 5.
come those obstacles and allow states to begin projects
sooner than they otherwise could. Those cash flow
tools are known as advance construction, partial con¬
version of advance construction, phased funding, and
tapered match.
Advance Construction Beyond the Authorization
Period. In some instances, the innovative financing
initiatives that the TE-045 program comprised ex¬
panded on techniques that FHWA was already using on
a limited basis. Advance construction is a case in point.
This cash flow tool allows states to finance highway
projects with their own money (cash or debt) and obtain
federal reimbursement later, which is useful if states
can tap their own sources of funds to provide money for
pressing projects but have exhausted the federal obliga-
tional authority that is available. Advance construction
is also helpful for a state that must borrow funds. With
such a mechanism in place, the state may be able to
borrow under more favorable terms because lenders can
be assured that future federal funding will be used to
help repay the loans.
In the years before the TE-045 program, reaching
the end of the authorization period (the federal-aid
highway program has usually been authorized for four
to six years at a time) effectively cut off aid for new
projects until reauthorization occurred.^® As a result,
states could preserve eligibility for federal aid only on
new projects that could be "converted" to draw on that
aid before the federal authorization expired. That con¬
dition made it difficult to start large, multiyear projects
toward the end of an authorization period, given the
uncertainties about when reauthorization might occur
and the lack of any commitment that more federal aid
would be forthcoming. Section 308 of the NHS Act
codified the funding mechanism developed under the
TE-045 program and modified advance-construction
provisions by allowing the Secretary of Transportation
to approve projects for federal aid beyond the authori¬
zation period. (The Secretary's approval was not a
guarantee of funding but rather a declaration that if aid
was available later, it could be applied to that project.)
The one proviso was that the project had to appear on
the state's approved transportation plan.''®
39. ISTEA authorized the program for the 1992-1997 period.
40. 23 U.S.C. 115(d), 109 Stat. 582.
14 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Partial Conversion of Advance Construction. Par¬
tial conversion of advance construction helps a state's
cash flow by allowing it to obtain federal reimburse¬
ment for advance-construction projects in stages. That
IS, states can obtain reimbursement over several years
rather than waiting until the project is completed and
ready to be "converted" to federal aid. The benefit of
partial conversion is that it helps free up funds for other
projects.
Phased Funding. FHWA has used phased funding in
the past to accomplish essentially the same objective as
partial conversion of advance construction. Both tech¬
niques involve a federal commitment of aid for a proj¬
ect before the full amount of federal funding is avail¬
able to be obligated. However, phased funding posed a
problem because it essentially required states to make
prospective commitments of federal funds that had not
yet been authorized. Highway officials therefore have
dropped the use of phased funding in favor of partial
conversion.'"
Tapered Match. With tapered funding, the federal
match varies by the stage of a projeet. It enables states
to draw on federal funding to finance the entire cost of a
project's development stage, for example, and then take
a lower amount of federal funding for later stages.
Overall, the project would still achieve an 80/20 fed¬
eral/state split. Tapered funding, then, helps a state's
cash flow by providing federal aid sooner than would
be possible under traditional financing and at the stage
of the project when it is most needed.
Leveraging Federal Dollars
The second category of TE-045 innovative financing
tools attempts to make more money available by lever¬
aging state and local funds. The options it offers are
more flexible matching requirements, toll investment
credits under section 1044 of ISTEA, loans under sec¬
tion 1012 of ISTEA, and debt financing.
41. For a more detailed discussion, see Miriam Roskin, Ann Sowder, and
JoAnne Carter, An Evaluation of the TE-045 innovative Finance
Research Initiative (prepared for the Federal Highway Administra¬
tion, November 1996), p. 26.
Flexible Match. Before FHWA's TE-045 initiatives, a
state's matching share (usually 20 percent) of a project
had to come from its own or local funds. The one ex¬
ception was donations of private property, which could
be counted toward the share under certain conditions.
The more flexible matching requirements of the TE-
045 program allow states to use the value of public or
private donations to meet part or all of their matching
share. By augmenting state sources of fimding in that
way, a project can qualify for more federal aid per dol¬
lar of state money.'^^ Section 322 of the NHS Act codi¬
fies the change; it explicitly allows states to credit to¬
ward their share of a project any donated funds or the
fair market value of any donated materials or services
that are incorporated into a project.''^
Section 1044 Toll Investment Credits. Section 1044
of ISTEA allowed states to count certain highway
spending that was funded by revenues from tolls toward
the matchmg share required by the federal highway pro¬
gram. But to qualify, a state had to meet a mainte-
nance-of-effort (MOE) test: its spending on highways
in the prior year had to equal or exceed the average of
expenditures for the three previous years. The TE-045
program relaxed the MOE requirement by allowing
states to use toll credits as part of their matching share
as long as they met the test prospectively. That is,
states could show that their anticipated expenditures for
the current year would be at least as large as the aver¬
age of the three previous years. In addition, the TE-045
program altered the disposition of credits earned in pre¬
vious years so that now they no longer lapse.
FHWA has implemented toll investment credits
through administrative action."'' However, the major
reauthorization proposals of the first session of the
105th Congress—S. 1173, H.R. 2400, and the Admin¬
istration's bill—would codify that change."' The lever¬
aging potential of toll credits can be realized only if
42. The state as a whole is still constrained by the total amount of federal
aid available each year.
43. Section 322, 23 U.S.C. 323(c), 109 Stat. 591.
44. Roskin, Sowder, and Carter, An Evaluation of the TE-045 Initiative,
p. 36.
45. The proposed changes are in section 1112 of S. 1173, section 120 of
H.R. 2400, and section 1025 ofS. 468.
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 15
using them prompts more toll projects than would oth¬
erwise be the case. Regardless of the leveraging effect,
however, toll credits offer cash flow benefits because
they reduce or eliminate the need for a state to use cur¬
rent cash to match federal funds.
Section 1012 Loans. Section 1012 of ISTEA permit¬
ted states to lend their federal-aid funds to toll projects
(after obtaining approval from FHWA)."® The federal
government still exercised control over such transac¬
tions, however, because under ISTEA, interest rates
were set by federal regulation. The NHS Act gives
states the flexibility to negotiate interest rates and other
terms of the loans and to offer loans to projects with
dedicated sources of revenue that do not include tolls.
As revenues from those projects repay the loans, states
may use the funds to make grants or loans for addi¬
tional projects without categorical restriction.
Bond Financing. Before the TE-045 initiatives, states
could use federal aid to repay the principal of bonds but
not to cover other bond-related costs (except for some
limited interest costs). The NHS Act ratified the ad¬
ministrative changes FHWA had made in relation to
bonds by allowing states to use federal aid to pay for
interest, issuance costs, and insurance, as well as the
bond principal, on eligible projects.''®
T0 sum up, most of the innovative financing tech¬
niques discussed in this section are now available to all
of the states. The NHS Act ratified most of them;
FHWA put the toll investment credits and partial con¬
version of advance construction in place through ad-
mimstrative action. The agency has dropped consider¬
ation of phased funding since partial conversion accom¬
plishes the same objective. The tapered match is still
under consideration.
46. These are also known as Section 129 loans because section 1012 of
ISTEA amended 23 U.S.C. 129.
47. Section 313, 23 U.S.C. 129(aX7), 109 Stat. 585.
48. Section 311,23 U.S.C. 122, 109 Stat. 583.
Evaluating FHWA's
Innovative Financing
Initiatives
The TE-045 innovative financing initiatives were in¬
tended to make more funds available sooner for high¬
way construction. Experience to date with the initia¬
tives is limited, which is only to be expected, given the
long time horizon necessary for planning and building
highway projects. Nevertheless, the record shows that
many states have availed themselves of the opportuni¬
ties the initiatives have presented.
Financing Potential
In what sense do the TE-045 initiatives make more
funds available to the states? They work primarily by
making funding available sooner than it would be with
the use of traditional methods of financing. But they
also have the potential to tap sources of funds that oth¬
erwise might not have been used for highways. In addi¬
tion, they give states greater flexibility in their selection
of which projects to finance first.
Making money available sooner can enhance a
state's resources in two ways. First, it can increase a
state's buying power when construction costs are esca¬
lating but federal aid remains fixed in nominal terms.
Second, it can generate more income and wealth for the
community through investment in highway projects.
The latter result assumes that states use the added flexi¬
bility to build the projects with the greatest returns on
investment first.
The cash flow tools that the TE-045 program pro¬
vides make federal aid to the states available sooner.
The leveraging tools make more money available
sooner through the use of debt financing. They also
attempt to generate more money in the aggregate by
attracting private equity investment.
A recent report commissioned by FHWA found
that the initiatives had generated a net increase of $ 1.15
billion in current and expected investment. Of that to¬
tal, $593 million was from private sources in the form
16 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Table 3.
Innovative Financing Tools Used by
the States Under the Innovative Financing
Initiatives of the Federal Highway Administration,
April 1994 to July 1996
As a
Financing Tool
Number of
Projects'
Percentage of
All Projects
Flexible Match
28
39.4
Advance Construction
15
21.1
Partial Conversion of Advance
Construction
14
19.7
Phased Funding
9
12.7
Tapered Match
5
7.0
Section 1012 Loans
5
7.0
Section 1044 Toll Credits
3
4.2
Bond Reimbursement
2
2.8
Other
2
2.8
SOURCE: Miriam Roskin, Ann Sowder, and JoAnne Carter, An
Evaluation of the TE-045 Innovative Finance Research
initiative (report prepared for the Fédérai Highway Ad¬
ministration, November 1996), p. 11.
a. A totai of 71 projects used at least one innovative financing tool;
some projects employed more than one.
of bond proceeds and equity contributions."® Those
funds provided about one-quarter of the financing for
the 71 projects that were part of the TE-045 program
(see Table 3). Most of the "new" funding was for two
projects: a toll project in California that raised $482
million in bond proceeds and equity contributions, and
the President George Bush Turnpike in Texas, whieh
was funded by $534 million in bond proceeds, interest
earnings, and a cash contribution from the Texas Turn¬
pike Authority. The FHWA report estimated that 60 of
the 71 projects had been accelerated; for the 43 projeets
for which the acceleration could be quantified, the aver¬
age acceleration was 2.2 years.^°
The tools that were most effective in advancing the
availability of funds were the flexible match, the Sec¬
tion 1044 toll credits, and the Section 1012 loans. As
noted earlier, the flexible match, which was used in 28
49. Roskin, Sowder, and Carter, An Evaluation of the TE-045 Initiative,
pp. 15-16.
50. Ibid., p. iii.
of the projects, enables states to draw on private and
sub state sources of funds (for example, local govern¬
ments or turnpike authorities) to meet the state share of
federal-aid projects. Using toll credits and revenues as
part of the match frees up state money, which may then
be spent on additional highway projects or other state
priorities. The tools discussed here do not generate
additional federal aid, but they may speed receipt of
that aid. They also free up state resources for use in
highway projects that do not qualify for federal aid.
Efficiency
The innovative financing mechanisms developed
through the TE-045 initiatives may contribute to more
efficient use of resources in several ways. First, by
merging several categories of federal aid into one pool,
states can select the highway projects they value most
without being constrained by the amount of federal aid
available for each individual category. That is, if a
state has reached the limit of federal funds in one cate¬
gory but still has a project in the category that would
yield greater net benefits than any project in any other
category, it has the flexibility to fimd the high-return
project.
Second, the irmovative mechanisms indirectly help
to ensure that the highway projects a state builds are
worthwhile. To attract private funds—both debt and
equity—projects must demonstrate the ability to repay
the investment. That requirement provides incentives
to build the right projects and to ensure that they are the
right size.
A project financed with user fees may or may not
stimulate greater efficiency in the use of that resource
than a project on which user fees are not imposed. To
achieve allocative efficiency, the price paid by the user
must reflect the marginal social cost of the resources
assoeiated with that use. On an imcongested highway,
the cost of one additional automobile may be quite
low—lower than the toll that would have to be charged
to cover debt repayment and other costs of building and
operating the road.^' On congested corridors, however.
51. There are efficient ways of covering total costs when marginal costs
are less than average costs. Some of them are discussed in Congres¬
sional Budget Office, Paying for Highways, Airways, and Water¬
ways: How Can Users Be Charged'!' (May 1992).
CHAPTER TWO
RELAXING RESTRICTIONS ON FEDERAL AID 17
tolls that reflect the cost of traffic delays (or, put an¬
other way, the value to users of avoiding delays) are
likely to generate enough revenue to cover the cost of
the road while also promoting efficiency in the use of
the roadway.
Administrative Feasibility
Thirty-nine states entered projects in the TE-045 inno¬
vative financing program. Because the program was
voluntary and no additional federal funding was avail¬
able for projects within the program, the large number
of participants suggests that the states found the inno¬
vative mechanisms both feasible and attractive. One
factor contributing to ease of use was that the initiatives
were simply a relaxation of rules with which state high¬
way officials were already quite familiar. No new bu¬
reaucracy or apparatus was needed to use them. The
mechanisms were attractive because they made more
financing available. The tools that were most popular
in terms of the number of projects on which they were
used were the cash flow mechanisms (see Table 3). In
terms of dollar amount, the two largest projects made
extensive use of leveraging.
The TE-045 program has generated interest in and
support for innovative financing mechanisms. The Na¬
tional Highway System Designation Act of 1995 under¬
scored that support by enacting into law most elements
of the program. The tools give states additional flexi¬
bility in meeting their highway needs. Not all states
will find the tools useful; those that do can pick and
choose from among the various options. The important
point is that use of these innovative mechanisms is
strictly voluntary.
Implications for the Federal Budget
What effects do the innovative tools have on federal
outlays? The annual amount of federal aid is fixed.^^
Greater use of TE-045 mechanisms could lead to a
somewhat faster spendout of the aid if states drew on
every federal dollar as soon as it became available in¬
stead of having to build up a large enough account to
52. The amount of contract authority is provided in legislation authorizing
the federal-aid highway program. Annual appropnation legislation
sets a limit on the amount that can be obligated.
begin a new project. But states are adept at juggling
projects to make the most of their federal aid. Conse¬
quently, the innovative tools may have less of an effect
on the flow of federal funding than on the timing of
individual projects. Even if the spendout rate of the aid
increased initially as states undertook projects with in¬
novative financing, eventually, federal outlays would
probably resume flowing at a steady rate.
Distributional Effects
The distributional effects of FHWA's innovative fund¬
ing mechanisms vary, depending on the type of project
being built and its financing. The cash flow tools basi¬
cally enable states to get federal aid for a project sooner
than they otherwise would. If that acceleration leads to
a faster spendout of federal funds, then taxpayers at the
federal level will bear a greater cost than before. Proj¬
ects that borrow from private lenders and repay them
with revenues from user fees shift the cost burden of
the project from current taxpayers to future users. Proj¬
ects that repay lenders with dedicated tax revenues shift
the burden to future taxpayers. Projects that use private
equity investment and nonrecourse bonds run the risk
that revenues will not be sufficient to repay the debt
and thus provide a sufficient return to investors.
Federal-Aid Rules and
Innovative Financing
Initiatives: Costs and
Trade-Offs
All of the innovative financing mechanisms discussed
here relax restrictions that the federal-aid program im¬
poses. Ironically, although the federal government pro¬
vides substantial funding for highways, the conditions it
imposes diminish the value of the funding to the states.
Those conditions are the basis of impediments that
have motivated the innovative financing initiatives. A
eentral element of the federal-aid highway program—
and a key condition at the heart of the TE-045 innova¬
tive financing reforms—is the matchmg-share require¬
ment. Not only the flexible match but all of the cash
flow tools (advance construction beyond the current
period, partial conversion of advance construction.
18 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
phased funding, and tapered match) and the Section
1044 toll investment credits represent attempts to re¬
duce the burden of the matching-share requirement (see
Appendix B for more discussion of the requirement).^^
From the point of view of state and local govern¬
ments, the conditions associated with federal aid can be
costly and can keep those governments from achieving
their objectives efficiently. But the federal government
may have good reasons for attaching strings to its aid.
One reason is accountability, to ensure that government
funds are spent properly and not squandered. Account¬
ability may have been a motivating factor in requiring a
state to set aside funding for a project before beginning
it. Without such assurances, so the reasoning may have
gone, a state might start a large project, run into finan¬
cial difficulties in completing it, and seek a federal bail¬
out.
Another purpose of conditions of aid is to promote
federal goals. The amount of funding for the various
categories of federal highway aid reflects federal priori¬
ties. In addition, some rules reflect Congressional inter¬
est in furthering environmental, labor, or other social
objectives. Such requirements may raise the cost of
building highways. For example, depending on condi¬
tions in the local labor market, the requirement that la¬
bor on federal-aid highway projects must be paid the
prevailing wage according to the Davis-Bacon Act may
cause labor costs to be higher than if a minimum level
had not been set.
Other federal constraints are directed at transporta¬
tion policy and have proved quite contentious. For ex¬
ample, highway user groups have complained about the
requirement that 10 pereent of the funds in the surfaee
transportation program must be used for transportation
enhancements, such as beautification, historic preserva¬
tion, and bieycle facilities. They argue that highway
53. The other TE-045 tools, Section 1012 loans for toll projects and cover¬
age of bond financing costs, relate to eligibility for federal aid.
funds should be spent on highways, not enhancements.
Supporters of enhancements reply that some expendi¬
tures are needed to redress past "sins"—such as destruc¬
tion of communities and the envirorunent—associated
with highway construction. They would also argue the
need to ensure a balance among the preferences of mo¬
torists, bicyclists, pedestrians, environmentalists, his¬
toric preservationists, and other interest groups.
Thus, meeting federal goals must be balanced
against the varying needs, conditions, and preferences
of the states. On the one hand, the federal government
may impose regulations as a way to achieve what it
considers worthwhile social goals and financial ac¬
countability. On the other hand, if those rules are
overly specific, they may impose higher costs on states
and their taxpayers and prevent states from using then-
highway funds most effectively. Balaneing competing
objectives has been a recurring theme in legislation con¬
cerning the federal-aid highway program. For example,
in the 1991 reauthorization, ISTEA provided additional
flexibility in some aspeets of the program, such as com¬
bining multiple categories of projects into the surface
transportation program and allowing greater use of toll
finaneing. However, in other areas, such as the trans¬
portation enhancements, it imposed more requirements.
During 1997, some proposals called for eliminating
or substantially reducing the federal-aid highway pro¬
gram and turning highway policies—and financing—
back to the states. Although a full critique of proposals
for devolution is well beyond the scope of this study,
locating responsibility for highways at the state and
local levels is the direction in which many of the inno¬
vative financing proposals are headed. Such proposals
give state and local officials greater authority and flexi¬
bility in funding their programs. But devolution eomes
with the drawback that states must raise all the money
they need for highways from their own sourees, without
federal aid. Faced with that proposition, state and local
officials may prefer to trade flexibility in financing for
federal assistance.
Chapter Three
State Infrastructure Banks
Around the same time that the Federal Highway
Administration was developing innovative
fmaneing initiatives under the Innovative Fi
nancing—Test and Evaluation Project, it was also
studying various organizational structures that might
facilitate debt financing. One result of those efforts
was a proposal for a program of state infrastructure
banks that would provide loans or other credit assis¬
tance for transportation projects. FHWA modeled the
program in part after state revolving loan funds for
wastewater treatment facilities.' Although such banks
or funds can take many different forms, they are gener¬
ally established at the state level with capitalization
from federal and state funding. They lend money to
municipalities or other entities for the purpose of build¬
ing infrastrueture.^
The Department of Transportation (DOT) defines a
SIB as "an infrastructure investment fund established to
facilitate and encourage investment in eligible transpor¬
tation infrastrueture projects sponsored by public
and/or private entities."^ Thus, a SIB is a financial in-
1. The program of federal assistance to states for construction of
wastewater treatment plants was authorized in the Clean Water Act.
The Water Quality Act of 1987 established the program in which the
federal government provides grants to capitalize state revolving funds.
The funds are designed to lend money to communities to build facili¬
ties to treat wastewater according to the requirements of the act.
2. This study uses the terms "bank" and "revolving fund" interchange¬
ably, although some academics prefer to use "revolving fund" when
the repayments of debt are lent out again and "bank" when the pro¬
ceeds are used simply to retire debt. The "revolving fund" terminology
is more common for entities that finance wastewater treatment facili¬
ties, following the terminology in the Clean Water Act that provided
for them. "State infrastructure bank" is the term more commonly em¬
ployed for transportation, following its use in the National Highway
System Designation Act of 1995.
termediary established by a state or group of states to
help finance transportation projects. SIBs can provide
financial assistance through loans and credit enhanee-
ment. Credit enhancement generally refers to some
form of guarantee that strengthens the quality of the
debt used to fmance transportation projects. It typically
includes such measures as bond insurance, loan guaran¬
tees, capital reserves, letters of credit, and lines of
credit (see Box 3).
The State Infrastructure
Bank Program
The SIB program has moved quickly from inception to
implementation. The President proposed the program
in his budget submission for 1996 and as part of sev¬
eral other bills. The Congress subsequently authorized
a pilot program for SIBs as part of the National High¬
way System Designation Act of 1995.'' The act autho¬
rized the Secretary of Transportation to enter into
agreements with up to 10 states for establishing state
and multistate infrastructure banks.
The 1997 appropriation act for transportation ex¬
panded the SIB pilot program. It authorized the Secre¬
tary of Transportation to select additional states (be¬
yond the original 10) to participate in the program. The
act also appropriated the first "new money," $150 mil¬
lion, to capitalize the initial pilot SIBs as well as any
new ones. States may also choose to contribute to a
SIB by using funding authorized by the Intermodal Sur-
3. Department of Transportation, "Participation in the State Infrastruc¬
ture Bank Pilot Pro^am," Federal Register, vol. 60, no. 249 (Decem¬
ber 28, 1995), p. 67159.
4. Section 350, 109 Stat. 618-622.
20 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Box 3.
Types of Financial Assistance SIBs Can Offer
Section 350(1)(3) of the National Highway System
Designation Act of 1995 authorizes SIBs to:
0
provide credit enhancements;
0
serve as a capital reserve for bond or debt in¬
strument financing;
0
subsidize interest rates;
0
ensure the issuanee of letters of credit and
credit instruments;
o
finance purchase and lease agreements with
respect to transit projects;
0
provide bond or debt financing instrument se¬
curity; and
0
provide other forms of debt financing and
methods of leveraging funds that are approved
by the Secretary [of Transportation] and that
relate to the project with respect to which such
assistance is being provided.'
The Secretary has approved the following "other
forms" of assistance:
o
lease guarantees for highway and transit capi¬
tal projects;
o
certificates of participation;
0
letters of credit;
0
lines of credit;
o
grant anticipation notes; and
o
standby guarantees.
1.
109 Stat. 622.
face Transportation Efficiency Act of 1991.^ The $150
million appropriation is from the general fund, not the
Highway Trust Fund.
The Administration has continued its support for
this type of organizational structure. The President's
5. States may contribute up to 10 percent of binds they receive for most
categories of federal aid.
budget for 1998 requested an additional $150 million
for capitalizing the SlBs. His proposal for reauthoriz¬
ing the highway program, the National Economic
Crossroads Transportation Efficiency Act (NEXTEA),
would continue funding for the SIB program at the
same levels through 2003. The Congress did not in¬
clude additional funds for SlBs in the Department of
Transportation's 1998 appropriation. Instead, it left the
issue for consideration in authorizing legislation.
Ground Rules
Shortly after the NHS Act was passed, the Department
of Transportation issued a notice in the Federal Regis¬
ter soliciting proposals by states to participate in the
SIB pilot program.® DOT encouraged states to be cre¬
ative in their applications; it gave them flexibility in
determining the structure of the banks, the types of as¬
sistance the banks could provide, their sources of fund¬
ing, and other features. However, the department laid
down rules for several aspects of the program. For ex¬
ample, states can capitalize infrastructure banks with
up to 10 percent of their federal funding from specified
categories under the federal-aid highway and mass tran¬
sit programs.' But they must contribute enough of their
own money to match 25 percent of the federal funding
(to yield the customary federal/state arrangement of an
80/20 cost split).
DOT also established certain restrictions on the use
of SIB funds. Some restrictions apply only to the initial
assistance that the SIB provides. The NHS Act prohib¬
its federal funds that are contributed to the SIB from
being used as grants. Instead, at least in the first round,
the SIB must use its money only for loans or credit en¬
hancements. As loans are repaid, however, the SIB can
use the proceeds to make grants, although the spirit of
the program is to keep recycling the funds into more
projects through the use of loans. In the initial round of
assistance, fimding must be used for projects that are
6. Department of Transportation, "Participation in the State Infrastruc¬
ture Bank Pilot Program," pp. 67159-67160.
7. Under the SIB pilot program, states may deposit up to 10 percent of
their 1996 and 1997 apportionments and allocations for most program
categories of federal highway funds. The categories of highway binds
eligible for use in SIB capitalization are National Highway System,
surface transportation program. Interstate maintenance, bridge replace¬
ment and rehabilitation, minimum allocation. Interstate reimburse¬
ment, hold harmless, 90 percent payment adjustments, and donor state
bonus (see Table 2 on page 8).
CHAPTER THREE
STATE INFRASTRUCTURE BANKS 21
eligible under SIB rules. In later rounds, funds that
have come to the SIB from the highway account can be
used for any project allowed under the overall federal-
aid highway program. (The same restrictions and flexi¬
bility apply to mass transit funds and projects.)
States' Responses to the Pilot Program
Fifteen states met the March 1996 deadline for submit¬
ting applications to participate in the initial phase of the
SIB pilot program. Of those, DOT selected eight in the
first round: Arizona, Florida, Ohio, Oklahoma, Ore¬
gon, South Carolina, Texas, and Virginia. For the other
applicants—California, Massachusetts, Michigan, Min¬
nesota, Missouri, Tennessee, and Washington—DOT
provided advice on how to revise their submissions to
improve their chance of being selected for one of the
two remaining slots. All of the states that were rejected
in the first round resubmitted their revised applications
or asked that their original submissions be reconsid¬
ered. DOT subsequently selected California and Mis¬
souri as part of the initial group.
The NHS Act authorized the Secretary of Trans¬
portation to enter into cooperative agreements with
each approved SIB state specifying the terms of the
SlB's funding and operation. In recent months, DOT
staff have provided technical assistance to states in de¬
veloping those agreements. Ohio and Oregon were the
first states to sign their agreements, and DOT sent
Ohio's form to the other states as a guide. By July
1997, DOT had signed cooperative agreements with all
10 pilot states.
In response to the 1997 appropriation act, which
opened the SIB program to all states, DOT solicited
additional applications. By December 1996, DOT had
received 26 additional proposals covering 27 states and
Puerto Rico. (There were two multistate proposals, and
a 28th state later joined a multistate agreement.) Thus,
by May 1997, a total of 38 states plus Puerto Rico had
sought approval for SlBs. In June 1997, DOT an¬
nounced its approval of all the proposals and allocated
the $150 million of new money appropriated for 1997
among the 39 participants in the SIB program.^
Department of Transportation, "29 Additional States Are Approved to
Participate in State Infrastructure Bank (SIB) Pilot Program" (press
release, DOT 93-97, June 19, 1997). Four of the 38 states received
only conditional approval from IX)T until they had passed legislation
The Initíal Set of Applications
The initial set of applications contained proposals for a
variety of projects, methods of funding, and sources of
repayment. However, the projects that the states pro¬
posed in their applications are not necessarily the ones
they will eventually carry out. There has already been a
great deal of flux—states have dropped some projects
and added others. Consequently, the descriptions that
follow should be considered illustrations rather than a
catalog of SIB projects.
Types of Projects. Twenty-five of the 32 projects pro¬
posed for SIB financing through 1998 are highway
projects.' Two proposals involve mass transit facili¬
ties—a van-pool leasing program in Oregon and reha¬
bilitation of a passenger rail bridge in Cleveland. Sev¬
eral projects are classified as multimodal: two parking
facilities, a facility for transferring freight between rail
and truck modes, and an at-grade highway/rail crossing.
Most of the highway projects involve new construc¬
tion, but a few involve major reconstruction. The new
construction projects include bypasses, cormectors,
beltway segments, and interchanges. Several road-
widening and realignment projects are also part of the
group.
Sources of Repayment. Most of the projects that are
being financed by SlBs will use dedicated tax revenues
to repay loans.'" A few will use special assessments.
Four plan to use tolls, and four will impose other fees
associated with the use of the project, such as fees for
parking.
Leveraging. One key matter each state has to decide is
whether to leverage SIB funding. One way of doing so
is to issue bonds backed by SIB capital. Issuing bonds
increases the amount of funding available in the near
term and is referred to as leveraging because each initial
allowing them to establish a SIB. The 1997 appropriation act directed
DOT to refrain from distributing the $150 million before April 1997,
180 days after the legislation was enacted. The Congress gave the
Secretary of Transportation the authority to decide how to distribute
the funds among the SIBs.
9. Department of Transportation, Evaluation of the U.S. Department of
Transportation State Infrastructure Bank Pilot Program: Status as
of February 28. ¡997 (June 1997). As of June 1997, two states were
still evaluating candidate projects.
10. Ibid.
22 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
dollar of SIB funding is used to back several dollars of
debt. In the absence of leveraging, only the initial
amount of funding (plus any later infusions of addi¬
tional aid) is available to lend. As initial loans are re¬
paid, the repayments (including interest) become avail¬
able to be lent again. A second means of leveraging
SIB capital is to offer credit enhancements, such as loan
guarantees, which enable the sponsors of projects to
borrow money at lower interest rates.
As of September 1997, Arizona, Missouri, and
Ohio had expressed interest in leveraging SIB funds by
issuing bonds. The other states simply intend to make
loans from their initial capital and, as the loans are re¬
paid, to make additional loans with the repaid principal
and interest. California plans to initiate a loan guaran¬
tee program under which its SIB will agree to guarantee
up to 25 percent of the debt used to finance a project.
Evaluating State
Infrastructure Banks
The principal goal of the SIB program is to speed up
the availability of funds for transportation projects.
SIBs accomplish that objective in part by attracting
additional funds from the private sector that otherwise
would have been invested elsewhere. Using those funds
for transportation purposes affects the allocation and
distribution of resources and has implications for the
federal budget. Establishing and operating a SIB also
place demands on the resources of state governments.
Financing Potential
How much additional money SIBs can make available
for highway and transit projects, and how soon, de¬
pends in large part on three main factors: the amount
and nature of federal aid; the way states respond, in¬
cluding how they structure their SIBs; and the success
SIBs have in attracting private capital.
Additional Federal Aid. As previously mentioned, the
Congress did not provide any additional federal aid
when it established the SIB pilot program in 1995; it
simply permitted states to deposit up to 10 percent of
most categories of their existing federal aid into SIBs.
In making appropriations for 1997, the Congress gave
DOT an additional $150 million to award to SIBs. Al¬
though that sum was small in relation to total appropri¬
ations for highways (about $20 billion that year), the
hope was that it would be enough to induce more states
to apply to participate in the SIB program. No addi¬
tional appropriation was provided for 1998. DOT ex¬
pects that by the end of 1998, states will have allocated
$324 million in federal aid to their SIBs to help fund
$1.6 billion worth of projects "
Opportunities Resulting from Fewer Restrictions on
Federal Aid. Using the SIB mechanism, states may be
able to make money available for some types of proj¬
ects sooner than would otherwise be the case because
the SIB accords greater flexibility in using federal aid.
After a state deposits the categorical federal aid into its
SIB and provides a 25-percent match from its own
sources, the categorical restrictions on the use of the
money within a given mode disappear or are greatly
diminished.'^ As a result, a project that is so large that
the state might ordinarily need several years to accumu¬
late enough funding in its category could become af¬
fordable more immediately by drawing on SIB funding
from multiple categories. In that way, SIBs can serve
as intermediaries that pool federal aid from more than
one category and thereby broaden the set of eligible
projects. States may then be able to deploy their high¬
way funds first on projects that have the greatest total
payoffs rather than be constrained to choose the best
projects within each category.
SIBs are also more flexible than traditional fmanc-
ing in another way. They can provide assistance for all
stages of projects without restrictions on the amounts
or percentages of funding that come from specific
sources. However, as noted earlier, capitalization
grants for the SIB require a nonfederal match of at least
25 percent.
A factor that could affect both the amount of
money available for a project and its efficiency and
distributional effects is the extent to which rules on fed-
11. Ibid., p. 13. DOT estimates that by the end of 1997, states will have
allocated $260 million in federal aid to their SIBs to help fund $940
million worth of projects.
12. According to DOT, highway and mass transit fiinds must be kept sepa¬
rate for tbeir initial use, but repaid funds can be blended subject to
state laws. See Federal Highway Administration, "SIB Update," Octo¬
ber 1996 (available at http://www.fhwa.dot.gov/innovativefmance/).
CHAPTER THREE
STATE INFRASTRUCTURE BANKS 23
eral aid govern the use of funds that are repaid after the
first round of SIB lending. For example, the Davis-
Bacon Act requires that the prevailing wage be paid to
workers on projects built with federal aid, a rule that
tends to increase costs. If that requirement was im¬
posed only on the first round of projects financed by a
SIB, the costs of later projects built with recycled funds
would be lower than if Davis-Bacon had been applied
to them as well. Environmental regulations imposed as
conditions of federal highway aid are another such ex¬
ample. Whether those requirements apply to later proj¬
ects has not been resolved, although the Administra¬
tion's proposal for reauthorizing ISTEA would apply
them to recycled funds. Other proposals, however,
would loosen federal requirements.
Additional Private Capital. The potential SIBs have
for raising money to finance transportation projects
depends on whether states leverage SIB funds by at¬
tracting additional sources of capital. Leveraging
makes more funds available immediately. By providing
credit enhancements, SIBs may be able to market bonds
at lower interest costs. Using SIB funds to provide
backing for projects may also help to draw private eq¬
uity investment to a venture, including the kind of pub¬
lic/private initiatives described in Chapter 5.
Yet why would a state use the SIB instead of float¬
ing a bond issue directly, as many states already do for
highways and other capital projects? By using a SIB, a
state may bypass its own constitutional or legislative
limits on debt, especially if the debt is backed only by
toll or other user fee revenues and not by taxes. (As
noted earlier, such debt is known as nonrecourse debt.)
Of course, a state may still be constrained by its own
laws restricting the amount of debt it can carry and by
conditions in the financial markets.
Recycling Loan Repayments. Even if SIBs do not
leverage funds to magnify the amount of funding avail¬
able initially, they can still provide additional money for
transportation projects during later periods. The reason
is that SIBs make loans, not grants, and the loans must
be repaid. The funds that are repaid, together with the
interest on them, can then be lent again to additional
projects. Keeping the funds within the SIB effectively
earmarks them for transportation projects and protects
future projects from having to compete with other state
and local government programs.
Additional Funds from Returns on Investment. If
the projects that states choose to finance through SIBs
are ones with large payoffs, they will generate addi¬
tional economic activity and wealth even beyond the
amount needed to repay the loans. At least some of that
activity is likely to result in additional revenues from
fuel taxes and other user taxes and fees that tradition¬
ally have been earmarked for transportation. Increased
economic activity will also generate additional revenues
for states and localities from income, sales, and prop¬
erty taxes. Those revenues may not directly finance
additional work on roads, but by helping provide for
other public needs, they can reduce pressures on gov¬
ernment budgets that restrict the total amount of funds
available for transportation.
Keeping the SIB Capitalized. To continue operating,
a SIB must maintain its capitalized value. Subsidizing
projects by making loans at below-market interest rates
can jeopardize a SIB's viability because repayments
may be insufficient to maintain that value. Defaults
could further reduce the SIB's value, as might inflation,
which can erode the purchasing power of SIB funds. If
the SIB's capitalized value shrinks, it will need an infu¬
sion of additional capital at some point to continue its
operations.
Regardless of whether loans from SIBs are funded
only by capital provided by the federal and state gov¬
ernments or also by borrowed capital, they must be re¬
paid. Typically, funds for repayment will come from
tolls or other user fees or taxes. Using those sources of
revenue to repay loans has implications both for effi¬
ciency and for the way the costs of this kind of mecha¬
nism are distributed among taxpayers and users. In
addition, the choice of repayment source could affect
the federal budget.
Efficiency
To the extent that the SIB mechanism allows states to
blend funds from several categories of federal aid, it
lessens some constraints on states' choice of projects.
Moreover, it enables them to choose projects with the
highest economic returns overall and not just the high¬
est returns within each category of federal aid. The
ability to move beyond categories to focus on the "big
picture" can increase efficiency in investment.
24 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Whether resources are allocated efficiently depends
largely on how loans from the SIBs will be repaid. Sev¬
eral SIB projects that states have proposed will repay
loans with tolls and other user fees, a strategy with two
beneficial economic effects. First, using tolls and fees
for repayment reinforces the need to select projects for
which benefits—and user-fee revenues—exceed costs.
It also provides incentives to make those investments
the "right size" and to keep costs in check. Under such
a repayment plan, gold-plated investments and invest¬
ments that are too large for the expected demand (fea¬
tures that have characterized some public works in the
past) would be discouraged.
Second, repaying loans with tolls and other user
fees can provide incentives for efficiency in the use of a
road project if tolls and fees are set in a way that re¬
flects the cost of use. For example, tolls that are higher
at peak periods encourage those motorists who can to
shift their use to less congested roads or times, reducing
traffic at rush hour on congested tollways. In addition,
the revenues raised from relatively high tolls at peak
hours can help fund projects that increase highway ca¬
pacity. Alternatively, they can support projects to im¬
prove mass transit facilities and equipment.
As Chapter 2 notes, user-fee financing may or may
not produce an efficient allocation of resources. To
achieve allocative efficiency, the price paid by the user
must reflect the marginal social cost of the resources
associated with that use. User fees that are set with
recovering costs or maximizing revenues as their pri¬
mary objective may not provide incentives for the most
efficient use of the resources.
In the early responses to DOT's call for applica¬
tions for the SIB program, 14 proposals listed tolls as a
source of loan repayment (although some proposals
combined revenues from tolls and from dedicated
taxes). By February 1997, only four projects cited tolls
as a principal means of repayment. Many more were
relying on state and local taxes on transportation and on
fees that were not directly related to the use of the spe¬
cific project that they were helping to fund. Using re-
13. In general, a government should not undertake a project unless benefits
exceed costs. But in the case of highways, recouping all the benefits m
the form of user fees may be difficult. The existence of extemal bene¬
fits is one reason roads in the United States have been publicly pro¬
vided and have not been required to cover costs with user fees.
payment sources that are not directly related to the proj¬
ect's use diminishes the prospects for increased effi¬
ciency resulting from user charges. But the goal of effi¬
ciency may have to give way in order to get the project
built. Dedicated tax revenues may be needed to attract
private investment in facilities that are desirable to
build because they pass a benefit-cost test. SIBs can
help reduce the cost of debt if they signal to private in¬
vestors that a project is especially creditworthy or if
they provide enough backing to enhance the quality of
the debt.
Some of the projects mentioned in SIB proposals
submitted by the states are projects to be done at the
state level; most, however, are projects sponsored by
municipalities, authorities, or other agencies created by
state governments. Those substate entities may find it
difficult to borrow money for projects even if the work
has a reasonable prospect of yielding competitive re¬
turns. In contrast, states generally have experience with
bond issues, together with the necessary apparatus
(staff, legal and financial advisors, legal authority, and
so forth). Furthermore, states are known entities to
credit-rating agencies and bond buyers.
Municipalities and other substate government bod¬
ies may be less experienced and little known in the
bond world, especially if they are relatively small and
have not engaged in much debt financing. Such entities
may have difficulty floating a bond issue, even if their
fmances are strong, because potential bond buyers may
know little about them and consider them too much of a
risk.^'' By making loans or offering credit enhancement
to lesser-known entities, SIBs may overcome those ob¬
stacles." A state infrastructure bank with expertise in
transportation projects might be willing to overlook
such problems to lend to a project with a strong poten¬
tial for return. In that way, a SIB could promote effi¬
ciency in investment.
14. Municipalities with poor underlying economic conditions or with a
history of default may find borrowing virtually impossible. Their in¬
ability to find lenders is not a failure of the market but rather a market
signal that lending under those conditions is a high-risk proposition.
15. Some states already make use of credit pooling. See Congressional
Budget Office, An Analysis of the Report of the Commission to Pro¬
mote Investment m America's Infrastructure, GEO Paper (February
1994), pp. 28-29.
CHAPTER THREE
STATE INFRASTRUCTURE BANKS 25
In addition, by aggregating the financing of several
projects into one bond issue, SIBs can reduce costs as¬
sociated with issuing bonds, such as hiring legal and
financial counsel and obtaining a credit rating.
For municipalities (or authorities or states) that
have excellent credit ratings, using the SIB to float a
bond issue may not be the wisest approach. Those enti¬
ties might do better to borrow on their own instead of
through an institution in which they are pooled with
higher-risk borrowers. Depending on the structure of
the debt (in particular, whether debt repayments can be
used to cover obligations of other members of the
pool), stronger borrowers might get lower interest rates
by "going it alone" than by pooling with "weaker links."
Over time, if SIBs prove successful in helping fi¬
nance projects at the substate level, they might encour¬
age more such projects. The same reasoning that un¬
derlies the federal-aid highway program—^that states
know their specific highway needs better than the fed¬
eral government—may apply at a lower level as well.
That is, local communities may know their needs better
than the state government. Consequently, a mechanism
such as the SIB, which would provide a form of state
assistance to local projects and shift more decision¬
making to lower levels of government, might yield more
efficient investments in infrastructure than financing
mechanisms at higher levels of government.
Distributional Effects
Debt fmancing by SIBs has two types of distributional
effects that differ from the effects of traditional fmanc¬
ing based on user taxes and fees. First, it shifts the bur¬
den of financing from current payers of user taxes to
future users and other beneficiaries of the projects it
funds. Depending on the source of revenues dedicated
to repayment, the debt financing that SIBs provide may
also shift the burden to future taxpayers in general.
Second, tax-exempt debt shifts some of the burden
from taxpayers at the state and local levels to those at
the federal level.
SIB fmancing may have other distributional effects
as well. For example, as mentioned earlier, federal re¬
quirements may not govern projects that states or other
entities build with recycled SIB funds. Thus, if the
rules of the Davis-Bacon Act regarding prevailing
wages were deemed not to apply, workers might receive
lower wages (unless labor markets were tight). But
taxpayers and users of roads would gain from lower
project costs.
Administrative Feasibility
Participation in the SIB program is voluntary. States
that want to form SIBs and capitalize them with federal
aid may apply to the Department of Transportation for
permission. Some states found SIBs attractive even
before the Congress appropriated additional funds for
them in 1997. Others waited until the Congress had
provided that money to submit their applications.
Whether the added funds were the primary motivation
or whether states just came to the conclusion that the
SIB mechanism could be useful to them is not clear,
although some state officials have suggested that the
$150 million was an influential factor.
The 15 states that initially applied for the pilot pro¬
gram indicated that they thought SIBs were a feasible
approach to highway financing. The second group of
26 applications by 28 states and Puerto Rico provided
further evidence. DOT permitted flexibility in the orga¬
nizational structure: a state could form a SIB within its
highway or transportation department, in a budget or
finance department, or in whatever organizational for¬
mat it chose. The NHS Act allowed states to use up to
2 percent of federal capitalization grants to cover the
costs of administering SIBs. All of those factors proba¬
bly encouraged more states to participate.
Some states may need legislation to be able to put a
SIB in place, especially if the bank is given authority to
issue bonds or make loans for private ventures.'® How¬
ever, most states have experience to draw on in autho¬
rizing and establishing such an entity because most
states have state revolving funds (SRFs) for wastewater
treatment, after which SIBs are partially patterned. The
NHS Act prohibits commingling of highway and mass
transit funds with other fimds. As a result, SIBs proba¬
bly could not be combined with SRFs unless mecha¬
nisms were established to keep separate accounts for
each program.
16. General Accounting Office, State Infrastructure Banks: A Mecha¬
nism to Expand Federal Transportation Financing, GAO/RCED-
97-9 (October 1996), p. 18.
26 INMUVATIVE- 'r'
"GHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Implications for the Federal Budget
The only direct effect of the SIB program on the federal
budget thus far is the $150 million in appropriations for
1997. The 1998 appropriation legislation did not in¬
clude additional funds for SIBs.
In authorizing the SIB program, the Congress di¬
rected that disbursements to capitalize the banks be at a
rate consistent with historical levels for the federal-aid
highway program. Funding the SIBs in that manner
assumes disbursements of 15 percent in the first year,
53 percent in the second, 16 percent in the third, and
the remaining 16 percent spread out over the fourth
through ninth years. The Federal Highway Administra¬
tion has advised the states that they must delay actual
cash deposits to their SIBs to conform to those dis¬
bursement rates. The Administration's proposal for
reauthorization of ISTEA would alter the disbursement
rate so that SIBs would receive 20 percent of the funds
each year for five years.
The Outlook for State
Infrastructure Banks
SIBs are a relatively new mechanism for funding trans¬
portation projects, and it remains to be seen whether
they will fulfill their intended purpose. The next few
years will determine how well the SIB structure works
or whether alternative mechanisms might be of greater
value.
one or more factors: a bias toward highways on the
part of state transportation officials, a judgment that
highways are a more pressing need than transit ven¬
tures, an absence of suitable mass transit projects, or an
inability to identify sources of repayment of loans for
transit projects. Identifying and reprogramming funds
with which to capitalize transit accounts are another
problem. The requirement contained in the NHS Act
that states keep separate accounts for highways and
mass transit may present an institutional obstacle in
that regard.
If all states deposited 10 percent of their eligible
funding for highways into SIBs, the total would be
about $2 billion a year. In a survey of 15 states, the
General Accounting Office found that SIBs would
probably be funding less than 10 percent of the trans¬
portation projects in those states over the next five
years." Apparently, states are being cautious about
SIBs, but if initial experiences are successful, they will
be receptive to greater use of the mechanism.
Without restructuring of the entire federal-aid pro¬
gram, SIBs are unlikely to become a major source of
highway financing in the next few years. Indeed, the
General Accounting Office cited an FHWA official
who said that only a small number of projects could
generate enough revenue to repay loans made by
SIBs.'® Nevertheless, SIBs may enable certain types of
projects—especially those for which tolls and other user
fees are the source of debt repayment—to go forward
sooner than they could under traditional financing.
Highway projects dominate the set of initial pro¬
posals for SIB funding. That circumstance could reflect
17. Ibid. p. 2.
18. Ibid, p. 16.
Chapter Four
Federal Credit Assistance
The federal government could help to finance
transportation projects by assisting the parties
involved to obtain credit. Federal loans, loan
guarantees, lines of credit, infrastructure banks, and
government-sponsored enterprises are all approaches
by which the federal government could aid in financing
public, private, and mixed public/private projects. In
making additional money available to lend or in prom¬
ising to assist with repayment, such mechanisms can
enhance the quality of the credit and enable the project's
sponsors to borrow at lower interest rates. However,
some approaches to credit enhancement would lead to
more tax-exempt debt and result in losses of revenue to
the federal government.
Federal Loans, Loan
Guarantees, and
Lines of Credit
Federal loans, loan guarantees, and lines of credit work
in similar ways to make the debt that the sponsor of a
project issues more attractive to lenders. The Congress
will be considering proposals for enhancements of that
kind in the near future. In addition, some of the mecha¬
nisms are being used now in a limited way for transpor¬
tation projects.
Transportation Infrastructure Credit
Enhancement Program
The National Economic Crossroads Transportation Ef¬
ficiency Act of 1997, the Administration's proposal for
reauthorizing the federal highway and transit programs.
would establish a federal program of credit enhance¬
ment for transportation infrastructure. The program
would aid transportation projects by making grants to
sponsors of projects for capitalizmg revenue stabiliza¬
tion funds that would back a project's debt.' The pro¬
posed legislation specifies that obligations secured by
the stabilization fimds are not to be considered federally
guaranteed under the tax code. As a result, the funds
can back both taxable and tax-exempt debt.
The intent behind the Administration's proposal is
twofold. First, the legislation is framed to encourage
development of projects of national significance that
otherwise might be delayed because their risk or scope
was too large for traditional financing. To be eligible, a
project would have to cost at least $100 million or at
least 50 percent of the state's annual apportionment of
federal-aid highway funds, whichever was less. Eligi¬
ble projects would also have to be supported at least in
part by user charges or other dedicated sources of reve¬
nue. The Department of Transportation's seeond pur¬
pose in proposing the credit enhancement program is to
encourage participation by the private sector in devel¬
oping and building highways. The combination of user
charges and private investment would put such projects
to a market test, in contrast to traditional transportation
projects. The Administration's proposal would autho¬
rize $100 million a year for the 1998-2002 period.
Transportation Infrastructure Finance
and Innovation Act of 1997
The Transportation Infrastructure Finance and Innova¬
tion Act of 1997 (TIFIA) would establish a pilot pro-
1. Strictly speaking, the proposed program is a federal grant program that
supports issuance of debt rather than a federal credit program.
28 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
gram of federal credit assistance for large highway and
mass transit projects.^ The bill would authorize $530
million in federal budget authority over six years for
secured federal loans, loan guarantees, and standby
lines of credit. Relevant provisions are as follows:
o Section 1314(a)(4): Credit assistance would be
available for projects that generated streams of rev¬
enue through user charges or other dedicated
sources of funding,
0 Section 1314(a)(4)(B): This section is an amend¬
ment to the original bill that would prohibit using
proceeds from tax-exempt financing for repayment.
o Section 1315(b)(4): The interest rate on secured
loans made through the program would be set to
equal the yield on marketable U.S. Treasury securi¬
ties of a maturity similar to that of the secured loan
on the date the loan agreement was executed.
o Sections 1315(b)(3) and (6): The federal govern¬
ment may assume junior-lien status for repayment
of secured loans (that is, it would be repaid after
the senior creditors). But in the event of bank¬
ruptcy, insolvency, or liquidation of the obligor, the
federal government would have the same claim on
assets as other creditors.
0 Section 1322: The Secretary of Transportation
must report to the Congress within four years about
whether the objectives of this program would be
best served by continuing it under DOT, by creat¬
ing a government-sponsored enterprise to adminis¬
ter the program, or by phasing it out and relying on
capital markets to fund the types of projects it had
assisted.
Loan for the Alameda Corridor Project
The federal government has agreed to lend money for
an intermodal project, the Alameda Corridor project.
2. TIFIA is Title I, Subtitle A, Chapter 2, of the Intermodal Transporta¬
tion Act of 1997 (S. 1173), the Senate's bill to reauthorize the
Intermodal Surface Transportation Efficiency Act. To be eligible for
the program, projects must cost at least $100 million or 50 percent of
the state's most recent apportionment of federal-aid highway funds,
whichever is less. However, for projects involving intelligent transpor¬
tation systems, the threshold is $30 million.
which will improve highway and rail links between the
ports of Los Angeles and Long Beach and the regional
rail distribution center near downtown Los Angeles.
The total cost of the project is about $2 billion; the fed¬
eral government is lending $400 million to the Alameda
Corridor Transportation Authority (ACTA) as part of
the financing. The loan will be disbursed over three
years: $140 million a year each in 1997 and 1998 and
$120 million in 1999. In 1997, the Congress appropri¬
ated $58.7 million, which is the estimated subsidy cost
of the loan.^
ACTA has agreed to repay the debt over a 30-year
period, which will start when the project begins to re¬
ceive revenues from the participating railroads and
ports. Currently, the project's expected start-up date is
2002. During the construction phase, interest on the
loan will accumulate at the same rate that the Treasury
pays on its 10-year securities and will be added to the
principal for repayment. Once revenues begin to flow,
interest will accumulate at the rate the Treasury pays on
its 30-year notes.
The loan agreement provides some flexibility in
repaying the federal government. If ACTA carmot meet
the repayment schedule, it may postpone repayment,
although interest will continue to accrue at the agreed-
upon rate. That flexibility, combined with the federal
government's junior-lien status, has provided key assis¬
tance to ACTA that has allowed the authority to get the
project started.
Lines of Credit for Orange County
Toll Roads
The federal government has provided assistance in the
form of standby lines of credit to the Orange County,
California, transportation corridor agencies (TCAs) for
the San Joaquin Hills and Foothill/Eastern Toll Roads.
The TCAs are agencies that the state of California au¬
thorized local governments to create for the purpose of
3. U.S. Department of Transportation, "President Clinton Says Alameda
Corridor Project Will Create Jobs and Build Trade Links to Pacific
Rim" (press release, DOT 09-97, January 17, 1997). See also U.S.
House of Representatives, Omnibus Consolidated Appropria
ations for Fiscal Year J 997, conference report to accompany H.R.
3610, Report 104-863 (September 28, 1996), pp. 527-528. The sub¬
sidy cost is defined under the Federal Credit Reform Act as the present
value of the cost of a loan or loan guarantee, factoring in the possibil¬
ity of defaults or interest rate subsidies.
CHAPTER FOUR
FEDERAL CREDIT ASSISTANCE 29
financing and building toll roads (see the fiirther discus¬
sion in Chapter 5). The agencies can draw on the lines
of credit if revenues from tolls are insufficient to repay
lenders and cover other costs. For potential lenders, the
Imes of credit enhance the quality of the debt the TCAs
issue; the credit assistance may also encourage lenders
to accept lower rates of interest, which would reduce
the TCAs' cost of borrowing. In addition, interest on
the debt the TCAs issue is exempt from federal income
taxes. As a result, the projects enjoy a tax subsidy.
The San Joaquin Hills and Foothill/Eastern projects
each obtained a line of eredit from the federal govern¬
ment of $120 million.'' Under the terms of the agree¬
ments, the agencies can draw on the lines over a 10-
year period, with a maximum draw of $12 million a
year. Any such draws must be repaid based on Trea¬
sury rates (the 30-year rate for debt service and the 3-
year rate for operations).
Although the TCAs do not expect to draw on their
lines of credit, the lines constitute a potential claim
against federal taxpayers. Thus, the cost of the subsidy
they entail must be factored into the federal budget.
The budget showed $8 million in budget authority in
1995 for the subsidy associated with the line of credit
of $120 million for the Foothill/Eastem project. It
showed $9.6 million in budget authority in 1993 for the
subsidy assoeiated with the San Joaquin project's line
of credit.
Federal Infrastructure
Financing Institutions
In a federal credit program, the federal government
would be directly involved in selecting and overseeing
projects and in providing loans or other credit assis¬
tance. Another approach would be to make federal
grants to capitalize an institution that would provide
credit for transportation projects. The Intermodal Sur¬
face Transportation Efficiency Act of 1991 created an
infrastructure investment commission to explore that
concept.^ Its mandate was to study the feasibility and
desirability of creating a type of infrastructure security
that would be attractive to pension funds. It was also
charged with examining other methods of encouraging
public and private investment in infrastructure facili¬
ties.
In its 1993 report, the Commission to Promote In¬
vestment in America's Infrastructure recommended sev¬
eral new options related to infrastructure financing. It
suggested establishing a new, federally chartered na¬
tional infrastructure corporation that would make loans
to infrastmcture projects with appropriated or borrowed
funds. It also recommended creating new investment
mechanisms for institutional investors, including secu¬
rities issued or guaranteed by the corporation.® The
commission further envisioned a subsidiary to the cor¬
poration that would insure debt issued by infrastructure
projects.'
An infrastructure fmancing entity could take any
one of several forms. It could be an on-budget federal
agency that made subsidized loans or grants; a private,
for-profit company capitalized initially with federal
funding; or a government-sponsored enterprise (GSE)
akin to Fannie Mae (the Federal National Mortgage
Association) or Freddie Mac (the Federal Home Loan
Mortgage Corporation).^ As a private company or a
GSE, the bank could be capitalized by issuing debt and
equity. The federal government could back the debt,
and the interest could be exempt from state and local
income taxes. However, the interest would still be sub¬
ject to federal income taxes unless a significant change
in federal tax policy occurred. Section 149(b) of the
Internal Revenue Code states that bonds that are guar¬
anteed in whole or in part, directly or indirectly, by the
federal government are taxable. The provision is in¬
tended to prevent so-called double-dipping—that is.
5. Section 1081, 105 Stat. 2020. The commission is also known as the
Flanagan Commission after its chairman, Daniel V. Flanagan Jr.
6. The Commission to Promote Investment in America's Infrastructure,
Financing the Future (February 1993), available through the U.S.
Department of Transportation, Office of Economics.
7. For a detailed description and critique of the commission's report, see
Congressional Budget Office, An Analysis of the Report of the Com¬
mission to Promote Investment in America's Infrastructure, GBO
Paper (February 1994).
4. Subsequently, the Congress increased the line of credit for the Foot¬
hill/Eastem project to $145 million.
8. Fannie Mae and Freddie Mac are analyzed in Congressional Budget
Office, Assessing the Public Costs and Benefits of Fannie Mae and
Freddie Mac (May 1996).
30 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
when an investor receives favorable tax treatment as
well as a federal guarantee.
Evaluating Federal Credit
Assistance
Certainly, federal credit assistance can generate addi¬
tional money for transportation projects. But will it
help to allocate resources more efficiently? And who or
what will bear the burden of costs under that kind of
approach? The distributional and efficiency effects
depend in large part on the way states or other entities
structure specific projects. The implications for the
federal budget depend on several things: the form of
the assistance but also whether credit assistance comes
in addition to funds made available in the federal-aid
program or whether it substitutes for federal aid. (See
Box 4 for some of the forms that a federal financing
institution for infrastructure could take.)®
Financing Potential
The amount of money federal credit assistance could
make available depends on the demand for and supply
of funds. In the case of the Alameda Corridor project,
the federal loan of $400 million, which required an ap¬
propriation of $58.7 million, helps to support a $2 bil¬
lion effort. The Federal Highway Administration has
calculated the financing potential of federal credit assis¬
tance proposed under TIFIA, the Senate's bill to reau¬
thorize ISTEA. FHWA estimates that the $530 million
in budget authority in the bill could provide as much as
$10 billion in credit assistance—because only the costs
of the subsidy would be recorded in the budget. That
amount of budget authority could conceivably support a
total investment of $30 billion, given that TIFIA would
limit the federal credit share to 33 percent of project
costs.'®
9. For more detailed analyses, see Congressional Budget Office, An
Analysis of the Report of the Commission to Promote Investment in
America's Infrastructure, 2n\á Assessing the Public Costs and Bene¬
fits of Fannie Mae and Freddie Mac.
10. Personal communication to the Congressional Budget Office by David
Seltzer, Senior Advisor to the Administrator, Federal Highway Admin¬
istration. See also Bryan Grote and David Seltzer, "Transportation
Infrastructure Finance and Innovation Act of 1997," FHWA's Innova-
The number of projects that would take advantage
of such a credit program is difficult to judge. Underly¬
ing the uncertainty is how states would respond as they
weighed the options of using federal credit assistance or
outright grants for transportation projects. Whether
states would use the credit to accelerate projects sched¬
uled for future years, when federal grants and matching
state funds became available, remains to be seen. The
credit option would be less appealing if federal grants
were sufficient to finance all the projects a state was
ready to build. But if federal funds were insufficient to
meet the state's demands, borrowing would enable it to
build a needed project sooner. Of course, any projects
built on credit would have to generate revenue for re¬
paying loans.
Any analysis of the financing potential of federal
credit assistance rests on whether federal credit repre¬
sents additional money available to states or whether it
comes out of funds from the federal-aid program. The
proposals described in this chapter generally refer to
new money. Under the program of state infrastructure
banks discussed in Chapter 3, most of the federal fund¬
ing that states could use for credit assistance would
come from the federal-aid program." Whether project
sponsors would prefer SIB financing or direct federal
credit would depend on the details of the project and the
financing options that were available when the project
was being planned. With more federal credit available,
more projects might be proposed; however, there is no
evidence to support that assumption. As for the supply
of funds, investors would probably behave as they have
always behaved: they would continue to buy whatever
securities offered the best combination of risk and re¬
turn for their situation, without regard to whether a
bond issue financed a highway, sewage treatment plant,
school building, or any other kind of facility.
Efficiency
How a financing option affects efficiency depends on
whether the option encourages efficient allocation of
tive Finance Quarterly, vol. 3, no. 1 (Summer 1997), p. 2. The arti¬
cle estimated that $800 million in budget authority, which was the
amount being considered at the time the article was written, could
provide as much as $16 billion in federal credit assistance and enable
total investments of $50 billion.
11. The exception is the additional $150 million that the Congress appro¬
priated for state infrastructure banks in 1997.
CHAPTER FOUR
FEDERAL CREDIT ASSISTANCE 31
Box 4.
Federal Infrastructure Financing Institution
An idea that was proposed in the early 1990s but that has
not been a focus of recent legislative debate is the estab¬
lishment of a federal entity to provide financing for high¬
way and other infrastructure projects. ' The Intermodal
Surface Transportation Efficiency Act of 1991 established
the Commission to Promote Investment in America's Infra¬
structure. In 1993, the commission recommended that the
Congress create two new corporations to provide credit
assistance for infrastructure projects. A National Infra¬
structure Corporation (NIC) would purchase and bear the
credit risk of municipal bonds issued by states and locali¬
ties to provide long-term financing for infrastructure proj¬
ects. It would also insure private firms against a portion
of the risk of developing new facilities. Aui Infrastructure
Insurance Company (IIC), which would initially be a sub¬
sidiary of the NIC, would insure infrastructure bonds is¬
sued for such projects. The commission also asked
policymakers to consider easing restrictions on tax-exempt
financing for infrastructure used for private aetivities and
giving a new tax break to participants in pension plans
that purchased qualified infrastructure securities.
By subsidizing the development and financing of new
projects, the NIC would increase investment in municipal
infrastructure. Because it would bear the credit and de¬
velopment risks of projects on subsidized terms, munici¬
palities could pay lower interest rates on their infrastruc¬
ture bonds and still attract investors. As a result, munici¬
palities might be encouraged to take up such projects.
The proposed changes in tax law would also provide sub¬
sidies for municipal borrowers.
The primary effect of the commission's proposals
would be to divert resources from investments in business
plant and equipment, housing, and other government
spending to investments in infrastructure projects. Would
diverting funds to the NIC and nC from alternative uses
improve the allocation of resources, thus producing more
benefits to society? That might happen if infrastructure
projects had spillover benefits that were not adequately
taken into account or if credit markets were not function¬
ing efficiently. But the spillover effects are unlikely to be
large enough to justify federal subsidies, and no major
problems with credit markets are apparent.
The new corporations the commission proposed
could be organized as on-budget federal agencies or as
private, federally subsidized firms. Setting up the NIC as
a federal agency would mean that policymakers could ob¬
tain complete information about its activities and directly
control the cost of the subsidies it provided to municipal
borrowers. As an on-budget agency, the NIC would also
require much smaller initial appropnations than if it was
established as an off-budget entity.
If the NIC was established ofT-budget as a private,
for-profit finance company and subsidized by a long-term
federal loan with a below-market interest rate, the cost of
the subsidy that the loan provided would be controlled in
the appropriation process and recorded in the federal bud¬
get. The company would be subject to less direct control
by policymakers than an on-budget agency and could op¬
erate as a revolving fund. But a finance company would
have to stand on its own after it repaid the government's
loan, which would subject the NIC to significant market
discipline. A private, for-profit status would also provide
a strong incentive for the infrastructure corporation to use
the limited, one-time subsidy to build its capital and estab¬
lish a track record rather than give ongoing subsidies to
municipal borrowers.
Organizing the NIC as either an on-budget agency or
a finance company would have fewer risks than establish¬
ing the corporation as a government-sponsored enterprise
(GSE). If the NIC was organized as a GSE, the federal
budget would not measure, and policymakers could not
directly control, the subsidies provided by the implicit
federal guarantee of its obligations.
The type of organizational structure chosen for the
nC would have particular ramifications, just as in the case
of the NIC. As a federal agency, the HC could not insure
tax-exempt infrastructure bonds unless policymakers re¬
versed the long-standing federal policy of not providing
explicit federal guarantees of tax-exempt debt. If the com¬
pany made loans at tax-exempt rates, the cost of the inter¬
est subsidies would have to be appropriated each year.
Organizing the HC as a private, for-profit bond in¬
surer that was partially owned by the federal government
would have other effects. The federal budget would re¬
cord the cost of purchasing stock in the company, but as a
private insurer, the HC would be subject to less direct con¬
trol than a federal agency, and it could insure tax-exempt
bonds. Moreover, it would have an incentive to manage
its resources prudently because investors would be un¬
likely to perceive an implicit federal guarantee of the
bonds it insured. By establishing a sunset date for the
company, policymakers could use this organizational form
to provide temporary federal support for insuring infra¬
structure bonds that existing insurers do not now cover.
1. This box is drawn from Congressional Budget Office, An Analy¬
sis of the Report of the Infrastructure Investment Commission,
CBO Paper (February 1994).
32 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
transportation resources and whether it makes efficient
use of capital. As noted in the discussion of state infra¬
structure banks, projects that are subject to market tests
are likely to be more efficient than those that are not.
The various proposals for federal credit subsidies re¬
quire the projects to gamer revenue to repay their debt.
Sponsors of projects could stmcture such revenues to
provide incentives for efficiency. For example, they
could choose tolls as the revenue mechanism and then
set them to vary according to congestion and other costs
imposed by users. But projects might also rely on taxes
that reduce efficiency. Economists generally agree that
taxes (with the exception of lump-sum taxes) distort the
way resources are allocated and thus impose costs on
society. The taxes that support the federal-aid highway
program have the favorable characteristic that they are
imposed on highway users; yet even those taxes give
rise to inefficiencies.'^
Efficiency in credit assistance would entail provid¬
ing only enough federal help to overcome any failure of
markets for municipal bonds. But those markets ap¬
pear to work as efficiently as other credit markets, and
proposals for federal credit subsidies do not seem to
address any sources of inefficiency that the markets
might have.'^ The types of projects that would be can¬
didates for federal credit assistance can offer no guaran¬
tees about their ability to repay loans. However, given
that both borrowers and lenders lack certainty about the
outcome, neither side appears to have information that
would give it an advantage and justify the federal gov¬
ernment's intervention in the market.
Are some projects too risky for private investors to
finance? That claim is sometimes made about large
ventures that employ novel methods or approaches.
Sponsors argue that the federal government is the only
entity with pockets deep enough to finance such proj¬
ects and with the ability to spread the risks over a num¬
ber of ventures that collectively have a reasonable prob¬
ability of success (although any individual project may
have as much probability of failure as of success). In
addition, the federal government has a longer time hori-
12. See Congressional Budget Office, Paying for Highways, Airways, and
Waterways: How Can Users be Charged? (May 1992).
13. For a detailed discussion of the efficiency effects of a federal infra¬
structure bank, see Congressional Budget Office, An Analysis of the
Report of the Commission to Promote Investment in America's Infra¬
structure, pp. 29-52.
zon, lower liquidity requirements, and fewer funding
constraints than most investors. Those characteristics
give the government the flexibility to restructure inter¬
est and debt repayments if the startup of a project is
slower than anticipated. But why not finance projects
of that type privately by issuing securities that pay a
high rate of interest? The venue for such activity is the
market for junk bonds—that is, bonds with high credit
risk. The junk bond market brings together sponsors of
high-risk ventures and investors who want to include
such securities in their portfolios for balance, diversifi¬
cation, or some other reason.
A distinctive feature as well as a component of the
risk of the types of projects discussed here is that they
face potential competition from government. Tradition¬
ally, such projects have been financed and owned by
governments. If a government decides to build a toll-
free road, financed with tax revenues, parallel to a new,
privately financed toll road, investors in the latter could
lose all or part of their investment. Government back¬
ing of debt for toll roads might be a way to reassure
private investors about the government's intentions and
persuade them to risk their money in toll projects.
Credit enhancements that make financing available
at below-market rates can distort resource allocation by
making subsidized projects more attractive to investors
than projects that might yield greater net benefits on
their own merits. But the subsidy associated with fed¬
eral credit assistance for transportation projects is likely
to be much smaller than the outright subsidy of the
federal-aid program. To determine whether projects
built with federal grants had effects on efficiency that
differed from those of projects built with federal credit
assistance would require knowing how the subsidies
compared with the external benefits of the projects.
Finally, in evaluating the efficiency of federal credit
enhancements, it is useful to consider them in relation
to a baseline. If the baseline was one of no credit en¬
hancements, the enhancements would appear inefficient
in comparison. But if one used the existing program of
federal highway grants as the baseline, the result would
be different. Drawing the money for credit enhance¬
ments from what otherwise would be given outright to
the states and usmg the credit assistance for projects
that included private investment and user fees could tip
the balance in favor of efficiency.
CHAPTER FOUR
FEDERAL CREDIT ASSISTANCE 33
Distributional Effects
Federal taxpayers bear some of the costs of projects
financed with credit subsidies from the federal govern¬
ment. Some of the costs are out-of-pocket costs; if a
project defaults, the federal government loses all or part
of the money it has made available for loans, loan guar¬
antees, or lines of credit. Even if a project succeeds,
there will still be opportunity costs if the federal gov¬
ernment has provided funding at a below-market inter¬
est rate. The reason is that the funding from the gov¬
ernment comes at the expense of something else. Thus,
it may crowd out other federal or private spending, or,
in the face of a budget deficit, it may mean more bor¬
rowing by the federal government. If the government
has to borrow to finance spending for the credit subsi¬
dies and if it has to pay more interest to U.S. Treasury
bondholders than it receives fi"om the loan it has made
to a transportation project, federal taxpayers foot the
bill.
Federal credit assistance could also have distribu¬
tional effects within credit markets. Assistance for cer¬
tain kinds of projects, such as transportation, would
make those projects more attractive to investors than
other equally worthy ventures that did not receive fed¬
eral assistance. In addition, the federal government
would be in a position of picking winners and losers
among the projects—and among the localities and states
in which they were located.
Administrative Feasibility
For the most part, the administrative structures for car¬
rying out federal credit assistance are not specified in
the various proposals. However, the federal govern¬
ment already has a number of credit programs, which
indicates that such programs are administratively feasi¬
ble. Depending on its size and scope, a credit program
for transportation projects might be handled by existing
offices within DOT or another federal agency, or it
might require establishing a new federal bureaucracy.
Regardless of the administrative structure that was fi¬
nally chosen, an independent oversight agency might
also be desirable.'''
14. For a discussion of the need for oversight, see Congressional Budget
OflSce, Controlling the Risks of Government-Sponsored Enterprises
(April 1991).
Implications for the Federal Budget
Loans, loan guarantees, and lines of credit are not free
of costs to the federal government. They bear some
risk of default, and even if they are eventually repaid,
they may return less interest to the government than the
opportunity cost of the funds (that is, if they were lent
at a below-market rate of interest). The Federal Credit
Reform Act of 1990 requires the budget to reflect the
budget authority and outlays needed to cover the value
of cash disbursements that are not expected to be repaid
to the federal government." Before the act was passed,
the budget counted loans, loan guarantees, and repay¬
ments on a cash basis. Thus, the full amount of a loan
was recorded as an outlay in the year it was lent, and
repayments were counted as receipts in the years they
were made. Loan guarantees were not counted unless
and until they were invoked. Those procedures made
loan guarantees appear less burdensome to the federal
budget and loans more burdensome than the net present
value of their cost. With the advent of the Federal
Credit Reform Act, loans and loan guarantees came to
be scored more realistically.
When compared with outright grants, credit en¬
hancements can have the same effect on the federal
budget and yet make more funding available for high¬
way projects. If the federal government shifted some of
its aid from grants to loans, loan guarantees, lines of
credit, or other enhancements, more projects could be
financed.
The tax status of debt backed by the federal gov¬
ernment (including indirect backing, such as lines of
credit) is an important consideration in any decision to
make more use of debt financing. If the Congress al¬
tered current tax policy to give such debt tax-exempt
status, the volume of outstanding tax-exempt bonds
would probably increase, which could lead to additional
losses of revenue by the federal government. That risk
is in large part responsible for the Congress's lack of
enthusiasm for modifying policies in place since the
early 1980s that attempt to limit the availability of tax-
exempt financing at below-market interest rates.
15. The Federal Credit Reform Act is part of the Budget Enforcement Act
(Title XIII of the Omnibus Budget Reconciliation Act of 1990). See
Congressional Budget Office, An Explanation of the Budgetary
Changes Under Credit Reform, CBO Staff Memorandum (April
1991).
Chapter Five
Private-Sector Participation
For financing and developing highways, transpor¬
tation officials are looking increasingly to
private-sector participation. At the same time,
the confluence of several factors—legislation relaxing
federal rules regarding the financing of highways, a
scarcity of funds at the state and local levels, and the
prospect of profitable toll projects—has encouraged the
private sector to explore such investments.
The structure of private participation varies de¬
pending on specific aspects of a project. No single
model has emerged as dominant; instead, the extent of
private debt and equity capital and the degree of in¬
volvement in building and operating toll roads differ
from project to project. A review of several ventures
illustrates the range of private-sector involvement and
the ways the public and private sectors have interacted.
Yet one feature is present in all cases: a belief in the
need for additional roadway capacity combined with
insufficient government funding to provide it in a
timely manner. Governments view private participation
as a source of additional capital that can be raised vol¬
untarily without resorting to taxes; private investors
view toll projects as potential moneymakers. As the
later discussion shows, however, those expectations are
not always realized.
Increased private-sector participation in highway
projects is part of a larger trend toward privatization of
government services, which became increasingly popu¬
lar during the 1980s. Proponents are inspired by both
philosophical and economic rationales. The philosophi¬
cal motivation is to reduce the size of government: the
smaller the amount of activity controlled by govern¬
ment, the better, in that view. The economic rationale
is that the profit motive of the private seetor will lead
firms to contain costs and use resources in the most
efficient way. Government deregulation of the trans¬
portation industries beginning in the late 1970s offers
evidence that less government control can lead to
greater economic efficiency in some circumstances.
The federal government's sale of Conrad to private in¬
vestors was also a clear success in privatization. A log¬
ical question is whether changes in federal policies
could induce greater private-sector investment in roads.
If private firms decide to enter the business of
building and operating highways, it will not be the fnst
time that has happened in the United States. In the late
18th and early 19th centuries, many turnpikes were
built by private investors looking for ways to make
profits by providing road links between interior agricul¬
tural markets and ports. The federal government also
built roads in that period, seeing them as a means of
bringing the nation together and promoting trade, com¬
merce, and manufacturing. By the middle of the 19th
century, however, railroads began to divert traffic from
highways. Private turnpike companies withered away,
and public investment in roads for the most part took
the form of local streets and access roads to rail and
port facilities.
The invention of the internal combustion engine
toward the end of the 19th century rekindled govem-
ment interest in road building.' State and local govern¬
ments began to build roads financed by tax revenues,
many of which linked farms to markets. Roads of that
1. The popularity of bicycling in the late 19th century also created a de¬
mand for paved roads with smooth surfaces.
36 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
kind were unlikely to thrive as toll facilities because
they did not serve enough traffic to cover their costs
with revenues from tolls.^
By the early 20th century, the federal government
had also become involved in fmancing highways. Be¬
ginning with the Federal-Aid Road Act of 1916, the
federal government established a policy of toll-free
roads.^ Only in the 1980s did that policy change as a
result of several circumstances; development of the
technology for collecting tolls electronically, which re¬
duced the costs and delays associated with toll booths;
the perceived need for more roadway capacity; and a
reluctance to raise taxes on fuel. The combination of
those factors led the Congress to relax restrictions on
the use of tolls on highways built with federal aid.
Of course, tolls are not the only way of charging the
users or beneficiaries of roadways. Some public/pri¬
vate efforts also raise money or obtain contributions of
land or other resources from developers or businesses
that benefit from improved access to their properties.
But tolls are the most common user fee for highways.
The history of public and private provision of roads
in this country reflects their dual nature: they have
some characteristics of public goods and some charac¬
teristics of private goods. On the public side, roads
provide benefits in many instances beyond those to di¬
rect users (motorists)—for example, by reducing the
cost of shipping goods to consumers. Moreover, col¬
lecting fees for use of a road is not always practical. As
a result, governments over the years have played a large
role in providing highways. But roads have some char¬
acteristics of private goods as well. Limited-access
highways can exclude motorists who are not willing and
able to pay tolls. Furthermore, when roads are con¬
gested, one motorist's use of the road imposes costs in
the form of delays on other users. Those mixed-good
characteristics of roads both motivate and explain the
formation of road-building partnerships between gov¬
ernment and the private sector.
2. José A Gómez-Ibáñez and John R. Meyer, Going Private: The Inter¬
national Experience with Transport Privatization (Washington,
D.C.: Brookings Institution, 1993), p. 165.
3. Section 1 ofthe Act of July 11, 1916 (popularly known as the Federal-
Ad Road Act of 1916), 39 Stat. 355. For an overview of federal poli¬
cies toward tolls, see Congressional Budget Office, Toll Roads: A
Review of Recent Experience, CEO Memorandum (February 1997).
Exploring ways to involve both sectors in investing
in highway projects can take advantage of what each
sector is in the best position to offer. Governments
may be better able to assume many of the risks associ¬
ated with building new roads. By bearing some of
those risks, governments can make investing in road
projects more attractive to private entities. And as the
following descriptions of projects suggest, achieving a
proper balance of risks between the public and private
sectors can contribute to economic efficiency and
equity.
Types of Private-Sector
Participation
Private-sector participation in highway projects can
take many forms (see Table 4). The most common
practice has been for state highway departments to con¬
tract with private firms to build roads. In recent years,
however, a trend has been growing to give private firms
more of a stake in the outcome of a project so that they
will have incentives to provide good highways effi¬
ciently. Within the traditional contracting relationship,
governments can encourage that behavior by introduc¬
ing fmancial incentives for the contractor to build roads
better and faster.
At the other end of the spectrum are primarily pri¬
vate ventures in road building. In those projects, pri¬
vate firms bear most or all of the risk and make most or
all of the decisions about construction and operation.
Governments typically get involved at some point,
however, because they must certify that a project has
fulfilled all applicable environmental requirements and
give permission for charging tolls. In addition, project
sponsors may need to draw on the government's power
of eminent domain to obtain land for rights-of-way.
In between the extremes of public and private pro¬
vision of roads are partnerships forged between govern¬
ment and private firms for building transportation proj¬
ects. The roles and responsibilities of each partner are
specified in contracts between the parties. In the major¬
ity of cases, the private sector risks some capital and is
rewarded if the investment is successful. The partners
often form a new entity—either a special-purpose gov-
CHAPTER FIVE
PRIVATE-SECTOR PARTICIPATION 37
Table 4.
Sponsors and Features of Highway Financing
Sponsor
Major Features of Financing
Examples
Private Equity investors
Private, Nonprofit Entity
Special-Purpose Public Agency
State Agency
State Agency
State Agency
Finance and develop the project us¬
ing private resources
Issues tax-exempt debt backed by
tolls (and without recourse to taxes)
and oversees the project under the
terms of the agreement between the
state and the private developer
Issues tax-exempt debt backed by
tolls (and without recourse to taxes)
and oversees the project under the
terms of the agreement with a private
developer
Issues tax-exempt debt backed by
tolls (and without recourse to taxes)
Issues tax-exempt debt backed by
taxes
Finances highway on a pay-as-you-
go basis using state taxes and fees
plus federal aid
Dulles Greenway (Virginia)
91 Express Lanes project (California)
TH 212 (Minnesota)
Southern Connector (South Carolina)
Interstate 895 (Virginia)
Tacoma Narrows Bridge (Washington)
Arizona toll projects
E- 470 (Colorado)
Orange County, California,
transportation corridor agencies
Some turnpikes
Most highway projects that are
financed by debt
Most highways
SOURCE; Congressional Budget Office.
emment agency or a private, nonprofit corporation—to
finance and oversee the project.
Another nontraditional arrangement is that of a
government contracting with a private firm to operate
and maintain a roadway that the government has built.
Great Britain is experimenting with such a form on a
limited basis, but the United States has yet to explore
its possibilities in any systematic way.
Traditional Contracting
Traditionally, the private sector has built public roads
under contracts with state or local governments. Typi¬
cally, a state highway agency enters into separate con¬
tracts for the design and construction of a project. Af¬
ter approving the design, the agency solicits bids
through an open competition for construction. It gener¬
ally awards the contract to the qualified bidder who of¬
fers to complete the road according to the exact specifi¬
cations at the lowest cost.
In 1988, the Federal Highway Administration
launched an initiative promoting higher-quality con¬
struction of highways while reducing costs over a road's
life cycle.'' That effort resulted in an innovative con¬
tracting program called Special Experimental Project
No. 14. FHWA identified four contracting techniques
for further testing and evaluation:
o cost-plus-time bidding, in which the firm bids on
both the cost of construction and the time needed to
4. Federal Highway Administration, Rebuilding America: Partnership
for investment, Innovative Contracting Practices, FHWA-PD-95-
028(1995), p. 1.
38 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
complete the project (the highway agency then multi¬
plies the number of days by the cost per day to users of
not having the road available);
o lane rental, in which the contractor is charged for
occupying or obstructing part of the roadway and
thus has an incentive to complete work expedi¬
tiously;
o design/build contracting, in which fums bid simul¬
taneously on both the design and construction
stages of the project; and
o warranty clauses, which hold the designer^uilder
responsible for meeting specified performance re¬
quirements for a period of years following the proj¬
ect's completion. A federal regulation prohibits the
use of warranties on federal-aid projects except for
electrical and mechanical equipment.^ The Inter-
modal Surface Transportation Efficiency Act of
1991 provided an exception to that restriction un¬
der certain circumstances.®
Each of those innovative techniques shifts some
risk from government to the private sector. In addition,
the techniques offer private firms more incentives to
reduce costs to motorists and the state and to improve
quality. Although such approaches do not add to the
available financial resources for highways—and hence
are not discussed further in this study—they enable ex¬
isting highway funds to go farther by decreasing costs,
accelerating construction, or improving quality. They
also provide a point of reference in considering new
options for financing.
Private Equity Investment
At the other end of the spectrum from traditionally
funded highways provided by the public sector are proj¬
ects developed primarily by private firms with an equity
stake. Two recently built roads would fall into that
category.
The Dulles Greenway. In response to growing interest
in private investment in transportation facilities, Vir¬
ginia's General Assembly in 1988 authorized private
development of toll roads in the commonwealth. A
group of investors, the Toll Road Investors Partnership
11, thought that a toll road linking Washington's Dulles
International Airport and Leesburg, Virginia, would be
a promising investment. Their judgment was based on
residential and commercial growth in the area, which
was causing increased congestion on existing arterial
roads serving the corridor.
The product of their investment is the Dulles
Greenway. The Greenway is a 14-mile, limited-access
highway extending from the state-owned Dulles Toll
Road, which carries traffic between Washmgton's Capi¬
tal Beltway and Dulles Airport, to Leesburg.' The two
roads connect at a toll plaza. Drivers pay one toll,
which the operators of the two facilities divide. Vehi¬
cles equipped with prepaid electronic tags may drive
through "FastoH" lanes without having to stop at a toll
booth; their tags are read and their accounts debited
automatically.
To fmance the Greenway, investors put up $40 mil¬
lion in cash and secured $310 million in privately
placed taxable debt.^ Ten institutional investors led by
Cigna Investments, Prudential Power Funding Associ¬
ates (a unit of the Prudential Insurance Company of
America), and John Hancock Mutual Life Insurance
Company provided $258 million in long-term, fixed-
rate notes (due in 2022 and 2026). Three banks
(Barclays, NationsBank, and Deutsche Bank AG)
agreed to provide part of the construction funding and
$40 million in revolving credit. Loans are to be repaid
with toll revenues, and the financing is secured by a
first mortgage and security interest in the developer's
right, title, and interest in the facility.®
7. The Dulles Toll Road was opened in 1984 to serve the rapidly grow¬
ing suburbs between the beltway and the airport. By serving local
traffic, it augmented the adjacent Dulles Access Road, which the fed¬
eral government built in 1962 to carry traffic to the airport. The origi¬
nal roadway had no outbound exits or inbound entrances between the
beltway and the airport because it was not intended to carry local traf¬
fic.
8. The investors are Maggie Bryant, a local resident, and her son, Mi¬
chael R. Crane; an Italian company, Autostrade International, which
operates the road; and Brown & Root, a Houston-based construction
company, which built the road.
5. 23 C.F.R. 635.413.
6. Federal Highway Administration, Rebuilding America, pp. 7-8.
9. Dulles Greenway home page, sponsored by Toll Road Investors Part¬
nership II (available at http://www.his.com/~cwealth/greenway/
index.html/).
CHAPTER nVE
PRIVATE-SECTOR PARTICIPATION 39
The Greenway opened to traffic in September
1995, Initially, the toll was $1.75 each way, but when
traffic fell short of projected levels, the toll was reduced
to $1.00. Lowering the toll attracted more users but not
enough to increase total revenues.'® Consequently, in
July 1997, the Greenway's operators raised the toll to
$1.15.
The shortfall in toll revenues from the project has
brought problems for its investors. They had projected
toll revenues for the first year at $27 million; $7 million
was to go for operating costs and $20 million toward
the $30 million in annual interest. When those reve¬
nues did not materialize, the investors began to miss
their quarterly interest payments of $7 million each.
However, they won approval from lenders to skip the
payments for the rest of the year, avoiding foreclosure
through the end of 1997." As of December 1997, the
sponsors were discussing a further extension of their
standstill agreement with the lenders.
The 1988 enabling legislation passed by the Gen¬
eral Assembly prohibits the state from bailing out the
Greenway or other such facilities. Nevertheless, after
the disappointing results of the first few months of the
toll road's operation, the Virginia legislature considered
such action. In the end, it rejected a bailout; however, it
voted to allow the speed limit on the Greenway to rise
(from 55 to 65 miles per hour) in hopes of attracting
more motorists.
The Greenway is a build/operate/transfer facility
and becomes the property of the state after 42.5 years.
Under that kind of arrangement, the people of Virginia
get a road financed through tolls, not taxes, that is built
sooner than otherwise would have been the case. The
developers receive the profits (assuming that the market
eventually provides profits) for a long enough period to
recoup their investment. Virginia's State Corporation
Commission limits the rate of return on the project to
18 percent, but profits appear unlikely to approach that
level anytime soon.
10. By way of comparison, the 10-mile Dulles Toll Road, with which the
Greenway connects, has a maximum toll of 85 cents.
11. Peter Pae, "Agreement Buys Time for Dulles Greenway," Washington
Post, May 29, 1997, p. A24. For an earlier report, see Peter Pae,
"Struggling Dulles Greenway to Raise Toll," Washington Post, No¬
vember 2, 1996, p. B5.
California State Route 91 Express Lanes. The sec¬
ond example of a road financed with private equity is a
10-mile toll section of State Route 91 (SR 91), the Riv¬
erside Freeway, in southern California. The toll por¬
tion, which is known as the 91 Express Lanes, is in the
median of the freeway. It is separated from other traffic
by a buffer zone.
The 91 Express Lanes project was developed under
a program authorized by the California legislature in
1989. The developer and operator is the California Pri¬
vate Transportation Company (CPTC), a limited part¬
nership led by the large construction company Peter
Kiewit Sons. Other partners are a French toll road
company, Cofiroute Corporation, and a large, locally
based construction company. Granite Construction.'^
The partnership raised $126 million in fmancing from
several sources: $65 million in variable-rate loans from
Citibank and two French banks; $35 million in a 24-
year loan from Cigna; $19 million in CPTC's equity;
and $7 million in subordinated debt to repay a local
agency's engineering and environmental studies.'^
Unlike the Greenway, the 91 Express Lanes are a
build/transfer/operate facility. Thus, on completion of
the project in 1995, the developer transferred owner¬
ship to the state. CPTC will operate the express lanes
for 35 years—and pay for maintenance, law enforce¬
ment, property taxes, and other operating costs. After
that period, control of the roadway reverts to the state.
The state does not directly regulate tolls, but it lim¬
its the company to a rate of return of 17 percent on the
project. The road has no toll booths to impede the flow
of traffic; instead, tolls are collected electronically and
vary by the time of day. The operators initially consid¬
ered varying the tolls instantaneously depending on the
flow of traffic. For example, if too many vehicles en¬
tered the express lanes, the toll would rise to moderate
the demand and prevent the toll portion from becoming
congested. However, research into attitudes about real¬
time adjustments in tolls revealed that commuters
wanted to know the cost before they started their trips.
To accommodate them, planners developed a toll struc-
12. Peter Samuel, "Highway Aggravation: The Case for Privatizing High¬
ways," Cafo/nsWuie PoZ/cy AnaZyj/s, no. 231 (June 27, 1995), p. 15.
13. Data supplied to the Congressional Budget Office by Gerald S. Pfeffer,
Senior Vice President, United Infrastructure Company, in March
1997.
40 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Table 5.
Motivation, Contributions, and Risks of Partners in Public/Private Ventures in Highway Financing
Motivation
Contribution
Risks
State
Speed up road construction
without having to issue more
debt
Power of eminent domain;
authority to create an entity
to issue tax-exempt debt
Private partner may seek
added contribution later, or
social benefits might prove to
be less than social costs
Private Developer Prospect of profit under the
construction contract; earlier
availability of the contract for
a large project
Expertise; pays for design
studies; assumes the risk of
a fixed-price construction
project
Costs of construction may be
underestimated and thus
money lost on fixed-price
contracts
Nonprofit Entity Management fees or fees for
bond issuance, or both
Bond Buyers
Attractive return on invest¬
ment in an asset
Authority to issue tax-exempt
debt without being subject to
state limits
Cash
Project may be terminated
and money spent on devel¬
opment lost
Revenues may be insuffi¬
cient to make interest pay¬
ments and repay debt
SOURCE: Congressional Budget Office.
ture with relatively low tolls in the middle of the night,
higher tolls at peak hours, and a series of steps leading
up to and down from the peaks.
Unlike the Greenway, the express lanes are adja¬
cent to a heavily congested highway. With a ready-
made demand from the day the lanes opened, projec¬
tions of revenues were less uncertain than they were for
the Greenway. As of February 1997, more than 80,000
vehicles had been equipped with electronic transpond¬
ers to pay tolls automatieally.''' Average weekday traf¬
fic in 1996 was 25,000 vehicles, and 20 percent or
more of peak-hour traffic used the express lanes.The
lanes have proved so popular that the road's operators
raised tolls twice in 1997 with little loss of business.'®
14. Ibid. The same devices are also accepted for payment on three pub¬
licly funded toll roads in Orange County.
15. California Private Transportation Company, Î996 Annual Report
(Anaheim, Calif; CPTC, 1996), p. 1.
16. "91 Express Lanes for Sale?" Public Works Financing, vol. 110 (Sep¬
tember 1997), p. 3.
Public/Private Partnerships
Several new or proposed highway projects are hybrids
built by partnerships between the public and private
sectors. What makes public/private partnerships attrac¬
tive to the participants? For state and local govem-
ments, they aehieve a quicker result than would be the
case if the majority of funding had to come from con¬
strained government budgets (see Table 5). Moreover,
the debt issued by the partnership is not considered
debt of the state. It is not backed by state tax revenues
and consequently does not jeopardize the state's ability
to issue bonds for other purposes.'^ Debt repayment is
typically through revenues from tolls, although the state
may use tax revenues to enhance the quality of the
credit or to cover other expenses. Bond buyers volun¬
tarily purchase bonds on the basis of the contribution
they expect the bonds to make to their portfolios, con-
17. A state's ability to issue bonds may be constrained by constitutional or
legislative limits on debt, how taxpayers view debt, and how the finan¬
cial community evaluates the state's ability to repay what it has bor¬
rowed.
CHAPTER nVE
PRIVATE-SECTOR PARTICIPATION 41
sidering returns, risk, diversification, maturity, tax sta¬
tus, and other factors.
As for the private participants, one must assume
that they are in the partnership to make money. Al¬
though firms may also be motivated by the desire to
produce goods or services that provide social benefits
to the community, they must keep their shareholders
happy, generally by earning a reasonable return on in¬
vestment.'^ On the plus side, that focus on returns
means private firms will probably scrutinize potential
investments carefully and not undertake projects that
appear unlikely to achieve an attractive payoff. Private
firms are also likely to be quite cost-conscious. On the
minus side, the focus on making money may cause the
interests of private firms and the general public to di¬
verge. Consequently, governments must be on their
guard and must structure agreements with private firms
to provide incentives to do what is good for the public
along with disincentives to waste resources.
Two models of public/private sponsorship of high¬
way projects have emerged: a special-purpose public
agency and a private, nonprofit corporation. Under
both forms, tax-exempt financing is available, and
sponsors provide general oversight of the project.
Initiatives that have used public-agency sponsor¬
ship are the E-470 toll road in Colorado and a group of
toll roads in Orange County, California, run by the
transportation corridor agencies.
Colorado E-470. One of the first proposals for a pub¬
lic/private partnership was the E-470 toll road on the
east side of Denver. The history of the project goes
back to 1987. In that year, the state of Colorado passed
the enabling statute that permitted the city of Aurora
along with Douglas, Arapahoe, and Adams Counties to
form an authority through an intergovernmental agree¬
ment. The jurisdictions established the E-470 Public
Highway Authority to finance, build, and operate the
highway.'® Its board of directors is composed of
18. Of course, stockholders have different ideas about what is reasonable.
Some want a high rate of return and are willing to assume a high risk;
others are willing to forgo a high rate of return to preserve a stable
value of capital.
19. Lehman Brothers and George K. Baum & Company, "E-470 Public
Highway Authority, Capital Improvement Trust Fund Highway Reve¬
nue Bonds" (remarketing statement, August 22, 1995), p. 3. David
Klinges, Senior Vice President of Lehman Brothers, provided addi¬
tional information for this section.
elected officials from the governmental jurisdictions
through which the 34-mile road will pass. The author¬
ity entered into a design/build construction contract
with a wholly owned subsidiary of the Morrison
Knudsen Corporation, Platte River Constructors. In
addition to assuming the risks of a fixed-price construc¬
tion contract, Platte River also contributed $16 million
in return for second-tier subordinate bonds (bonds on
which interest is paid only after more senior bondhold¬
ers are satisfied).^"
The E-470 project consists of four segments. Seg¬
ment 1, a 5.3-mile stretch between Interstate 25 (1-25)
on the south side of Denver and Parker Road, opened to
traffic in June 1991. Segments 2 and 3, which extend
about 29 miles Ixom Parker Road to 120th Avenue on
the east side of Denver, are scheduled for completion in
June 1999, with some portions to be opened to traffic in
June 1998. A fourth segment, which would fill the 12-
mile gap between 120th Avenue and 1-25 on the north
side of Denver, is not part of the project as currently
defined.^' However, in 1997, the authority obtained
additional financing to begin the initial design work on
Segment 4 and acquire the necessary rights-of-way.
The E-470 project was initially undertaken by
Arapahoe County, which issued bonds to finance the
road in 1986. Shortly thereafter, the project ran into
problems when a recession hit the local economy, re¬
ducing projections of toll revenue, and citizens voted
against using taxes to make up for shortfalls.^^ Their
action led to the creation in 1988 of the E-470 author¬
ity, which assumed primary responsibility for the bonds
marketed in 1986 as well as for getting the road built.
In 1995, the authority remarketed $654 million of
the 1986 bonds and released the proceeds for construc¬
tion of Segments 2 and 3. Of the total, $50.8 million in
bonds is backed by revenues from motor vehicle regis¬
tration fees, and $587.6 million in senior bonds is
backed by net revenues from tolls. Neither type of
20. The contract holds Platte River Constructors liable for liquidated dam¬
ages for each day of delay in completing the project, except in the case
of certain "force majeure" events including delays associated with
abnonnally inclement weather conditions, endangered species, and
hazardous substances.
21. Lehman Brothers and George K. Baum & Company, "E-470 Public
Highway Authority," p. II-l.
22. Federal Highway Administration, Implications of Changes in Proce¬
dures and Laws to Advance Public-Private Partnerships, FHWA-
PL-95-026 (April 30, 1995), p. 5.
42 INNOVATIVE FINANCING OF HIGHWAYS; AN ANALYSIS OF PROPOSALS
January 1998
bond is backed by tax revenues. (The remaining $16
million in bonds was sold to Platte River Constructors,
as noted earlier.) The interest is not subject to federal
or Colorado ineome taxes, aceording to bond counsel,
nor is it subject to the federal alternative minimum tax.
In addition to the bond issues, the project expects
to receive loans from the state and local governments.
The Colorado Department of Transportation has agreed
to lend up to $20 million (subject to armual appropria¬
tions) to match contributions from local governments.
The authority also has the power to assess and collect
fees from developers of land within 1.5 miles of the
center line of the highway.^^ For businesses, the fee is
based on square footage and distance to an interchange;
for residential property, the fee is based on an estimate
of relative use of the projeet by residents.
Such charges, whieh are known as development
impaet fees, are a common tool for financing new high¬
ways. The one-time charges are generally levied on
both residential and nonresidential development within
the established area of benefit of a roadway.
Orange County, California, Transportation Corri¬
dor Agencies. In 1986, the California legislature au¬
thorized local governments to create "joint-powers"
agencies with the right to fmance and build roads and,
collect tolls and development impaet fees.^'' Orange
County responded by creating two transportation corri¬
dor agencies, the San Joaquin Hills TCA and the Foot¬
hill/Eastern TCA. The agencies consist of elected rep¬
resentatives from 15 cities and three supervisorial dis¬
tricts within the county.^^
The San Joaquin Hills TCA has built a 15-mile,
six-lane toll road linking Newport Beach and San Juan
23. Lehman Brothers and George K. Baum & Company, "E-470 Public
Highway Authority," p. 30. For some portions of the roadway, the
area subject to development impact fees extends 2.5 miles from the
center line.
24. Federal Highway Administration, Implications of Changes in Proce¬
dures and Laws, p. 32.
25. The information presented here is drawn largely from the TCAs' web
site (available at http://www.tcagencies.com/). Additional information
was provided in a briefing by TCA officials and their advisors on No¬
vember 25, 1996.
Capisfrano.^® About half of the corridor was opened to
traffic in July 1996, and the remainder was opened in
November of that year.^^ As each segment was fin¬
ished, the TCA transferred ownership of it to the state
highway system, along with the responsibility for oper¬
ation and maintenance. However, the TCA retains
ownership of the toll collection system until all debt is
retired.
The Foothill/Eastern TCA has opened 7.5 miles of
roadway and is working on another 44.5 miles. The
segment opened to traffic runs from Portola Parkway
North near Irvine to Antonio Parkway in Rancho Santa
Margarita. When completed, the Foothill corridor will
extend from North Irvine to Interstate 5 south of San
Clemente. The Eastern corridor connects State Route
91 and Irvine, where it splits into two legs, the eastern
one connecting with the Lagima Freeway south of Inter¬
state 5 and the western one merging with Jamboree
Road south of Interstate 5 in Irvine.^^
The San Joaquin Hills and Foothill/Eastern TCAs
have identical organizational structures, powers, and
staff, and they are involved in similar financing ar¬
rangements. They are separate agencies with separate
books, however, because they cover different geo¬
graphic areas and hence have different areas for which
they can levy development fees. The debt issued by the
agencies is also separate.
The two agencies combined have raised a total of
about $3.6 billion to cover project costs. About 77 per¬
cent of that financing is from bonds, 7 percent from
development impact fees, 9 percent from interest, 5
percent from the state, and 2 percent from other
sources.^' The bonds are nonrecourse bonds: bond¬
holders can look only to toll revenues, development
fees, and interest earnings for repayment. The bonds
are not backed by local or state government, although
they qualify as municipal bonds, the interest on which
26. San Joaquin Hills Transportation Corridor Agency, The Journey Be¬
gins (undated brochure).
27. TCAs' web site (available at http://www.tcageneies.com/).
28. Ibid.
29. Ibid.
CHAPTER nVE
PRIVATE-SECTOR PARTICIPATION 43
is exempt from federal ineome taxes. As discussed ear¬
lier, the federal government gave each of the transporta¬
tion corridor agencies a standby line of credit of $120
million, which enhanced the marketability of the bonds.
Current traffic and revenue projeetions suggest that the
TCAs will not have to draw on those lines of credit.
Development impact fees have played a key role in
financing the Orange County toll roads. The fees are
based on the number of trips on the toll roads that de¬
velopment is projected to generate. Geographic loca¬
tions close to the roads carry higher impact fees than
those that are farther away. For residential develop¬
ment, rates are higher for single-family houses than for
multiple-unit buildings. Commercial development fees
are based on square footage.
The development impact fees have provided seed
capital for the projects, a responsibility that private in¬
vestors have been reluctant to assume. Private inves¬
tors consider the initial stages of highway projects—
especially the time spent getting the necessary environ¬
mental permits—to be a risky period. They tend to shy
away from committing capital until the project gets all
the approvals it requires from government regulators.
Like the 91 Express Lanes project, the Orange
County toll roads are transferred to the state of Califor¬
nia once they are opened to traffic. The state, however,
gives the TCAs the toll franchise until the debt is paid
off (the bonds have 40-year maturities). The state as¬
sumes tort liability as well, but unlike the arrangements
for the 91 Express Lanes, it also assumes responsibility
for all operations and maintenance (except as related to
the collection of tolls).
Several other initiatives in states across the country
are using private, nonprofit entities rather than public
agencies to oversee the financing, building, and opera¬
tion of roadways.
South Carolina's Southern Connector. The South
Carolma Department of Transportation is working with
a nonprofit community association known as the Con¬
nector 2000 Association and a private development
team, the Interwest Carolina Transportation Group, on
a new toll project. The 16-mile, four-lane road, the
Southern Connector, will serve the Greenville area.
Most of the financing will come from tax-exempt bonds
issued by the association and backed by revenues from
tolls.^° The project will also receive $20 million from
the state; the association plans to use it for traffic stud¬
ies, valuation of rights-of-way, and financing of early
construction work.^' The state expects the federal gov¬
ernment to approve a loan of federal-aid funds to the
project as allowed under section 1012 of ISTEA.^^
Both the private and public sectors stand to gain a
great deal from this arrangement. In return for incur¬
ring the costs and risks of developing the project, the
development team will receive a substantial fee. The
state benefits because it gets the road built sooner than
would be possible with conventional financing and be¬
cause it will not be liable for any of the project's debt.
Two legal issues have delayed development, how¬
ever. The first question was whether the bond issue
required approval by county voters in a referendum.
The courts ruled that approval was not needed because
the relevant cost of the project was less than the amount
that would trigger the referendum requirement.^^ The
second issue was the constitutionality of a 1995 state
law that gave affected counties the power to veto trans¬
portation projects with statewide effects. In August
1997, the state supreme court ruled in favor of the proj¬
ect's sponsors, stating that the law represented an illegal
delegation of authority.^'*
The sponsors hope to complete their financing ar¬
rangements in January 1998. If they are successful, the
project will be the first to proceed under that model of
public/private sponsorship.
Minnesota Trunk Highway 212. The Minnesota De¬
partment of Transportation (Mn/DOT) has established
an Office of Altemative Transportation Financing to
explore opportunities available under ISTEA and the
National Highway System Designation Act of 1995.^^
30. "First '6320' Nonprofit Toll Road," Public IVorks Financing, vol. 110
(September 1997), p. 7.
31. "Good News in South Carolina," Public IVorks Financing, vol. 104
(February 1997), p. 8.
32. "First '6320' Nonprofit Toll Road," p. 6.
33. "Good News in South Carolina," p. 8.
34. "First '6320' Nonprofit Toll Road," p. 7.
35. Minnesota Department of Transportation web site (available at
http://www.dot.state.mn.us/).
44 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
In July 1995, Mn/DOT issued requests for proposals
for developing toll facilities. It received five responses
and in May 1996 selected one, a 20-mile highway run-
nmg from Eden Prairie to Cologne on the southern side
of the Twin Cities and known as Trunk Highway 212
(TH 212).^® Over the summer of 1996, Mn/DOT
worked on an agreement with the developers—a local
not-for-profit association, the 212 Community High¬
way Association, and private, for-profit firms led by
Interwest/DLR Group Infrastructure Corporation,
One of the ground rules under which the public/
private coalition operates is that a project can be vetoed
by any community involved with the road. In Septem¬
ber 1996, one of those communities exercised its veto.
As of December 1997, the sponsors were still holding
out hope that the opposition could be overcome. The
opposing community's primary concern appears to be a
"not-in-my-backyard" objection that is common to new
highways.^' Some complaints have also arisen about
subjectmg motonsts in one part of the metropolitan
area to tolls when other commuter corridors are toll-
free.
If backers of the TH 212 project can address the
concerns of its opponents, the highway is likely to be
Minnesota's first toll road. The plan is for the $220-
million project to issue tax-exempt bonds backed by
toll revenues. Again, that kind of innovative financing
would allow the road to be built sooner than it would be
if it had to rely on a conventional, pay-as-you-go ap¬
proach. As for the proposals that were rejected in the
first round, Mn/DOT left the door open to consider
modifications to them. First, however, the projects must
solve several problems that were responsible for their
initial rejection: lack of support by affected communi¬
ties, environmental concerns, and questions about the
projects' financial and technical feasibility.
Arizona. In 1991, the Arizona legislature enacted leg¬
islation authorizing four transportation projects to be
financed by the private sector. It amended the law in
1995 in an attempt to increase the likelihood that such
projects would be developed. Several toll roads have
36. "Minnesota is Set to Get its First Toll Highway," Engineering News-
Record, McGraw-Hill (May 27, 1996), p. 16.
37. Adee! Z. Lari, "Minnesota Toll Road Public-Private Partnership: The
TH 212 Preliminary Development Agreement" (presentation at the
aimual meeting of the Transportation Research Board of the National
Research Council, January 13, 1997).
been proposed but have faltered for lack of public sup¬
port.^® The most recent casualty is MetroRoad, a proj¬
ect in the Phoenix area that would have added express
lanes with tolls to congested East Valley freeways and
accelerated construction of the San Tan Freeway.
MetroRoad had been structured as a nonprofit corpora¬
tion and had plaimed to finance construction by issuing
tax-exempt bonds. In November 1997, the project's
sponsor, HDR, Inc., decided to withdraw its proposal
because of concerns about the project's financial viabil¬
ity. A key factor in that decision was the changes the
state had made in its plans for roads in the area.^®
One venture remains: the South Mountain Toll
Road, which will be about 25 miles long and provide an
altemative to Interstate 10 through Phoenix. The Inter-
west group is working with the Arizona Department of
Transportation on that $360-million effort.'*® To ad¬
vance, the project must undergo scrutiny at public hear¬
ings and secure approvals from the Arizona State
Transportation Board.
Virginia. In Virginia, the Public-Private Transporta¬
tion Act of 1995 authorizes the state government and
substate entities that meet certain qualifications to enter
into agreements with private firms to acquire, build,
improve, maintain, and operate qualifying transporta¬
tion facilities. Private sponsors may submit their pro¬
posals to the state, which reviews and considers them in
a process that is open to public scrutiny and competi¬
tive bidding.
One of the proposals submitted in response to the
act is a toll road between 1-95 and 1-295 south of Rich¬
mond. The Virginia Department of Transportation
(VDOT) began planning the link, known as the 1-895
connector, about 20 years ago. The Commonwealth
Transportation Board endorsed the nine-mile corridor
in 1983 and approved the major design features in Au¬
gust 1997. VDOT has allocated $12 million for de¬
signing the road. The construction and engineering
38. Federal Highway Administration, Implications of Changes in Proce¬
dures and Laws, p. 5.
39. HDR, Inc., "Update: MetroRoad Project Team Withdraws Project,"
November 12, 1997 (available at www/hdrinc.com/metroroad/
update.htm/).
40. "Allstate Sues interwest for Fraud," Public Works Financing, vol. 104
(February 1997), p. 6. The fraud referred to in the title refers to a
sewer project.
CHAPTER nVE
PRIVATE-SECTOR PARTICIPATION 45
firms of Fluor Daniel and Morrison Knudsen have
formed a partnership, FD/MK, to build the facility they
call the Pocahontas Parkway. The partnership is nego¬
tiating an agreement with VDOT that would establish a
not-for-profit association akin to those in South Caro¬
lina and Minnesota. The association would sell about
$300 million in nonrecourse tax-exempt bonds to pri¬
vate investors and borrow $15 million from Virginia's
infrastructure bank. The debt would be repaid through
tolls of up to $2. The proposed schedule calls for exe¬
cuting an agreement between the partnership and
VDOT in early 1998, beginning construction in late
1998 or early 1999, and opening the roadway in mid- to
late 2001.''
Another project that was proposed in response to
the Public-Private Transportation Act has been with¬
drawn.'^ The venture was a toll road on an uncom¬
pleted section of Route 288 between the Powhite Park¬
way Extension in Chesterfield County and 1-64 in
Goochland County. The project's sponsor was the
James River Parkway Associates, a joint venture of
Brown and Root Civil of Houston, a large engineering
and construction company, and Dewberry and Davis, an
engineering firm in Fairfax, Virginia. The group pro¬
posed building the 17.5-mile road as a limited-access
highway and unposing a toll of $2. The cost of the pro¬
ject was estimated at $320 million.
Tacoma Narrows Bridge, Washington. The Tacoma
Narrows Bridge on State Route 16 is the primary link
between the Seattle/Tacoma metropolitan area and the
Olympic Peninsula. In 1994, the Washington State
Department of Transportation, using authority provided
by the state legislature the previous year, selected the
United Infrastructure Company, a partnership of the
Bechtel Group and Peter Kiewit Sons, to develop plans
for improving the bridge.'^ The firm is conducting
technical, financial, and environmental studies of alter¬
native ways to relieve congestion.
41. Virginia Department of Transportation web site on its 1-895 project,
September 1997 (available at www.vdot.state.us/proj/895x.html/),
supplemented by additional information from VDOT.
42. The reasons for withdrawal were not made public by the project's
sponsors. Subsequently, VDOT announced that it would proceed with
the proj-ect on its own using state funding.
43. Statement of Gerald S. Pfeffer, Senior Vice President, United Infra¬
structure Company, before the Subcommittee on Transportation and
Infrastructure of the Senate Committee on Environment and Public
Works, March 6, 1997.
Toll roads have been controversial in Washington
State. Thus, what might in other instances have been
the approach of first choice —a self-supporting toll
bridge with enough profit potential to attract private
investment—may not be a viable option. Voters in the
affected area will have an opportunity to express their
views in an advisory election in the fall of 1998."
Contracting for Construction,
Operations, and Maintenance
Another form of private-sector participation occurs
when the government owns a toll-free road but eon-
tracts with a private firm to build, operate, and maintain
it. Typically, an agreement of that kind is designed to
impose some risks on the firm and thus create incen¬
tives to provide good service. Although "contracting
out" has found favor among state and local officials
when applied to trash collection and even public bus
service, contracting for road operations has not received
much attention in the United States.'^
In the 1980s, the British government began explor¬
ing such contracting to build, finance, and operate high¬
ways.'® Although the specifics of the model may vary,
in essence, the government gives a private entity the
franchise to build and operate a road and contracts to
pay the operators from general governmental revenues
according to the volume of traffic. The charges per ve¬
hicle are known as shadow tolls because users do not
see them (or pay them) directly. Indeed, from the
standpoint of users, the roads are as free as traditional
toll-free highways that are financed by taxes and owned
and operated by some level of the government.
44. Washington State Department of Transportation, "The Public Advi¬
sory Elections," SOLVE 16 Fact Sheet, no. 8 (available at
http://wvnv.wsdot.wa.gOv/solvel6/8.html/).
45. For discussions of contracting for municipal services, see John C.
Weicher, "Making Cents; Better City Services for Less," Outlook:
Ideas for the Future from Hudson Institute, vol. 1, no. 2 (February
1997). Privatization Watch, a monthly newsletter published by the
Reason Foundation in Los Angeles, is another source of information.
46. URS Consultants, Inc., in association with McDermott, Will & Emery,
The Applicability of Shadow Toll Concepts in the United States (pre¬
pared for the Federal Highway Administration, October 1995).
46 D4N0VATIVE FINANCING OF FHGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
The British government is working with private
contractors on several projects that use shadow tolls/'
Whether the British experience can provide a useful
model for the United States is not yet clear, however.
The numerous differences between Great Britain and
the United States (and its state and local governments)
in the institutional arrangements for projects sponsored
by government and the private sector may prevent a
direct transfer of approaches from one nation to the
other.
Evaluating Private-Sector
Participation in Highway
Financing
The approach to highway financing that taps private-
sector resources can be evaluated according to several
criteria: the amount of additional money that can be
generated to finance the projects, implications for using
resources efficiently, distributional consequences, ad¬
ministrative feasibility, acceptability to users and tax¬
payers, and effects on the federal budget.
Financing Potential
A primary motivation for encouraging private-sector
participation in building and operating roads is to at¬
tract additional sources of funds for investment. Pri¬
vate investors in a highway project provide capital at
the beginning of the venture and expect to recoup their
investment during the project's useful life. Thus, to the
extent that private participation occurs, resources can
be brought to bear sooner than with traditional pay-as-
you-go financing based on user taxes and fees. Be¬
cause private firms do not have the power to tax, they
must recover their investments through tolls or other
fees. Eventually, then, roads built with private funds
are paid for by future users (or by investors, if a project
fails to earn enough revenue to cover costs).
So far, the amount of private equity investment in
road projects has been limited. In recent years, it has
47. Gabriel Roth, Roads in a Market Economy (Brookfield, Vt. : Ashgate
Publishing Company, 1996), pp. 208-209.
been at most a few billion dollars annually, which is
dwarfed by the roughly $80 billion spent on roads by
federal, state, and local governments each year. And
private investment in roads will remain limited as long
as most highways are provided by government and fi¬
nanced with taxes rather than tolls. To be attractive to
private investors, a project must carry the expectation
of profitability. The number of projects that have the
potential to offer enough value to motorists that they
will be willing to pay tolls when toll-free alternatives
are available appears relatively small. Such projects
face substantial obstacles, which are discussed later in
this chapter.
Efficiency
Involving the private sector in building and operating
roads has potential benefits from the standpoint of eco¬
nomic efficiency. The profit motive leads private firms
to invest in projects that they think will be valued
highly enough by users to recoup costs. Private or pub¬
lic/private road projects generally charge tolls; if the
tolls are set at the right level, they can help allocate re¬
sources among those who are willing and able to pay
for them. Contracts between governments and private
providers can be structured to provide incentives for
keeping costs under control and to assign risks to the
party with the most information about and ability to
control them. The form of privatization that a highway
project embodies and the specifics of agreements be¬
tween private investors and governments have a bearing
on efficiency.
Efficiency of Investment. Before a private or pub¬
lic/private road project is undertaken, it is put to more
of a market test—in that it must attract investors—than
a public road is ordinarily subjected to. Potential inves¬
tors will be unwilling to take the risk of financing and
building a road for which the expected demand is insuf¬
ficient to recoup the investment. Yet a private road is
not unambiguously more efficient than a public one.
From the standpoint of society, there would probably
be underinvestment if all roads were private because
many road projects yield benefits beyond the amount
paid for by users.
In principle, benefit-cost analysis forms the basis
for selecting public road projects (although other fac¬
tors may influence those decisions as well). Carrying
CHAPTER nVE
PRJVATE-SECTOR PARTICIPATION 47
out such analysis requires estimating the demand for a
project, which is no easy matter in the case of a good—a
public road—whose price is zero. Users quite under¬
standably tend to want more of that good because they
do not perceive it as costing them anything. As a result,
highway offieials may overestimate the demand for
such projects, and public roads may be overbuilt—that
is, built larger than the actual demand might warrant.
However, mistakes in estimating demand are not eon-
fined to the public sector. Incorrect judgments about
either private or publie projects can lead to an ineffi¬
cient amount of investment, and given the characteris¬
tics of roadways, that investment is essentially sunk—
the resources cannot readily be put to alternative uses.
When private road builders are also responsible for
subsequent operations and maintenance, they have in¬
centives to build roadways that are designed to meet the
specific demands and characteristics of users. For ex¬
ample, if a highway is going to be used by heavy vehi¬
cles, a private firm that has to maintain as well as build
the road will probably ehoose a stronger, more durable
pavement. Private operators may also be more respon¬
sive than government to demands for better mainte¬
nance and smoother operations, since their incentive is
to attract more traffic and hence more revenues.
Efficiency of Use. As noted earlier, most private and
public/private road projects impose tolls, in contrast to
public roads that generally are toll-free and fmaneed
with tax revenues."*^ On the one hand, if the sponsor of
a project sets tolls at a level that reflects the marginal
cost of use, sueh charges can help allocate resources
efficiently. On the other hand, a toll set at an efficient
level may not raise enough in revenues to pay for the
investment. Unless a road is congested, the marginal
cost of use by an automobile is relatively low. Conse¬
quently, a toll equal to the marginal cost would be un¬
likely to yield enough revenues to cover the cost of the
investment."® If the toll was set high enough to recover
the cost, it would impede efficiency by discouraging
48. Even though many of those revenues come from taxes associated with
the use of roads, such as taxes on motor fuels, the taxes are not tied to
use of a specific road at a specific time. Therefore, they do not serve
the same pricing function as tolls.
49. One solution to that problem is a two-part tariff that combines a ilxed
fee and a marginal-cost price. That and other pricing options are dis¬
cussed in Congressional Budget Office, Paying for Highways, Air¬
ways, and Waterways: How Can Users Be Charged? (May 1992).
users who were willing to pay as much as the marginal
cost but not as much as the toll.
When a road is congested, each additional motorist
imposes costs related to delays on all other motorists.
By acting as price signals, tolls can moderate the de¬
mand for use of the road, especially when premium
rates are charged at peak hours. Sponsors of the 91
Express Lanes project in California designed its pricing
structure with that in mind. The aim of the road's oper¬
ators is to keep traffic flowing without delays. To do
so, they are trying to set tolls at levels that reflect dif¬
ferent demands at different times of the day and that
help to moderate demand during peak hours.
Shadow tolls do not serve as a price mechanism.
Instead, such charges make a road appear free to motor¬
ists, encouraging them to use or overuse it. As long as
the road is not congested, overuse poses no problem
because the marginal social cost of use is essentially
zero. However, shadow tolls, unlike tolls that a user
pays directly, do not discourage use of eongested road¬
ways.
Minimizing Costs. Advocates of privatization believe
that private fmns can operate less expensively than
governments. Part of their reasoning involves the in¬
centives faced by private fmns, and part is based on the
checks placed on governments. In many situations, pri¬
vate firms can react more nimbly than governments,
which are constrained by procurement regulations, em¬
ployment policies, and other red tape; hence, fmns can
be more responsive to users' demands. To provide in¬
centives for effieient behavior by private firms, a gov¬
ernment must design contracts with care to ensure that
they reward socially beneficial behavior and discourage
detrimental behavior. That is no easy task, as difficul¬
ties with other large-scale government procurements
show. Procurements for major weapons systems or
modem computer and communications systems for air
traffic control, for example, seldom proceed without
problems.
Handling Risks. Investments in road projects have
many uncertainties for private developers, who may
need to see a great potential for profits to induce them
50. For example, see General Accounting Office, Aviation Acquisition: A
Comprehensive Strategy is Needed for Cultural Change at FAA,
GAO/RCED-96-159 (August 1996).
48 INNOVATIVE FINANCING OF fflGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
to assume the financial and liability risks that accom¬
pany building and operating highways. With more
roads and resources, governments may be able to
achieve a reasonable overall level of risk by spreading
their risks among individual projects. As a government
designs contracts with private firms, it must ensure that
risks and returns are balanced and that it does not bear
all of the risks and leave all of the returns to the private
sector.
For example, who should be held liable if a design
flaw contributes to accidents? For the sake of effi¬
ciency, the party in a position to affect the outcome
should be held accountable because the liability will
induce that party to take appropriate preventive mea¬
sures. With respect to financial risks, a contractor that
is only building a roadway and is not responsible for its
subsequent upkeep and maintenance might stint on de¬
sign and materials to keep the initial cost low. In doing
so, however, the total cost of the road (including opera¬
tions and maintenance) may increase over the highway's
life cycle.^'
Distributional Effects
Private investment in roads shifts some or all of the
cost and risk from governments and taxpayers to pri¬
vate investors and users of the roads. Because invest¬
ment costs must be recouped with tolls or other user
fees and not with taxes, road projects financed by the
pnvate sector shift the eventual burden of costs to users
of the projects and away from other motorists. The
taxes and fees related to highway use that those other
motorists pay go to support other highways. Of course,
if a road built with private financing does not attract
enough users to cover its costs, equity investors and
bondholders are left to bear those expenses.
For projects that are structured as public/private
partnerships and that make use of debt financing, tax¬
payers at the federal level as well as those of the state in
which the project is located bear a share of the cost.
The costs are distributed in that way as a result of "lost"
revenues associated with the federal income tax exemp¬
tion of interest on municipal bonds. The exemption
51. Section 303 of the National Highway System Designation Act of 1995
established a program that requires states to analyze the life-cycle costs
of projects on the National Highway System with price tags of $25
million or more. The measure is codified at 23 U.S.C. 106(e).
helps agencies to market debt for their projects at a
lower rate of interest than would be possible if inves¬
tors had to pay income taxes on that interest.
The amount of tax-exempt debt issued for pub¬
lic/private road projects has been relatively small thus
far and the resulting loss of federal tax revenue negligi¬
ble. Still, from the standpoint of the federal govern¬
ment, the tax exemption is open-ended. If that model
of road financing became popular, the amount of reve¬
nue forgone would mount.'^ The loss of revenue is lim¬
ited, however, by two factors: the capacity of the mar¬
ket to absorb additional tax-exempt issues and the debt
ceilings established by state and local governments.
Hence, instead of reducing federal tax revenues, new
bonds for toll roads might crowd out bonds for school
construction or other public works.
Administrative Feasibility
Builders of roads—public and private—face a number
of obstacles, but those impediments are generally
greater for private firms than for governments In addi¬
tion to obtaining financing, sponsors of road projects
must acquire land for rights-of-way and the applicable
environmental permits. They also face potential tort
liability from accidents. That vulnerability adds to the
difficulty and cost of obtaining financing.
Such obstacles indicate that strictly private road
development—without any governmental involvement
—is exceedingly difficult. They also suggest that some
roles are better suited to government and some to the
private sector. In particular, the early stages of a proj¬
ect are probably best undertaken by public entities.
Enjoying the power of eminent domain, governments
face fewer obstacles to acquiring rights-of-way. And
obtaining environmental permits can be easier for gov¬
ernments than for private developers, although still a
daunting task. Without such permits, a project cannot
go forward. Thus, the risks to investors m the early
stages of a project are great and may be borne more
easily by governments. Specifically, if a government
body is undertaking a large number of projects, it can
52. The open-ended tax exemption would apply if bonds qualified as
public-purpose instruments. In contrast, federal law limits the volume
of bonds issued by state or local agencies for pnvate purposes. In ad¬
dition, interest on private-purpose obligations is subject to the alterna¬
tive minimum tax.
CHAPTER mVE
PRIVATE-SECTOR PARTICIPATION 49
spread the risks among them. However, governmental
resources are not unlimited, and consequently, govern¬
ments must focus on projects that have the best chance
of going forward.
In the case of shadow tolls, the government needs a
mechanism for monitoring the performance of a private
road contractor to ensure that the government is not
overpaying. One possibility is high-tech sensors, which
could help count vehicles. In addition, however, the
government would still need an effective way of audit¬
ing the accounts.
Acceptability
Will government officials, road users, and the general
public accept the idea of private or public/private road
projects? As the efforts described in this chapter sug¬
gest, federal, state, and local transportation officials are
interested in giving them a chance as a way of augment¬
ing governmental resources in road building. Yet the
lack of experience with private roads makes officials
cautioús about proceeding too quickly. For example,
state and local officials must be convinced that they will
not be left "holding the bag" if private ventures are un¬
successful. If a public/private road has design or con-
stmction flaws, the government may be held responsi¬
ble or forced to use highway funds of its own to repair
the problem. Therefore, government officials must be
assured that if they participate in a public/private proj¬
ect, they will not lose control over public resources.
They must also be confident that those resources will be
used productively and prudently.
For example, consider the kinds of issues that arise
regarding a public/private toll road that is due to be
turned over to the government for operation and main¬
tenance when it is opened to traffic. For one thing, the
government must anticipate the need for funding those
activities in the future. With relatively few such roads,
the costs should not present a large burden, but if nu¬
merous roads were built under those conditions, they
could impose a noticeable demand on the government's
budget for maintaining roadways. To decide whether a
build/transfer/operate agreement was worthwhile, of
course, the government would have to compare its costs
with the benefits of substituting private capital for pub¬
lic funds in highway construction.
The government also has an interest in seeing that
roads meet certain standards so that they do not crum¬
ble prematurely or otherwise impose greater burdens on
public budgets In that regard, operators of toll roads
share some of the same incentive because they want the
roads to be in good condition to attract motorists and
keep toll revenues high.
Private and public/private projects also need accep¬
tance by motorists. Most such projects will be sup¬
ported by tolls. But the motoring public is accustomed
to toll-free use of limited-access highways as well as
local roads and may object to paying tolls. Still, motor¬
ists always have an altemative; they can take the toll-
free routes that they used before the new roads were
built and avoid the problem of suddenly facing high
tolls.
Recent experience with toll roads suggests that mo¬
torists are willing to pay tolls if they see a clear bene¬
fit—such as having additional capacity available that
enables them to avoid congestion and save time getting
to their destinations. A recent study funded by the
Texas Department of Transportation found that the
public supported the use of tolls on newly built road¬
ways as an altemative to boosting fiiel taxes.^^ How¬
ever, most Texans in the survey were against imposing
tolls on roads that currently do not have them, and they
also opposed tolls as a mechanism for easing conges¬
tion. The last finding is surprising because the existing
toll roads in Texas are m the Houston and Dallas areas,
where higher tolls during peak periods could reduce
congestion and save travel time for commuters who
were willing to pay.
Some observers complain that toll roads are unfair
to motorists with low incomes who may not be able to
afford them, a concem that intensifies in considering
trips to work by motorists with few alternatives. Yet
the Texas study reported that the level of income of the
people surveyed did not affect their preference for fuel
taxes versus tolls, although people with higher incomes
tended to use toll roads more often than people with
low incomes. On the 91 Express Lanes in California,
the income distribution of users mirrors the overall in¬
come distribution in the area.
53. Chungwon Lee and others, "A Surrey Approach for the Acceptability
of Highway Tolling and Congestion Pricing in Texas," Journal of the
Transportation Research Forum, vol. 36, no. 1 (1996), pp. 43-58.
50 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Technology is helping to overcome one long-stand¬
ing complaint about toll roads: wasting time waiting in
line to pay the tolls. The new toll roads make use of
electronic toll collection, which reduces the costs of
collection and delays. Electronic tags placed in vehicles
can be read by roadside sensors and the amount of the
toll debited automatically.'" Although some observers
have expressed concerns about privacy, most motorists
do not find that a problem. One way of ensuring pri¬
vacy is to give motorists the option of paying a cash toll
at a traditional toll booth (with the traditional delays) or
buying a debit card that does not identify the user.
In the end, private and public/private road ventures
must be acceptable to voters and taxpayers. If private
funding substitutes for or augments public funding
from tax revenues, taxpayers would be likely to wel¬
come it.
Implications for the Federal Budget
Strictly private ventures in highway financing should
affect the federal budget no differently from any other
private investment. Tax effects associated with earning
income, taking tax credits or deductions allowed under
the tax code, and incurring capital gains or losses are
likely to be about the same as those of alternative in¬
vestments and thus would offset them.
How public/private ventures affect the federal bud¬
get depends on how they are structured. If, for exam¬
ple, the federal government provides loans, loan guar¬
antees, or lines of credit for a project, the budget will
record the risk of default and any interest subsidies as
federal outlays." Reduced tax receipts are another po¬
tential effect; the federal government can lose revenues
if a public/private partnership sells tax-exempt bonds
to investors who otherwise would have purchased tax¬
able bonds.
54. For additional discussion of electronic toll collection, see Congres¬
sional Budget OtTice, High-Tech Highways: Intelligent Transporta¬
tion Systems and Policy (October 1995).
55. For detailed discussions of the way the federal budget treats loans, loan
guarantees, and lines of credit, see Congressional Budget Office, An
Explanation of the Budgetary Changes Under Credit Reform, CBO
Memorandum (April 1991 ) and Congressional Budget Office, Budget¬
ing for Administrative Costs Under Credit Reform (January 1992).
The Outlook for Private-
Sector Participation
What are the prospects for increased participation by
the private sector in road building? The projects de¬
scribed in this chapter suggest that both government
officials and private firms are interested in exploring
new ways to tap private-sector resources—^with the ex¬
pectation of reasonable returns on investment. The
projects also illuminate a number of obstacles that must
be overcome to make private or public/private ventures
successful.
Attraeting private investment requires the prospect
of profit. A key question in highway projects is
whether revenues from tolls or other user fees will be
sufficient to repay debt to bondholders and provide an
attractive return on investors' equity. The most promis¬
ing candidates for tolls in the near term appear to be
new roads or additional lanes on existing roads in con¬
gested eorridors, as m the case of California's 91 Ex¬
press Lanes. (Motorists are more willing to pay for
tolls when the new capacity is readily identifiable. Oth¬
erwise, they may complain about having to pay for
something that previously was "free. ")
A study by J. P. Morgan Securities of 14 urban toll
roads financed over the past 12 years offers some in¬
sight into the types of projects that are most promising
and what sponsors look for in deciding whether to un¬
dertake such an effort.'® The projections of revenue
and traffic for most of the projects were overly optimis¬
tic. That finding may prompt potential lenders and eq¬
uity investors to scrutinize such projections with more
care and to require government funding or financial
guarantees to reduce the risk of investing, especially at
the earliest (and riskiest) stages of the project. In gen¬
eral, the projections that proved most accurate were
based on conservative assumptions about economic
activity in the traffie corridor served by the toll road.
Toll revenues were most likely to meet projections in
corridors that were already congested—that is, where
substantial potential demand already existed.
56. Robert H. Muller, "Examining Tollroad Feasibility Studies," A/umc/-
pal Market Monitor, J.P. Morgan Securities, Inc. (March 22, 1996).
CHAPTER nVE
PRIVATE-SECTOR PARTICIPATION 51
Finding new toll projects to interest private inves¬
tors may prove difficult. If state and local governments
have chosen highway projects according to their rates
of return on investment, it means they have already un¬
dertaken the projects with the highest net benefits,
which would leave only lower-priority projects that
would be less able to recoup costs through tolls. But
factors other than net benefits—such as the geographic
distribution of highways across a state—also enter into
states' priorities. Thus, some good candidate projects
may remain.
Highway funds in the near future may well remain
tight at the federal and state levels, reflecting higher
priorities for projects other than highways and a reluc¬
tance to raise taxes on motor fuels. If funding contin¬
ues to be constrained, governments may look increas¬
ingly to ways of raising money voluntarily from the
private sector (that is, not through taxes). From the
taxpayer's standpoint, letting private investors bear the
risk and financial burden of building new roads is at¬
tractive. Private investors can take advantage of a vari¬
ety of institutional arrangements for obtaining fimds,
usually involving equity or debt that is repaid through
toll or tax revenues in the future. Of course, to induce
private investors to invest in roads, the anticipated re¬
turns must exceed those of alternative investments.
The projects described in this chapter illustrate the
gradations in balancing the public and private sectors.
For example, the Dulles Greenway used only private
sources of funding (although it required state approvals
of environmental permits, toll levels, and other mat¬
ters). As noted earlier, the Greenway is a build/oper¬
ate/transfer facility; it will be operated and maintained
privately for 42.5 years and then turned over to the
state. In comparison, the 91 Express Lanes project is
also privately financed, but it follows a build/transfer/
operate model. (The private firm will operate the ex¬
press lanes for 35 years—and collect tolls to pay for
maintenance, law enforcement, and other operating
costs—after which all interest in the roadway will revert
to the state.)
The public/private ventures in Arizona, Minnesota,
Virginia, and South Carolina make use of nonprofit
entities set up by private developers and the govern¬
ment to serve as sponsors of the projects. The entities
can issue debt that is exempt from federal income
taxes, hence, they can borrow at interest rates that are
lower than the rates private firms can generally obtain.
Federal taxpayers help to subsidize those projects as a
result of the tax exemptions. The tax effect is the same
for the pubhc/private ventures operated by public agen¬
cies (for example, the E-470 authority in Colorado and
tlie Orange County, California, transportation corridor
agencies).
Even the toll roads with the least amount of govern¬
ment participation entail some involvement by govern¬
ment. Acquiring land for rights-of-way is virtually im¬
possible without tlie right of eminent domain. And the
strong opposition to new highways that frequently re¬
sults from environmental concerns may also encourage
investors to seek a partnership with government. From
the standpoint of private developers, the difficulties and
risk involved in obtaining environmental permits may
lead many participants in road projects to require some
kind of governmental backing in the early stages of a
venture. Backing could be in the form of financing or
loan guarantees for the planning and initial engineering
phases. Once a project receives the needed permits and
the fmancial risk drops substantially, private invest¬
ment (either equity or debt) will then be more attractive.
Another large risk for investors is tort liability.
Accidents involving deaths, injuries, and damage to the
environment (as might happen in multivehicle collisions
or accidents with trucks that are carrying hazardous
materials) may result in sizable financial losses. Pri¬
vate investors may also face financial risks when the
state regulates the level of tolls or the rate of retum on
investment.
Under the right circumstances, private investment
can augment public investment in transportation. The
challenge for policymakers is to prevent public/private
partnerships from becoming a proposition in which the
public sector pays while the private sector profits.
Appendixes
Appendix A
Highway Financing:
Sources and Trends
Traditionally, highways have been financed by
user taxes and fees at the federal and state lev¬
els. Although states have made some use of
debt financing, the bulk of highway spending has been
funded on a pay-as-you-go basis in which annual
spending roughly equals annual revenues. In 1994,
federal, state, and local governments together raised
nearly $91 billion in revenues that were used for high¬
way purposes.' Of that total, 56.5 percent came from
user tax revenues (primarily from fuel taxes), and 4.2
percent came from road and crossing tolls (see Table
A-1). The remainder was derived from general sources.
Highway Financing at
the Federal Level
Funding provided by the federal government for high¬
ways comes from taxes imposed on highway users.
Those taxes flow into the federal Highway Trust Fund.
From there, the government apportions funds to states
according to complicated formulas and subject to an¬
nual limits imposed in the Congressional appropriation
process.
The primary source of federal revenues for high¬
ways is taxes on motor fuels (see Table A-2). In 1995,
federal motor fuel taxes provided $15.7 billion for
highways.^ Those taxes also brought in $2.2 billion for
1. Federal Highway Administration, Highway Statistics ¡995 (January
1997), Table HF-1, p. lV-13.
2. Federal Highway Administration, Highway Statistics 1995, Table FE-
210, p. IV.20,
mass transit and $7.7 billion for the general fund of the
U.S. Treasury. Excise taxes on tires, trucks, buses, and
trailers provided $2.4 billion. The heavy-vehicle use
tax, which is imposed atmually on vehicles whose gross
weights exceed 55,000 pounds, raised $682 million.
Highway Financing at
the State Level
At the federal level, highway users are the source of all
revenues that go to fmance highways. Revenues from
highway user taxes go primarily for highways but also
for mass transit.^ In contrast, state and local govern¬
ments fmance some of their highway needs with general
funds, and many use debt financing in addition to cur¬
rent taxes, tolls, and fees. Some revenues from state
highway user taxes fmance mass transit projects, and
some are used for general purposes.
Motor Fuel Taxes
All states and the District of Columbia impose taxes on
motor fuels. In 1995, tax rates on gasoline ranged from
7.5 cents a gallon (in Georgia) to 34 cents a gallon (in
Connecticut), with a weighted average of 18.5 cents a
gallon. (The average is very close to the federal rate.)
The tax rates on diesel fuel ranged from 7.5 cents a gal-
3. From December 1, 1990, to September 30, 1995, 2.5 cents per gallon
of federal taxes on motor fuels went to the general fund to reduce the
deficit. From October 1, 1993, to September 30, 1997, an additional
4.3 cents per gallon went to the general fund.
56 INNOVATIVE FINANCING OF HIGHWAYS; AN ANALYSIS OF PROPOSALS
January 1998
Ion (in Georgia) to 29 cents a gallon (in Rhode Island),
with a weighted average of 19.8 cents a gallon. Most
states also taxed gasohol and other motor fuels.
In 1995, the states grossed $26.8 billion from mo¬
tor fuel taxes. Not all of those receipts went for high¬
way purposes, however. Some went for collection ex¬
penses, some for mass transit, and some for general
purposes, leaving $24.1 billion for highways (see Table
A-3).
Registration, Titling, and Other Fees
and Taxes on Highway Users
In addition to taxing fuel, states impose fees for titling
and registering vehicles and for operator's licenses.
Some states also impose taxes on motor carriers.
Those fees and taxes provided $20.5 billion in receipts
in 1995. Compared with receipts for fuel taxes, a much
larger share of these receipts went for nonhighway pur¬
poses. About 25.0 percent went for general purposes,
13.4 percent for collection expenses, and 4.3 percent
for mass transit, leaving 57.3 percent ($11.8 billion) for
highways.
Tolls
In 1995, 29 states collected $3.8 billion in toll reve¬
nues. Most of the revenues ($3.5 billion, or 92.5 per¬
cent) were designated for highway purposes. About 5.1
percent went for mass transit, and 2.4 percent went for
general purposes.
Bonds
Many states finance some of their highways or bridges
through bonds. At the end of 1995, all but 11 states
had debt outstanding for highways and bridges. Their
combined indebtedness (including that of the District of
Columbia) totaled $37.4 billion.
In 1995, states issued bonds with a total of $4.3
billion in par value to finance highways and bridges."
The bonds are to be repaid from a variety of sources.
About 13.1 percent of the debt is to be repaid through
motor fuel taxes; of that amount, about two-thirds is
backed by the full faith and credit of the state (see
Table A-4). States plan to cover 23.4 percent of the
debt by using tolls. Another 45.8 percent is to be cov¬
ered by highway user revenues, including fuel taxes,
tolls, motor vehicle titling and registration fees, and
other taxes and fees paid by highway users. Of that
amount, about two-thirds is further backed by the full
faith and credit of the state. Those sources of repay¬
ment together amount to 82.4 percent, indicating that
future users of highways and bridges will pay for them
—that is, unless the taxes and fees are insufficient. In
that case, general revenues will cover bond issues that
are backed by the full faith and credit of the state, and
bondholders will bear the burden of the rest of the non¬
payment. The remainder of the debt, 17.6 percent of
the total, is to be repaid from general revenues and is
backed by the full faith and credit of the state.
Aid from Other Levels of Government
States receive intergovernmental transfers from both
the federal and local levels of government. In 1995, the
federal government provided about $18.0 billion in aid
to the states for highways. Local governments paid
$1.2 billion to state governments for highways (see
Table A-5).
Trends in State Financing of Highways
In the decade from 1973 to 1983, state revenues for
highways nearly doubled (in nominal terms). From
1983 to 1993, they nearly doubled again. The sources
of funding remained much the same, although revenues
from motor fuel taxes dropped from 41.5 percent of the
total in 1973 to 32.2 percent in 1983, and then edged
up to 35.0 percent in 1993. Much of that difference
was made up by a variety of general fund sources.
4. The $4.3-billion figure excludes debt issued by toll authorities and
other transportation agencies as well as that issued by municipalities.
APPENDIX A
HIGHWAY FINANCING: SOURCES AND TRENDS 57
Highway Financing at
the Local Level
In 1994, local governments received $35.4 billion in
revenue that was used for highways (see Table A-6).^
Nearly 60 percent of the money came from general
sources, including property taxes, special assessments,
and other general fund revenues. Twenty-five percent
came in the form of aid from state governments; most
of those funds were from taxes and fees imposed on
users of highways. About 5 percent was from user rev¬
enues, including motor fuel taxes, motor vehicle fees,
and tolls. Just under 10 percent of the money came
from bond proceeds.
5. Federal Highway Administration, Highway Statistics 1995, Table
LGF-l, p. IV-99. Statistics for local governments lag behind those of
the federal and state governments by one year.
Lxxal governments had $23.7 billion in outstanding
obligations for highways at the end of 1994. Original
issues of bonds for that year totaled nearly $5 billion.
Compared with 20 years ago, local governments
today receive far less assistance (on a percentage basis)
from the federal and state governments. Although state
aid for highways increased in nominal terms from $1.8
billion in 1973 to $10.4 billion in 1993, its share of the
total fell from 43.5 percent to 26.6 percent. Aid from
the federal government rose from $295 million to $815
million over that period, but its share dropped from 7.5
percent to 2.3 percent. Local governments made up the
difference, primarily through general fund sources.
They also increased revenues from users and from
bonds.
In summary, local highways receive much more
support from general sources than from taxes and fees
on highway users. That trend contrasts sharply with the
federal and state governments. At those levels, all or
most of the funding for highways is from users.
58 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 19
Table A-1.
Types of Revenue Used for Highway Financing, 1994
Category In Billions of Dollars As a Percentage of Total
Highway User Taxes
Federal
15.83
17.3
State
34.39
37.7
Local
1.33
1.5
Subtotal
51.55
56.5
Road and Crossing Tolls
3.84
4.2
Appropriations from General Funds
12.43
13.6
Property Taxes
4.83
5.3
Other Imposts
4.33
4.7
Miscellaneous Receipts'
7.03
7.7
Bond Receipts"
7.30
8.0
Total
91.31
100.0
SOURCE: Federal Highway Administration, Highway Statistics 1995 (January 1997), Table HF-1, p, IV-13, as updated on January 10, 1997.
a. Includes interest earned on Highway Trust Fund reserves.
b. Exciudes short-term notes and refunding bond issues.
APPENDIX A
HIGHWAY FINANCING: SOURCES AND TRENDS 59
Table A-2.
Revenues Flowing to the Highway Trust Fund, 1995
As a Percentage As a Percentage
In Billions ofTotal, of Total,
Category of Dollars Excluding Interest Including Interest
Motor Fuel Taxes
To highway account
15.74
74.9
70.9
To transit account
2.19
10.4
9.9
Excise Taxes on Tires®
0.40
1.9
1.8
Excise Taxes on Trucks, Trailers, and Buses®
2.01
9.6
9.1
Heavy-Vehicle Use Tax®
0.68
3.2
3.1
Subtotal
21.02
100.0
94.7
Interest
To highway account
0.55
n.a.
2.5
To transit account
0.62
n.a.
2.8
Subtotal
1.17
n.a.
5.3
Total
22.19
n.a.
100.0
SOURCE: Federal Highway Administration, Highway Statistics 1995 (January 1997), Table FE-210, p. IV-20.
NOTE: n.a. = not applicable.
a. Revenues from these taxes go to the highway account only.
60 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Table A-3.
Disposition of Revenues from State Highway Users, 1995
Category
In Billions of Dollars
As a Percentage of Total
Motor Fuel Revenues Used for;
Highways
24.08
90.0
Mass transit
1.40
5.2
General purposes
1.07
4.0
Collection expenses
0.21
0.8
Total
26.76
100.0
Motor Vehicle Revenues Used for:
Highways
11.75
57.3
Mass transit
0.88
4.3
General purposes
5.13
25.0
Collection expenses
2.76
13.4
Total
20.52
100.0
Toll Revenues Used for:
Highways
3.49
92.5
Mass transit
0.19
5.1
General purposes
0.09
2.4
Total
3.77
100.0
All Revenues
51.04
n.a.
SOURCE: Federal Highway Administration, Highway Statistics 1995 (January 1997), Table SDP, p. IV-63.
NOTE: n.a. = not applicable.
APPENDIX A
HIGHWAY FINANCING: SOURCES AND TRENDS 61
Table A-4.
Revenues Available for Repaying State Highway Bonds
Source
In Millions of Dollars
As a Percentage of Total
Tolls
1,006.2
23.4
Motor Fuel Taxes
Backed by the state's full faith and credit
Not backed by the state's full faith and credit
Subtotal
381.0
183.2
564.2
8.9
4.3
13.1
Highway User Revenues'
Backed by the state's full faith and credit
Not backed by the state's full faith and credit
Subtotal
1,363.0
605.0
1,968.0
31.7
14.1
45.8
General Revenues
756.7
17.6
Total
4,295.1
100.0
SOURCE: Federal Highway Administration, Highway Statistics 1995 (January 1997), Table SB-1, pp. IV-82 to IV-83.
NOTE: The numbers in the table represent obligations Issued or assumed during 1995.
a. Some states combined all highway user revenues and did not report fuel taxes and tolls separately.
62 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS January 1998
Table A-5.
Sources of State Financing for Highways, 1995
Source
In Billions of Dollars
As a Percentage of Total
Revenues from Highway Users
Motor fuel taxes
24.08
35.1
Motor vehicle and motor carrier taxes
11.75
17.1
Road and crossing tolls
3.49
5.1
Subtotal
39.32
57.4
Revenues from General Sources
Appropriations from general funds®
1.61
2.3
Other state imposts
1.82
2.7
Miscellaneous
1.91
2.8
Subtotal
5.33
7.8
Bond Proceeds
Original issues
4.32
6.3
Refunding issues
0.35
0.5
Subtotal
4.67
6.8
Payments from Other Governments
Federal Highway Administration
17.59
25.7
Other federal agencies
0.46
0.7
Local government payments
1.16
1.7
Subtotal
19.21
28.1
Total
68.53
100.0
SOURCE: Federa! Highway Administration, Highway Statistics 1995 (January 1997), Table SF-1, p. iV-65.
a. This amount represents gross general fund appropriations for highways minus the amount of highway user revenues placed in state general fundi
APPENDIX A fflGHWAY FINANCING: SOURCES AND TRENDS 63
Table A-6.
Sources of Local Financing for Highways, 1994
Source
In Billions of Dollars
As a Percentage of Total
Revenues from Highway Users
Motor fuel and motor vehicle taxes
1.33
3.8
Road and crossing toils
0.54
1.5
Subtotal
1.87
5.3
Revenues from General Sources
Appropriations from general funds
9.64
27.2
Property taxes and special assessments
4.83
13.7
Other local imposts
2.03
5.7
Miscellaneous
4.28
12.1
Subtotal
20.78
58.7
Bond Proceeds
Original issues
3.11
8.8
Refunding issues
0.12
0.3
Subtotal
3.22
9.1
Payments from Other Governments
State highway user imposts
7.46
21.1
Other state payments
1.38
3.9
Federal Highway Administration
0.28
0.8
Other federal payments
0.39
1.1
Subtotal
9.51
26.9
Total
35.38
100.0
SOURCE: Federal Highway Administration, Highway Statistics 1995 (January 1997), Table LGF-1, p. lV-99.
Appendix B
The Matching-Share Requirement
Designing a program of federal aid to state or
local governments is a complicated endeavor.'
Even when federal policymakers are in com¬
plete agreement about the objectives of a program,
structuring it so that it does not have unintended and
undesired consequences can be a challenge. The debate
leading to the recent overhaul of federal welfare policy
is an example of the difficulty of fashioning a program
that achieves certain goals without providing incentives
for people to undercut the program's purpose.
A key feature of the federal-aid highway program is
the matching-share requirement. In general, states must
match with their own resources the federal aid they re¬
ceive for highway projects. For most projects, the fed¬
eral government pays 80 percent and the state 20 per¬
cent of the costs. Several iimovative financing initia¬
tives have been designed to make that requirement
more flexible, which raises questions about the require¬
ment's purpose and effects.
1. A number of elements go into designing such a program. For compre¬
hensive discussions of federal aid programs, see J. Richard Aronson
and John L. Hilley, Financing State and Local Governments, 4th ed.
(Washington, D.C.: Brookings Institution, 1986); George F. Break,
Financing Government in a Federal System (Washington, D.C.;
Brookings Institution, 1980); Edward M. Grämlich, "Intergovernmen¬
tal Grants: A Review of the Empirical Literature," in Wallace E.
Gates, ed.. The Political Economy of Fiscal Federalism (Lexington,
Mass.: D.C. Heath and Company, 1977), pp. 219-240; Richard A.
Musgrave and Peggy B. Musgrave, Public Finance in Theory and
Practice, 4th ed. (New York: McGraw-Hill, 1984), pp. 537-545; and
Gates, The Political Economy of Fiscal Federalism.
Rationale for the
Requirement
Several federal grant programs—úicli-ámg the highway
program—require state o-' local governments to match
the federal funds pt ovided.'^ The usual rationale for a
match requirement is to ensure that states have an in¬
centive to provide an economically efficient level of
services in the area that the program eovers. A state
comparing the costs of a project with the benefits to its
own citizens might underinvest, ignoring the benefits to
people living in other states. To overcome that prob¬
lem, the federal government may want to subsidize
state spending on highways with tax dollars collected at
the national level.^ Theoretically, the federal share
should reflect the extent to which benefits of the pro¬
gram spill over to citizens of other states. (That is, the
relative shares paid by the federal and state govern¬
ments should reflect the relative benefits to out-of-state
and in-state users.) Another part of the argument for
requiring that states put some of their own money at
stake through a match is that it may give them incen¬
tives to pick their projects more carefully tlian they
would if the money were completely "free."
Judging the relative benefits of a project to in-state
and out-of-state motorists is not an easy task. One
2. According to the Council of Economic Advisers, most federal grant
programs do not require a match. See Council of Economic Advisers,
Economic Report of the President (February 1996), p. 114.
3. Federal spending on highways is financed by user taxes. Conse¬
quently, there is less of a distributional effect than if it were fmanced
by general tax revenues.
66 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
Study estimated that on an average interstate highway,
only 30 percent of the traffic was from out of the state.''
If that share held for all roads built with federal aid,
theoretically, the federal share of highway funding
should be 30 percent and the state share 70 percent.
The principal economic effect of a matching re¬
quirement is to change the price of the good or service
being subsidized relative to the prices of other goods
and services. Consider a program in which the federal
government pays an 80 percent share, as is the case for
much of the federal-aid highway program. The state
spends one dollar, which is matched by four dollars
from the federal government (as long as the spending
cap that applies to federal-aid funds has not been
reached), to buy five dollars' worth of highways. The
federal contribution makes highways appear relatively
cheaper than unsubsidized programs and is supposed to
induce greater spending on highways by the states.
In theoîw, programs that require matches should be
open-ended if tliey .are to mak e the most use of the price
effect (that is, the effect on the distribution of resources
that results from changes in the relative prices of high¬
ways and other goods). If a state or local government
reaches the amount needed to get the maximum federal
aid available under the cap, any additional dollar spent
by the state buys only that dollar's worth of goods or
services. The federal highway program does not adhere
to that design principle, however, because budgetary
control requires a ceiling on spending. Therefore, the
program combines a match and a cap.
In addition to a price effect, a matching-grant pro¬
gram has an income effect (that is, an effect on the dis¬
tribution of resources that results from a change in pur¬
chasing power). The additional federal aid leaves the
state with more money for other purchases—or for re¬
ducing taxes.
Expected Effects of a
Matching Grant
Since the late 1960s, when federal grant programs be¬
gan to proliferate, economists have tried to determine
the effects of federal grants on spending by state and
local governments. They have questioned how the
amount of aid affects state and local spending decisions
and tried to assess what role the structure of that aid
plays—that is, whether the aid is open-ended or capped;
whether it requires state or local governments to match
federal funds; and whether it comes in the form of a
grant, loan, or credit. Does federal aid entice states to
spend more money on a program, or does it substitute
for the money that the states would otherwise have
spent? Does aid cause states to reduce taxes or shift
money to other programs? Economists have advanced
various theories about the expected effects of federal
grants.
Government as Economic Agent
One line of reasoning views state and local govern¬
ments as rational economic entities that, like house¬
holds, make spending decisions based on their prefer¬
ences and income constraints. The median voter within
the community determines the amount of taxes and
government spending. The community would consider
an unrestricted federal grant (net of taxes needed to fi¬
nance the federal government) to be additional income
and would allocate it according to the median voter's
preferences. That hypothetical voter would probably
want to devote some of the grant money to additional
government spending and return some to taxpayers
through tax reductions.'
Even if the grant was for a specific purpose, the
recipient could use the money to replace its current
spending for that purpose (that is, out of its own funds)
and reallocate the money that was made available. If
one assumed that a state or local government's budget
was initially allocated so that at the margin, the benefits
4. Edward M. Grämlich, "How Should Public Infrastructure be Fi¬
nanced?" in Alicia H. Munnell, ed., Is There a Shortfall in Public
Capital Investment? Proceedings of a Conference Held at Harwich
Port, Massachusetts, June 1990, Conference Series No. 34 (Boston:
Federal Reserve Bank of Boston, 1990), p. 227.
5. For a discussion of the effect of accounting for federal taxes, see
J. Richard Aronson and Vincent G. Munley, "(Non) equivalence in a
Federalism; Dual Tax Shares, Flypaper Effects and a Leviathan," Pub¬
lic Choice, vol. 89 (1996), pp. 53-62.
APPENDIX B
THE MATCHING-SHARE REQUIREMENT 67
from the last dollar spent on each program were equal
to each other, spending all of the increase in income on
one program would probably not yield the most addi¬
tional benefits. Instead, a rational economic player
would distribute the additional income among various
goods and services and return some to citizens in the
form of reduced taxes. The exact distributional effects
would depend on the income elasticities associated with
the various government programs.®
Additional intergovernmental aid increases the in¬
come to the community. If the community had a rela¬
tively small economic base, it might use the additional
money to meet such basic needs as improving the safety
of its drinking water or repairing schools to comply
with fire codes. A wealthier community might already
be meeting such needs. Therefore, it would be able to
spend the additional resources on, for example, a sports
stadium.
Bureaucratic Model
Another line of reasoning, the "bureaucratic model,"
begins with the premise that governments do not func¬
tion like households. Rather, they are run by officials
who have their own preferences, one of which is to ex¬
pand the programs they administer, either because of a
commitment to those programs or because nmning a
larger program will give them more power, prestige,
and pay.^ Under this model, additional federal aid for a
given program area would probably be spent on that
area.
Fiscal Illusion
Still a third model theorizes that voters suffer from "fis¬
cal illusion" in that they fail to perceive the true mar¬
ginal price of public expenditures.® Money received
6. The income elasticity of a program is a measure of how much the de¬
mand for that program increases in response to an increase in income.
7. See, for example, William A Niskanen Jr., "The Peculiar Economics
oi Buremiciicy," American Economic Review, vol. 58 (May 1968),
pp. „93-305.
8. Pf- N. Courant, Edward M. Grämlich, and Daniel L. Rubinfeld, "The
Stimulative Effects of Intergovernmental Grants: Or Why Money
Sticks Where it Hits," in Peter Mieszkowski and William H. Oakland,
Fiscal Federalism and Graníí-;n-/liíJ (Washington, D.C.; Urban
Institute, 1979), p. 6.
fi-om the federal government appears to be free. There¬
fore, such aid is easier to spend than money raised from
local taxes (an example of the "easy come, easy go"
philosophy).
Interactive Model
Yet another set of theoretical models attempts to char¬
acterize the interrelationships among the many entities
involved in decisions about government spending. In
such models, which economist George Break calls in¬
teractive, "no single group dominates, there are numer¬
ous goals and instruments to be considered, and deci¬
sions must be made in spite of a high degree of uncer¬
tainty about the effects of different policy actions.'"
Interactive models place greater weight than the other
models on the specific institutional arrangements asso¬
ciated with a given federal grant program and the recip¬
ient community; as a result, interactive models offer
fewer generalizable predictions. One conclusion, how¬
ever, is tliat states might imd budgeting on an incre¬
mental basis mere feasible than using a grand zero-
based allocation, as suggested by tire govemment-as-
economic-agent models.
Empirical Findings
To answer the question of how states respond to federal
aid, researchers need to estimate how much states
would spend on a program in the absence of federal aid
and compare that with spending in the presence of fed¬
eral aid. Although the task sounds easy m principle, it
is more difficult in practice. In principle, one holds
constant all other factors to isolate the effect of federal
aid. But in the real world, other factors are continually
changing. Researchers use multiple regression analysis
to handle that problem. They set up models in which
state spending on a program is a function of numerous
variables, such as income, population, factors related to
demand for the program (for example, miles of high¬
way, number of registered motor vehicles, number of
vehicle-miles traveled), and the amount of federal
grants for the program in question and for other pro-
9. Break, Financing Government in a Federal System, p. 93.
68 INNOVATIVE FINANCING OF HIGHWAYS: AN ANALYSIS OF PROPOSALS
January 1998
grams. They then use statistical tests to determine the
size and significance of the variables.
Empirical results on the effect of federal grants in
general are mixed, and results with respect to highways
specifically are meager. Several econometric studies
have found evidence that state and local governments
treat federal grants differently from other additions to
their resources. Economists would expect rational enti¬
ties to allocate federal aid among their programs in the
same way they would allocate an increase in general tax
revenues. But both casual observation and empirical
evidence suggest that states tend for the most part to
spend federal aid on the intended program. That is,
grant money tends to "stick where it hits," giving rise to
the term "flypaper effect."" Nevertheless, research
findings are mixed: some find evidence of a flypaper
effect, whereas others do not. In instances in which an
effect is found, questions may arise about whether it is
real or whether it results from incorrect specification of
the econometric model or problems witli the data.'^
Very few studies have focused on the effects of the
federal-aid highway program on decisions about state
and local spending. The two most relevant studies are
of limited use because they apply to earlier versions of
the program before the Intermodal Surface Transporta¬
tion Efficiency Act of 1991 was passed. Nevertheless,
10. States, of course, must comply with the rules imposed by the federal
government on the use of federal aid. Education grants must be spent
on education, highway grants on highways, and so on. But over the
long run, states may come to expect to receive federal grants for spe¬
cific purposes. As a result, they may reallocate their general funds
under the assumption that federal aid will provide resources for some
activities and they can then devote a larger share of their general fimds
to unaided programs.
11. The term is attributed to Ar'Jiur Okun. See Grämlich, "Intergovern¬
mental Grants: A Review of the Empirical Literature," p. 226.
12. For a recent review of econometric studies of the effects of federal
grants, see James R. Hines Jr. and Richard H. Thaler, "Anomalies:
The Flypaper Effect," Journal of Economic Perspectives, vol. 9, no. 4
(Fall 1995), p. 219.
their findings are interesting, and they indicate how the
program as it is currently structured may affect states.
In a 1987 study of the non-Interstate part of the
highway program, Harry Meyers concluded that federal
aid largely displaces state spending on highways.'^ He
found that an additional dollar of federal aid would dis¬
place about 63 cents of state spending. That is, states
would use 63 cents of the federal dollar to substitute for
spending from their own resources, leaving a net in¬
crease in spending on highways of 37 cents associated
with the additional dollar of federal aid. The study sug¬
gested, however, that states allocated some of the sav¬
ings at the state level to local road projects.
In another study of the non-Interstate highway pro¬
gram published in 1974, Edward Miller compared the
amounts states spent on federal-aid highways with the
amount they would have had to spend to obtain the
maximum amount of federal aid.^'' Miller found that
only eight states and the District of Columbia were
within the limits of the caps. The rest of the states were
spending more than the amount needed to maximize
federal aid. For them, the marginal dollar spent on
highways was entirely from their own sources; it was
not matched by federal aid. Hence, no priee effect oc¬
curred at the margin. Those fmdings indicate that fed¬
eral funds were being used for what states would have
spent money on anyway, even in the absence of federal
aid. During the study period, 1959 to 1969, the federal
match was 50 percent for most of the program compo¬
nents that were the subject of the research. Whether
state spending decisions changed as the federal match
rose is a matter for future inquiry.
13. Hany G. Meyers, "Displacement Effects of Federal Highway Grants,"
National Tax Journal, vol. 40, no. 2 (June 1987), pp. 221-235.
14. Edward Miller, "The Economics of Matching Grants: The ABC High¬
way Program," National Tax Journal, vol. 27, no. 2 (June 1974),
pp. 221-229.
ISBN 0-16-049413-3
9 780160 494130
90000