The first paper demonstrates the important role of job displacement in the household bankruptcy decision. Consistent with predicted filing behavior under persistent income shocks, I find that households in the NLSY are four times more likely to file in the year following job loss, with a smaller but significant response persisting two to three years. Aggregate patterns are also consistent with the model: At the county level, 1000 job losses are associated with 8-11 bankruptcies, the effects also last two to three years, and manufacturing job loss is more likely to induce bankruptcy than non-manufacturing job loss. The results suggest that providing credit counseling to vulnerable households at the time of displacement may be more effective than providing it at the time of bankruptcy.Theories of financial intermediation suggest that securitization reduces financial intermediaries' incentives to screen borrowers. The second paper, co-authored with Tanmoy Mukherjee, Amit Seru, and Vikrant Vig, examines this question using a unique dataset of securitized subprime mortgage loans. We exploit a rule of thumb in the lending market to generate exogenous variation in the ease of securitization and compare the composition and performance of lenders' portfolios around this threshold. Conditional on securitization, a portfolio that is more likely to be securitized defaults by 20% more than a similar risk profile group. Crucially, these two portfolios have similar observable risk characteristics and loan terms. Our results suggest that securitization can adversely affect lenders' screening incentives.The third paper, co-authored with Brian C. Cadena, uses insights from behavioral economics to offer an explanation for a surprising phenomenon: Nearly 20 percent of undergraduate students who are offered interest-free loans turn them down. We model the financial aid process and show that students facing self-control problems may optimally decline subsidized loans to avoid excessive consumption during school. Using the NPSAS, we investigate students' financial aid situations and subsidized loan take-up decisions and find that students who would receive their loans in cash are significantly more likely to reject the loan. These results suggest that consumers limit their liquidity in economically meaningful situations, consistent with the predictions of the behavioral model.
