Are corrupt acts contagious? Evidence from the United States

摘要

This paper adds to the literature by shedding new light on the causes of corruption. Specifically, we provide evidence on the extent to which corruption might be contagious. In other words, what is the extent to which a demonstration effect is at play at inducing corrupt acts? Using state-level U.S. data over the 1995-2004 period, the results show that the effect of neighboring corruption is positive and statistically significant in all cases, implying that corruption does appear to be contagious. Specifically, a 10% increase in corruption in neighboring states appears to increase corruption in a state by about 4-11%. Of the different types of government activity, the size of defense and non-defense federal sectors in a state seem to have opposite effects on corruption, with the former contributing to corruption and the latter serving as a deterrent. The size of the state and local governments does not seem to be relevant. Of the variables controlling for detection and punishment of corrupt individuals, greater corrections employment reduces corruption, while greater judicial employment seems to increase corruption. Changes in the police force do not seem to have a statistically significant impact. These findings generally hold when we control for the disproportionate effects of the Washington, DC area and when a pooled data set is estimated. Contagion effects are also found for other crimes. Policy implications are discussed. Published by Elsevier Inc. on behalf of Society for Policy Modeling