In this paper we investigate the effects of labour market policy on several types of criminal offences for fifteen European countries. The main results are the following: Firstly, the results change markedly if we control for unobserved heterogeneity. In the context of criminal offences the estimates seem to be reliable only if we apply fixed effects instead of simple pool specifications. Secondly, the effects of labour market policy vary considerably with respect to the different types of criminal offences and cannot be subdivided into unambiguous effects on property crimes and violent crimes, respectively. Thirdly, the proxy variables for labour market policy we consider have different importance with respect to their effect on criminal offences. Benefit replacement rate, benefit duration, and average years of schooling seem to be important, whereas active labour market policy appears not to be linked to crime. The combination of a shorter benefit duration and higher replacement rate, like in the Nordic countries, seems to be a 'crime reducing' combination.
Unemployment labour market policy illegal behaviour time allocation