Models for forecasting highway revenues

Bismark Richard Delaidem Kofi Agbelie, Purdue University

Abstract

Throughout the United States, state highway revenue, which is dominated by receipts from fuel taxes, has failed to keep up with expected investments required for infrastructure preservation and improvement. The reasons for this trend include the increasing fuel efficiency of vehicles, slowing of the growth in vehicle-miles of travel, and the erosion of the purchasing power of the dollar due to inflation. This development motivates highway agencies not only to seek new funding sources but also to make predictions of the expected future revenues from existing and potential sources. To establish and implement an effective and efficient financing strategy that incorporates potential new funding sources, it is necessary to model the possible outcomes of these sources in terms of their impacts on revenue stream and to study the sensitivity of these outcomes with respect to changes in key revenue factors such as vehicle-miles of travel and fuel price. In addressing this issue, this study utilizes data on amounts of travel, fuel price, and other primary information to enhance the existing models for state highway revenue forecasting in Indiana. To facilitate implementation of the study results, the existing revenue forecasting software package has been enhanced to include traditional and new revenue sources, to estimate revenue under several different scenarios, and also to analyze sensitivity of revenue to changes in input factors such as fuel price, per capita income, gross domestic product, driving age population, and traffic growth rate. The package provides long range annual forecasts for both existing and alternative highway revenue sources in Indiana, and short range forecasts for existing highway sources.

Degree

M.S.C.E.

Advisors

Labi, Purdue University.

Subject Area

Civil engineering|Public administration|Transportation planning

COinS