We study how political turnover in mayoral elections in Brazil affects public education provision. Exploiting a regression discontinuity design for close elections, we find that municipalities with a new party in office subsequently have test scores that are .05–.08 standard deviations lower. Party turnover leads to a sharp increase in the replacement rate of head-masters and teachers in schools controlled by the municipality. In contrast, turnover of the mayor’s party does not impact local (non-municipal) schools. These findings suggest that political turnover can adversely affect the quality of public services when the bureaucracy is not shielded from the political process.
Revise and Resubmit, American Economic Review