Court delays and crime deterrence. An application to crimes against property in Italy

Using Italian data in the period 1999-2002, we estimate the impact of trial delays on the willingness to commit crimes against property. However, the endogenous relationship that links the former to the latter could generate serious problems of inconsistency in the estimation procedure. Since geographical distance can be considered an exogenous determinant of the probability of belonging to peripheral courts, which are typically considered less efficient than main ones, it should represent a valid candidate instrument for trial delay. Estimates obtained by means of Two-Stage Least Squares show a significant positive effect of trials duration on crimes, supporting the hypothesis that some criminals are either sensitive to the discounting process of punishment or aware of the probability of prescription, or both. As a side result, we also find a relationship between courts' fragmentation and trial duration. This suggests that an optimal dimension of courts is likely to exist, and that policy makers should take this into consideration in the design of the jurisdictional geography. © 2008 Springer Science+Business Media, LLC.