Acquisitive Crime, Sentencing and Detection : An Analysis of England and Wales

We provide the first detailed econometric analysis of the impact of sentencing on various types of acquisitive crime in England and Wales. We examine (a) whether sentencing reduces crime and (b) whether short sentences are more effective than long sentences. Detection is an important explanatory variable whose effect needs to be controlled in order to identify the impact of sentencing. To address the potential endogeneity of detection, we instrument using lagged values of police expenditure and detection. Our results show that detection is significant and negatively affects all crime types while the impact of sentences is negative and significant for all crime except robbery in a linear specification. A quadratic specification shows that the linear term is positive while the square term is negative suggesting short sentences may be counterproductive in reducing robbery. We also control for a number of socio economic variables whose effects significantly affect crime