Abstract:

This dissertation consists of three essays that study housing markets, migration patterns, and higher education choices. The first essay, "Identifying substitute cities: observable similarity and home price correlation", studies the demand for housing across U.S. cities. The paper identifies cities that are substitutes in consumer demand using two types of proxies: (1) measures of observable similarity such as average temperature, industrial composition, and per capita income and (2) the pairwise correlation in annual home price changes as a priced-based measure. The paper validates these proxies empirically using migration data from the U.S. Census. Results suggest that both types of proxies can identify substitute cities but that home price correlation is a particularly useful measure.

The second essay, "Home price risk, local market shocks, and index hedging", asks how local housing markets are defined and estimates how much individuals would be willing to pay to hedge local market risk. The paper uses home transaction data and locally weighted regression techniques to construct price indices across three dimensions of similarity: geography, price, and home type. Results suggest that city indices explain a large fraction of home price variation already but that a typical homeowner would be willing to pay 5-10% more to hedge with a local index relative to a city index.

The third essay, "Quality, tuition, and applications to in-state public colleges", studies the student demand for in-state public higher education, and in particular, the student willingness-to-pay for college quality. The paper uses score report data from over 500,000 SAT and ACT test-takers and variation across states in college quality and tuition in order to empirically estimate the elasticity of student demand. Results suggest that the student willingness-to-pay for quality is quite large, as measured by the quality versus tuition trade-off in individual student regressions.