Credit registers collect, store and share information regarding borrowers� past and current credit relations. Interestingly, such data is typically erased from the public records after a number of years, in accordance with privacy protection laws, which aim at providing individuals with a fresh start from past events. In order to secure credit-worthy but unlucky borrowers with a new beginning, however, these provisions end up removing all of the public information, including that possibly still relevant for screening purposes. This paper assesses such trade-off, by studying the impact of limited records on borrowers� behavior and market outcomes in a stylized credit market for unsecured loans. In this setup, limited records endogenously give rise to beneficial reputation effects in the form of higher equilibrium effort, which alleviate, rather than worsen, the distortions caused by asymmetric information. Further, we demonstrate that when moral hazard is high, 1-period records can achieve higher welfare and lead to a lower default rate than records that show all, or nothing, of the past history.
