Replication materials: Rooting Out Corruption or Rooting for Corruption?

This github repository contains the replication materials for the paper "Rooting Out Corruption or Rooting for Corruption? The Heterogeneous Electoral Consequences of Scandals," forthcoming in Political Science Research and Methods, authored by Pablo Fernandez-Vazquez, Pablo Barbera, and Gonzalo Rivero.

Abstract: Corruption scandals have been found to have significant but mild electoral effects in the comparative literature (Golden 2006). However, most studies have assumed that voters punish all kinds of illegal practices. This article challenges this assumption by distinguishing between two types of corruption, according to the type of welfare consequences they have for the constituency. This hypothesis is tested using data from the 2011 Spanish local elections. We exploit the abundance of corruption allegations associated with the Spanish housing boom, which generated income gains for a wide segment of the electorate in the short term. We find that voters ignore corruption when there are side benefits to it, and that punishment is only administered in those cases in which they do not receive compensation.

This README file provides an overview of the replications files for the article.

• dataset.dta contains the dataset used in the study, in .dta format

• replication.do is a Stata do-file that generates quantities reported in the paper as well as the regression analyses

• figures.R is an R script that generates the figures included in the paper.