Abstract

Information and Stigma in Household Bankruptcy Decisions

Ethan Cohen-Cole and Burcu Duygan-Bump, Federal Reserve Bank of Boston

Abstract

Using a very large national sample from one of the biggest US credit bureaus, this paper investigates the empirical relevance of stigma and other social factors on household bankruptcy decisions. Following the dramatic rise in personal bankruptcies, and the associated public policy debate, many researchers have attempted to analyze the micro dynamics of household bankruptcy decisions. Many potential explanations have been identified, and there is a growing consensus that households have become, or are becoming, more willing to default. Most observers have conjectured that the increased willingness to default reflects a diminution of the traditionally associated stigma. In this paper, we exploit sociological evidence and draw on a new methodology to disentangle stigma and social learning—-two of the most important social factors affecting default. Our results show that both factors significant; although on average societal stigma dominates the role of information. We also find that social stigma has indeed decreased somewhat on a national basis, as many have speculated. However, we also show that this aggregate trend disguises enormous heterogeneity. Exploiting the large size of our dataset, we evaluate differences in social patterns by community income and education levels. Our results show that while social factors appear quite important among the poor and less educated, the social stigma attached to bankruptcy has increased and information costs have decreased among these very groups. On the contrary, we show that it is among the rich and well educated that stigma has declined. These findings suggest that the overall increase in the bankruptcy rates cannot be explained by a decrease in social stigma, because it has fallen only for a small group of the population and one that has not been greatly impacted by bankruptcy. This implies that the key driver of the recent increases in bankruptcy is due either to changes in transaction and information costs or to changes in individual exposure to macroeconomic risk.