Classical Criminology

IV. Recent Studies

This possibility of reinvigoration of classically inspired thinking via rational choice in the form of a sophisticated, integrated, and formal theory of criminal behavior is signified by a study of high-risk youth in Denver (Matsueda, Kreager, & Huizinga, 2006). Matsueda et al. (2006) drew on longitudinal data to assess how perceived certainty of arrest is affected by individual characteristics (control variables, including age, race, residential stability, poverty) as well as theoretically inspired rational choice variables of experienced certainty (the ratio of experienced arrest or police questioning to crimes committed), unsanctioned offenses (the number of crimes committed by persons not questioned or arrested), and delinquent peers. Lagged perceived risk of arrest and self-reported risk preference (enjoyment of daring things) also were included. Perceived risk of arrest for theft and violence was regressed onto these variables.

Matsueda et al.’s (2006) results revealed that individuals who have not offended or been arrested overestimate the certainty of arrest. Experience of arrest certainty also leads to increased certainty estimates, revealing that offenders learn a desired lesson from being arrested for the crimes they commit; those who are arrested most regularly resemble the naive in their high assessments of risk. Those who offend but go unsanctioned perceive that the certainty of arrest is low and tend to move farther and farther away from naïve offenders with more criminal successes. Delinquent peers lead to decreased estimates of the certainty of arrest. On the whole, these findings show that assessments of the rewards of crime are acquired by Bayesian learning; individuals begin with a prior subjective probability estimate of an event based on accumulated information and update probability estimates as new information is gained directly or vicariously. Perceptions are formed by experiences and approximate rational outcomes.

In the next part of their article, Matsueda et al. (2006) showed the effects of certainty estimates on offending. They predicted violence and theft commission in one wave of data with previous wave reporting of rational choice/ perceptual variables including the positive experiences found in excitement from crime, seeing crime as cool, and committing crime and getting away with it. The downside of the criminal calculation is conceptualized as perceived risk of arrest (measured as perceived certainty by perceived utility) and lost investment in employment and grades. The authors assessed these effects, controlling for the effects of five measures of (1) neighborhood disadvantage, (2) basic individual control variables, (3) police contact, (4) risk preference, and (5) impulsivity. They also controlled prior delinquency to ensure that the temporal specification is clear and that current crime is a function of theoretically important variables instead of stable criminal behavior across waves.

Considerable support was found for a rational choice perspective on offending. This support is seen in significant effects for perceived risk of arrest as a cost and being seen as cool as significant benefits for both theft and violence; excitement from crime predicted theft but not violence. Perceived opportunities to commit theft and violence and get away with it also were significant predictors of both theft and violence in the expected direction.

This litany of findings is presented here only because it instructs on contemporary rational choice criminology and the attractions of the perspective today. The abstract lessons are that rational choice theory can guide the formation of intuitively valid, creative, and testable hypotheses that add to understanding of crime. The prospects for the perspective include understanding how experience and conditions shape thinking that makes some people more likely to offend than others, and this is the most promising line of inquiry for understanding why people commit, persist, or desist from committing crimes. More important for its future is that rational choice theory is completely compatible with dynamic and developmental perspectives that measure movement into, out of, and within criminal careers. Therefore, rational choice theory is commensurate with the core of contemporary criminological research: life course and developmental research. Any reader intrigued by studying crime over the life course or the prospects for interfering in the sequences of thinking that contribute to higher chances of crime commission can find guidance in rational choice theory as well as in the more commonly used control theories.

As with many theories, however, the features that distinguish rational choice and classical perspectives from other camps can be difficult to see when they are used in integrated pictures of crime. It is safe to say that the degree to which adherents view crime as an outcome of a choice that is subject to general economic principles and that results from maximizing behavior on the part of decision makers distinguishes the camps from other areas of criminology. From this perspective, offending rates result from optimizing individuals’ reactions to their estimates of incentives (referred to as perceptual deterrence, i.e., the role of an individual’s perceptions of costs because these deter crime) and are aggregate results of individuals’ assessments of costs and benefits. The offender traditionally is portrayed as a normal person (with no peculiar motives or individual defects that are thought to predispose crime) who responds predictably such that if probable costs exceed the probable benefits, he or she will not commit crime. Reflecting a tradition of methodological individualism, individuals’ combined decisions to commit or abstain from crime on the basis of judgments of its net payoff are thought to explain variation in crime rates across space and time.

Gary Becker’s work best represents the purist economic camp and thus may best represent a contemporary application of traditional classical thinking; it is grounded on the assumption that offenders are by and large normal and for the most part are responding to measurable incentive structures. Becker has been awarded a Nobel Prize and a Medal of Science, as well as a Presidential Medal of Freedom, and he is Distinguished Professor of Economics, Business and Sociology at the University of Chicago. Becker proposed what is now known as the subjective expected utility model of crime, which has proven to be surprisingly controversial. According to this theory, crime will be more likely if the individual’s perceived expected utility (expressed in monetary terms) for criminal behavior is greater than the expected utility of some legal alternative. This model is often represented mathematically with the following formula: EU = pU (Y – f) + [1 – P_ U (Y)], where EU represents the expected utility from the behavior, p is the probability of punishment, and U represents the utility (the severity) of costs or benefits. Becker contends also that people who might consider crime calculate the following: (a) practical opportunities of earning legitimate income, (b) amounts of income offered by these opportunities, (c) amounts of income offered by various illegal methods, (d) probabilities of being arrested for illegal acts, and (e) probabilities of punishment if caught. The act or occupation with the highest discounted return is likely to be chosen (Sullivan, 1973). As an economic purist, Becker asserts provocatively that there is little reason for theorizing that treats offenders as if they have a special character that leads them to crime. Becker extends his logic to conclude that the most prominent theories of motivation in criminology are not needed; basic economics addresses their problem sufficiently. Criminal offenders are normal, reasoning economic actors responding to market forces. This dogma understandably leaves him out of favor with many sociologists of crime who specialize in other criminological theories and, as one might imagine, he is not a favorite of criminal psychologists either. If classical criminology theory as used by Becker is accurate, then one should expect that rates of crime will be stable if the costs and benefits of it remain so; however, in modified versions that are endorsed most often among scholars who claim rational choice as their perspective, allowance is made for the fact that the population of likely criminals and prevalence of criminal thinking might also vary.