The Effects of Supervision on Bank Performance: Evidence from Discontinuous Examination Frequencies

Previous research has found little evidence that banking supervision improves bank performance, possibly because supervision is endogenous to performance. This paper estimates causal effects of supervision on performance using discontinuities in the minimum frequency of examinations imposed by regulation. The bank asset size threshold at which these discontinuities occur changes over time, providing important variations both across banks and across time for identification. In particular, the time varying threshold allows us to remove confounding factors that may be present at certain asset sizes. We find that more frequent examinations increase profits and decrease loan losses and delinquencies. This is consistent with the hypothesis that regulators limit the risks that banks are exposed to and, consequently, limit their losses on risky assets. Our findings suggest that banking supervision improves bank performance.