Estimating the Number of Crimes Averted by Incapacitation: An Information Theoretic Approach

This paper presents an information theoretic approach for estimating the number of crimes averted by incapacitation. It first develops models of the criminal history accumulation process of a sample of prison releasees using their official recorded arrest histories prior to incarceration. The models yield individual offending trajectories that are then used to compute the number of crimes these releasees could reasonably have been expected to commit had they not been incarcerated—the counterfactual of interest. The models also afford the opportunity to conduct a limited set of policy simulations. The data reveal a fair amount of variation among individuals both in terms of the number of crimes averted by their incarceration and the responsiveness of these estimates to longer incarceration terms. Estimates were found not to vary substantially across demographic groups defined by offender race, gender, or ethnicity; variations across states and offense types were more pronounced. Implications of the findings and promising avenues for future research are discussed.

Avinash Singh BhatiEmail: