The Certainty versus the Severity of Punishment, Repeat Offenders, and Stigmatization


There is a widely held presumption that the certainty of punishment (p) is a greater deterrent than the severity of punishment (s). This presumption is at odds with evidence from recent experimental work suggesting the contrary, and the implication of simple law enforcement models that risk-averse individuals must be deterred more by an increase in s than a comparable increase in p. This article demonstrates that this discrepancy may be the result of subtle differences in the effects being investigated. In particular, when repeat offenders are punished more severely than first time offenders, a change in p can have a greater effect than an increase in s on the aggregate offense level, even when each individual offender is more responsive to s than p. This is because an increase in p corresponds to moving some offenders from the first time offender category to the repeat offender category, which reduces the crime rate by causing a discrete increase in the sanctions that these individuals face. This effect is reversed when a first conviction results in stigma that more than off-sets the difference between the formal sanction for repeat offenders and first time offenders, because, then, the total sanction for repeat offenders is lower than the total sanction for first time offenders. In these cases, stigmatization can cause criminogenic effects. However, these negative effects are generally off-set by a second effect that emerges when stigmatization is present: an increase in p results in greater expected formal as well as informal sanctions, whereas an increase in s only affects the expected formal sanction. Finally, all results are derived by assuming that individuals are risk-neutral, implying that individuals need not be risk-seeking for deterrence to be more responsive to p than s, which is a point claimed in Becker (1968).