Political corruption and accounting choices

We examine how political corruption affects firms’ accounting choices. We hypothesize and find that firms headquartered in corrupt districts manipulate earnings downwards, relative to firms headquartered elsewhere. Our finding is robust to alternative corruption measures, alternative discretionary accrual measures, alternative model specifications, the instrumental variable approach and difference-in-differences analyses based on firm relocation and high profile cases. We find that firms headquartered in corrupt districts prefer income-decreasing accounting choices and exhibit higher conservatism. Finally, we find that the effect of corruption on earnings management is more pronounced for geographically concentrated firms, for firms without political connections, for firms in politically sensitive industries, for firms with lower transient institutional investor ownership and for firms with less analyst coverage. In sum, our findings suggest that firms respond to corruption by lowering their accounting earnings.