Ethnic fragmentation and police spending: social identity and a public good

We present evidence that more ethnically fragmented communities spend, all else equal, more on police services than less fragmented communities. We introduce a model of spending on police services which we use to interpret the data. In this model, we assume that the decision to commit a crime is a rational consideration of the costs and benefits and that spending on police services reduces the attractiveness of committing a crime. We also assume that being a victim of crime affects a loss in utility. However this victimization cost, if victim and perpetrator are a different ethnicity, is greater than or equal to that if the perpetrator is the same ethnicity. A consequence of the model is that a higher level of spending on police services is associated with more ethnically fragmented communities only when agents suffer this differential cost of victimization. These results contribute to our understanding of the stylized fact that spending on police services is increasing at a time in which crime rates are falling. Further, our results provide empirical support for the contention that people have a larger cost of victimization when the perpetrator is a different ethnicity.