Prática esportiva e prevenção criminal: uma análise para os municípios brasileiros

Abstract

The aim of this dissertation is to measure the sport-crime relationship for Brazilian municipalities. The hypothesis is that crime is negatively correlated with sports. It is argued that sport can reduce crime through three channels: i) occupation, due to the potential of sporting activity to occupy lazy time, especially for younger people; ii) increase in the moral cost, mainly due to the positive effects related to the social capital of sports activity; and iii) increase in expected legal income, linked to possible gains in productivity and human capital stimulated by physical activity and opportunities in the legal sports market. The panel estimation method with Fixed Effects is applied using the municipal data for the years 2002 and 2010, estimating the model with the dependent variable homicide rate and youth homicide rate. Five variables are used as a proxy for sport: i) municipal spending per capita on sport and leisure; ii) time of existence of the municipal sports council; iii) accumulated sports capital; iv) number of municipal sports institutions; and v) presence/absence of any municipal sports institution. The results show that there is no positive relationship between crime and sport; the relationship is negative in two estimates and not significant in three. Furthermore, the sport-crime relationship is negative with the youth homicide rate.