Does Exposing Corrupt Politicians Reduce Corruption?*

Abstract: Does the disclosure of information about corruption practices induce a sustained reduction in corruption levels? We use publicly-released routine audit reports to study this question. The government of Puerto Rico has established a mechanism to routinely conduct municipal government audits, whose findings are then made publicly available and disseminated to media sources. Using a longitudinal dataset of corrupt violations constructed from the audit reports during the period 1987-2006, we compare the subsequent term governments ’ levels of reported corruption for municipalities audited at different points in time around an election. The pre-election release of the audit reports led to significant short-term reductions in municipal corruption levels and an increase in incumbent mayors’ electoral accountability. However, municipal corruption levels in the subsequent term are higher in municipalities audited preceding the previous election and these are concentrated among municipalities shown to have refrained from rent-seeking activities in the first audit. These findings are consistent with a political agency model of reputation dynamics in whic