Research

Abstract:  This paper examines the role of geographical diversification in the U.S. homebuilding industry. Despite being traditionally fragmented, the US homebuilding sector has seen significant expansion of national builders over the past two decades. Using the universe of housing transaction data from the second half of the housing busts, I examine how interdependent pricing decisions of national builders of single-family homes are influenced by the firm’s financial frictions and the local market structure. I design an equilibrium model for new housing markets wherein national builders set prices in each market to offload existing inventories and increase cash flow. I recover novel estimates of market-specific values of finished home inventory and convex external financing costs. The model estimates suggest that financial frictions decreased profit margins for the average public builder by 20% in the sample period. Counterfactuals indicate that builders operating across diverse geographical markets are better protected against local housing shocks, due to their capacity to adjust pricing strategies across local markets with varying economic conditions.

Abstract:  Using the universe of housing transaction data from 2001 to 2015, this paper documents how local housing shocks propagate across US regions through homebuilders' internal capital market. Consistent with a model of inventory depletion under capital market imperfections, I find that national homebuilders that experience a decrease in the value of the sale in one area lower the sales price in another area, causing the nearby competing sellers to lower the sales price. A 10 percent price shock from other regions transfers to a 3 percent decline in sales prices in unaffected areas, and a 0.6 percent decline in sales prices of nearby competing properties. The distressed builder's discount and spillover effects persist in States that are less affected by the Great Recession shocks, suggesting that the price declines in some geographical markets are attributable to the national builders' financial distress. While such spillover may harm local builders and individual sellers, distressed properties are more likely to be purchased by buyers who use non-traditional mortgages with a higher loan-to-value ratio, indicating potential reallocation effects. 

Abstract:  This paper assesses the extent of monopsony power in the labor market for highly skilled immigrants, focusing on employment concentration and its impact on wages. We obtained the universe of petitions to the H1-B visa program, the largest high-skilled immigration program in the U.S., through a FOIA request. We first discover that the H-1B labor market is 30% more concentrated than the broader U.S. labor market, with a steadily increasing trend over the past decade. We then examine the causal impact of high employment concentration on H-1B workers’ wages by leveraging the lottery system for over-the-cap H-1B applications. The random lottery win rates in each labor market alter employment concentration, as different-sized firms vary in demand for first-time H-1B applicants, the only group subject to the lottery. Our results indicate that transitioning from the 25th to the 75th percentile of employment concentration results in a 12.2% wage decline, equivalent to a $10.5k salary loss for the median H-1B worker. The wage effects of concentration are primarily borne by first-time H-1B applicants, whose visa status is tied to a single employer. As a suggestive mechanism to these wage effects, we also find evidence that higher concentration leads to less job mobility between H-1B employers. These findings collectively suggest that H-1B policy reform should consider lifting the employer tie for first-time applicants.

Abstract: This study investigates the impact of teardowns and renovations on housing supply and gentrification dynamics in a city. Using single-family housing transaction records, we use a novel nonparametric approach to identify instances of housing upgrading activity by analyzing relative changes in resale prices within specific zip codes. Our findings show that counties with high land prices and stringent land regulations, as well as zip codes with strong appreciation rates and higher income levels, are more likely to see significant renovations or teardowns of pre-existing residential properties. Using a theoretical model of housing assignment with endogenous supply choice, we quantify the costs of teardowns, renovations, and new constructions. Finally, we examine how a range of policy interventions to enhance housing affordability, such as incentivizing affordable housing construction and change in land use regulations affect the distribution of housing qualities and demographics in the long run.