Short- and Long-Term Effects of Land Use on Reducing Personal Vehicle Miles of Travel

While voluminous empirical studies have examined the impact of land use on travel behavior, few have relied on longitudinal data and an analytical approach. With data from two activity travel surveys (1997 and 2006) conducted in Austin, Texas, this paper develops a longitudinal multilevel model for estimating the change in the effect of land use on vehicle miles of travel (VMT) over time and the long-range land use effect on VMT reduction. Results suggest that the influences of land use mixture and street density on VMT would vary between 2 years. The effects of VMT reduction on land use policies by raising population and street densities are salient in the short run but insignificant in the long run, whereas those of mixed-use policies are even larger in the long run. These findings validate the importance of longitudinal data and analysis in land use–travel studies and suggest that the short-run elasticity of land use derived from cross-sectional analyses may be inappropriate for assessing the long-run effect of land use–based mobility strategies for reducing VMT.