Rational Choice Theory

3. Empirical Support for Deterrence and Rational Choice Theory

Deterrence and rational choice are simply theories about how we think crime is brought about, and they may or may not provide accurate understandings of crime. One of the important ways that professionals in the field of criminology and criminal justice determine the worth or value of a theory is by evaluating the extent to which it fits the facts. A good theory is one that enjoys empirical support when a researcher empirically tests in the real world the propositions or hypotheses that can be derived from the theory. For example, if deterrence theory is true, one could make the following hypothesis about homicide rates and subject this hypothesis to empirical testing:

Deterrence hypothesis: Because capital punishment is a more severe punishment than life imprisonment, then states that have the death penalty as a possible punishment for murder should have lower murder rates than states that punish murder with life imprisonment.

This hypothesis is consistent with deterrence theory because it argues that would-be offenders are affected by punishments or costs. Because the death penalty is more costly to the offender than life imprisonment, would-be murderers in states with the death penalty face a more severe punishment than would-be murderers in states that administer only life imprisonment. One way to find out if deterrence theory is correct, then, would be to compare homicide rates in states that have the death penalty with states that have life imprisonment. A large number of empirical studies have been undertaken over the past 60 years with this goal in mind (Paternoster, Brame, & Bacon, 2008). Unfortunately, the evidence with respect to this hypothesis is not very convincing. Although the research has generated a great deal of controversy and the findings have been subject to considerable dispute, it seems that the safest conclusion is that there is no unequivocal evidence to date that the death penalty is a more effective deterrent to murder than life imprisonment. Does this mean that deterrence theory is false? Of course not; there are other ways to test the predictions of the theory.

Other empirical tests designed to examine deterrence theory have shown more convincingly that the fear of formal and informal punishment may indeed act as an effective general and specific deterrent to crime. With respect to formal sanctions, there is evidence that, although its effect is not large, the use of imprisonment serves as an effective deterrent and the national increase in the use of imprisonment in the 1990s may have been responsible for some part of the decline in crime (Levitt, 2001). Police studies have shown that when high-crime areas, known as “hot spots,” are saturated with extra police patrols (in light of the fact that increased police presence in a high-crime area increases the certainty of punishment), crime generally goes down (Sherman, 1990). Other kinds of police crackdowns, such as highly publicized roadside sobriety tests, have shown that the incidence of crime—in this case, drinking and driving— goes down, at least in the short term (Ross, 1984).

Over the years, numerous studies have tested whether not only sanctions that are actually imposed deter crime (e.g., arrest, conviction, imprisonment) but also whether both formal and informal sanctions that people think or perceive will be imposed act as a deterrent to crime (Paternoster, 1987; Pratt, Cullen, Blevins, Daigle, & Madensen, 2006). For example, people can be asked the following question to capture how certain they think they will be caught for a given crime: “On a scale of 1 to 10, where 1 is ‘I’d get caught every time’ and 10 is ‘I’d get away with it every time,’ how often do you think you’d get caught by the police if you used marijuana?” This measures the perceived certainty of formal punishment; people with low scores have a high certainty of punishment, and those with higher scores have a lower certainty of punishment. If deterrence theory is true, one would expect the former group to be less likely to use marijuana than the latter group, because they have been deterred. If one added the phrase “caught by your parents/ dorm advisor/employer,” one would have a measure of the perceived certainty of informal punishment and could predict that people who think they will not get caught by another would be at a higher risk of using marijuana than those who think they will more likely get caught. To measure the perceived severity of informal punishment, they could also be asked questions that pertain to “how much of a problem it would be” if someone found out about them committing a crime. Research studies that have used this method and various other similar approaches have generally found that the perceived certainty (but not generally the severity) of formal sanctions has a weak deterrent effect, whereas the perceived certainty and severity of informal punishment have a stronger deterrent effect (Pratt et al., 2006). In sum, although there may be some dispute as to how large an effect formal sanctions may have on crime, there is little dispute in the field that formal punishments either by the police, the courts, corrections, or some other criminal justice agency does seem to lower crime. There is also less disputed evidence indicating that the kinds of informal sanctions imposed by others in our lives act as an effective deterrent to crime. These findings from deterrence theory regarding formal and informal sanction threats provide indirect support to RCT as well, because RCT would make the same prediction that a high cost of crime would tend to reduce crime. Furthermore, this implies that people do respond to the expected consequences of their actions and that they are rational enough to be deterred from committing some crime when they think the penalties are certain and perhaps high.

In recent years, numerous attempts to measure the predictions of the rational choice method have been made. One stream of research is policy oriented and examines what is called situational crime prevention. In some of this research, active and “retired” criminals are extensively interviewed about what factors entered their decision to commit crime in general and/or a specific offense (Cornish & Clarke, 1986; Kubrin et al., 2009). Supportive of RCT, offenders frequently cite factors related to the costs and benefits of offending, such as the ease with which they can enter and leave a possible crime site, what the expected payoff for the crime is, the perceived ability to avoid detection, and their capacity to get what they want legally rather than illegally (Clarke & Cornish, 1985). More quantitative research also generally supports the rational choice model of offending. These studies have asked would-be offenders about the expected costs and benefits of criminal activity, and the results indicate that the decision to commit a crime is based at least in part on the expected costs and rewards of offending (Grasmick & Bursik, 1990).

It should also be noted here that whereas it may be intuitive to think that a rational choice view of offending might apply to property or instrumental crime, it might not also apply to crimes of violence, such as armed robbery, murder, or sexual assault; neither would it be able to explain what might be thought of as compulsive kinds of behavior, such as drug addiction. In point of fact, RCT has been used to adequately explain a variety not only of what might be thought of as instrumental crimes, such as robbery, burglary, and shoplifting but also expressive crimes, such as sexual assault and crimes of compulsion (e.g., drug addiction; Cornish & Clarke, 1986). A rational choice model has also been applied to the study of terrorism, and terrorists have been shown to rationally respond to increases in the costs of their crimes (Dugan, LaFree, & Piquero, 2005). RCT is, then, a very broad and general theory of crime that is able to explain property, violent, drug, sexual, and politically motivated offenses (Weisburd, Waring, & Chayet, 1995).