Research in Criminal Deterrence: Laying the Groundwork for the Second Decade

Deterrence theory has been developed primarily by economists, who have viewed potential criminals as rational decision-makers faced with an array of illicit opportunities characterized by costs (time, possible adverse legal consequences, and so forth) and payoffs. The crime decision is thus characterized in a way that fits the well-developed theoretical framework of decision-making under uncertainty. Herbert Simon and others have questioned the descriptive accuracy of this theory, and are beginning to uncover systematic patterns in decision-making that violate the predictions of the economic theory: this work could usefully be incorporated into the crime choice framework. One of the most important issues for further research in this area is the way in which potential criminals acquire information about criminal opportunities and the effectiveness of the criminal justice system. A simple "realistic" model of threat communication can be outlined that yields deterrence-like effects, even though no one is well informed concerning the true effectiveness of the system. Three other questions that have been of great interest to deterrence theorists are discussed: (1) what factors influence the rate at which active criminals commit crimes; (2) which dimension of the threat of punishment has a greater deterrent effect-likelihood or severity; and (3) what effect does the threat of punishment for one type of crime have on involvement in other criminal activities? Much of the recent empirical work on deterrence has used a fundamentally flawed approach to estimating the responsiveness of crime rates to sanction probability and severity. The flaw is that the measures of "probability of punishment" used in these studies reflect the choices made by criminals as well as the intrinsic effectiveness of the criminal justice system. Therefore, these measures do not serve as appropriate indices of criminal justice system effectiveness. The empirical approach that appears most productive is the evaluation of discrete changes in law and policy-"natural experiments," that can tell us a good deal about the deterrence process.