General Deterrence, Unobserved Heterogeneity and Conventional Simultaneity in a Panel of Italian Regions
Posted: 27 Apr 1998
Date Written: June 1996
In this paper we test for the general deterrence hypothesis using Italian regions as units of observation. This can be interesting for a number of reasons: (i) a large variability of crime rates across time and space; (ii) a massive presence of organized crime with strong regional roots; (iii) a criminal justice system based on a codified criminal law.Using panel data techniques, we control for unobservable region-specific characteristics that might be correlated with the included regressors. This type of endogeneity, usually neglected in the applied literature on crime, could not be skipped in our sample due to the evident role played by organized crime in the determination of regional crime rates. We also check the robustness of our findings by addressing conventional simultaneity, i.e. endogeneity arising from the possibility that crime rates and criminal justice variables mutually affect one another.Using a recent dataset, provided by the Antimafia Commission and the Ministry for Justice, concerning the regional distribution of police and magistrates, we are able to estimate a simultaneous model of crime along the lines of similar works in the literature, and to reach some preliminary conclusions on the effectiveness of general deterrence taking account of both sources of endogeneity.
JEL Classification: K42
Suggested Citation: Suggested Citation