Measuring the impacts of local land-use policies on vehicle miles of travel: The case of the first big-box store in Davis, California
Concerns over climate change have brought new impetus to the goal of reducing vehicle travel through land-use policy. In California, for example, Senate Bill 375 (2008) led to the establishment of regional targets for reducing greenhouse gas emissions from passenger vehicles and requirements that metropolitan planning organizations adopt land-use policies, among other strategies, that help meet these targets. While empirical evidence shows that residents drive less in communities with greater densities and mixes of land uses, local governments have little basis for knowing how much less their own residents will drive if they succeed in increasing densities or the land-use mix. At the local level, changes are usually incremental, occurring one project at a time within the context of the existing community, and neither one-size-fits-all elasticities based on cross-sectional studies (Ewing and Cervero, 2010) nor regional travel-demand models (Rodier, 2009) are likely to offer more than rough predictions of the potential reduction in driving.
To determine the degree to which land-use policies are effective in reducing vehicle travel, a National Academies report recently called for more before-and-after studies (National Research Council, 2009). Before-and-after studies, considered the “gold standard” for accurately measuring project impacts, are rare in the extensive literature on the links between land use and travel behavior. Rooted in program evaluation, these studies measure the behavior of interest before an intervention is implemented and compare it to behavior measured after the intervention. Such studies approximate true controlled experiments to the extent possible in the real world. A body of such studies would provide decision makers with more credible and reliable evidence on the effectiveness of alternative land-use policies in reducing driving than is available from cross-sectional studies or travel-demand models.
In this paper, we present a before-and-after study of the impact of the opening of the first “big-box”1 store in Davis, California, on vehicle-miles travelled (VMT) for shopping. We use a repeated cross-sectional design to capture residents’ shopping travel before and after the store’s opening, allowing us to more directly identify a possible causal relationship between a change in land use and VMT than is possible with studies that simply compare behavior across communities. Although the observed change in VMT is specific to the Davis context, the findings support the general proposition that bringing retail destinations closer to residences could help reduce vehicle travel, thus contributing one piece of evidence to the body of studies needed to inform land-use policy in California and beyond. The work also offers important insights, potentially useful to other researchers, into the challenges of conducting before-and-after studies of the impact of local land-use changes.