National Academies Press: OpenBook

Estimating Toll Road Demand and Revenue (2007)

Chapter: Chapter One - Introduction

« Previous: Summary
Page 2
Suggested Citation:"Chapter One - Introduction." National Academies of Sciences, Engineering, and Medicine. 2007. Estimating Toll Road Demand and Revenue. Washington, DC: The National Academies Press. doi: 10.17226/23188.
×
Page 2
Page 3
Suggested Citation:"Chapter One - Introduction." National Academies of Sciences, Engineering, and Medicine. 2007. Estimating Toll Road Demand and Revenue. Washington, DC: The National Academies Press. doi: 10.17226/23188.
×
Page 3
Page 4
Suggested Citation:"Chapter One - Introduction." National Academies of Sciences, Engineering, and Medicine. 2007. Estimating Toll Road Demand and Revenue. Washington, DC: The National Academies Press. doi: 10.17226/23188.
×
Page 4
Page 5
Suggested Citation:"Chapter One - Introduction." National Academies of Sciences, Engineering, and Medicine. 2007. Estimating Toll Road Demand and Revenue. Washington, DC: The National Academies Press. doi: 10.17226/23188.
×
Page 5
Page 6
Suggested Citation:"Chapter One - Introduction." National Academies of Sciences, Engineering, and Medicine. 2007. Estimating Toll Road Demand and Revenue. Washington, DC: The National Academies Press. doi: 10.17226/23188.
×
Page 6

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3CONTEXT Throughout the United States, traditional public-sector fund- ing sources for transportation projects are unable to meet the growing demand for new highway infrastructure and maintain an aging infrastructure. Motor fuel taxes—the primary source of transportation finance in the United States—have not kept pace with the demand for travel and, in turn, for capital invest- ment, owing to inflation, improved fuel efficiency, and increased vehicle usage (1). The shortfall can lead to economic impacts on highway construction, as well as its operation, maintenance, and expansion. As a result, some state depart- ments of transportation (DOTs) and transportation authorities are relying increasingly on tolling as an alternative means of financing new and expanded highway infrastructure. A related problem is the need to improve the management and utiliza- tion of existing facilities, because many of the urban centers that require additional capacity to relieve congestion have lim- ited space for expansion. With budget shortfalls and an increasing demand for road- way facilities to alleviate the congestion problems that are common to many highways, state DOTs are turning to user- based fees or tolling as a means of financing roadway improvements and expansion and managing growing traffic demand for both interurban and urban facilities. The Texas DOT, for example, has determined that any new highway project in the state must be evaluated as a toll road (2). As state DOTs turn to tolling, increasing attention is being focused on the performance of the underlying revenue fore- casts and the projected ability of the facility to service debt. This is because the performance of the revenue forecasts, which are derived largely from forecasts of traffic demand, has varied among projects. In some cases, the lower-than- anticipated revenues were addressed through alternative sources of revenue to pay debt service, such as other toll roads, gas taxes, or government guarantees, or, where available, through sufficient reserve funds. This ensured that no strug- gling project was entirely dependent on traffic revenues for debt payment. As a result, between 1985 and 1995, forecast- ing errors or inaccuracies, which were known to exist, did not result in a single default in payment or any serious payment difficulties with new toll road projects in the United States (3). However, concern has since been expressed that these alternate sources of revenue may not be available to protect more recent projects or even future projects. For example, the privately held Dulles Greenway in Virginia went into default in 1996, as a result of toll revenues being less than projected (achieving only 20% of projected revenues in 1995, its first year of operation, and still only 35% of projected revenues in its fifth year). Other toll roads have also struggled; for exam- ple, revenues from the Southern Connector in South Carolina have been sufficient to cover operating costs but only a por- tion of the debt service, because traffic projections have not been met (just over half of the projected demand was realized in its third year of operation). Similarly, traffic on the Poca- hontas Parkway located southeast of Richmond, Virginia, has been just under half of the projected demand in its second year of operation. One result is that the credit ratings for the bonds for both facilities were lowered (4). The Foothill/Eastern toll road in Orange County, California, was refinanced in 1999 (3). Contributing factors to the financial problems of the Pocahontas Parkway and the Foothill/Eastern toll roads (and others) have been attributed to various inaccuracies in the demand and revenue forecasts, which included the unantici- pated affects of a recession, actual ramp-up volumes being less than projected, and the expected extension of a connect- ing road not occurring (5). For toll facilities to be financially viable and/or attractive to potential investors (public–private partnerships, etc.) in the future, the facility must be seen to be able to generate suffi- cient revenue from operations to cover debt service cost and potentially other project and maintenance costs over the life- time of the facility, as well as providing a reasonable return on equity. This requires a reliable and credible forecast of the expected revenues, which are functions of the estimated traf- fic demand and toll rates for the facility. However, industry experience in tolling forecasts and the associated recoverable benefits historically have been quite varied, in that demand (and the accompanying revenues) has ranged from frequently overestimated to occasionally underestimated. Also, the accu- racy of when specific levels of demand are projected to occur has been mixed, with problems being particularly acute in the short-term facility ramp-up. The resultant variations have had significant impacts on both the actual revenue streams and on the facility’s debt structuring and obligations. This has led to concerns among facility owners and the financial community (which rates and insures and/or invests the bonds that are issued for the facility’s implementation) about the accuracy, reliability, and effectiveness of the demand forecasts upon which the revenue projections are based. In addition, the CHAPTER ONE INTRODUCTION

growing use of Intelligent Transportation Systems and other technologies provides DOTs with the ability to implement variable pricing, HOT (high-occupancy toll) lanes, and other innovations that require a greater level of accuracy in the depiction of demand and revenues. It is of critical importance that the forecasted toll revenue targets be accurate, based on the ability of the toll road to achieve its forecasted traffic targets, if debt service on bonds or a return on equity for private operators are to be paid (6). The fiscal feasibility of new toll road projects, or the revenue side of the benefit–cost relationship, is based on models that forecast traffic demand. Therefore, the reli- ability, accuracy, and effectiveness of these forecasts are important, because they are critical in determining the credit quality of the projects. Accordingly, to maximize the prospects of a project’s financial viability—that is, the likelihood that the forecasts match the actual revenues—it is necessary to look at the early stages of the process; namely, the models that are used to forecast travel (traffic) demand and their resulting successes or failures to improve the forecast results. To understand the reasons for unrealized traffic demand forecasts, it is neces- sary to review not only the model as a whole, but to separate it into its constituent structure, inputs, calibration, and appli- cation and examine the underlying assumptions that drive the forecasts. This synthesis explores the subject in a manner that takes into account viewpoints and experiences from both the engineering and planning communities and the financial community. PURPOSE AND SCOPE Purpose The purpose of this synthesis is to report the state of the prac- tice in toll road demand forecasting models that are used as the basis of revenue forecasts for these facilities. The syn- thesis had four specific goals: • Develop a profile of the current state of the practice in toll road demand forecasting through a survey of the forecasting community and a literature review. • Based on this profile, identify the technical modeling issues that affect the accuracy, effectiveness, and relia- bility of the forecasts. • Make recommendations regarding ways to improve the state of the practice. • Identify areas for potential future research. Subject This synthesis focused on the models that are used to fore- cast the demand for travel on tolled facilities and their reli- ability. It also took into account the application of these 4 models to project revenues as a function of the demand estimates. Several different aspects of the models were considered. • Purposes of toll demand forecasting (i.e., how they dif- fer from other travel demand forecasts). • Methods for forecasting toll facility demand and rev- enues, ranging from simple sketch-planning tools to investment grade studies; also taking into account how these have changed over time. • Products provided from forecasting, and how they were used. • Input data used, such as surveys, and data sources and key assumptions. • Methods of estimating the value of time. • Treatment of different horizon years, with particular attention to short-term (ramp-up) forecasts. • Peer review procedures. • Comparison of forecasts with the actual experience. • Comparison of the accuracy and effectiveness of various toll facility demand and revenue forecasting methods. • Variables associated with forecasting successes and failures, such as the type of proposed investment (i.e., stand-alone projects or expansions to existing projects), financing methods, modeling techniques, etc. • Assessment of risk for demand and revenue forecasts. • Transparency (i.e., availability and understanding of detailed information on modeling procedures, the underlying assumptions, etc.). • Innovative techniques used to improve the quality of forecasts. Scope Given the subject, the scope of what was included in this syn- thesis is defined by the following: • Tolled roads and highways and related infrastructure, including bridges and tunnels. In practice, these facili- ties generally have limited access; that is, they gener- ally are not accessible from adjacent properties. • HOT lanes (i.e., limited access lanes that provide free or reduced cost access to qualifying high-occupancy vehi- cles (HOV), but also provide access to other paying vehicles that do not meet the occupancy requirement). • Road-based traffic forecasts for the aforementioned facilities; typically light vehicles (i.e., personal auto- mobiles, vans, pick-up trucks, or motorcycles) and heavy vehicles (typically trucks), but potentially including commercial, fleet, or service vehicles and public transit buses (i.e., road-based transit, but not including rail that operates in mixed traffic). • Urban as well as interurban tolled facilities. • Different types of facility ownership. • Different types of toll collection, pricing schemes, level of service and capacity differences, etc. Distinctions were noted where they were appropriate to the forecasting

5process (e.g., the incorporation of bias factors regarding the type of toll collection). • Toll roads and highways anywhere in the United States, although experiences elsewhere could be considered if they were deemed relevant to the situation in the United States. • Different types of proposed investments, ranging from network-wide feasibility planning studies to facility- specific forecasts. • Forecasts up to 30 years old (although virtually the entire literature addressed facilities that were 20 or fewer years old), with a focus on the most recent expe- rience to appropriately represent the prime objective of the state of the practice. The following topics were beyond the scope of the syn- thesis and so were not included: • Non-tolled roads and highways of any kind. • Transit-only infrastructure or facilities for other modes of transportation. Study Process The synthesis focused on the practice as opposed to aca- demic or theoretical considerations, although the latter were noted if they were relevant. The synthesis is based on infor- mation that was acquired either through a review of the liter- ature or by means of a web-based survey of state DOTs, toll authorities and their consultants, bond rating agencies, and bond insurance agencies. Thus, the synthesis considered the perspectives of both the engineering and planning communi- ties (the builders and owners) and the financial community (the financial backers), assembling factual information as well as opinions and interpretations. Where appropriate, information from actual cases was used to illustrate the different approaches: these were drawn from the literature or the survey. Finally, it is important to note that this synthesis focused on the technical aspects of the state of the practice in model- ing. It was not intended to be judgmental with respect to the policies of individual states, tolling authorities, bond rating agencies, or any other parties, or how these organizations have used the resultant forecasts. ORGANIZATION This synthesis is organized into five chapters. The remainder of chapter one describes the balance of the report, identifies the intended audiences for the synthesis, and presents vari- ous definitions for terms that are used in the text. Chapter two describes the methods used for the litera- ture review and for a survey of practitioners. This chapter is not intended to be a detailed discussion, but rather pre- sents an understanding of how the information was gath- ered and any caveats or comments that may be associated with the information. Chapter three is the core of the synthesis. It begins by describing the state of the practice and the emerging devel- opments in travel demand forecasting, and then describes the applications to toll road traffic and revenue forecasts. The chapter then describes the problem at hand by documenting and commenting on the “performance” of toll facility fore- casts and the reasons for this performance. The discussion is based on the information collected in the literature search and in the survey of state DOTs, toll authorities, bond rating agencies, and bond insurance agencies. Although the state of the practice largely reflects that of the United States, prac- tices elsewhere also are considered, where appropriate. Taking into account the identified problems and how oth- ers have addressed them, chapter four synthesizes the state of the practice in terms of “checklists” and indices that practi- tioners can use to identify and account for the relevant issues and needs when they develop and apply travel demand mod- els for toll road traffic and revenue forecasts. Chapter five offers conclusions drawn from the findings and makes suggestions for future research in the area of toll road demand and revenue forecasting. A list of references, a bibliography of sources, and a glossary of selected terms are also included. Four appendixes complement the synthesis. Appendix A describes the development and administration of the survey, as well as the response to the survey. Appendix B is a copy of the original survey as sent to practitioners. Appendix C tabulates the survey results. Appendix D describes the char- acteristics of the toll facilities that were included in the chap- ter three comparison of projected and actual revenues. AUDIENCES This synthesis is intended to serve as a resource for several types of organizations. • State DOTs, which are in various stages of considering, planning, implementing, and operating tolled facilities, either within the organization specifically, through a dedicated authority, and/or some type of arrangement with a private-sector owner. The DOTs also set policies (pricing, funding, enforcement, etc.) and are responsible for other competing or complementary transportation infrastructure. • Metropolitan planning organizations (MPOs), which are responsible for planning an urban area’s long-range transportation plan, within which a tolled facility must be planned, and for gaining a consensus on priorities for

funding and implementation (if federally funded proj- ects; not required for privately funded). • Tolling authorities and operators, who are charged with the actual planning, implementation, financing, opera- tion, maintenance, and expansion of a specific facility. • Potential investors, who might provide the financial backing for the facility. • Bond rating agencies, which must assess the underlying credit quality of the bonds. • Bond insurance agencies, which underwrite the bonds. • Consultants, who prepare the models and the forecasts on behalf of the DOTs and facility owners. • Academia and researchers, who might seek ways to improve the modeling process, algorithms, structure, inputs (e.g., through improved quality control or risk management), and the application of the actual model. Although it is based largely on practice in the United States, it is expected that the synthesis also would be of inter- est to audiences outside the country for adoption to local needs, and because of the growing involvement of foreign financiers (and their consultants) in owning and operating facilities in the United States. Finally, it is important to note that the broad list of audiences in the synthesis requires that both the technical modeling community and those who ben- efit from or participate in the use of these models are addressed. DEFINITIONS This section identifies and explains key travel demand fore- casting and modeling terms. Explanations of other terms (e.g., HOT lanes) are provided only if they are pertinent to travel demand and revenue forecasting. Alternative definitions may exist for some of these terms; the definitions presented here represent how the terms were understood in the context of this synthesis. These definitions are complemented by a glossary of terms found at the end of the report. • Congestion pricing (or value pricing)—Use of pricing as a means of managing traffic, through the imposition of a premium fee to road users who choose to drive dur- ing peak periods, such as rush hour or holiday week- ends. Tolls vary according to the level of congestion; for example, higher tolls are charged during peak hours of operation or peak direction of travel (7). (See also variable pricing.) • Critical review (or audit or second opinion)—Process by which an independent review is conducted of the toll demand and revenue forecasts, the model on which these are based, and the input data and assumptions. This is distinguished from a peer review, which is con- ducted during the calibration and development of the model, thus providing input to the actual development 6 of the model (and not necessarily examining any resul- tant forecasts). (The terms are used loosely and inter- changeably in practice). • Feasibility study—Examination of the feasibility of implementing a proposed toll facility or a network of toll roads. It is generally not used as the basis for fund- ing, but rather as the basis for determining whether the subject warrants further investigation. The feasibility study might test alternate facility or network configura- tions, different assumptions of other competing routes or modes, demographic and economic forecasts, etc. • High-occupancy toll (HOT) lanes—Limited access roads that provide free or reduced-cost access to quali- fying HOVs, but also provide access to other paying vehicles that do not meet the occupancy requirement. By using price and occupancy restrictions, the number of vehicles that travel in these lanes can be managed (8). • Investment grade traffic and revenue forecast—A more detailed estimate of the traffic demand and revenues for a specific proposed facility. The term is used in differ- ent ways in the practical literature (which may reflect, in part, some of the observed problems regarding relia- bility, accuracy, and credibility of the forecasts). How- ever, a general definition that has been accepted in the financial community is that an investment grade traffic and revenue forecast represents a forecast that can form the basis for credit ratings, financing approval, and the sale of capital markets debt. • Ramp-up period—Time for traffic volumes to reach their full potential, without considering growth, after the opening of a new toll facility. The ramp-up period, which can last for several years, is the time it takes for users to become aware of the new toll road, change their travel patterns accordingly, and recognize the potential time-savings of using the new toll road (9). • Revealed preference survey—Quantitative survey of observed travel behavior in the study area. These sur- veys record how people actually travel over a certain time period (e.g., a 24-h period on a “typical” weekday) or at a certain point in time. The surveys capture the trip origin and destination, its purpose (e.g., the home-to- work commute and going shopping), the mode used and, in some cases, the start and end time of the trip. Related questions are also often asked (e.g., whether the driver paid for parking or whether a transit rider could have used a personal vehicle). The surveys can be con- ducted with a representative sample of homes in an urban area, generally by face-to-face interview, tele- phone, or mail (household origin–destination survey, capturing all the trips made by household members over a period of time), by interviewing drivers “intercepted” along a road or highway of interest (the interview typi- cally covers the current trip), goods movement surveys (of truck drivers and dispatchers, etc.), and so on. • Risk assessment—Quantitative or qualitative estima- tion of the incidence and magnitude of an adverse effect on a given population (10). In the context of toll

7road revenue forecasts, this often refers to the values of various inputs (e.g., forecasts of population or of the value of time); assumed configurations of the trans- portation network, such as the timing of planned com- peting routes or modes; or the treatment of specific components of the forecasting process (e.g., the meth- ods used to estimate ramp-up traffic). It also can refer to inherent uncertainties in the modeling process and structure. • Risk management—Uses the results of a risk assess- ment to develop options for addressing or mitigating the identified risk, and subsequently evaluating and imple- menting these options (10). In the context of toll road revenue forecasts, this could refer to the development of alternate scenarios (e.g., a range of population pro- jections) or sensitivity tests to be run in the model to test the variability of the toll road demand with regard to alternate scenarios. • Stated preference survey—Survey that attempts to quantify how travelers would behave in a situation that is new to them. These surveys are typically used to esti- mate the value of time for proposed toll facilities, which generally cannot be captured in revealed preference sur- veys. (Thus, a revealed preference survey provides a general quantification of the distribution, magnitude, and characteristics of a region’s or corridor’s travel activity; whereas a stated preference survey is used to estimate the impact of the imposition of pricing on the routes that the travelers who generate this activity would take.) Stated preference surveys are designed to present different options to respondents; for example, to determine not only the value of time but also how their perceptions of that value would vary by time of day (i.e., by congestion level). • Value of time (VoT)—Monetary value given by travel- ers to travel time. VoT is used in the forecasting process to relate how the value of tolls influences route choice (i.e., the driver’s decision to use the tolled facility rather than a non-tolled alternate route). VoT varies by indi- vidual, trip purpose, mode, average income levels, or time of day. • Variable pricing—User charge that varies by time period as a way to manage travel demand and reduce congestion. This is the basis of congestion pricing. Such fees are higher during peak periods when the conges- tion is most severe and lower during off-peak periods when there is minimal congestion. This concept is sim- ilar to many services, such as telephone service, electric utilities, and airlines that use time-variable pricing to encourage more efficient use of system capacity and allow users to save money by shifting their consump- tion to off-peak periods (11). • Willingness to pay (WTP)—Value of time that accounts for how much travelers value different attri- butes of the proposed facility, as opposed to simply its availability. The average WTP can be greater than the actual value of time, and can vary by time of day (i.e., as congestion increases). The WTP can reflect drivers’ expectations of what the tolled facility offers, such as improved safety and reliability.

Next: Chapter Two - Method for Literature Review and Survey »
Estimating Toll Road Demand and Revenue Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's National Cooperative Highway Research Program (NCHRP) Syntheses 364: Estimating Toll Road Demand and Revenue examines the state of the practice for forecasting demand and revenues for toll roads in the United States. The report explores the models that are used to forecast the demand for travel and the application of these models to project revenues as a function of demand estimates.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!