How crime affects the economy: evidence from Italy

Italy represents an interesting case study for an empirical analysis of the impact of crime on economic performance. The country, in fact, presents considerable disparities among its regions in terms of economic outcomes and crime rates. The economic performance is measured using data on regional GDP per capita, while the crime variable is given by the homicides rate (attempted and committed) per 100,000 inhabitants. Data shows that crime is higher in regions characterized by low GDP per capita. This may be due to reverse causality. GDP per capita may be lower because of the large presence of crime and the high presence of crime can be due to the economic conditions of these regions. Baseline analysis with fixed effect model shows a small negative effect of crime on GDP per capita. Results change drastically when the potential endogeneity issue is addressed. Instrumenting crime with the effective abortion rate (similar to the one developed by Levitt, 2001) and using the Two Stage Least Square method, it emerges that if the homicides rate increases by 1% than the GDP will be lower of 0.32%. System GMM estimator is used to capture the effect of crime on economic growth. Results show that the growth rate will be reduced by 0.13% if homicides rate increases of 1%. Concluding, results indicate that crime substantially affects the level of GDP per capita and economic growth across Italian regions, especially in Southern Italy, thus finding one of the possible factors explaining the Italian dualism.