Labor Market Polarization and Student Debt with Elena Loutskina (UVA) and Constantine Yannelis (Cambridge)
Abstract: This paper shows that labor market polarization—the shift in employment from middle-skill jobs toward high and low skill jobs—sharpens incentives to take on student debt to pursue higher education. We find that consumers experiencing deeper labor market polarization incur higher student debt balances through greater participation rates in this credit market. This relationship cannot be attributed to their inability to repay existing student debt and is about three times stronger for credit-constrained consumers. We confirm the economic mechanism by showing that deeper labor market polarization is associated with higher college enrollment rates among graduating high school students.
Who Struggles After Buying? Disparities in New-Homeowner Financial Distress with Carlos Avenancio-Leon (UCSD)
Abstract: Discussions of homeownership and mortgage-access policy often flaunt its benefits while giving limited attention to its costs. Using linked property transactions and credit records, we document race and gender gaps in financial fragility after home purchase. Black and female borrowers show larger increases in 30+ and 90+ day delinquencies than White and male borrowers, even after controlling for individual and neighborhood fixed effects, property characteristics, credit score, and income-at-purchase. They also accumulate more student debt, while credit card balances rise for all borrowers but faster for Black borrowers. These post-purchase disparities inform our understanding of benefits to homeownership.
Evaluating Mortgage Renegotiation Strategies: A Data-Driven Framework for Investors Management Science (2025) | Working paper version
Abstract: This paper offers a novel framework to quantify the expected gains from renegotiating delinquent loans. The framework accounts for important trade-offs between concessions to borrowers, post-delinquency loan performance and expected collateral values. The framework's parameters are calibrated using data on renegotiated 30-year residential Fixed-Rate Mortgages that went delinquent during the Great Recession. Our model-implied expected gains increase during the 2007 to 2009 period coinciding with an increase in the rate of loan renegotiation. Counterfactual analyses show that larger expected gains can be generated from employing principal forbearance and extensions of the term-to-maturity, compared to principal write-downs and interest-rate reductions. On the other hand, principal write-downs can be a powerful tool when borrowers are deeply underwater. Our analyses illustrate how lenders or policy-makers might deploy this framework when faced with another delinquency crisis.
This paper previously circulated under the title 'The Limited Benefits of Mortgage Renegotiation'.
Information in Financial Contracts: Evidence from Securitization Agreements , Journal of Financial and Quantitative Analysis (2024) with Brent Ambrose (Penn State), Yiquiang Han (Roku Inc.) and Lily Shen (Clemson))
Abstract: We bring to bear a novel application of machine learning (ML) to compare the Pooling and Servicing Agreements (PSA) that govern asset-backed securitization deals in order to shed new light on the operation of these securities. The PSA is often viewed as mostly boilerplate legal text and thus may appear similar across deals despite heterogeneity in the underlying collateral and deal structures. After documenting variation in PSAs, we explore how the differences and similarities between PSAs governing commercial mortgage-backed securities (CMBS) relate to the performance of the mortgages contained in the security pool as well as to the performance of the individual securities issued as part of the securitization deal. Our findings suggest that underwriters tailor the governing documents to reflect the macroeconomic environment prevailing at the time of origination as well as differences in the underlying collateral pool.
Working paper version
The Agency Costs of Tranching: Evidence from RMBS, Journal of Financial Intermediation (2023)
Abstract: This paper documents the agency costs resulting from the deeper tranching of subprime residential mortgage pools. Mortgage servicers are less likely to renegotiate delinquent loans collateralizing a greater number and variety of tranches. We find that an interquartile increase in tranching reduces mortgage servicers’ probability of loan renegotiation by 14% relative to the mean. This effect is concentrated in mortgages with greater ambiguity surrounding the loan value maximizing action. Overall, our results support the notion that tranching worsens agency frictions by increasing coordination costs among investors and impeding their monitoring of the agent.
Working paper version
Partial Deregulation and Competition: Effects on Risky Mortgage Origination, Management Science (2019), with Marco Di Maggio (HBS) and Amir Kermani (Berkeley)
Abstract: We exploit the preemption of national banks from state laws against predatory lending by the Office of the Comptroller of the Currency (OCC) as a quasi-experiment to study the effect of deregulation and its interaction with competition on the supply of complex mortgages. Following the preemption ruling, national banks significantly increased their origination of loans with prepayment penalties relative to national banks in states without anti-predatory-lending laws. We highlight a competition channel: in counties where OCC-regulated lenders had larger market shares, non-OCC-regulated lenders responded by increasing their supply of contracts with riskier features, such as deferred amortization, which were not restricted by the state anti-predatory-lending law. Non-OCC lenders were able to lure borrowers into these higher-margin contracts by increasing their appeal with teaser rates.
Multiple Tranches, Information Asymmetry and the Impediments to Mortgage Renegotiation
Presented at: London Business School Trans-Atlantic Doctoral Conference 2016, ABFER 2019, AREUEA National Meeting 2019
My paper 'The Agency Costs of Tranching: Evidence from RMBS' partially absorbs results from this paper.
Funding Shocks and Competition in the Mortgage Market
Abstract: How do primary and secondary mortgage markets interact? This paper shows that funding shocks to mortgage originators interact with the degree of local credit market competition to increase lending growth. Specifically, I use a shift-share approach to estimate the causal effect of the growth in private-label securitization related funding. This effect is stronger in more competitive mortgage markets; with less-regulated non-bank lenders being most responsive to this competition channel. The results emphasize the interaction between the two layers of the mortgage market and document how credit market structure can amplify, or dampen, shocks to the economy.