The Design of Credit Information Systems

Abstract

We examine large credit markets with borrower moral hazard and bounded records. Defaulters should be temporarily excluded in order to incentivize repayment, but lending to defaulters who are the verge of rehabilitation is profitable. With perfect bounded information, defaulter exclusion unravels and lending cannot be sustained. By pooling recent defaulters with those nearing rehabilitation, coarse information disciplines lenders, since they cannot target loans towards the latter. Equilibria where defaulters get a loan with positive probability also improve efficiency, by raising the proportion of likely re-offenders in the pool of defaulters. Thus, endogenously generated borrower adverse selection mitigates moral hazard.