UC Berkeley

This dissertation explores three ways in which consumers' choices about which vehicle to purchase and how much to drive affect others. Chapter 2 seeks to understand how consumer preferences for new vehicles, which may vary in ways that are correlated with consumer demographics, can lead different demographic groups to pay different prices for new vehicles. By estimating consumer preferences for new vehicles by demographic group, I show that dealers do use the distribution of preferences within a demographic group to engage in third-degree price discrimination based on consumer demographics. Additionally, controlling for this third-degree price discrimination, I find that women and single buyers pay more for the same new vehicles than male and married buyers, suggesting that either there are differences in negotiating ability that are correlated with demographics or that dealers are engaging in taste-based discrimination that is not apparent when demographic groups' preferences are not controlled for. Chapter 3, which is coauthored with Nathan Miller, shows that manufacturers adjust new vehicle price incentives in response to changes in gasoline prices in a way that suggests that manufacturers believe that consumers care about vehicle operating costs. We show that these price adjustments would bias earlier estimates of consumer demand for fuel economy that assume that vehicle prices are constant, which implies that consumers have a higher demand for fuel economy than earlier estimates and that the optimal gasoline tax may be lower than earlier estimates. Finally, Chapter 4, which is coauthored with Clifford Winston, estimates the relationship between congestion, urban land use, and home prices in order to understand how congestion tolling would affect urban land use. We find that congestion tolling would lead to denser city centers and subcenters, which would reduce urban sprawl and substantially increase the social benefits of congestion tolling relative to the costs.