What ‘bad apples’ removed from barrels can tell us about corruption

At first glance, identifying who has been sanctioned for corruption may reveal little about how corruption operates in an organization. In an environment where corruption is very likely to be systemic, institutionalised, and/or tolerated to a certain extent, as it happens in Brazil – used here as a case study –, there are indeed those who were caught but are not punished, those who have never been caught, and those who may still be engaging in misconduct no matter the risks of being sanctioned. However, pinpointing certain characteristics of those who were implicated in acts of corruption may reveal much about how accountability systems work. By asking who is sanctioned for corruption and which features differentiate them from those punished for other serious offenses, this article explores a unique database 5,000 administrative penalties imposed against civil servants between 2003 and 2014 in the Brazilian federal executive. The analysis here is performed for the subsample of corruption and non-corruption expulsive sanctions together with variables gender, job tenure, status (career civil servant or occupant of a position of trust), and gross earnings. The analysis and discussion integrate qualitative data gathered through 24 semi-structured interviews with civil servants directly involved with disciplinary sanctions. What emerges is a complex picture that defies easy stereotypes. Those punished most often for corruption are men and those who are not in the first five years of their careers. Data also suggest that holding a special position or earning a bonus is correlated with being more sanctioned for corruption as opposed to other misconducts. The primary purpose of this paper is to discover whether there are important differences in the demographic characteristics among those civil servants who were caught and punished in the Brazilian federal executive, categorizing them by sanctions against corruption and against other serious offenses. In doing so, this paper aims to contribute to corruption and public management literature, there is still a lack of studies exploring who those punished in public organizations are and how effective sanction enforcement is as a response to bureaucratic misconduct, especially to corruption. The paper also engages in the discussion of what we can learn from disciplinary punishment data, mainly about methodological challenges and empirical contributions to the anti-corruption field. Although we agree that official crime statistics in general, or disciplinary sanctions as it is the case of this study, cannot be considered a reliable measure of corruption, data on punishment may offer new possibilities of analysis that are still largely unexplored. This study suggests that the analysis of disciplinary sanctions allows, for example, identifying and discussing whether – and, if so, why – disciplinary and internal affairs departments have been punishing more men than women, more experienced bureaucrats than inexperienced ones, and more career civil servants than those in positions of trust. In this regard, this study aims to open new avenues for exploring the issue of punishment for corruption at a more individual level.