A time-series analysis of gasoline prices and public transportation in US metropolitan areas

Bradley W. Lane, University of Texas at El Paso

Abstract

This research examines the temporal aspects of the relationship between public transportation and gasoline prices in US cities from January 2002 through March of 2009. Data are collected at monthly intervals for transit ridership, service, gasoline price, and price variability for thirty-three metropolitan areas. These data are analyzed using time-series regression to estimate the presence of lagged effects of price and service on transit patronage.

The results indicate a small but consistently significant amount of transit ridership fluctuation is due to gasoline prices. Repeated lags of gasoline prices of up to thirteen months are influential on ridership. There is considerable variability across cities in the magnitude of the effect on transit ridership, and the impact also appears to vary by mode. The results are discussed in light of their implications for transit operations, using cost to influence travel behavior, and transportation sustainability.