We examine the causal effects of a policy allowing early withdrawal of pension funds on individuals’ investment behavior. Upon turning 55, eligible individuals may withdraw a portion of their pension savings. Using detailed brokerage data, we find that this liquidity access triggers increased trading of 9% to 33%, especially in riskier, leveraged assets, without improving investment performance. The resulting increase in trading costs and portfolio volatility, particularly among males and lower-income investors, ultimately diminishes retirement wealth.

We study a controlled experiment in which a bank's loan officers were incentivized based on originated loan volume to encourage prospecting for new business. While treated loan officers did attract new applications, both extensive and intensive margins of loan origination expanded (+ 31% new loans; loan size + 15%). We find that loan officers gave greater weight to hard information in approval decisions. Despite no change in the observable characteristics of approved loans, their default rate increased (+ 24%). Finally, the bank's imputed credit default model lost its predictive power. Overall, loan prospecting incentives led to unfavorable soft information being overlooked in the origination process.

Using several large data sets of mortgage loans originated between 2004 and 2007, we find that in the prime mortgage market, banks generally sold low-default-risk loans into the secondary market while retaining higher-default-risk loans in their portfolios. In contrast, these lenders retained loans with lower prepayment risk relative to loans they sold. Securitization strategy of lenders changed dramatically in 2007 as the crisis set in with most unwilling to retain higher-default-risk loans in return for lower prepayment risk. Contrary to the prime market, the subprime market does not exhibit any clear pattern of adverse selection.

We analyze an experiment conducted by a large U.S. bank that offered consumers achoice between two credit card contracts, one with an annual fee but a lowerinterest rate and one with no annual fee but a higher interest rate. We findthat on average consumers chose the credit contract that minimized their costs.A substantial fraction of consumers (about 40%) still chose the suboptimalcontract. Nonetheless, the probability of choosing the suboptimal contractdeclines with the dollar magnitude of the potential error, and consumers withlarger errors are more likely to subsequently switch to the optimalcontract.

This paper shows how liquidity infusions affect loan modification in the mortgage market. The design of pooling and servicing agreements leads mortgage servicers to prefer foreclosure over modification when they are liquidity constrained. Therefore, a positive liquidity shock is expected to boost modification rates. Using a residential mortgage dataset that includes loan-level information, we find that the Troubled Asset Relief Program significantly increased the modification rate. Our findings help us better understand the economic consequences of government intervention and have important policy implications for the renegotiation of distressed mortgages.

Using unique consumer financial transactions of more than 56,000 consumers, we study the consumption response to a housing policy experiment in Singapore that resulted in a decrease in access to home equity. Using difference-in-differences analysis, we find a significant negative consumption response to the policy shock. Moreover, the consumption response is concentrated in credit card spending and is stronger among individuals with limited access to credit market or with a high precautionary saving motive. These results suggest that a decrease in access to home equity reduces the role of housing as a self-insurance mechanism for consumption smoothing.

Given the growth in home equity lending during the 1990s, it is imperative that lenders and regulators understand the risks associated with this segment of the residential mortgage market. Using a unique panel data set of over 135,000 homeowners with second mortgages, our analysis indicates that significant differences exist in the prepayment and default probabilities of home equity loans and lines, providing insights into bank minimum capital requirements. We find that households with equity loans are relatively more sensitive to changes in interest rates. By contrast, households with equity lines are more sensitive to appreciation in property value.

We examine whether relative income differences among peers can generate financial distress. Using lottery winnings as plausibly exogenous variations in the relative income of peers, we find that the dollar magnitude of a lottery win of one neighbor increases subsequent borrowing and bankruptcies among other neighbors. We also examine which factors may mitigate lenders’ bankruptcy risk in these neighborhoods. We show that bankruptcy filers obtain more secured, but not unsecured, debt, and lenders provide additional credit to low-risk, but not high-risk, debtors. In addition, we find evidence consistent with local lenders taking advantage of soft information to mitigate credit risk. Received October 12, 2016; editorial decision January 15, 2019 by Editor Philip Strahan.