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, restatement-based...