This thesis explores the possibility of reducing transportation emissions by reducing the growth of demand for travel in the United States light-duty vehicle fleet. Many government agencies seek to reduce the environmental and social ills associated with excess travel demand (e.g. congestion, reduced safety during travel, local pollution and noise, energy consumption, and climate change). These agencies have many tools at their disposal to reduce vehicle miles traveled (VMT) per capita - including encouraging compact mixed-use development, providing alternatives to single occupancy vehicle travel such as transit and biking and walking infrastructure, and restricting/regulating driving alone for instance by providing less parking. But the fastest way to reduce travel demand is through higher pricing that accounts for the externalities that drivers impose upon each other and society more broadly. The degree to which higher pricing can reduce travel demand is a function of two interrelated factors: 1) how high of a price increase is politically feasible to implement, and 2) the degree to which the driving public responds to the higher cost of driving. Both these factors vary over time. Given that carbon pricing and/or higher gas taxes are likely to take years to gain broader political acceptability, the future price elasticities of travel demand are just as relevant as today's elasticities. Therefore, this thesis focuses on the variability of price elasticity, the factors that explain this variation, and how these factors might change in the future. Using a diverse set of methods including literature review, semi-structured interviews, and odometer data, I find evidence that the magnitude of price elasticity is lower for vehicles of higher fuel economy, for vehicles further from the urban center, and for vehicles in lower income zipcodes. This is the first analysis I am aware of that evaluates the variation of the price elasticity of travel demand within a metro area, an approach that is important to the understanding the political feasibility of pricing and as a lens to the future effectiveness of pricing. It suggests that gas price increases will affect certain households in very different ways, with the most inelastic households simply paying more to maintain their lifestyle and the most elastic households pushed to make significant changes to their daily travel patterns and opportunities. These two types of impact may lead to different types of resistance to the policy. As for the future, the findings regarding fuel economy and distance to the urban center are particularly relevant, as we foresee society continues to become more metropolitan and the vehicle fleet is increasingly comprised of high fuel economy vehicles. Finally, the magnitude of price response suggested by both my interviews and my odometer data analysis suggests that price is still a significant determining factor in distance driven; therefore, policy that increases the cost of driving remains an important emissions reduction strategy.

Co-supervised by Prof. John Heywood and Prof. Jinhua Zhao