Life is now! Time preferences and crime: Aggregate evidencefrom the Italian regions

tThis paper tests the relationship between time preferences and crime rates as posited by Davis (1988),whose theoretical analysis suggests that individuals’ attitude towards the future significantly affectstheir propensity to commit crime. Our empirical analysis is based on a panel of Italian regions from2003 to 2007. Various proxies for time preferences are considered: the consumer credit share out ofthe total amount of loans to households, the share of obese individuals out of the total population, therate of marriages out of the total population, and the teenage pregnancy rate. Controlling for a greatnumber of factors suggested by the scientific literature on the determinants of crime, adding to themodel also time and regional fixed effects, and clustering standard errors to account for both serial andpanel correlations, our results basically provide support to the ‘Davis’ hypothesis’ for property crimes,while for violent crimes there seems to be less evidence that these are higher where people discount thefuture more heavily. Moreover, there is no evidence of a reverse effect from crime to time preferences atthis aggregate level.