Academic journal article Political Research Quarterly

The Corruption-Enhancing Role of Re-Election Incentives? Counterintuitive Evidence from Brazil's Audit Reports

Academic journal article Political Research Quarterly

The Corruption-Enhancing Role of Re-Election Incentives? Counterintuitive Evidence from Brazil's Audit Reports

Article excerpt

Traditional wisdom on the effect of re-election on incumbent performance is that first-term politicians refrain from rent extraction because they want to be re-elected. The authors find no evidence to support this belief. The corruptionreducing effects of re-election may be offset when the expected utility for rent extraction is very high. This occurs when the likelihood of detecting corruption is very low. Mayors who committed irregularities are less likely to be re-elected when this information is released in the electoral year only. Although citizens hold mayors accountable for bad behavior, they do not do so to the point of discouraging misbehavior.

Keywords: corruption; re-election; accountability; political competition

In 2006, in the State of Pernambuco, Brazil, eight former mayors were put in jail. They were imprisoned for a variety of crimes, including malfeasance, serious irregularities in bid procurement, overinvoicing, and forgery of payments to so-called phantom (i.e., existing in paper only) firms belonging to themselves or relatives while in office. It is surprising that all of them had been mayors for two terms of office and had committed crimes in their first term. More significantly, although most of these crimes were detected by the state audit court in the first term of office, it did nothing to impede die mayors' continued perpetration of these crimes or running for re-election. This represents a puzzle contrary to me traditional wisdom on the governance-enhancing role of re-election incentives. Why would they run for office again if their crimes were made public by the audit court through a variety of mechanisms? Why does it seem that first-term mayors engaged in corrupt practices more often than "lame ducks" in their second and final term (the Constitution bans running for re-election for a third term)? Are citizens more likely to reward corrupt practices when these are disclosed by watchdogs?

Since Schumpeter's [1942] (1972) attack on the classical notion of public interest, democracy has come to be evaluated not by the ability of governments to express the general will - or in modern jargon, a social welfare function - but by the ability of citizens to hold governments accountable. Elections are important because citizens are able to hold incumbent politicians accountable by either punishing or rewarding their performances (Powell 2002). In the case of poor performance, citizens would be able to "throw the rascals out" (Riker 1983, p. 244). The possibility of re-election provides politicians with incentives to align their preferences with that of the citizens and to refrain from rent-seeking behavior. Ferejohn's (1986) highly influential model predicts that incumbent politicians avoid maximum rent extraction in their first electoral term to get re-elected and enjoy future rents. Besley and Case (1995) propose an alternative model in which incumbent governors' opportunistic behavior may be curtailed through the possibility of re-election.

When the re-election incentives argument is extended to encompass the factor of corruption, it is expected that incumbents seeking re-election are supposedly less corrupt than those who are ineligible to re-election. Corruption behavior can be distinguished from mere rent-seeking and opportunism. However, we can expect a similar deterrence effect in both caused by the election process. Admittedly, due to measurement problems and data, few empirical studies have tested the impact of re-election incentives on corruption. The illicit nature of corruption practices and the corresponding difficulty in finding evidence and measuring the extent of corruption renders studies on corruption problematic. Ferraz and Finan (2007b) provide one of the few empirical tests of the governance-enhancing role of re-election incentives. Their test uses the term limits in Brazil's municipal elections to determine whether re-election incentives affect the level of rent extraction of incumbent politicians. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.