Criminal Choice, Nonmonetary Sanctions, and Marginal Deterrence: A Normative Analysis

This paper develops a normative model of optimal sanctions in the Becker Tradition which emphasizes the role of marginal deterrence. The paper complements Shavell's 1987 American Economic Review paper, the essential difference being that Shavell's model concentrates on variations in the sanction imposed within a single category of acts (a specific crime) while the model in this paper concentrates on variations in the sanction imposed across categories of acts (different crimes). In their most general formulations, neither Shavell's model nor the model developed in this paper yields the result that acts with greater social harm should receive greater sanctions. But special cases, which readers may or may not find reasonable, do yield that result, within crimes for both models and across crimes in the model developed in this paper. This paper also identifies the necessary condition of jointness in the cost of law enforcement in the case of comparisons across crimes.