The Local Curse: Spillover Effects of Mayoral Victories in Competitive Autocracies

Abstract

This paper investigates whether controlling local offices in competitive autocracies results in a better local electoral performance at the national level, as previous literature has suggested for democracies. Using newly coded electoral data from Venezuela and a regression discontinuity design in close elections, the study examines the causal effect of controlling the municipal executive on the national (and local) electoral performance of the pro-regime party. We find evidence of a negative spillover effect, where the election of a pro-regime mayor leads to a 4 percentage point decrease in their party's vote share in the subsequent national elections and a 24 percentage point reduction in the probability of electing a pro-regime mayor in the next election. To explain those results, we build a model of political selection in competitive autocracies, which suggests that a relevant channel may be a negative selection effect of pro-regime mayoral candidates.