Socioeconomic Determinants for Regional Variation of Crime in Germany

This work explores which spatial patterns of crime exist in Germany and what their socio-economic determinants are. This is done using the PKS crime statistic for six types of crime on the county level for the years 2003-2014. To the authors knowledge, this is the first time crime in Germany has been studied with such granular data on such a broad scale. Due to the granularity of the data set, spatial dependencies of crime and its covariates became very apparent, which made it necessary to use a random effects spatial durbin model (SDM) to model regional variation of crime. With this, I was able to show that the factors explaining crime in one county also spill-over and affect crime in a neighboring one. To my knowledge, this hasn't been done before either for Germany. Also, I was able to confirm earlier findings that the Becker-Ehrlich model works well explaining property crime in Germany and the social disorganization theory explaining violent and drug crime, with the routine activity mostly working in both cases. Also as a first for Germany, there is some evidence that the broken-window theory applies and that disorder can drive crime to an extend.