Abstract
Considerable debate has surfaced over the past decade as to whether police contribute to crime control and, if so, what relationships exist between police resources and apprehension rates (arrest production) and between apprehension rates and crime rates (deterrence production). This paper compares the implications for both of these processes by measuring criminal activity and apprehension risk with alternative data sources: FBI Uniform Crime Reports (UCRs) and Bureau of Justice Statistics National Crime Surveys (NCSs). An analysis of the error structures in each statistical series finds that both series produce biased estimates of deterrence parameters in simple regression models, the greater biases generated by the Uniform Crime Reports. In a more comprehensive deterrence model, UCR-based measures of apprehension risk and criminal activity uncover no relationship between apprehension risk and crime rates while comparable NCS-based measures find strong deterrent effects. Both series find strong associations between police resources and apprehension risk for burglary and total property offenses, but not for larceny. Determinants of law-enforcement demand were also similar except for the effect of crime-rates on demand. The fact that UCR crime-rates were significant determinants of demand and NCS rates were not was interpreted as evidence that official statistics, not subjective estimates of community victimization risk, determined police resources.
Get full access to this article
View all access options for this article.