PDF version

Politicians, Publicly-Released Audits of Corruption, and Electoral Outcomes in Brazil

Location: Brazil
Sample: 669 Municipalities
Target Group: 
Outcome of Interest: 
Intervention Type: 
AEA RCT Registration Number: 

Although citizens are responsible for the election of political leaders in most democracies, voters often lack access to relevant information about politicians’ past performance and preferences which could help them make informed voting decisions. Researchers evaluated how exposing information about corrupt politicians impacted election results in Brazil. Releasing information about politicians’ corruption levels prior to an election significantly decreased reelection rates of corrupt incumbent municipal mayors. Having a radio station in the municipality further decreased these reelection rates, whilst improving the chances of non-corrupt politicians.

Policy Issue 

Although citizens are responsible for the election of political leaders in most democracies, voters often lack access to relevant information about politicians' past performance and preferences which could help them make informed voting decisions. In situations where the interests of voters and politicians are not aligned, politicians may be incentivized to exploit their informational advantage for personal gain, often resulting in corrupt practices. Transparency is valuable to democratic governments because it can provide citizens, who have the best understanding of their own needs and preferences, with information on how officials are working to meet those needs. Transparency is often considered to be the most effective and powerful method of monitoring politicians and preventing corruption, but there is limited evidence on how public awareness of corruption affects political accountability.

Context of the Evaluation 

In May 2003, the federal government of Brazil started an unprecedented anti-corruption program based on the auditing of municipal government' s expenditures. This program, implemented through the Controladoria Geral da União (CGU), a federal agency responsible for overseeing the implementation of federal resources, aims to discourage the misuse of public funds among public administrators.

Under this program, each year federal auditors from the CGU audit a random 1 percent of Brazil' s 5,500 municipal governments. The auditors examine accounts and documents for any irregularities in federally-sponsored projects and public works. After approximately one week of inspections, a report describing all the irregularities is released to public prosecutors and the municipal legislative branch.

Women discussing the merits of politicians.

Women discussing the merits of politicians.

Details of the Intervention 

Researchers in Brazil sought to investigate whether making information on politician corruption publically available would affect the electoral outcome of incumbent mayors. Political corruption is defined as any irregularity associated with fraud in procurements, diversion of public funds, and over-invoicing.

Prior to the October 2004 municipal elections, the Federal government audited 676 municipalities. In a random sub-set of 376 municipalities the results of the audit were released prior to the election. The remaining 300 municipalities served as a comparison group, and did not have the results of the audit released until after the election. The randomized assignment provided an opportunity to observe whether voter-access to information about a politician' s corruption level prior to the election impacted the average vote share and re-election rate for incumbent mayors.

Data about political outcomes and mayoral characteristics are drawn from the Tribunal Superior Eleitoral (TSE) which provides vote totals for each candidate by municipality, and data about the socioeconomic and demographic characteristics of each municipality are drawn from national population censuses.

Results and Policy Lessons 

For every additional corrupt violation reported, the audit policy reduced the likelihood of re-electing an incumbent by approximately 20 percent. The effect of the policy was similar for other measures of electoral performance, such as the change in vote share and margin of victory. These results suggest that voters not only care about corruption, but once empowered with the information, they update their prior beliefs and punish corrupt politicians at the polls.

In municipalities with local radio stations, the effect of disclosing corruption on the incumbent' s likelihood of re-election was more substantial. Results indicated that for municipalities that released audit results prior to the election and revealed at least one count of corruption, the presence of an additional radio station decreased the incumbent's probability of re-election by 10.7 percent. Not only did radio stations increase the effect of the audit when corruption was revealed, it also promoted the re-election of non-corrupt incumbents. When corruption was not found in a municipality with a local radio station, the audit increased the likelihood that the mayor was re-elected by as much as 20 percentage points.

These results indicate that the disclosure of information enhances political accountability.

Ferraz, CLaudio and Frederico Finan. "Exposing Corrupt Politicians: The Effects of Brazil's Publicly Released Audits on Electoral Outcomes." The Quarterly Journal of Economics123 (2): 703-745

Avis, Eric, Claudio Ferraz, and Frederico Finan. "Do Government Audits Reduce Corruption? Estimating the Impacts of Exposing Corrupt Politicians." Working Paper, May 2017.