Concerns about the environmental impacts of transportation have turned reducing vehicle-miles traveled (VMT) into a policy priority. One way to decrease VMT is to reduce the length of commuting trips. Unfortunately, the average U.S. commute keeps getting longer. Prior research has investigated the determinants of commuting, but few have analyzed the linkage between housing costs and the length of commuting. This problem is especially salient in California given the state’s perennial housing shortage and the high costs of housing, which have forced many lower- and middle-class households to move inland in search of more affordable housing at the cost of longer commutes. Most of those commuting trips are by car. This project investigates these linkages using Generalized Structural Equation Modeling and analyzing 2012 CHTS data for Los Angeles County – the most populous county in the U.S. The model, which jointly explains commuting distance and time, accounts for residential self-selection and car use endogeneity, while controlling for household characteristics and land use around residences and workplaces. Better understanding the determinants of commuting is critical to inform housing and transportation policy, improve the health of commuters, reduce air pollution, and achieve climate goals.

This study is addressing local travel forecasting models through a fundamental and comprehensive examination of the state of the practice, focused on data, methodologies, software, and operational practice elements. The study will identify modeling practice relating to local and region-wide impacts and benefits of land use location, density, design, and configuration. Sensitivity analysis will assess how the elements of modeling practice affect model quality and applicability for testing land use and multimodal strategies. The introduction of state of the art modeling techniques for local travel demand models will be considered to identify potential improvement measures.