Cost minimizing sequential punishment policies for repeat offenders

Abstract

This article discusses optimal sanctions for repeat offenders. We analysed a multi-period decision problem, where the regulator’s main objectives are to block any violations of law and to minimize the costs of crime control. We conclude that, when offenders are identical and wealth constrained, the government is resource constrained, can perfectly observe illicit gains and commits to a certain policy throughout the whole planning horizon, forward-looking solution implies that cost minimizing deterrence is decreasing in the number of offenses. This analysis is relevant in case when imprisonment is not commonly used, only monetary sanctions are allowed and limited liability of offenders plays an important role. The examples are tax evasion, violations of environmental regulations and violations of competition law.

Keywords:

• crime and punishment
• crime prevention
• repeat offenders

JEL Classification:

• K14
• K42
• C61

Acknowledgements

The author thanks Eric van Damme, Peter M. Kort, Dolf Talman, Winand Emons, Quan Wen, Jacco Wielhouwer and anonymous referees for stimulating discussions and valuable comments.

Notes

¹ Although, Emons (2007) finds partial support for both hypotheses.

² Without restriction on probability of detection being sufficiently small.

³ This assumption may seem to be restrictive. In most of the cases, for example, in the case of tax evasion or illegal price-fixing activities, the penalty takes into account not only initial wealth of the firm but also accumulated rents from illegal activities. However, this assumption is adopted here in order to focus on obtaining analytical results with respect to establishing an optimal sequence of sanctions.

⁴ This may also be applicable for other ‘economic crimes’ and especially in the corporate environment. For extensive discussion of those examples see Burnovski and Safra (1994).

⁵ The fact that only constraint (n + 1) on the benefits from crime is binding when can be proven by contradiction. Detailed proof is available from the author upon request. The intuition is as follows. Assume, for example, that constraint is binding for some then it follows that the LHS of the constraint has to be strictly positive, which is impossible by construction of the problem.