Measuring Public Corruption in the United States: Evidence From Administrative Records of Federal Prosecutions

Abstract

A growing empirical literature examines public corruption in the United States, but most of these studies use survey data on federal corruption convictions that is aggregated, sparse, and of dubious quality and provenance. This article describes an alternative data source from administrative records that provides more detailed and reliable information on corruption prosecutions and convictions by type of state official and by lead charge. Contrary to conventional wisdom, corruption convictions are not increasing and mostly involve low-ranking federal or local officials. Administrative data on public corruption are not highly correlated with survey data in state-year observations. Consequently, prior empirical findings based on survey data may not be replicated when reexamined with administrative records on corruption prosecutions. The availability of more reliable and detailed data on corruption prosecutions is a boon to scholars interested in the causes and consequences of public corruption in the United States.