Avaliação do impacto do INFOCRIM sobre as taxas de homicídios dos municípios paulistas: uma aplicação do método de diferenças em diferenças espacial

Abstract

The reversal of the trend of crime rates in the state of São Paulo has arisen the interest of researchers to verify the possible causes for the reduction of 67.1% in homicide rate from 2000 to 2010. Several factors have been suggested to explain this phenomenon, but it was not found in the literature, so far, researches to evaluate the Criminal Information System – INFOCRIM, adopted by 67 municipalities in São Paulo until 2010. Thus this work is aimed at testing the hypothesis that INFOCRIM has contributed to the reduction in homicide rate between 2000 and 2010, controlling for determinants of crime and other public security actions. To do so, a treatment evaluation is performed using the spatial difference-in- differences (SDID) approach, a pioneering method able to model the crime migration from the treated regions to their neighboring regions. The treatment group is defined as being the 67 municipalities with INFOCRIM whereas the control group is defined as being 548 municipalities remaining in the sample. Using two-period panel data (2000 and 2010), the spatial lag model (SDID-SAR) indicated the existence of spatial spillovers (crime migration) among municipalities in the São Paulo state and a crime reducing effect of INFOCRIM. As to total effects, INFOCRIM reduced homicide rate by 6.183 over the period 2000-2010. In terms of direct terms, the program decreased homicide rate by 3.745, while, in terms of indirect effects, INFOCRIM reduced homicide rate by 2.437 during this period. . This implies that the spatial interaction between agents enhanced the average effect of INFOCRIM. These findings confirmed the positive relation between crime and unemployment and the inverse relation between crime and education. The improvement of social equity contributed to the reduction of violent crime while the population density showed a sign contrary to theoretically expected. There is an indirect effect on homicide rate due to the existence of municipal guard. In terms of lives saved, the Government of the State of São Paulo was able to minimize the social loss from offenses, since the implementation of INFOCRIM has avoided 2,546 homicides over the period 2000-2010. Therefore, the INFOCRIM can be considered successful in fighting lethal crime.