Printer Friendly

Role of toll financing in the state programming process: models and results.

Interest among the states in toll financing has risen and waned over five decades. Changes in federal funding relative to needs and variations in key financial market parameters affecting the financial viability of toll financing largely account for the sometimes keen, but more often scant, interest in financing highway improvements from tolls.

Recent developments suggest a modest reversal in this attitude towards toll financing. Reasons for this change have been amply documented elsewhere.(1) Generally summarized, they include a deteriorating condition of the highway infrastructure under traditional methods of financing, an inability to cope with growing congestion through traditional supply strategies, and an increasing interest in privatization initiatives. Additionally, technological advances which enable better toll collection and administration and impose less traffic congestion have improved the relative attractiveness of toll financing. The last major barrier to possibly greater use of toll financing, a seven-decade federal policy discouraging toll roads, has been substantially lowered under the new Intermodal Surface Transportation Efficiency Act (ISTEA) enacted in December 1991.

The new federal act provides unprecedented flexibility to states and cities in the use of federal highway trust funds for highway and transit improvements. In the toll financing area, ISTEA permits the use of federal funds at a 50 percent matching rate on non-interstate intercity highways (80 percent on bridges and tunnels) for new construction or reconstruction. Furthermore, ISTEA allows states to use privatization or public-private partnership structures with federal aid in the building, financing, operating, and maintenance of the toll facilities.(2)

Table 1 compares various provisions of ISTEA with prior federal policies. Also included in Table 1 are the provisions of the Toll Road Pilot Program, enacted in 1987, which is described later. Clearly, federal policy towards toll highway projects has evolved to a posture of greater tolerance and flexibility. But the federal policy only sets the boundary conditions; the highway infrastructure activities themselves are largely undertaken by state transportation agencies.

How will states use the toll financing opportunities created by ISTEA? Answering this question first requires understanding how toll financing might fit into the planning and programming process in states, and what key factors influence the states to utilize a combination of toll financing and conventional user-based funding mechanisms, rather than relying solely on the latter. This paper provides an initial assessment of these issues by a combination of case studies and statistical models.


Previous studies related to toll financing have explored such aspects as privatization, economic and social policy issues, federal policy TABULAR DATA OMITTED issues, and technology, including the use of automatic vehicle identification (AVI) and electronic toll and traffic management (ETTM).(3) Others have examined the economic principles of toll pricing and the merits of congestion tolls.(4) However, due largely to federal policy that generally discouraged state DOTs from undertaking toll projects, there has been little work exploring the role of toll financing in the states' planning and programming process. Planning and programming literature suggests that the nature of financing plays a critical role in the commitment to and implementation of transportation projects.(5) Therefore, it is reasonable to infer that toll financing will also play a role in the planning and programming process as states take advantage of the opportunity and flexibility provided in ISTEA.

The research methodology in this study consisted first of detailed case studies of the projects in the eight states involved in the federal Toll Road Pilot Program, authorized by Congress in 1987 to determine if permitting the use of federal aid on toll projects would assist the states in addressing the nation's congestion and infrastructure problems. A survey instrument was prepared, pilot tested, and used in lengthy phone interviews with officials from the eight states. The Federal Highway Administration and the U.S. General Accounting Office were consulted for the names of key contacts in each state for the pilot projects. Generally, these individuals were the project chief engineers or individuals responsible for highway programming. In some states, a separate office had been established in the state DOT to administer the pilot program. In several cases, key contacts included individuals in state turnpike authorities, particularly on questions addressing electronic toll collection and toll pricing schemes, as well as in the state DOT. In addition, in some cases, representatives of the state's congressional delegation were also contacted to clarify issues.

The key facets of the programming process explored in the case studies from the eight pilot program states were as follows:

* How were the pilot program projects selected and how did they surface in the programming process?

* How was the project financing mix determined in each case, and how critical was the availability of federal aid and the federal matching ratio?

* What project modifications were made due to the toll nature of the project, particularly the use of advanced technology for toll collection and demand management?

* Did the projects receive higher priority in the state programming process over non-toll projects, and if yes, why and what were the benefits of fast tracking?

* If the federal restriction on use of federal TABULAR DATA OMITTED aid on toll projects were removed, are there other viable toll financing candidates in the state?

Analyses of the state responses indicate that the undertaking of toll-financed projects at a micro (individual project) level is a complex undertaking influenced by several factors. However, there are some general inferences which can be made and these are useful in assessing the direction which future federal and state toll policies might take.

Some of the factors which influence decisions to use toll financing or simple conventional user charges are common to all projects--they reflect the demand for highway projects and the financial capabilities of the state to meet this demand. These common factors were analyzed through an aggregate fifty-state statistical model designed to uncover those factors that will most influence the future role of toll financing. The statistical models utilized both published highway data and a data base collected as part of another study?(6) The combination of case studies and statistical models provides useful insights on what might happen under ISTEA with respect to toll-financed projects.


The Toll Road Pilot Program was created by the Surface Transportation and Uniform Relocation Assistance Act of 1987 (P.L. 100-17). The program marked a major turning point in federal toll road policy because it permitted, for the first time, the use of federal highway trust fund monies in the financing of new toll roads. It was limited to single projects (California was granted an exception in the 1988 Budget Act) in eight states. A ninth state, Colorado, was selected for the program but subsequently declined to participate. Eligible pilot program projects included construction of a new toll highway, bridge, or tunnel on any federal aid system except the Interstate system or reconstruction of any non-Interstate federal aid highway, bridge, or tunnel that would expand capacity. Table 2 shows the projects selected by each state, estimated project cost, and current status.

Other key statutory constraints on the program were that federal aid was limited to a maximum of 35 percent of project costs, each project must be publicly owned and operated, all toll revenue from the demonstration facility must be used on that facility (i.e., excess revenues cannot be directed to other transport improvements), and toll revenues may be used only for construction, reconstruction, operation, maintenance, and debt service.(7)

Toll mileage added under the pilot program did not increase a state's apportionment under any apportionment formula. Also, the '87 act did not add any new federal monies for the pilot program; if states were going to use federal aid on the pilot projects, that aid had to be taken from the state's regular apportionment.

As shown in Table 2, most of the projects are very large investments and most involve significant contributions from all three sources: toll revenues, federal aid, and state DOT funds. Six of the eight participating states currently have projects active and under way. West Virginia has linked the progress on its pilot project to completion by Pennsylvania of the southern extension of Pennsylvania's pilot project, and therefore it is now dormant. Progress on the Texas project may be resumed under the more favorable federal aid provisions of ISTEA.

Project Selection Process

Table 3 shows a summary of the key variables influencing project selection in each state under the pilot program. There are several important commonalities. All of the chosen projects have a long history of appearing on state candidate project lists and/or have had toll feasibility studies conducted prior to the inception of the pilot program.(8) This phenomenon indicates that the chosen projects are of relatively high priority, but other factors such as the cost of the project and availability of funds precluded their construction.


Congestion and traffic volume are also obvious project selection criteria, although they are not, by themselves, sufficient to warrant selection. In California, where many projects are advanced for congestion relief, the Orange County projects were selected for the pilot program because they were unique in one important aspect: They grew out of local initiatives as toll undertakings, and thus seemed to be natural candidates for the pilot program.

Another common thread in Table 3 is the status of candidate projects in the state programming process. Overcoming early hurdles in the process such as preliminary engineering and environmental assessment can favorably influence decision makers. Planners also feel more comfortable with the cost estimates of projects which have progressed further along the programming process. Cost figures for projects in early stages could change substantially as changes are made in project type and scope to accommodate community objections or environmental problems.

In some of the states, project selection was also influenced by the fact that a candidate project would cross or connect another toll facility, thereby providing continuity or eliminating a missing link problem. This was the case in both Florida and Texas.

The provisions of the pilot program allowed federal aid for just one toll project in each of the pilot states. Under ISTEA, states may undertake many additional toll projects with federal assistance. Are there many additional candidate projects for such undertaking? The response from the eight states was that there might be a few, typically one or two, qualifying candidates. The primary reason for this response is that most high-volume traffic corridors already contain high-class, toll-free highways and converting them to toll highways, even after significant improvement, may be difficult from a public acceptance viewpoint. However, this survey response was obtained prior to the enactment of ISTEA, before the full dimensions of the toll flexibility contained in the act were known.

Project Priority, Fast Tracking, and Scope

Table 4 shows results pertaining to project priority, fast tracking, and scope. With the exception of West Virginia, which dropped its project from the pilot program because it was able to obtain federal funding to build it as a TABULAR DATA OMITTED toll-free road, all states gave their pilot projects high priority.

High priority in the state programming process implies fast tracking. Projects of lesser priority are often delayed due to budgetary and/or engineering (resource) reasons. Even a project with secure planning and construction funds might still be delayed, in the absence of offsetting increases in resource commitment, because of unexpected problems in environmental assessment or preliminary engineering. However, high priority projects get the necessary follow-through to insure a high degree of schedule adherence. Bond-financed toll projects often are under pressure from investors to get projects completed so that revenues begin flowing, thereby reducing investors' risk. Most of the pilot projects, being of high priority, received such fast tracking attention. This led to estimated time savings in project completion, as shown in Table 4.

Table 4 also shows changes in the scope of the projects due to their being toll undertakings. Each toll project will require a toll collection system. (The exception is in Florida, where the project is a short extension to an existing toll expressway.) In most cases, the toll collection system will add 5 to 10 percent to total capital costs of the project. The negative traffic flow/delay impacts traditionally associated with toll barriers will be mitigated with new toll collection technology on those projects serving a high proportion of daily home-based work trips. California, Georgia, and Texas plan on employing Automatic Vehicle Identification (AVI) technology, enabling traffic to proceed through toll barriers without stopping. The need to reduce queues and congestion at toll barriers and pressures for improved productivity were the most frequent reasons given for this trend. These three states also had decided on employing Electronic Toll & Traffic Management (ETTM) systems, which would enable motorists to pay tolls electronically in conjunction with AVI. Some states were also designing Variable Message Signs (VMS) to provide advance notification of traffic conditions to motorists. However, few states were planning to implement peak demand pricing strategies. The most common reason cited for this was lack of public support for congestion pricing.

Project Financing and Importance of Federal Funds

With the potential to combine with conventional user-based financing sources, toll financing provides leverage for available federal and state aid to expedite selected, high-cost capacity additions that will help ease steadily worsening metropolitan traffic congestion. Projects in five states (California, Delaware, Florida, Georgia, and Pennsylvania) would not be moving forward without the benefits of such leverage. In each case, toll revenues are expected to cover a high percentage of project cost. For example, in California greater investment in highway infrastructure has been made possible by the state's willingness to depart from a long-standing, no-toll-road policy by allowing the construction of a limited number of toll roads.

In Pennsylvania, the Turnpike Commission issued $1 billion in bonds over four years beginning in 1986 to fund a variety of new construction and expansion projects, while the size of the state DOT's traditionally financed (toll-free) long-range program was shrinking. The existing turnpike, a mature but high-revenue generating resource, acted as the "cash cow" for financing additional toll projects, which, in their early years at least, would not be able to support debt service. One of these turnpike projects was subsequently designated as the federal pilot program project.

Toll financing played similar critical roles in Delaware, Florida, and Georgia. In Texas, additional leverage available in ISTEA is expected to stimulate progress on the project as a toll facility. In the remaining state, South Carolina, the project apparently would have gone forward as a toll-free project in the absence of the pilot program, funded in the conventional way, that is, out of general highway funds.

Turning now to the influence of the federal match ratio for toll projects, the nature of the programming process must first be understood. In many states, the development of the long-range capital program is subject to various funding allocation constraints, imposed according to highway program category, geographic location, and year. Allocation schemes that significantly diffuse available funds have made it difficult for some states to pool sufficient monies to undertake large-scale projects within reasonable time frames. This problem is particularly acute when there is a shortage of state or federal funds and a project's cost is large relative to the size of the highway program.

Thus on a large, high-cost toll project, a state is less likely to commit the full 35 percent federal share than if a smaller project is chosen for the pilot program. For example, the 35 percent federal share on a relatively modest size new construction project of $215 million is $75 million. Assuming a three-year construction period, this magnitude of federal funds exceeds 25 percent of the annual non-Interstate federal appropriations in 27 states over the period 1987-89.

Five of the six states proceeding with projects in the pilot program are using or are expecting to use the maximum 35 percent federal share permitted in the program (DE, FL, GA, PA, TX) as shown in Table 5. In four of these states, the estimated federal share of the annual capital cost of the project relative to the size of the non-Interstate annual apportionment was less than 15 percent. In California, a state with a highly structured, formula-based funding allocation scheme that diffuses funds geographically, the pilot project will receive a small federal share due to prior commitments of available federal monies to projects not eligible for the pilot program. In South Carolina, where less than 10 percent federal money is expected for the pilot project, a full 35 percent federal share would require nearly 30 percent of the state's annual non-Interstate apportionment over the construction period--an amount that would be politically impossible to commit to one project.

One extreme example of the interaction between the funding allocation mechanism and federal funding was found in Colorado. Although originally authorized for the pilot program, Colorado declined to participate because with the state-federal funding situation already tight and no additional federal funds being made available, the state could see no way to allocate scarce federal obligation authority towards a major toll project. Colorado was proceeding with the building of E-470 on its own, with a combination of toll revenue, tax-increment financing, and vehicle-related taxes; project financing may now change under the provisions of ISTEA.

In Georgia, recent administrations have been very reluctant to increase the state gasoline tax. Consequently, Georgia has the lowest motor fuel tax rate (7.5 cents/gallon) in the nation. Over the period 1987-1989, federal allocations for the state's highway capital improvement TABULAR DATA OMITTED program have exceeded state expenditures by a two-to-one margin. As a result of state dollars being relatively more scarce than federal, the state's programming practice has been to leverage available state monies to the maximum extent possible so as to maximize the use of federal aid.

On the other hand, Delaware has recently experienced a shortage of federal obligation authority relative to state funds for highway capital improvements. During the 1987-89 period, the state spent five state dollars for every three federal dollars. The state DOT's programming strategy has been to concentrate its federal funds on the largest projects and fund smaller projects from 100 percent state monies. This puts the smaller projects on a faster track to completion.

Other evidence of the influence of programming practice on the federal share was found in Florida and Texas. Florida's response is shaped by the state's practice of removing federal aid for special programs from the total amount of federal-aid available for distribution to each of the engineering districts. Like Delaware, federal aid for highways is relatively more scarce than state funds in Florida. Consequently, the districts generally dislike having funds removed from the top of the federal allocation and would have looked even less favorably on the pilot project if the federal aid match had been higher than 35 percent.

In Texas, current statutes prohibit the use of state general highway revenues on toll projects. The only circumstances under which more federal funds would be used on the project was if projected toll revenues did not cover the necessary non-federal share. Even in this circumstance, however, there may have been considerable reluctance to use more federal dollars because the federal contribution to the project is coming out of the Houston engineering district's share of federal funds and the district has many other priorities.

If the amount of federal obligation authority is increased, then clearly more states would be willing to take advantage of higher federal participation in toll projects. States indicated that the participation rate would need to be higher than 35 percent in order to have significant impact, with the lowest recommended rate of 50 percent and the highest being 90 percent. The new federal surface transportation act helps in both regards, increasing annual authorizations by an average of 40 percent to states, as well as increasing permitted federal share to 50 percent.


The pilot program experience provides some clues to how states might take advantage of the toll financing provisions of ISTEA. Following are several inferences on the consequences that revised federal toll policies have for various aspects of state highway planning and programming processes. As will be seen later, findings from the pilot program experience complement the statistical analysis of the future role of toll financing.

First, the pilot program experience suggests that states resort to toll financing only when conventional methods of toll-free financing are unavailable. In other words, there is a strong policy bias in favor of toll-free roads, due at least in part to past congressional policy.

Second, an overwhelming majority of states will not notably increase their use of toll financing without significant federal assistance (such as that provided in ISTEA). The pilot program projects, all high-priority projects in the subject states, would have been decimated in the absence of the pilot program; only projects in California and South Carolina might have moved forward as toll projects on their own.

The reason for this is straightforward. Studies have shown that average daily traffic during the first year of operation would have to be in the neighborhood of 90,000 for toll financing to be viable.(9) According to the Congressional Budget Office, there are very few candidate projects not already built which meet this traffic criterion.(10) This coupled with the attractiveness of federal aid for toll-free projects at levels of 75 to 90 percent of capital costs likely means that only a select few of the strongest candidate toll projects could proceed without federal aid.

Third, when financially feasible, toll projects receive higher priority in the planning and programming process and tend to be implemented faster than toll-free projects. This is due to several factors, including the need to generate toll revenues (bondholder pressure) and the high public and political visibility of these projects.

Fourth, all states will not use the maximum 35 percent federal share available for each pilot toll project, and by extension, not all states will use the maximum 50 percent aid permitted under ISTEA. Some states will use a lower participation rate, depending upon the mechanism for allocating state and federal funds, the magnitude of project cost, the expected revenues from tolls, the priority and public commitment to competing toll-free projects, and the relative scarcity of state and federal funds.

Of course, the greater the revenues expected from tolls, the less the need for federal aid. Also, states that traditionally have committed more state relative to federal dollars to their highway programs will benefit from a lower federal share if their practice is to use up to the maximum federal contribution on each project. The lower federal share allows these states to get more leverage out of their relatively scarce federal dollars.

Fifth, toll projects will accelerate innovation, such as the adoption of AVI and ETTM technology. In addition to the experience in the pilot states, research is under way by a New York City metropolitan area interagency traffic management consortium, including several toll agencies, to develop an intelligent vehicle highway system in the New Jersey to Staten Island corridor. This system will help in the rapid detection of traffic incidents and will contain an advanced traveller information component to communicate advice to highway users on traffic conditions and alternative routes.

Finally, although toll projects provide a natural setting for experimenting with demand-based pricing systems to improve the efficient utilization of highway capacity, the pilot program experience provides scant evidence that such pricing schemes will be widely adopted. Indeed, most toll agencies are wary of breaking dramatically new ground in the pricing area, due both to the highly political and visible nature of these highways and bridges and to concerns about equity effects of congestion charges.


While the pilot program experience affords some indication of the future role of toll financing under a liberalized federal aid toll policy, it does not provide a quantitatively based foundation prerequisite to an examination of future trends. One such foundation is established in this section; it is formed from a statistical analysis of past and prospective relationships influencing the role toll financing has or may assume in the state's programming processes. A well established model of these relationships could provide the necessary platform for exploring hypotheses concerning those factors most determinative of the future role of toll financing.

Conceptually, one can approach the issue of toll financing's role in the planning and programming process from a macro and a micro perspective, although the two are intertwined in practice. The macro perspective refers to the overall approach towards toll funding in a state's long-range transportation plan, including the size of the toll program and components thereof and criteria for the use of toll financing in undertaking highway projects. The micro perspective refers to the individual decision in a specific project whether or not to employ toll financing. Clearly the macro perspective is a statement of policy while the micro perspective reflects the implementation of that policy. The model discussed below deals with the former.

A model of toll financing's role in the planning and programming process can begin with the following framework:

TFR = f {Adequacy of general highway funds, candidate projects availability, institutional capability, and federal aid terms}

where "f' denotes a function and the terms are as follows:

TFR: Toll financing's role; this factor may be measured in several ways, including number of toll projects in the long-range program, dollar value of toll projects, and toll projects as percent of the total program, among others. TFR reflects the end result of the toll financing's influence in the planning and programming process.

Adequacy of general highway funds: This term is obviously important because it is the scarcity of conventional highway user funds, such as fuel tax and registration fee revenues, relative to capital and reconstruction needs which may make toll financing attractive. In the programming process, states have the choice of undertaking highway and bridge projects solely with state funds, state and federal funds, or state, federal, and toll revenues (and possibly other sources as well), depending on project characteristics and planning objectives. This factor may be measured by several variables such as current expenditure levels, normalized on a per lane-mile basis, and size of the highway network which must be built and maintained.

Candidate projects availability: This term represents the availability of projects which meet various toll financing criteria. Important criteria include:

* projects of high priority

* projects which exceed minimum toll revenue threshold

* no significant opposition

To the extent there are few projects which meet such criteria, as one might expect in predominantly rural or low-density states, toll financing will not be a significant factor in the long-range plan. The availability of appropriate candidate projects meeting established criteria could be reflected in several measures such as the number of projects, dollar value of projects, percent value relative to total number of all projects, relative miles of capacity addition, etc. In addition, various surrogate variables could also be used, such as average daily traffic and population density.

Institutional capability: This term refers to the legal and organizational ability to undertake and manage toll projects and having the access to necessary capital markets. In some states, the state DOT is prohibited by either the state constitution or statute from issuing debt for highway projects. Such restrictions remove one attractive feature of toll financing, viz. the ability to issue bonds backed by toll revenues and speed up project completion.

Existence of separate turnpike authorities or commissions may enable a state to take quick advantage of opportunities for toll financed projects. Pennsylvania and Florida, for example, were quickly able to move an already authorized turnpike project into the pilot program to take advantage of federal funds. Other states, not equipped to handle toll- and bond-financed projects, may have to explore privatization or public-private partnerships, as is occurring in California and Colorado. This raises new issues of how well the state DOT can manage these structures.

The existence of separate, independent toll authorities or commissions, apart from the state DOT responsible for managing toll-free roads and bridges, could, however, raise other thorny issues. Since federal funds flow through state DOTs, an unusual degree of coordination would be necessary among all affected organizations to employ toll financing as an integrated component of the state's long-range plan. In some states, both the state DOT as well as other quasi-independent authorities and commissions may be empowered to undertake toll-financed projects on the portion of the highway and bridge network under their jurisdiction. There may be greater toll activity in these states compared to states where only one (or none) is so empowered.

Federal aid terms: This factor refers to such terms as the match ratio and lapsing provisions. Possible measures for this factor include effective match and ratio of state to federal aid, described below.

Capital programs of some states are driven largely by the amount of federal aid available and match terms. If a state's effective match ratio is 80 percent, then it must provide at least one state dollar in its capital program for every four dollars of federal aid to avoid lapses. If its maintenance and other non-capital requirements are two dollars, then it must generate at least three dollars of state highway revenues, in which case its state-to-federal ratio, SFRATIO, would be 0.75. Ratios higher than this figure would enable the state to undertake projects not eligible for federal aid, such as non-federally aided toll projects, and to alter its project mix.


Two variables were selected to represent the dependent factor TFR: (1) The percentage of total highway user revenues derived from tolls, TOLPER3YR; and (2) rating by state highway administrators on a ten-point scale of the likelihood of greater use of toll financing in their state, TFF. Data for the first variable were obtained from Highway Statistics, published annually by the Federal Highway Administration.(11) A three-year average for the period 1987 to 1989 was used for TOLPER3YR.

Data for the second variable came from a survey conducted in 1990, prior to the introduction of the highway authorization legislation in Congress by the Bush administration. (The survey methodology is described elsewhere; briefly, 48 out of 50 states responded to the survey, and follow-up on the non-respondents failed to show non-response bias.(12)) This is important because the administration bill departed from earlier federal policy in proposing greater flexibility for states in using federal aid on toll projects; and ISTEA increased this flexibility even more. However, survey responses were given prior to public knowledge about the administration bill and of course, prior to ISTEA, and thus represent state highway administrators' outlooks on the role of toll financing in the future speculating on the evolution of federal policy. The specific question in the survey was "How would you rate the importance of toll financing as a source for additional revenues two years from now if Congress relaxes restrictions...." TFF was measured on a ten-point scale, with 1 representing not important and 10 indicating very important. TFF thus represents a future outlook while TOLPER3YR represents the past.

The first explanatory factor, adequacy of general highway funds, was represented by two variables: (1) current capital expenditure on roads and bridges per lane-mile, RDEXPLM; and (2) size of the highway system under the jurisdiction of the state DOT, in thousands of miles, KSTMILES. Data for both variables were obtained from Highway Statistics. The three years of data for RDEXPLM from 1987 to 1989 were averaged in order to smooth out year-to-year random fluctuations.

The second factor, availability of candidate projects for toll financing, was also represented by two variables: (1) population density, DENSITY; and (2) responses by state highway administrators indicating the different opportunities in their state for using toll financing, V6TOT. Data for the first variable came from Statistical Abstract of the United States(13) while the second came from the aforementioned survey. The variable V6TOT was the total number of "yes" responses to the toll opportunities presented in question 6 of the survey. This question asked state highway executives to indicate whether toll financing was likely to be very useful in their state for twenty-one categories of highways and bridges. The total number of affirmative responses was taken as an indication of the availability of projects with the potential for toll financing.

Institutional capability was represented by two factors: (1) a combination of three dummy variables indicating the type of agency, if any, that has toll financing authority within a state, ORGANZ1-3; and (2) another dummy variable indicating which states had been active during the late 1970s and 1980s on toll financing, TOLLACTIV. A value of 1 was assigned to ORGANZ1, ORGANZ2, or ORGANZ3 if there was no separate toll authority or commission of major revenue significance, if there was such a separate authority or commission responsible for toll activity, or if both the state DOT and a separate authority undertook toll projects, respectively. If there was no toll activity in the state, then 0 was assigned to each of the ORGANZ variables. Data to construct the ORGANZ variables came from the 1987-89 Highway Statistics and the Federal Highway Administration's publication Toll Facilities in the United States.(14)

The second variable, TOLLACTIV, focused on recent toll activity because this was expected to be a better indicator of future toll activism than states which had exhibited toll activity only in the remote past, such as prior to the inception of the federal Interstate Highways Program. Data to construct this variable came from selected years of Toll Facilities in the United States.

SFRATIO was used to represent the fourth factor in the model, federal aid terms for toll-financed highway improvements. This factor was included only in the TFF model, however; past federal policy generally prohibiting federal aid for toll-financed projects makes SFRATIO applicable for the TOLPER3YR model.

Two specific model equations were estimated using stepwise regression, one with TOLPER3YR as the dependent variable and TABULAR DATA OMITTED the other with TFF. The equation for TFF included a few additional independent variables described below. Table 6 shows the results of the two estimations.

In the equation for TOLPER3YR, all of the variables have the correct signs, although only DENSITY, KSTMILES, and the ORGANZ variables are statistically significant at the 95 percent level, while V6TOT and TOLLACTIV were significant at the 90 percent level. The equation suggests that high population density states derive a greater proportion of their revenues from tolls. The type of institutional capability does not appear significant, although the presence of the legal and organizational ability to construct and manage toll facilities is obviously important. Greater availability of candidate projects and recent increase in toll activity are positive indicators of toll revenues. States with relatively low mileage under state jurisdiction, including both state DOT and separate authorities if they exist, and states expending relatively lesser amounts for capital investment, are likely to be more active in toll financing. State DOTs with relatively large highway systems are likely to depend more on general highway funds because traffic volumes on the bulk of their network, which would largely be secondary and feeder roads, are unlikely to support toll financing. In contrast, small state highway systems, which consist largely of high-volume roads and bridges, are likely to derive a good portion of their total income from toll funds, ceterus paribus.

With respect to the second equation, TFF, the results suggest that availability of candidate projects (V6TOT) and large road networks (KTOTMILE, total mileage in state in thousands), both reflecting the demand for transport funding, have a positive influence on the future likelihood of toll-financed projects. Large requirements for maintenance expenditures, as reflected in total expenditure per lane-mile, TOTEXPLM, also increase the need for the toll revenue source, although, as before, high capital expenditure reduces this need. The influence of federal funds in the programming process is brought through the term SFRATIO, ratio of state user revenues to federal receipts. States with small ratios are more governed by federal constraints; indeed in such states the capital program is largely determined by the state funds required to match federal aid. States with largest SFRATIO have a greater degree of freedom to pursue toll financing. Finally, TOLLACTIV and MAJREVINC (a dummy variable indicating whether states had any major revenue increases in the previous two years--increases in fuel tax rates or vehicle fees) have the correct signs but were not statistically significant.

While the results in Table 6 may be intuitive, they confirm that in order for states to take advantage of the generous provisions of ISTEA, they must tackle the issues of candidate project availability and organization capability. States where neither the state DOT nor a separate authority has toll financing powers will either need to enact legislation enabling such undertakings or pursue privatization options permitted under ISTEA. The results also suggest that federal aid has played a strong role in influencing state programming policy in the past. However, this factor may become less influential in the future because of the greater flexibility bestowed by ISTEA, not only with respect to toll financing and privatization, but also in terms of ability to transfer funds between federal aid categories.

Table 6 results also suggest that increases in the availability of general highway funds would lessen the need for toll financing (sign of variable MAJREVINC). Many states have recently enacted significant revenue increases, including California (eight-cent fuel tax increase in 1990-91) and Pennsylvania (55-mill oil franchise tax increase in 1991), and many other states are considering increases during 1992.(15) These increases, coupled with ISTEA provisions markedly raising federal funds (approximately 40 percent more funds over two years compared to FY 1991), may stifle toll activity in the immediate future.


The experience of the eight states which participated in the pilot program and results of the aggregate statistical models provide some clues to the future role of toll financing in highway planning and programming in the U.S. The study findings confirm some likely directions. Toll financing coupled with federal aid has given states an added tool to finance very expensive projects that otherwise might not be built. Secure funding and the impetus to generate revenues have moved many of the toll projects on a fast track. Most states have also decided to adopt AVI or are considering its adoption.

The statistical models establish an initial framework for examining the role of toll financing in the state programming process. Factors found to be important included the availability of candidate projects, an institutional capability, the adequacy of highway funds from conventional user-based sources, and federal aid terms for toll-financed projects. The framework provides the necessary foundation that, in combination with the state programming experiences gained under ISTEA, creates future opportunities for exploring additional hypotheses and modifications to the models.

By making the terms of federal aid in ISTEA more generous than contained in the pilot program, Congress has increased the flexibility of states to undertake toll-financed projects. In the long run, this may influence the state planning and programming process in several ways, including: (1) the composition of the long-range program will likely contain more toll projects and over time may assume more of an intermodal character by combining toll roads, transit, and commuter parking facilities, as is being investigated for the I-95 corridor in the PA-NJ-DE tri-state area;(16) (2) the inclusion of more participants in the process, particularly private sector parties putting complex project syndications together; and (3) the modification of the way project priorities are assigned and managed. However, there are some countervailing trends. The significant new aid contained in ISTEA coupled with recent and planned increases in general state highway levies may mean that in the short run, there may be few toll projects which are undertaken.


1 See, for example, Congressional Budget Office, Toll Financing of U.S. Highways (Washington, D.C., December, 1985); American Association of State Highway and Transportation Officials, Understanding the Highway Finance Evolution/Revolution (Washington, D.C., January 1987); R.C. Schaevitz, "The Private Sector Role in U.S. Toll Road Financing--Issues and Outlook, Transportation Research Record 1197, 1988, pp. 1-8; J.T. Berg, "Taxation and Revenue Policies for Future Federal Highway Programs," Transportation Research, vol. 24A, no. 4, pp. 251-264, 1990; K. Rao, G. Gittings and S. Sriraman, "Evolution or Revolution in Federal Toll Highway Policy? Issues and State Views for the 1990s," Journal of the Transportation Research Forum, vol. 32, no. 2, 1992 (forthcoming).

2 Public Law 102-240, December 18, 1991, Statutes at Large 105, sec. 1012, 1936.

3 See, for example, Transportation Research Board, "Private-Sector Involvement and Toll Road Financing in the Provision of Highways," Transportation Research Record 1107, 1987; M.E. Beesley & D.A. Hensher, "Private tollroads in urban areas: Some thoughts on the economic and financial issues," Transportation, vol. 16, pp. 329-341, 1990; J.A. Gomez-Ibanez, J.R. Meyer and D.E. Luberoff, "The Prospects for Privatising Infrastructure; Lessons from U.S. Roads and Solid Waste," Journal of Transport Economics and Policy, vol. XXV, no. 3, September 1991, pp. 259-278; T.F. Humphrey, A.Y. Kanaan and R.F. Cunningham, "Implementing Electronic Toll Collection and Traffic Management Systems in New England," paper presented at the 71st Annual Meeting of the Transportation Research Board, January, 1992.

4 See, for example, C. Sharp, K. Button and D. Deadman, "The Economics of Tolled Road Crossings," Journal of Transport Economics and Policy, vol. XX, no. 2, May 1986, pp. 255-274; K.A. Small, C. Winston and C.A. Evans, Road Work: A New Highway Pricing and Investment Policy (Washington, D.C.: The Brookings Institution, 1989).

5 T.F. Humphrey, NCHRP Synthesis of Highway Practice 84: Evaluation Criteria and Priority Setting for State Highway Programs (Washington, D.C.: Transportation Research Board, National Research Council, November 1981); T.D. Larson and K. Rao, NCHRP Synthesis of Highway Practice 151: Process for Recapitalizing Highway Transportation Systems (Washington, D.C.: Transportation Research Board, National Research Council, December 1989).

6 Rao, et al.

7 Public Law 100-17, April 2, 1987, Statutes at Large 101, sec. 120, 157.

8 For additional background on these projects and public reaction to them, see United States General Accounting Office, Highway Financing: Participating States Benefit Under Toll Facilities Pilot Program, (Washington, D.C., December 1990).

9 N.H. Wuestefeld, "Toll Roads," Transportation Quarterly, vol. 42, no. 1, January 1988, pp. 5-22.

10 Congressional Budget Office, pp. 24-30.

11 Federal Highway Administration, U.S. Department of Transportation, Highway Statistics (Washington, D.C.; U.S. Government Printing Office, published annually).

12 Rao, et al.

13 U.S. Department of Commerce, Statistical Abstract of the United States, 1989 (Washington, D.C.: U.S. Government Printing Office), see Table 22, p. 19.

14 Federal Highway Administration, U.S. Department of Transportation, Toll Facilities in the United States: Bridges-Roads-Tunnels-Ferries (Washington, D.C.: U.S. Government Printing Office, published periodically).

15 Highway Users Federation, "States Gear Up for New Highway Funding," Washington, D.C., February 4, 1992.

16 Pennsylvania Department of Transportation, "The I-95 Intermodal Mobility Project for the 21st Century: Design Concepts Competition," Harrisburg, Pa., August 1990.

Mr. Rao, EM-AST&L, is associate professor of business administration, and Mr. Gittings, EM-AST&L, is assistant professor of business logistics, The Pennsylvania State University, University Park, Pennsylvania 16802; Mr. Sriraman is reader (associate professor) in economics, Gokhale Institute of Politics and Economics, Pune, India.
COPYRIGHT 1992 American Society of Transportation and Logistics, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rao, Kant; Gittings, Gary L.; Sriraman, S.
Publication:Transportation Journal
Date:Jun 22, 1992
Previous Article:Transportation characteristics and performance in maquiladora operations.
Next Article:Motor carrier cost function comparisons.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters