Research
Papers Under Review
(with Stuart Rosenthal)
In large urban areas, home prices appreciate faster in city centers, in part because of risk-return tradeoffs that vary in response to differences in housing supply constraints and volatility. This echoes a similar pattern across cities. Within urban areas, location-specific risk is most important, while across cities, systematic risk dominates. A one standard deviation increase in the dominant source of risk increases total housing returns by 22.7% and 12.2% within and across cities, respectively. It is well-known that home price levels vary spatially. Our findings indicate that spatial differences in home price appreciation rates can also persist in equilibrium.
Working Papers
This paper shows that for college-trained individuals, congestion and related uncertainty about how long a commute may take adversely affects the labor market outcomes of married women. This holds even after controlling for average commute times and other MSA-level attributes (including MSA size). These patterns are also present for both labor force participation (LFP) and the degree to which working individuals with professional training are employed in occupations for which they trained. Evidence affirms that larger MSAs enhance labor market opportunities, consistent with previous literature. New to this paper, average commute time has little effect on labor market outcomes. Instead, it is the volatility of commute times as proxied by rush hour congestion that discourages highly trained women from participating in the workforce, and which reduces the quality of their labor market match for those who are employed.
How Centralized is U.S. Metropolitan Employment?
(with Jason Brown, McKenzie Humann, Jordan Rappaport, and Aaron Smalter Hall )
Centralized employment remains a benchmark stylization of metropolitan land use. To address its empirical relevance, we delineate ``central employment zones'' (CEZs)---central business districts together with nearby concentrated employment---for 168 metropolitan areas in 2000. To do so, we first subjectively classify which census tracts in a training sample of metros belong to their metro's CEZ and then use a learning algorithm to construct a function that predicts our judgment. On average, the predicted CEZs account for only 16 percent of metropolitan employment in 2000. However, the distribution of shares is positively skewed. Moreover, employment centralization is considerably higher for agglomerative occupations---those that arguably benefit most from face-to-face contact and account for 24 percent of metro employment.
[This paper was originally posted as a Federal Reserve Working Paper. New draft coming soon.]
Research in Progress
Evidence on the Tradeoff Between Neighborhood and House Quality Among Low-Income Families
(with Stuart Rosenthal, Alex Rothenberg, and Samuel Saltmarsh)