Effects of COPS Office Funding on Sworn Force Levels, Crime, and Arrests: Evidence from a Regression Discontinuity Design [open pdf - 2MB]
"This paper presents new evidence about one particularly promising approach to controlling crime that does not rely (at least directly) on incarceration and, unlike many other approaches, would seem to be relatively easy to scale: hiring more police. The canonical economic model of crime from Becker (1968) predicts that police deter criminal behavior by increasing the expected costs of punishment to potential offenders. Police may also reduce crime by making arrests that result in the incarceration and hence incapacitation of active offenders, or by engaging in preventive problem-solving activities. While the United States has increased spending on police over the past several decades, the growth in police per capita pales in comparison to the growth in corrections expenditures [...]. The prospect of diminishing marginal returns raises the possibility of shifting resources away from imprisonment towards police to reduce crime at no extra cost. And one thing that almost every police department in the country knows how to do is hire more police. [...] In this paper we present new estimates for the effects on crime from adding more police, which we believe have an unusually strong claim to identification."
Community Oriented Policing Services (COPS): http://cops.usdoj.gov/