Spatial analysis of criminality in Rio Grande do Sul State, Brazil

Abstract

This paper does a spatial analysis of crime in Rio Grande do Sul. Itpresents a spatial econometric model for criminality determinants. The modelfollows main contributions to the economics of crime, but it includes contributionsfrom the ecological approach and social learning. In the model, the criminality in cities can be explained by local characteristics, and the environment, theneighborhood and the individual’s history affect the criminality. The paper usesaggregate city data for homicides, robberies and thefts in a cross-sectional modelfor the year of 2000. The findings show the existence of spatial dependence inrobberies and thefts and spatial independence in homicides. The paper emphasizesthe role of income inequality and of urban agglomerations as factors thatincrements the criminality in cities. It also discusses the importance of familyand schooling on explaining criminality. It shows that problems in the familystructure and in schooling affect the crimes positively. The benefits of crime andopportunity costs are separated, which allows concluding that economic growthdoes not increase criminality. This happens if there is an increase in the income ofthe poorest, leading to a fall in criminality.