Robert Apel. 21st Century Criminology: A Reference Handbook. Editor: J Mitchell Miller. Sage Publications. 2009.
Employment has long been observed to be a correlate of criminal behavior. For example, Belgian criminologist Adolphe Quetelet, in an 1831 publication analyzing French crime statistics titled Research on the Propensity for Crime at Different Ages (cited in Beirne, 1987), remarked that individuals who were unemployed or employed in “lowly occupations” were more likely to commit crimes (Beirne, 1987, pp. 1153-1154). Thus, the study of crime and the economy is a long-standing tradition in criminology.
To maintain a sufficiently narrow scope, this chapter focuses on individual-level theories of, and observational research on, the relationship between employment and crime. It thus omits a review of employment-crime studies at the macro level and experimental or quasi-experimental evaluations of employment interventions. The first section in this chapter comprises a theoretical overview of the relationship between employment and crime. The second section reviews the empirical literature on the employment—crime connection, the third section identifies empirical challenges that must be overcome in employment-crime research, and the final section offers some concluding remarks and outlines future directions.
Theoretical Relationship between Employment and Crime
A number of theories rooted in labor economics and sociological criminology consider legitimate, remunerative employment to be an important causal factor in the prevention of criminal behavior. Conversely, unemployment is believed to genuinely cause an increase in criminal activity. Several of the more prominent theories of the employment—crime relationship are described in this section.
Economic choice theory is rooted in the neoclassical idea of utility maximization, which presumes that people are responsive to incentives and choose behavior by maximizing their utility from a stable set of preferences, subject to opportunities and other constraints on their resources (Becker, 1968). Distilled to its basics, the economic choice theory of crime is concerned with how self-interested individuals allocate their time and resources between legal and illegal activities when the returns to the latter set of activities in particular are uncertain. Prominent in this tradition is the expected utility model, according to which a person decides to commit crime when the expected returns from illegal behavior, discounted by punishment risk, exceed the expected returns from law-abiding behavior such as employment. All else equal, individuals faced with current or future unemployment or low wages experience lower costs of committing crime. To be precise, they experience lower opportunity costs of engaging in illegal activity, and thus they find illegal income generation to be an attractive and rational alternative compared with legal income generation.
Social control theory proposes that strong attachment to the institution of work constitutes a potent source of informal social control over criminal behavior (Hirschi, 1969; Sampson & Laub, 1993). Such attachment encourages a strong “stake in conformity” that can overcome the temptation to violate the law, in part because attached individuals fear putting their future careers in jeopardy. The acquisition of a stable job of high quality can also be a turning point for individuals with a history of criminal behavior because it fosters social capital, or investments in conventional institutional relationships (Hagan & McCarthy, 1997; Sampson & Laub, 1993). According to control theory, then, the mediating role of social capital implies that work quality is more salient than the mere presence of work, because higher quality jobs promote stronger interpersonal connectedness and institutional embeddedness.
Social control theory is also friendly to the notion that “idle hands are the devil’s workshop,” in the sense that employed individuals simply have fewer opportunities to commit crime because they are too busy working (Hirschi, 1969). This is the involvement hypothesis of the theory: “Many persons undoubtedly owe a life of virtue to a lack of opportunity to do otherwise” (Hirschi, 1969, p. 21). If the allocation of time is a zero-sum game, then one more hour spent in the workplace is one less hour available for criminal activity outside the workplace. In a recent elaboration of this idea, Laub and Sampson (2003) proposed that attachment to work not only constrains opportunities to commit crime but also leads to fundamental changes in how individuals spend their leisure time outside of work. The imposed structure of the workplace may permeate nonwork settings and thus foster changes in routine activities that lure individuals away from crime by channeling them into conventional behavior with law-abiding companions.
Strain theory presumes that lack of success in the legitimate labor market motivates individuals to “innovate” in the most expedient or technically efficient manner, usually through criminal behavior (Merton, 1938). Underlying this theory is the presumption that the desire for wealth is universal (it is a culturally approved goal) and therefore blocked access to legitimate opportunities to acquire this valued goal results in anger, frustration, desperation, or other forms of negative affect (see Agnew, 1992). Criminal behavior is one way to alleviate the negative feelings associated with the strain of unemployment or low-quality employment. Unemployed individuals thus commit crime as an income substitute; individuals employed in low-wage or low-quality occupations commit crime as an income supplement.
According to various strands of learning theory, the workplace provides a context for differential associations with conventional employers and coworkers that tip the balance of definitions favorable to law violation (Sutherland, 1947), a general process proposed to operate through modeling and differential reinforcement of law-abiding behavior (Akers, 1985). Steady work in a good job puts individuals in close proximity with a conventional social circle for a nontrivial number of hours each week. As such, they have exposure to colleagues who espouse prosocial beliefs toward the law, who act on these beliefs, and who therefore provide positive role models and reinforcers for behavior both inside and outside the workplace.
To summarize thus far, all of the foregoing theories provide support for two basic propositions. First, individuals who are employed are less likely to commit crime, on average, compared with individuals who are not employed, who are unemployed, or who are underemployed. Second, individuals who are employed in stable, high-quality jobs (e.g., high-paying, primary-sector occupations) are less likely to commit crime than their counterparts in unstable, low-quality jobs. Each of the foregoing theories—economic choice, social control, strain, and learning—presumes that the inverse correlation between gainful employment and crime is causal; however, according to at least one other prominent theoretical tradition, the employment—crime correlation is entirely spurious.
Self-control theory posits that individuals sort themselves into certain institutional settings on the basis of a differential tendency to consider the long-term consequences of their actions, what Gottfredson and Hirschi (1990) referred to as self-control. Because individuals with low self-control seek immediate gratification of their desires with minimal effort or long-term planning, they are less likely to be employed or, if they are employed, will have difficulty holding down a steady job: “People who lack self-control tend to dislike settings that require discipline, supervision, or other constraints on their behavior” (Gottfredson & Hirschi, 1990, p. 157). It so happens that these same personal qualities increase the likelihood that desires will be satisfied through criminal activity. Simply put, unemployment, low-wage employment, and crime are all manifestations of the versatility of individuals with low self-control. In statistical terminology, low self-control is a source of unobserved heterogeneity that is responsible for an artifactual (i.e., spurious) inverse correlation between employment and crime.
The Empirical Relationship between Employment and Crime
More than two dozen empirical studies among a variety of adult and young adult populations consistently confirm that labor market success in the form of employment, high wages, job stability, and occupational prestige are correlated with reduced criminal involvement (Crutchfield & Pitchford, 1997; Farrington, Gallagher, Morley, St. Ledger, & West, 1986; Good, Pirog-Good, & Sickles, 1986; Grogger, 1998; Hagan & McCarthy, 1997; Horney, Osgood, & Marshall, 1995; Laub & Sampson 2003; Sampson & Laub, 1993; Thornberry & Christenson, 1984; Uggen, 1999, 2000). Instead of reviewing each study in detail, a handful are selected that are representative of the wider literature and offer valuable insight into the employment-crime connection.
Thornberry and Christenson (1984) studied the relationship between unemployment and crime among a cohort of males born in 1945 in Philadelphia. In an analysis of yearly crime and unemployment during the 21-24 age range, unemployment duration was positively correlated with officially recorded arrest frequency (weighted by seriousness). They also found that the correlation grew stronger with age and that the correlation was more pronounced among the less advantaged individuals in the sample, including delinquent persons, African Americans, and individuals from blue-collar families.
Farrington et al. (1986) assessed the impact of unemployment on crime among a sample of 16to 18-year-old working-class London men. They found that rates of officially recorded convictions were higher during periods of unemployment. When they administered a prediction scale of delinquency at age 10 (e.g., low income, poor parental child rearing, low intelligence, parental conviction), they found that unemployment was significantly related to crime only among participants with the most risk factors. This finding suggests that unemployment is criminogenic only among individuals with a high propensity for crime and therefore may not cause crime among generally lowrisk individuals. Stated differently, employment may be associated with the largest crime-preventive benefits among high-risk individuals, but it may have little or no impact on crime among low-risk persons.
Sampson and Laub (1993) used data from a sample of young males sentenced to a Boston-area reform school and matched them with a sample of schoolgoing youth. They constructed a measure of job stability that was a composite of employment status at the time of the interview, duration of the most recent employment, and work habits as indicated by reliable and effortful work performance. They found that job instability during the 17-25 age range was correlated with higher probability, frequency, and hazard of arrest during the 17-25 and 25-32 age ranges, net of official and unofficial juvenile delinquency. A follow-up of a subset of the reform school sample to age 70 revealed that arrest frequencies were significantly higher during months in which the participants were unemployed compared with months when they were employed (Laub & Sampson, 2003).
Grogger (1998) assessed the relationship between wages and crime among nonenrolled males (i.e., those not in school) in a national probability sample. He reported that higher wages corresponded with a substantially lower probability of criminal participation, controlling for prior criminal justice involvement. Further inspection of the data led Grogger to conclude that the African American—white wage gap accounted for about one quarter of the racial differential in crime participation. Moreover, he found that the age-earnings profile could plausibly explain the age distribution of crime from the late teens to the early 20s, leading him to conclude that “the growth in market opportunities with age is largely responsible for the concomitant decrease in crime” (p. 786).
Uggen (1999, 2000) has studied the employment-crime relationship among a sample of males who were part of a larger study of supported work for high-risk individuals. In one study, he found that job quality (measured objectively by aggregate job satisfaction scores on the Quality of Employment Survey) was inversely associated with selfreported crime among a sample of ex-offenders who were successful in finding work (Uggen, 1999). This was true even when he controlled for prior criminality and substance abuse and when he considered both economic and noneconomic crime as outcomes. In a second study, he found that a work opportunity was a significant turning point in the criminal careers of individuals with an arrest history (Uggen, 2000). Securing employment—even marginal employment—through a random assignment process was associated with a lower hazard of illegal earnings and arrest. He also found that older offenders (over age 26) benefited the most from this work experience.
By way of summary, empirical studies confirm the expectation from a variety of theories that having a job is associated with less crime than not having a job and that being unemployed is associated with more crime than being employed or out of the labor force. It also appears to be the case that having a good job—more stability, higher wages, better quality—is associated with even less crime than having a bad job, although even a bad job is still associated with less crime than unemployment, at least among high-risk samples (e.g., Hagan & McCarthy, 1997; Uggen, 2000). However, it should be noted that the strength of the correlation between employment and crime is not as impressive as one might anticipate from theoretical arguments. The correlation is often quite weak once other characteristics are controlled. Two other noteworthy findings are that the employment-crime connection tends to be stronger among older individuals as well among high-risk individuals. On the other hand, employment is not so strongly associated with crime among young persons and generally low-risk individuals (e.g., Farrington et al., 1986; Thornberry & Christenson, 1984; Uggen, 2000).
The Special Case of Adolescent Employment and Delinquency
Almost all U.S. adolescents gain employment experience before they graduate from high school, with as many as 90% of teenagers entering the labor market at some point during their high school careers (National Research Council, 1998). A nontrivial proportion of employed adolescents also work at high intensity—a label denoting employment of more than 20 hours per week (Greenberger & Steinberg, 1986). Folk wisdom would suggest that intensive exposure to the world of adult work provides a number of positive benefits for adolescents because of the way that it structures a youth’s leisure time, increases exposure to adult authority figures, fosters independence and maturity, teaches responsibility in the use of money, and promotes balancing of multiple responsibilities. Surprisingly, however, empirical research has consistently demonstrated that “the correlates of school-year employment are generally negative” (Steinberg & Dornbusch, 1991, p. 309). This is especially true where delinquent behavior is concerned.
In the 1980s and 1990s, a generation of studies of youth employment and antisocial behavior emerged that gave more sustained attention to the developmental consequences of adolescent employment (Agnew, 1986; Bachman & Schulenberg, 1993; Mortimer, 2003; Steinberg & Dornbusch, 1991; Wright, Cullen, & Williams, 1997, 2002). The seminal work of this new generation of research was a book by Greenberger and Steinberg (1986), titled When Teenagers Work: The Psychological and Social Costs of Adolescent Employment. Their unambiguous conclusion was that “extensive commitment to a job may interfere with the work of growing up” (p. 7). Research by Greenberger and Steinberg and others has consistently found that working during high school was associated with higher rates of school misconduct (e.g., truancy, cheating, suspension), substance use (e.g., cigarettes, alcohol, marijuana), minor delinquency (e.g., theft, vandalism), and serious delinquency (e.g., interpersonal aggression, assault). Moreover, researchers discovered that these negative side effects of employment were generally a function of work intensity, or the number of hours per week devoted to working. Specifically, intensive employment of more than 20 hours per week was associated with the most negative outcomes.
By the late 1990s, the scientific consensus was that work of moderate intensity (1-20 hours per week) has few adverse effects and in some cases is more developmentally beneficial than not working at all. However, beyond this 20-hour threshold employment appeared to be associated with more costs than benefits for the social and emotional development of young people. The finding that intensive employment during adolescence increases the risk of delinquency is puzzling in light of the research on adult employment reviewed earlier, which consistently indicates that adults who are strongly attached to work and who acquire full-time (read: intensive) employment are less likely to be criminally involved. These contradictory results have forced researchers into the awkward position of suggesting that the sign of the work effect changes at some point during the transition to adulthood, that is, that strong attachment to work (as measured by the number of hours per week) is criminogenic for adolescents but prophylactic for adults (e.g., Uggen, 2000, p. 530; Wright et al., 2002, p. 10).
Fortunately, employment-crime theories are sufficiently flexible to accommodate the apparent anomaly of adolescent work. One set of explanations appeals to the job quality thesis of traditional economic and sociological theories. Teenage employment is concentrated in the retail and service industries, in occupations that are universally regarded as low quality. These jobs pay barely more than minimum wage, involve little in the use or acquisition of any notable skills, offer few or no benefits or opportunities for upward mobility, and suffer constant turnover. They are often derided as teenage “McJobs” that do not engender any significant degree of attachment on the part of adolescent workers. Moreover, they tend to involve stressful working conditions and are often a stopping point for high school dropouts. Thus, it does not require theoretical acrobatics to explain why adolescent employment may be criminogenic. Low-quality jobs lead to crime among adolescents and adults alike; it just happens to be the case that the typical job for the typical adolescent is a low-quality one and thus a criminogenic one (see Staff & Uggen, 2003, for evidence on “good jobs” in adolescence).
Another set of explanations focuses attention on adolescence as a life stage and is more firmly rooted in developmental psychology and theories of precocious development. Put simply, intensive employment is one symptom of a latent, stage-specific propensity to expedite the transition to adulthood before adolescents have acquired the maturity to do so. The underlying issue for precocious development theory thus has to do with early timing of work, or developmentally “off-time” entry into the work role and especially an intensive work role. According to this perspective, the family and school are the primary socializing institutions in adolescents’ lives, with the workplace taking on secondary importance until the postsecondary years. With respect to family relationships, intensive employment disrupts healthy parent-child relationships, because these youth spend less time with, are less emotionally close to, engage in more disagreements with, are less closely monitored by, and exercise greater decision-making autonomy vis-à-vis their parents than nonworkers or moderate workers (Bachman & Schulenberg, 1993; Steinberg & Dornbusch, 1991). In schooling domains, intensive employment is associated with disinvestment in and disengagement from school, because it is correlated with less time spent studying and doing homework, cutting class and absenteeism, lower educational aspirations, a nonacademic track curriculum, negative school attitudes, and lower scholastic performance (Agnew, 1986; Steinberg & Dornbusch, 1991). Therefore, embeddedness in a developmentally unproductive work role competes with family and school as the dominant influence in the lives of adolescents and leads to a variety of delinquent and deviant adaptations (Wright et al., 2002).
Irrespective of the explanatory mechanism, until recently there was virtual unanimity that youth employment was criminogenic (with notable exceptions, e.g., Good et al., 1986). However, a new round of youth employment research has emerged in the 2000s that strongly challenges the interpretation of the employment-delinquency association as causal (Apel, Paternoster, Bushway, & Brame, 2006; Apel, Bushway, Paternoster, Brame, & Sweeten, 2008; Apel et al., 2007; Brame, Bushway, Paternoster, & Apel, 2004; Paternoster, Bushway, Brame, & Apel, 2003). This research has been attentive to the fact that adolescent workers (especially high-intensity workers) are different from moderate workers and nonworkers, often well before they begin working. For example, youth tend to enter the labor market in part as a result of weak emotional attachment to their parents, academic underperformance and school disengagement, and early delinquent and antisocial behavior (see Apel et al., 2007; Bachman & Schulenberg, 1993; Mortimer, 2003; Steinberg et al., 1993). In other words, youth with a higher propensity for crime are precisely those most likely to work intensively while in school. This implies that the apparent criminogenic effect of youth employment may be a selection artifact instead of the true causal effect of employment on delinquent behavior.
Empirical Challenges to Studying the Employment-Crime Relationship
Any study of the causal effect of employment on crime must confront at least two empirical challenges: (1) endogeneity and (2) simultaneity. These are threats to causal inference that can seriously bias empirical estimates of the employment—crime association. Each is discussed in turn, and recent efforts to overcome these challenges are described.
Endogeneity: The Selection Problem
One of the most serious challenges to existing studies of employment and crime is the selection problem. It is the problem of endogeneity of employment effects on crime, meaning that individuals who are employed (or are employed in high-quality jobs) differ fundamentally from individuals who are not employed in a way that accounts for their lower crime involvement. One may conceive of such person-level characteristics as ability, planfulness, and agreeableness that might individually or jointly increase the likelihood that an individual will be gainfully employed and simultaneously reduce the likelihood that the person will commit crime. The selection problem arises when these traits are difficult or impractical to observe and measure. The consequence is systematic bias in the estimated effect of employment on crime. Moreover, the direction of the bias under this scenario is predictable: The impact of employment on crime will be overestimated.
Sampson and Laub (1993) found that weak occupational commitment and job instability from ages 17 to 32 were predicted by official delinquency, unofficial delinquency (self-, parent, and teacher report), and early temper tantrums (parent report) during childhood. Caspi, Wright, Moffitt, and Silva (1998) linked youth unemployment (ages 15-21) with a variety of factors that reach far back into childhood. As measured in early childhood (ages 3-5), longer duration of unemployment was predicted by low family occupational status, low intelligence, an unmarried mother at birth, and difficult temperament. As measured in late childhood (ages 7-9), youth unemployment was predicted by these same variables in addition to family conflict and behavior problems. An important contribution of these studies is that they directly address the selection problem and identify underlying factors responsible for the differential sorting of individuals into the labor market, oftentimes long before they do so.
Under these circumstances, causal inference about the nature of the employment-crime relationship is aided by the availability of longitudinal data, which allow researchers to overcome endogeneity of the employment effect on crime attributable to time-stable individual differences, so-called unobserved heterogeneity (e.g., Horney et al., 1995). Such studies have examined the way in which change in employment affects change in crime and have found that the employment-crime relationship (at least among adults) does withstand these more rigorous selection controls and is not seriously biased by endogeneity. However, it is worth noting that the strength of the correlation tends to be weak compared with other time-varying factors, such as drug consumption and living arrangements (e.g., marital living and cohabitation).
The consequences of the selection problem have been brought into sharp focus in recent youth employment research. Paternoster et al. (2003) and Apel et al. (2006) have addressed the selection problem using longitudinal data on employment and antisocial behavior for 3 years. Both studies replicated the positive correlation between intensive employment during the school year and delinquent behavior using conventional methods. However, both also found that intensive work was positively correlated with delinquency only when examined across individuals but that within-individual change in work involvement was not correlated at all with change in delinquent behavior and substance use. They concluded that the criminogenic effect of intensive work among adolescents was driven by a process of selection rather than causation and could be best understood as a spurious correlation.1
In one of the most recent statements on the subject of adolescent employment, Apel et al. (2008) exploited interstate variation in child labor laws at the 15-to-16 transition as a source of causal identification. They found in their analysis that work intensity was actually inversely correlated with delinquent behavior; that is, the increase in work involvement from age 15 to 16 attributable to a loosening of child labor restrictions (the magnitude of which varied across states) was actually associated with a substantial decline in delinquent involvement. Once the problem of endogeneity was addressed through the use of longitudinal data and instrumental variables, then, the employment-delinquency association was found to be inverse after all, contrary to most previous youth employment research but well in line with employment-crime research among adults. Moreover, the techniques that Apel et al. used allowed them to interpret this as a causal association.
Simultaneity: The Feedback Problem
The contemporaneous, inverse correlation between employment and crime is usually interpreted as the causal effect of employment on crime. However, the correlation may in fact represent the causal effect of crime on employment, which is the feedback problem. This is the problem of simultaneity of causal effects, in that employment and crime mutually influence one another. The practical consequence of simultaneity bias is to systematically overestimate the effect of employment on crime, because the simultaneous inverse effect of crime on employment will be erroneously attributed to the effect of employment on crime.
Labeling theory, for one, anticipates just this sort of feedback effect from crime to employment. This is the notion of secondary deviance, or deviance amplification, among persons toward whom a sanction has been directed. An arrest or conviction, for example, constitutes a social stigma that might lead to exclusion from legitimate employment (Pager, 2003). Many prospective employers may be disinclined to hire individuals with a criminal record because it serves as a signal of sorts about what kind of employee one is likely to be. For example, employers may be sensitive to liability for negligent hiring (Bushway, 2004), or they may perceive offenders as untrustworthy (Waldfogel, 1994). A criminal record may also relegate individuals to the secondary labor market, or to what Nagin and Waldfogel (1995) referred to as “spot market jobs” as opposed to “career jobs.” This effect may be attributable, in part, to state-imposed restrictions on employment in certain industries (e.g., government employment), catering to vulnerable clientele (e.g., children), and professional licensing in certain occupations (Burton, Cullen, & Travis, 1987).
Empirical research confirms that a criminal record in the form of arrest, conviction, or incarceration does indeed hamper an individual’s future employment prospects (e.g., Nagin & Waldfogel, 1995; Waldfogel, 1994; Western, 2002). A criminal record reduces employment, increases unemployment, lowers earnings, slows wage growth, diminishes job tenure, and exacerbates job turnover. Thus, the feedback problem is real, and research that examines the contemporaneous effect of employment on crime must be attentive to simultaneity bias that overstates the preventive effect of employment on crime.
One way that researchers have addressed the feedback problem is through estimation of reciprocal models of employment and crime. Simultaneous equation studies have confirmed that the cross-sectional association between employment and crime is a combination of the effect of employment on crime as well as the effect of crime on employment (e.g., Good et al., 1986; Thornberry & Christenson, 1984). In these studies, isolation of causal effects requires the use of exclusion restrictions (i.e., instrumental variables) or other modeling constraints that are capable of identifying the simultaneous effects in the model. Thornberry and Christenson (1984) imposed cross-time equality constraints on model parameters to identify the reciprocal effects of unemployment and arrest. Good et al. (1986) used the number of job rejections as an instrumental variable for employment and gang affiliation and police enforcement (specifically, police contact) as instrumental variables for arrest. Each of these studies found that the effect of (un)employment on arrest was stronger than the contemporaneous feedback effect of arrest on (un)employment. In fact, both studies discovered that the contemporaneous effect of arrest on (un)employment was not statistically significant, although Thornberry and Christenson discovered that the influence of arrest was lagged one period, and Good and colleagues noted that the total number of prior police contacts was more salient. These studies thus suggest that the influence of criminality on employment operates through the accumulation of an arrest record that impedes the acquisition of stable employment.
The question of the relationship between employment and crime has a long history in criminology and dates back to the earliest studies of crime beginning in the mid-19th century. Empirical criminology has repeatedly confirmed the presence of an inverse correlation between employment and crime, and research that has addressed the selection and feedback problems in a compelling way points to the correlation as a causal one. However, as noted earlier, the strength of the employment-crime correlation is not nearly as impressive as a number of theoretical accounts would suggest. Neither has research successfully pinpointed the precise theoretical mechanism for the correlation. Nevertheless, there is sufficient evidence to date that continued exploration of this relationship is justified that would illuminate the causal pathway.
A handful of more recent studies have endeavored to do just that by considering heterogeneity in the employment—crime relationship. These studies are based on the presumption that employment may not have the same crimecontrol benefits for all members of a population. The population average treatment effect of employment on crime will not be meaningful if it is not representative of the group average treatment effect for any identifiable subgroup in the target population. It could be, in other words, an average over a possibly wide range of subgroup averages. Relatively more recent studies have found that the strength of the employment-crime correlation varies as a function of the aggregate labor market context (Crutchfield & Pitchford, 1997), specific characteristics of the job (Staff & Uggen, 2003), and individuals’ offending history (Apel et al., 2007; see also Farrington et al., 1986; Thornberry & Christenson, 1984). Studies such as these identify important pathways for further empirical and theoretical exploration.