Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime

Abstract

Previous empirical studies have typically uncovered little evidence that police reduce crime. One problem with those studies is a failure to adequately deal with the simultaneity between police and crime: while police may or may not reduce crime, there is little doubt that expenditures on police forces are an increasing function of the crime rate. This paper attempts to use the timing of mayoral and gubernatorial elections to identify the effect of police on crime. It is first demonstrated that increases in the size of police forces are disproportionately concentrated in mayoral and gubernatorial election years. Having controlled for the state of the economy and other types of social spending, however, there is little reason to think that elections will otherwise be correlated with crime, making elections plausible instruments. Using a panel of large U.S. cities from 1970-1992, it appears that additional police reduce crime: the elasticity of violent crime with respect to sworn officers is found to be approximately -1.0; for property crimes the elasticity is -0.2. Because of the weakness of the instruments, however, the impact of police is imprecisely estimated. Consequently, the null hypothesis that the marginal social benefit of reduced crime equals the costs of hiring additional police cannot be rejected.

JEL Classification: H79, K14