Full Citation

Title: Three Essays on Urban/Regional Economics and U.S. Housing Markets

Citation Type: Dissertation/Thesis

Publication Year: 2022

ISBN:

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: Rosen (1974, 1979) and Roback (1982) imply that wage differentials across locations between similar workers arise from the differences in local amenities and cost of living. If the valuation of amenities varies across individuals (or groups of individuals), this affects the geographical sorting of households (Bayer and Timmins, 2007; Bayer et al., 2009; Lee, 2010; Moretti, 2013; Couture et al., 2019; Baum-Snow and Hartley, 2020). As a result, although the free mobility assumption ensures that the indirect utility of each worker type within the same group should be equalized across areas, this does not imply that interregional wage differentials should be the same for all groups of workers. The first essay proposes an approach to measuring interarea wage differentials by matching individuals with similar demographic and education characteristics to adjust for the effects of differences in preferences for local amenities and cost of living on wages. Given that the preferences for amenities, housing, and local public goods vary across individuals, workers are disaggregated into highly detailed subgroups based on their observable demographic characteristics under the hypothesis that people in the same group share similar utility functions and have similar preferences. The results show that disaggregating workers can overcome measurement problems in estimating amenity and house price variables commonly found in previous literature and can also adjust for preference heterogeneity across worker types. The results confirm the theoretical prediction that a commonly used aggregate Mincer equation underestimates the compensating variation required by the average worker, suggesting that disaggregating workers into highly detailed subgroups can be an effective method of explaining differences in intercity real wage differentials. The second essay applies the method developed in the first essay to investigate the effects of potential monopsony power on interregional wage differentials for a particular occupation. Current research investigates the effect of monopsony on wages by constructing a concentration measure (Herfindahl-Hirschman Index, for example) or estimating labor supply elasticity for a particular occupation or in a specific region. However, the problem with conducting direct tests for lower wages for a particular occupation that is susceptible to monopsony is that living cost and amenities also cause wage differentials. Higher wages for a given category of worker in city A than in city B might not arise just from a lower degree of monopsony, but from a higher cost of living or lower amenities in city A, and thus workers are paid more as compensation. The potential effect of monopsony on wages cannot be identified separately without adjusting for the effect of local cost of living and amenity differences, and thus classical Mincer-type wage equations with monopsony measures added could be subject to substantial omitted variable biases. The second essay matches each potentially monopsonized occupation with comparable occupations that are less prone to monopsony. Then, the effect of interregional differences in monopsony power for a particular occupation on interregional relative wage differentials with respect to a similar occupation is analyzed using five proxy measures of monopsony power taken from previous literature. The third essay concerns measurement of intercity differences in housing cost. Intercity housing price indexes that rely on median house price, pooled hedonic regressions, or repeat sales adjust imperfectly for differences in housing characteristics. In addition, intercity house price indexes that rely on asset value are a biased measure of differences in the rental price of housing, because capitalization rates vary dramatically across cities. To mitigate these shortcomings, the third essay creates a Fisher Ideal intercity housing price index for rental rates and asset value using a twofold Blinder-Oaxaca decomposition. The original method introduced in the third essay improves upon current house price indexes by allowing implicit prices to vary across locations.

Url: https://www.proquest.com/docview/2713709409/B0CEE37A3A21468CPQ/1?accountid=14586

User Submitted?: No

Authors: Chung, Hyung Joon

Institution: The George Washington University

Department: Economics

Advisor:

Degree:

Publisher Location:

Pages: 1-232

Data Collections: IPUMS USA

Topics: Housing and Segregation, Population Mobility and Spatial Demography

Countries: