Making the Punishment Fit the Crime or Taliban Justice? Optimal Penalties without Commitment

This paper argues that graduated penalties observed in most legal systems may be an attempt to direct law enforcement efforts towards crimes that are socially more harmful, thereby achieving better deterrence overall. The critical assumptions are: the state cannot commit to monitoring strategy, and has mixed motives (other than deterrence). However graduated penalties arise only in the presence of secondary motives that value punishment in itself. Other motives that are unrelated to the size of the punishment will also lead to distortions, but those cannot be corrected by restructuring penalties. The overall harshness of a criminal justice system and the retributive instincts of its designers may be related in counter intuitive ways, and law enforcement may be improved through strategic delegation. [CDE-DSE WP no.175]