While in public citizens express their clear rejection of corruption and their intentions not to support corrupt politicians, empirical evidence shows that voters around the world only mildly punish corrupt politicians. From a theoretical point of view it is puzzling how corrupt governments survive in democratic societies, as one of the main functions of free elections, a fundamental to every democracy, is to hold governments accountable. According to both the sanctioning and the selecting model of voting behavior, it is perplexing why an informed and free citizen would vote for a corrupt incumbent. Corruption is a clear signal of a harmful government that will not act in the voters’ best interests, and a distinct motive to sanction the ruling government.
In this study, I use a multidimensional survey experiment to assess under what circumstances voters decide to choose a crooked politician. I randomly assigned 2275 respondents to different pairs of politicians’ profiles and asked them their probability to vote for each of the mayors if they were running in the next municipal elections. Each profile has randomly assigned information about the candidate’s integrity, gender, party affiliation, educational level and managerial experience and economic performance in the previous mandate. Providing respondents different sets of information I overcame two issues that have been bothering researchers. (I) I mimic the informational scenarios voters face in a real election better, as respondents have to consider many different aspects when casting a vote. (II) I diminish the pressure for respondents to answer what they think is more socially accepted.
The results of this study provide clear-cut evidence that, even when obtaining highly credible information, citizens might choose to overlook corruption if they like other characteristics of the candidate. Respondents care about corruption but in line with previous findings, partisanship clearly leads voters to condone corruption. Co-partisanship strongly moderates the negative effect corruption has on the likelihood of voting for a politician. Furthermore, partisanship is together with corruption the attribute that most determines the vote. The economic performance of a politician can also condition whether or not a voter chooses a corrupt politician, however the results for this moderating effect are softer and less robust.
Sofia Breitenstein is a PhD candidate at the Universitat Autònoma de Barcelona (UAB) Political Science Department. She holds a FPI scholarship to study citizens’ attitudes towards corruption and their political consequences, and her main areas of research are: corruption accountability, political behavior, public opinion, political attitudes and experimental methodology. She is a member of the Democracy, Elections and Citizenship research group and the POLEXP network for experimental Political Science. Follow her on Twitter here: @sofiabrei.
First Published: 1 January 2019
From Research & Politics
Featured image credit: images used with permission of the author.