Dutch-Belgian Political Science Conference edition:7 location:Nijmegen date:29-30 May 2008
In this article, we investigate the occurrence of criminal acts at the community level in Belgium for the 2001-2005 period. Multivariate analysis shows that income levels do not have a significant effect on the occurrence of property crime, violent crime or homicide. Rather unemployment figures prove to be the strongest determinant of crime rates in general. Contrary to what has been shown by US research, we did not detect any patterns for homicide, an act that seems to be randomly spread across the territory. Further analysis suggest that male unemployment has a stronger effect on violent crime than female unemployment rates, with a standardized regression coefficient (at the community level) of .55. We speculate that the relation between unemployment and crime cannot just be explained by the effects of a loss of income but should also take into account other factors associated with the experience of unemployment. We close with some theoretical and policy considerations on the relation between unemployment and crime.