Determinants of VMT in Urban Areas: A Panel Study of 87 U.S. Urban Areas 1982 – 2009
This paper uses econometric techniques to examine the determinants of vehicle miles traveled (VMT) in a panel study using data from a cross section of 87 U.S. urban areas over the period 1982-2009. We use standard OLS regression as well as two-stage least squares techniques to examine the impact of factors such as urban density, lane-miles, per capita income, real fuel cost, transit mileage, and various industry mix variables on per capita VMT. We use a distributed lag model to estimate long-run elasticities and find that the long-run price elasticity of demand for per capita VMT is approximately five times larger than in the short run. Preliminary empirical results show the per capita demand for VMT in urban areas is positively and significantly impacted by lane miles, personal income, and the percent of employment in the construction and public sectors. Fuel price and transit use and the percent of employment in manufacturing, retail, and wholesale sectors are all found to be statistically significant and negatively related to VMT per capita. After correcting for endogeneity, urban population density exerts a negative, but not always statistically significant, impact on per capita VMT. Finally, per capita VMT is found to differ significantly by geographic region, being higher the more western and the larger the population size of an urban area.
by B. Starr McMullen and Nathan Eckstein