Deterring repeat offenders with escalating penalty schedules: a Bayesian approach

What is it about?

Penalty schemes typically involve higher penalties for repeat offenders. This paper offers an efficiency rationale for these schemes when punishment is costly (e.g. prison). It does this by assuming that criminal ability is unobservable by both the legal system and an individual criminal. Courts do not know if those with no prior convictions are truly first offenders or if they committed undetected prior offences. If they are truly first offenders, they are more likely to conclude that they have no aptitude for crime and will not reoffend, making harsh (and costly) punishments unnecessary. This implies that those with multiple convictions are likely high-ability criminals who have committed more crimes than they have convictions. If so, they need to face higher penalties to be deterred.

Featured Image

Why is it important?

Several theoretical models have found that penalty schemes should involve lower penalties for repeat offenders. Yet, it is difficult to think of any such penalty scheme in the ‘real world’. Given the large sums spent on incarceration, it is clearly important to understand whether actual penalty schemes possess efficient characteristics.

Perspectives