Criminal choice, nonmonetary sanctions and marginal deterrence: A normative analysis

Normative models of the optimal use of sanctions, monetary as well as nonmonetary, that employ the assumption of precommitment on the part of social authorities provide an important and useful benchmark for the evaluation of existing policies. Ever since Gary Becker published his classic article on the economics of crime (Becker, 1968), however, a conundrum has plagued the literature: if law enforcement is costly but crimes are socially undesirable and potentially deterrable, then efficiency requires that for all crimes the probability of apprehension be set arbitrarily low and the sanction arbitrarily high (see, e.g., Carr-Hill and Stern, 1979). This solution imposes no costs on society as long as the expected sanction is high enough to deter all crime; since no crime is ever committed, the sanction never need be imposed. Hence, even if sanctions are costly to impose, as in the case of nonmonetary sanctions, this is the efficient solution.