Abstract
Econometric studies of crime have typically assumed that crime rates and sanction levels are determined simultaneously. and to achieve identification they have assumed that particular exogenous variables affect sanction levels but not crime rates. In this article it is suggested that individuals'decisions on whether to participate in criminal activity depend on perceived sanction levels that depend in turn on sanction levels realized in past periods. With decisions made in this manner, crime rates and sanction levels are not determined simultaneously, and the identification problem in its usual form does not arise. Our econometric results are consistent with this view; arrest and imprisonment sanctions have little contemporaneous effect on burglary, larceny, and robbery rates. Lagged effects, although greater than contemporaneous effects, are statistically significant only in the case of robbery.
Get full access to this article
View all access options for this article.
