Credit Scoring and Competitive Pricing Default Risk: Positive and Normative Implications

University Of Texas At Austin, Austin TX

Investigators

Abstract

The goal of this project is to understand consumer bankruptcy. As is clear from the meltdown in the subprime mortgage market beginning in late 2006, consumer bankruptcy has important implications for the health of the U.S. economy and in turn government responses to the crisis. This research provides an economic framework to understand the reasons households default and how financial institutions price their consumer loans when there is risk of default. In practice, a consumer's credit history includes detailed records of the individual's past credit payments and adverse events such as bankruptcy. This detailed information is summarized by an individual's credit score (a number between 300-850) which gauges the likelihood that an individual will default. The higher the score, the less likely an individual will default. Subprime borrowers typically have credit scores of 620 or lower. Interest rates on loans to people with low scores are typically high in order to compensate the lender for the risk of default. To understand these practices, this project provides a model of unsecured consumer credit where borrowers have the legal option to default and lenders learn from an individual's borrowing and repayment behavior about his unobservable characteristics in order to price loans in a competitive market. The model is used to shed light on consumer welfare in the presence of earnings uncertainty and to study the consequences of variations in regulation affecting consumer debt including bankruptcy law. Specifically, the model helps answer three questions. First, is it possible for competitive credit markets to support lending to subprime borrowers given that it is difficult to assess their creditworthiness? Second, can the model reproduce key bankruptcy and credit scoring facts when confronted with data? Third, are legal restrictions like the Fair Credit Reporting Act (which requires adverse credit information like a bankruptcy to be stricken from one's record after a certain number of years) welfare improving? To see why a model is necessary to answer the third question, there is already empirical evidence, provided by David Musto of the Wharton School, that the removal of the bankruptcy flag leads to excessive credit, increasing the eventual probability of default. This is concrete evidence that the Fair Credit Reporting Act has real economic effects. As Musto suggests, his findings indicate market efficiency in reverse. On the one hand, in a world of imperfect information, the legal removal of a bankruptcy flag provides insurance or a "fresh start" to households who have experienced adverse events in situations where competitive intermediaries cannot provide such insurance. On the other hand, extending the length of time that a bankruptcy flag remains on one's credit record provides the right incentives not to default. In the end, a model calibrated to real world data is necessary to assess the welfare gains associated with these tradeoffs. The intellectual merit of the research is the development of what may be the first quantitative model linking general equilibrium theory with data on credit scoring and consumer bankruptcy. The broader impact of the proposal may be of use to policymakers trying to understand how bankruptcy law affects the extension of credit to households and to lending institutions trying to assess household creditworthiness.