Published Papers


American Economic Review, September 2014, 104(9): 2830-57.

with Christopher Mayer, Edward Morrison, and Tomasz Piskorski

We investigate whether homeowners respond strategically to news of mortgage modification programs. We exploit plausibly exogenous variation in modification policy induced by U.S. state government lawsuits against Countrywide Financial Corporation, which agreed to offer modifications to seriously delinquent borrowers with subprime mortgages throughout the country. Using a difference-in-difference framework, we find that Countrywide's relative delinquency rate increased thirteen percent per month immediately after the program's announcement. The borrowers whose estimated default rates increased the most in response to the program were those who appear to have been the least likely to default otherwise, including those with substantial liquidity available through credit cards and relatively low combined loan-to-value ratios. These results suggest that strategic behavior should be an important consideration in designing mortgage modification programs.

Journal of Legal Studies, June 2016, 54(2):571-505

with Christopher Hansman and Ethan Frenchman

Roughly 450,000 people are detained awaiting trial on any given day, typically because bail has not been posted. Using a large sample of criminal cases in Philadelphia and Pittsburgh, we analyze the consequences of bail assessment and pretrial detentions by exploiting the variation in bail setting tendencies among randomly assigned bail judges. Our estimates suggest that the assignment of money bail leads to a 6 percentage point rise in the likelihood of pleading guilty, and a 4 percentage point rise in recidivism. We also find evidence for racial bias in bail setting. Our results highlight the importance of credit constraints in shaping defendant judicial outcomes and point to important fairness considerations in the institutional design of pretrial detention programs. 


Working Papers


Revise & Resubmit, Journal of Finance

I analyze the existence of default spillovers in the residential mortgage market. I focus on shocks to interest rates paid by borrowers resulting from two administrative details in ARM contract terms: the choices of financial index and lookback period. I find that a 1 percentage point increase in interest rates at the time of ARM reset results in a 2.5 percentage rise in the probability of foreclosure in the following year. Instrumenting for mortgage foreclosure using these reset characteristics, I find evidence that each foreclosure filing leads to an additional 0.3 to 0.6 completed foreclosures within a 0.10 mile radius. I document that the price effects of foreclosures on neighboring home prices are unlikely to completely ac- count for the magnitude of this effect. Complementing the price channel, I emphasize two additional mechanisms: a bank-supply channel resulting in a one-third drop in refinancing activity after a foreclosure, and a borrower response channel arising from peer effects. Neighboring borrower payment responses are linked to the timing of local mortgage default, are not associated with defaults on revolving debts, and are concentrated in areas with few nearby local foreclosures—consistent with an information channel based on learning about the costs of default. I also provide suggestive evidence on the macroeconomic impact of fore- closure spillovers: I find that counties and zip codes experiencing greater intensity of reset among borrowers in adverse conditions experience subsequent drops in house prices and higher foreclosure volumes. These results shed light on an important amplification channel of shock transmission during the recent foreclosure crisis. 

with Kunal Sachdeva

Using a comprehensive and survivor-bias free dataset of U.S. hedge funds, we document the role that inside investment plays in managerial compensation and fund performance. We find that funds with greater investment by insiders outperform funds with less ``skin in the game'' on a factor-adjusted basis; exhibit greater return persistence; and feature lower fund flow-performance sensitivities. These results suggest that managers earn outsize rents by operating trading strategies further from their capacity constraints when managing their own money. Our findings have implications for optimal portfolio allocations of institutional investors and models of delegated asset management. 

with Edward Morrison, Lawerence J. Cook, Heather Keenan, and Lenora M. Olson

This paper assesses the importance of adverse health shocks as triggers of bankruptcy filings. Using a sample of all auto crashes in Utah during a thirteen year period, we report two findings: (i) There is a strong positive correlation between an individual's financial condition and his or her likelihood of suffering a health shock, an example of behavioral consistency, and (ii) After accounting for this simultaneity, we are unable to identify a causal effect of health shocks on bankruptcy filing rates. These findings emphasize the importance of risk heterogeneity in determining financial fragility, raise questions about prior studies of "medical bankruptcy," and point to important challenges in identifying the triggers of consumer bankruptcy.

with Edward Morrison, Catherine R. Fedorenko, and Scott Ramsey

This paper tests whether housing wealth mitigates the effects of health shocks on financial stress and mortality. We link cancer records to mortgage, bankruptcy, foreclosure, and credit report data. We find that cancer diagnoses are financially destabilizing even for households with health insurance, but the effect is driven by households without home equity. Households with equity extract it (by refinancing a mortgage or taking out a second). They are also more likely to accept recommended therapies and have higher post-diagnosis survival rates. Our findings show that housing wealth plays an important role in understanding how individuals buffer idiosyncratic shocks.

with Christopher Hansman

We estimate the presence of asymmetric information in adjustable rate mortgage (ARM) mortgage markets using detailed mortgage and credit data. Exploiting the distinction in ARM contracts between known initial "teaser" rates and ex-ante unknown reset rates, we distinguish between selection and moral hazard effects to identify both effects separately. We further utilize novel forms of exogenous variation in reset rates based on administrative differences in contracts. We find strong evidence of moral hazard or strategic default effects, and suggestive evidence of selection effects. Our effects are strongest for liquidity constrained and underwater individuals, but remain substantial for other borrowers. Our results emphasize the importance of informational issues in mortgage markets and point to unique challenges in policies aimed at reducing foreclosures.