Thieves, Thugs, and Neighborhood Poverty

This paper develops a model of criminal behavior that analyzes how such behavior may be associated with both individual and neighborhood poverty. Importantly, this model distinguishes between basic property crimes such as burglary and larceny, and interpersonal violent crimes such as robbery and assault. The model shows that even under relatively mini-mal assumptions, a connection between individual poverty and both types of crime will arise, and moreover, neighborhoode¤ects can develop, but will di¤er substantially in nature across crime types. A key implication of the model is that greater economic segregation in a city should have no e¤ect or even a negative e¤ect on basic property crime, but a positive e¤ect on violent crime. Using Instrumental Variable methods, I show this implication to be consistent with the empirical evidence