The traditional answer is affirmative, based on the idea that wrongful conviction lowers a person's pay-off for being innocent without affecting the pay-off for being guilty. However, this view fails to distinguish between mistake about identity and mistake about the act. For mistake about identity the traditional view is incorrect, since a person who commits a criminal act will not thereby eliminate the risk of being convicted of someone else's crime. Conviction of the wrong person may still have an effect on deterrence, but the effect can be positive as well as negative, and will tend to be small, mainly because the risk of wrongful conviction is shared between many people. When the court wrongly assesses the act committed by the defendant, the traditional view is correct when the choice set of the offender is binary, while the effect on deterrence is more likely to be positive when the choice set is continuous.

The traditional view is that wrongful conviction lowers deterrence by lowering a person’s payoff for being innocent without affecting the payoff for being guilty. However, this view fails to distinguish between mistake about identity and mistake about the act. For mistake about identity, the view is incorrect, since a person who commits a criminal act will not thereby eliminate the risk of being convicted of someone else’s crime. Conviction of the wrong person may still have an effect on deterrence, but the effect can be positive as well as negative and will tend to be small, mainly because the risk of wrongful conviction is shared among many people. When the court wrongly assesses the act committed by the defendant, the traditional view is correct when the choice set of the offender is binary, while the effect on deterrence is more likely to be positive when the choice set is continuous.

Abstract:It is shown in this paper that there may exist an intrinsic advantage in focusing law enforcement effort on a subgroup of possible violators of law, rather than applying law enforcement effort uniformly over the relevant population of potential violators. For example, it might be desirable for the tax audit rate to be higher for individuals whose names begin with the letters A-M than for those whose names begin with N-Z. This may be desirable even though, as is assumed, the frequency of violations does not differ between subgroups.

No abstract

Abstract:It is shown in this paper that there may exist an intrinsic advantage in focusing law enforcement effort on a subgroup of possible violators of law, rather than applying law enforcement effort uniformly over the relevant population of potential violators. For example, it might be desirable for the tax audit rate to be higher for individuals whose names begin with the letters A-M than for those whose names begin with N-Z. This may be desirable even though, as is assumed, the frequency of violations does not differ between subgroups.