Camille Hainnaux : camille.hainnaux[at]univ-amu.fr
Daniela Horta Saenz : daniela.horta-saenz[at]univ-amu.fr
Jade Ponsard : jade.ponsard[at]univ-amu.fr
Nathan Vieira : nathan.vieira[at]univ-amu.fr
Typically, anti-corruption campaigns are believed to promote accountability, strengthen democracy, and establish the rule of law. However, this paper demonstrates how corruption efforts can be manipulated to persecute political opponents and consolidate the incumbent's power. Through regression discontinuity design estimates, I show that opposition politicians charged with corruption in Pakistan's Anti-Corruption (NAB) courts are more likely to be convicted if they narrowly win the election compared to if they narrowly lose. This targeting of opposition politicians seems to displace the prosecution of corruption cases against non-politicians. In courts that are particularly congested, corruption convictions of opposition politicians crowd-out corruption convictions of non-politicians. In contrast, associates of the incumbent government who narrowly win the elections are less likely to be convicted. Analysis of mechanisms reveal that career concerns are likely important to explain these results. The judges who convict opposition politicians are more likely to be promoted. When the power to appoint and promote judges is taken away from the politicians, the career concerns mechanisms break down and the effect on convictions attenuates. The results cannot be explained by alternative mechanisms of judge selection or peer effects. Overall our findings underscore the need for caution when implementing anti-corruption campaigns. While such campaigns are typically viewed as a means to promote accountability, they can also be manipulated to target political opponents and maintain the government's hold on power.