Employment and crime are connected through some of the most robust causal relationships in applied criminology — a body of evidence accumulated across decades of longitudinal research, natural experiments, and randomized trials that establishes employment not merely as a correlate of reduced criminal involvement but as a genuine causal mechanism through which labor market conditions shape criminal behavior at both the individual and aggregate levels. The relationship operates through multiple theoretical pathways simultaneously: through the rational choice mechanism by which legitimate employment raises the opportunity cost of criminal activity and reduces economic motivation for property crime; through the social control mechanism by which stable employment builds the conventional bonds and routine structures that suppress criminal behavior independent of any economic calculation; and through the identity and desistance mechanism by which employment supports the narrative reconstruction of self that life-course criminology has identified as the cognitive foundation of sustained behavioral change. Understanding how these pathways operate, and which labor market and employment policy interventions most effectively engage them, is among the most practically consequential questions in criminological policy research.
Criminology and Public Policy engages the employment-crime relationship as both an empirical question — what does the evidence establish about how labor market conditions affect crime? — and a policy design question — which employment programs, wage policies, and labor market institutions most effectively reduce crime through the employment pathway? The policy implications of this research extend well beyond the criminal justice system to encompass minimum wage legislation, workforce development programs, occupational licensing reform, fair chance hiring, and the economic development policies that determine the availability of legitimate employment in the communities where criminal risk is highest. Criminal justice practitioners who ignore the labor market determinants of crime are addressing only the downstream consequences of conditions they cannot themselves change; policymakers who treat employment as irrelevant to crime control are leaving the most cost-effective available interventions unused.
Introduction
The theoretical architecture of the employment-crime relationship has been developed across three major intellectual traditions that generate different but complementary predictions about which employment interventions most effectively reduce crime. The rational choice and deterrence tradition, drawing on Gary Becker’s economic model of crime, treats employment’s crime-suppressing effect as operating primarily through the opportunity cost mechanism: legitimate employment provides legal earnings that raise the cost of criminal activity by placing legal income at risk if caught and convicted. On this account, policies that increase legitimate earnings — minimum wage increases, earned income tax credits, job training programs — reduce property crime by improving the financial returns to legal work relative to crime. The prediction is specific and testable: employment interventions that raise wages and stabilize employment should reduce economically motivated property crime, with larger effects among individuals for whom the legitimate-to-criminal earnings ratio is most unfavorable.
The social control tradition, drawing on Sampson and Laub’s life-course theory, treats employment’s crime-suppressing effect as operating primarily through the bond mechanism: stable employment builds conventional attachments, creates routine structures that reduce exposure to criminal opportunity, and provides stakes in conformity that make criminal activity incompatible with the legitimate identity and life trajectory that employment supports. On this account, employment reduces crime not primarily because it raises earnings but because it builds social bonds — and accordingly, the quality and stability of employment matters as much as its wage level. Precarious employment — unstable hours, low wages, limited benefits, no path to advancement — may not build the conventional bonds that suppress crime even when it provides some legitimate income. The life-course prediction is that turning point employment experiences — jobs that provide not only income but social integration, identity, and future orientation — are what most effectively redirect criminal trajectories.
The identity and desistance tradition, drawing on Maruna’s narrative research and the emerging literature on cognitive transformation in desistance, treats employment’s crime-reducing effect as operating through the identity mechanism: employment supports the construction of a prosocial self-narrative that makes continued criminal involvement incompatible with who the person understands themselves to be and who they are becoming. On this account, the most crime-preventive employment interventions are those that support not only job placement but the identity transformation that sustained desistance requires — programs that treat participants as capable of conventional success rather than as risks to be managed, and that provide the social support and recognition that identity reconstruction needs. This framework generates specific predictions about program design: employment programs that combine job placement with mentoring, peer support, and explicit attention to identity and narrative are more effective than those that focus exclusively on skill development and job matching.
The Aggregate Evidence: Labor Markets and Crime Rates
Business Cycles, Unemployment, and Crime
The aggregate relationship between labor market conditions and crime rates has been extensively studied using time-series methods that track the co-movement of unemployment rates and crime rates across years and decades. The relationship is real but empirically modest at the aggregate level — the correlation between unemployment and property crime is positive and statistically significant in most analyses, but the magnitude is smaller than either the theoretical predictions or the popular intuition about the relationship would suggest. Research by economist Steven Raphael and colleagues established through careful time-series analysis that a 1 percentage point increase in the unemployment rate is associated with roughly a 1–2% increase in property crime rates, with weaker and less consistent effects for violent crime (Raphael & Winter-Ebmer, 2001). The modest aggregate relationship reflects both the multiple competing causal factors simultaneously affecting crime rates and the heterogeneity of unemployed individuals — many unemployed workers do not have elevated criminal propensity, and the subset with elevated propensity may respond differently to unemployment than the aggregate statistics can detect.
The more striking and more policy-relevant evidence comes from studies focusing specifically on the crime consequences of labor market conditions for high-risk populations — young men with limited education and prior criminal justice involvement — for whom the ratio of legitimate to criminal earnings is most unfavorable and the employment-crime relationship is most theoretically expected to be strong. Research by David Card and colleagues using variation in local labor market conditions as instruments for employment and wages among low-skill workers has found larger employment-crime effects for this population than aggregate analyses suggest, consistent with the rational choice prediction that employment’s crime-reducing effect is concentrated in the population for whom legitimate and criminal earnings are closest substitutes (Gould et al., 2002). The implication for policy is that minimum wage increases, local employment programs, and economic development investments targeted at communities with high concentrations of low-skill workers should produce crime reduction benefits beyond those that aggregate employment analyses would predict.
The dramatic crime decline of the 1990s — one of the steepest and most sustained crime reductions in American history — provides the most vivid historical case study of the labor market-crime relationship’s practical importance. The 1990s economic expansion produced sustained declines in unemployment, particularly for low-skill workers, alongside rises in wages at the bottom of the wage distribution that are consistent with labor market tightening in tight labor market conditions. Research attributing a meaningful portion of the 1990s crime decline to the labor market improvement — alongside other contributing factors including the crack cocaine market’s stabilization, the effects of increased incarceration on incapacitation, changes in policing strategy, and demographic shifts — generally finds that labor market improvement accounts for 10–30% of the crime decline, making it one of the most important single contributors even if not the dominant one (Blumstein & Wallman, 2000).
Table 1. Employment Interventions and Crime: Evidence Summary
| Intervention | Mechanism | Study Design | Crime Outcome | Effect | Key Finding |
|---|---|---|---|---|---|
| Minimum wage increases | Human capital; opportunity cost | Quasi-experimental (state variation) | Property crime; youth crime | Positive | Moderate property crime reductions; concentrated among low-wage workers and youth |
| Earned Income Tax Credit expansion | Human capital; income supplement for workers | Quasi-experimental (policy variation) | Property crime; domestic violence | Positive | Consistent property crime reductions; domestic violence reductions documented |
| Job Corps | Vocational training; employment | RCT | Arrest rates | Positive (older youth) | Reduced arrests for 17–20 year-olds; weaker effects for younger participants |
| Transitional jobs programs | Immediate subsidized employment | RCT (multiple MDRC trials) | Recidivism | Mixed | Employment effects consistent; recidivism effects mixed and context-dependent |
| Supported employment (serious mental illness) | IPS model; competitive employment | Multiple RCTs | Criminal justice involvement | Positive | Reduced criminal justice contact for participants with mental illness |
| Occupational licensing reform | Remove employment barriers for record holders | Quasi-experimental (state variation) | Recidivism | Positive (emerging) | States removing licensing barriers show reduced recidivism; evidence developing |
| Fair Chance / Ban the Box | Remove record disclosure barriers in hiring | Quasi-experimental (policy variation) | Employment; recidivism | Mixed | Increases employment applications and callbacks; crime reduction effects modest |
Individual-Level Evidence: Employment and Desistance
Longitudinal Research on Employment and Criminal Careers
The individual-level evidence on the employment-crime relationship — drawn primarily from longitudinal cohort studies that track individuals’ employment histories alongside their criminal trajectories — provides the most detailed available account of how employment affects criminal career development and the timing of desistance. Sampson and Laub’s reanalysis of the Glueck data established that stable job attachment in adulthood predicted desistance even among men with extensive adolescent delinquency records, with the effect of employment operating independently of prior criminal history and individual characteristics that independently predicted recidivism (Sampson & Laub, 1993). The turning point mechanism they identified — stable employment as a life event that builds conventional bonds, creates routine structures, and supports identity reconstruction — remains the most influential theoretical account of employment’s role in the desistance process.
Research by Robert Apel and Gary Sweeten using data from the National Longitudinal Survey of Youth documented that employment quality — not just employment status — matters for the employment-crime relationship: good jobs that provide stable hours, reasonable wages, and conventional work environments reduce crime, while bad jobs that involve unstable hours, low wages, and criminal opportunity structures do not reliably suppress criminal involvement and in some cases may even provide criminal opportunity (Apel & Sweeten, 2010). This finding has important implications for employment policy: programs that place formerly incarcerated individuals in low-quality contingent employment may not produce the crime reduction benefits that the employment-crime relationship would predict if quality employment were the counterfactual. The policy implication is not merely job placement but job quality — supporting access to employment that builds the conventional social bonds and future orientation that desistance requires.
The Criminal Record Employment Barrier
The structural consequence of mass incarceration for labor market participation among Black men in particular has been documented by Bruce Western and colleagues in research establishing that incarceration has become so common in some demographic groups that it functions as a life stage — a near-universal experience that brands its participants with the permanent labor market stigma of a criminal record regardless of subsequent conduct (Western, 2006). By his estimates, approximately one-third of Black men without a high school diploma have been incarcerated, creating a population whose collective labor market disadvantage is not merely an individual-level problem but a structural feature of the American labor market that shapes the economic conditions of entire communities. The crime consequences of this structural labor market disadvantage are predictable from both the rational choice and social control accounts of the employment-crime relationship: when legitimate employment is systematically unavailable, the legitimate-to-criminal earnings ratio deteriorates and the social bonds that conventional employment provides cannot be built.
The labor market consequences of criminal justice contact are among the most extensively documented forms of collateral consequence in the reentry research literature. Criminal records reduce employment prospects through multiple mechanisms: direct employer discrimination against applicants who disclose criminal records; occupational licensing restrictions that formally bar record holders from credentialed occupations in healthcare, education, financial services, and other sectors; background check requirements that make the record permanently accessible regardless of the age of the conviction or evidence of rehabilitation; and the human capital depreciation that incarceration produces through disconnection from legitimate labor markets during the incarceration period.
Devah Pager’s experimental audit studies — sending matched pairs of testers to apply for entry-level jobs, with one tester in each pair disclosing a criminal record while the other did not — documented the magnitude of criminal record discrimination with experimental precision (Pager, 2003). White applicants with criminal records received callbacks at about half the rate of white applicants without records; Black applicants without records received callbacks at a rate comparable to white applicants with records, illustrating both the employment penalty of criminal records and the compounding disadvantage of racial stigma in the labor market. The intersection of racial stigma and criminal record stigma — both experimentally documented — helps explain why employment barriers for formerly incarcerated individuals are not only economically damaging but racially concentrated in the communities where criminal justice contact is itself most concentrated.
Employment Programs for Justice-Involved Populations
Transitional Employment and Subsidized Work
Transitional jobs programs — which provide immediate subsidized employment to recently released prisoners or others with significant employment barriers, typically in public service or nonprofit settings, as a bridge to unsubsidized competitive employment — represent the employment policy response most directly targeted at the reentry employment barrier. Multiple randomized controlled trials conducted by MDRC — the policy research organization that has conducted the most rigorous evaluations of transitional jobs programs — have found consistent evidence that transitional jobs substantially increase employment during the program period, with the employment effects persisting for several months after program exit. The recidivism effects of transitional jobs have been more mixed and more context-dependent: some evaluations have found significant reductions in recidivism, while others have found null effects, with the variation apparently reflecting differences in program quality, follow-up support, and the labor market conditions participants face after the transitional period ends (Redcross et al., 2012).
The conditional nature of transitional jobs’ recidivism effects is itself theoretically informative: it suggests that temporary employment alone, without the sustained social bonds and permanent labor market integration that the life-course model predicts are necessary for desistance, does not produce lasting crime reduction. Programs that combine subsidized employment with intensive case management, vocational training, and connections to permanent employment opportunities show more consistent recidivism reductions than those that provide employment without complementary support. The policy implication is that the transitional jobs model is most effective when it functions as a bridge to stable conventional employment rather than as an end in itself.
Occupational Licensing Reform
The National Reentry Resource Center, administered by the Council of State Governments Justice Center with support from the Bureau of Justice Assistance, has developed technical assistance resources for jurisdictions seeking to improve reentry employment outcomes through a combination of workforce development services, employer engagement, and policy reform. Its research on the characteristics of successful reentry employment programs identifies several consistent features: strong relationships with employers willing to hire individuals with records; individualized assessment of participants’ employment barriers and assets; skills training connected to occupations with genuine labor market demand; wage subsidies or other employment incentives during the initial employment period; and sustained case management support through the first year of employment when the employment instability risk is highest. These program features are consistent with the theoretical predictions of the life-course model — that it is stable quality employment with sustained support, not merely any employment, that produces lasting crime reduction through the bond and identity mechanisms that desistance requires.
Approximately one-quarter of American workers require an occupational license to practice their trade, and many licensing statutes contain provisions that bar or restrict licensure for individuals with criminal records — often with limited connection to the specific public safety concerns that occupational licensing is theoretically designed to address. Healthcare licensing statutes may permanently bar individuals with drug convictions from careers in home health care; financial services licensing may bar those with financial crimes from certain positions; education licensing may bar those with any felony conviction from teaching regardless of the relevance of the conviction to their capacity to teach safely. These restrictions collectively limit the employment opportunities of the approximately 19 million Americans with felony convictions and the many millions more with misdemeanor records, with crime prevention consequences that the research is beginning to document.
Research exploiting variation across states in occupational licensing restrictions for individuals with criminal records has found that states that have reformed their licensing statutes to require case-by-case review rather than categorical bar — mandating that licensing boards assess the relevance of specific convictions to specific occupational risks rather than applying blanket exclusions — show improved employment and reduced recidivism among formerly incarcerated individuals relative to states with more restrictive categorical approaches. The National Conference of State Legislatures and the Institute for Justice have both documented the landscape of occupational licensing restrictions and the reform efforts underway, with bipartisan support for reform reflecting both the conservative concern about government-imposed employment barriers and the progressive concern about collateral consequences as drivers of recidivism.
Labor Market Policy and Crime Prevention
Research by Robert Sampson and colleagues on the community-level consequences of male joblessness — documenting that neighborhood rates of male unemployment predict subsequent violence rates even after controlling for poverty and other measures of disadvantage — established that employment’s crime-reducing effect operates at the community level through mechanisms that individual-level research cannot capture (Sampson & Wilson, 1995). When large proportions of working-age men in a community are chronically unemployed, the resulting deficits in institutional participation, social network connectivity, and the normative authority of conventional institutions weaken the community-level informal social control that the collective efficacy literature identifies as the primary mechanism through which neighborhood social organization suppresses violence. Community-level employment investment — manufacturing retention, living wage ordinances, targeted hiring requirements in public procurement — is thus a crime prevention investment at the community level that individual job placement programs cannot substitute for.
Minimum Wage, Earnings, and Property Crime
The relationship between minimum wage levels and crime has been examined through multiple quasi-experimental studies exploiting variation in state minimum wage laws to estimate the causal effect of wage increases on crime. The consistent finding is that minimum wage increases reduce property crime, with the effect concentrated among the low-wage workers and youth populations for whom the minimum wage is most binding and for whom the legitimate-criminal earnings ratio is most relevant (Beauchamp & Chan, 2014). The mechanism is consistent with the rational choice account: higher wages increase the opportunity cost of criminal activity by raising what offenders would sacrifice if apprehended and convicted. The effect size is modest but consistent — a 10% minimum wage increase is associated with approximately a 1–3% reduction in property crime in most studies — and represents a meaningful crime prevention benefit that should be incorporated into minimum wage policy assessments alongside the labor market and distributional effects that typically dominate policy debate.
The Earned Income Tax Credit — which supplements the wages of low-income workers with children through the federal tax system — has accumulated evidence of crime reduction alongside its well-documented effects on poverty reduction and labor force participation. Studies by Maxfield and Scheuble, Lindner and Nichols, and others exploiting variation in EITC benefit levels across states and over time have found consistent associations between EITC expansions and reductions in property crime, with the effect concentrated in the population of low-income working parents for whom the EITC provides its most substantial supplements. Research by Anna Aizer and colleagues has documented reductions in domestic violence following EITC expansions, consistent with the stress reduction and income security mechanisms through which economic improvement may reduce violence in intimate partner relationships as well as the rational choice mechanisms through which it reduces property crime.
Place-Based Employment Policy and Crime
Place-based employment policy — investment in economic development, job creation, and labor market institutions in the specific communities where criminal risk is most concentrated — represents a crime prevention strategy that neither individual-level employment programs nor broad labor market policies can achieve alone. Research documenting the geographic concentration of both criminal involvement and unemployment in specific urban neighborhoods — the same neighborhoods where public housing concentration, school quality deficits, and limited institutional resources compound each other’s negative effects — suggests that the crime prevention benefits of employment investment are concentrated when that investment reaches the highest-crime, highest-unemployment neighborhoods rather than spreading its effects diffusely across metropolitan areas.
The New Markets Tax Credit program, Promise Zones designated by the Department of Housing and Urban Development, and Opportunity Zones created by the 2017 Tax Cuts and Jobs Act all represent federal policy mechanisms for directing private investment toward economically distressed communities — mechanisms whose crime prevention benefits have received less systematic evaluation than their housing and economic development effects. Research on enterprise zones — earlier incarnations of place-based economic development investment — found modest employment effects in targeted communities with limited evidence of crime reduction, suggesting that investment incentives that reduce the cost of capital in distressed areas are insufficient without complementary workforce development and employer-community partnership investments that connect local residents to the employment opportunities that capital investment creates. The most productive place-based employment strategies are those that create genuine connections between local hiring, living wage employment, and the communities most affected by both unemployment and crime.
Gig Economy, Precarious Work, and Crime
The structural transformation of the American labor market toward more contingent, gig-based, and precarious employment arrangements raises important questions about whether the employment-crime relationship documented in research conducted during the era of more stable employment holds under the conditions of 21st-century labor markets. If the crime-reducing effect of employment operates primarily through the bond and identity mechanisms that stable, quality employment builds — as the life-course evidence suggests — then the expansion of precarious employment without stable hours, benefits, or advancement may not produce the crime reduction benefits that employment transitions historically generated. Research on the crime consequences of gig work specifically is limited, but the theoretical predictions are not encouraging: the flexibility and autonomy of gig platforms may provide income without the routine structure, organizational belonging, and stake in conventional employment that the social control account of employment’s crime-reducing effect requires.
The increasing role of criminal records in algorithmic background check systems that serve gig economy employers adds a further dimension: platforms that require background checks for participation effectively extend the criminal record employment barrier into the gig sector, which has sometimes been proposed as an alternative employment pathway for individuals with records excluded from conventional employment. Research documenting the expansion of background check requirements across economic sectors — from traditional employment through professional licensing to rental housing, educational enrollment, and volunteer positions — suggests that the aggregate burden of criminal record employment barriers has increased substantially in recent decades, compounding the crime consequences of mass incarceration through the employment pathway.
The Second Chance Act grant programs, funded through the Bureau of Justice Assistance since 2008, provide federal support for reentry employment services across hundreds of jurisdictions — connecting recently released individuals with vocational assessment, job training, placement services, and employer engagement programs that the research evidence identifies as components of effective reentry employment support. Evaluations of Second Chance Act-funded programs have generally found positive employment outcomes alongside modest recidivism reductions, with the strongest effects in programs that provide sustained support through the first year of employment and that address the full range of reentry barriers — housing, transportation, mental health, substance abuse — alongside employment needs. The program’s explicit evidence-based requirements — funded programs must demonstrate use of validated risk assessment, evidence-based practices, and outcome measurement — have contributed to a gradual improvement in program quality across the reentry employment field that the grant program’s scale and sustained funding have enabled.
Research on prisoner reentry and employment by Jeremy Travis and colleagues at the Urban Institute established the foundational empirical picture of the reentry employment challenge: most released prisoners are unprepared for the labor market, face severe employment discrimination, and lack the support networks and professional credentials that successful job search requires. Travis’s But They All Come Back: Facing the Challenges of Prisoner Reentry (2005) synthesized the research in a form that influenced both the Second Chance Act’s design and the broader policy conversation about reentry as a public safety challenge requiring workforce development investment alongside criminal justice supervision. The subsequent expansion of reentry employment research — through the Second Chance Act evaluation portfolio, through randomized trials of specific program models, and through the growing use of administrative data linkages connecting employment and recidivism records — has produced a substantially more rigorous evidence base than existed when the Second Chance Act was enacted, supporting continued investment in the program models that evidence identifies as most effective.
Conclusion
The employment-crime relationship is one of the most thoroughly documented causal relationships in applied criminology, supported by theoretical frameworks that generate specific and testable predictions alongside a diverse body of empirical evidence ranging from longitudinal cohort studies to randomized trials to natural experiments. The evidence supports a policy-consequential conclusion: employment — particularly stable, quality employment that builds conventional bonds and supports identity reconstruction — reduces crime through mechanisms that operate independently of and in addition to the effects of criminal justice interventions. Policies that raise legitimate wages, reduce employment barriers for individuals with criminal records, provide transitional employment bridges for recently released prisoners, and reform occupational licensing restrictions that foreclose legitimate careers all contribute to crime reduction through the employment pathway.
The translation challenge for employment-crime policy is institutional: employment policy is made by labor departments, workforce development agencies, and legislatures focused primarily on economic outcomes, while criminal justice policy is made by agencies focused primarily on public safety outcomes. Creating the cross-agency collaboration and integrated policy frameworks that would allow employment interventions to be designed and evaluated explicitly for their crime reduction benefits — and criminal justice interventions to be evaluated for their employment consequences — requires institutional innovation that the conventional separation of policy domains impedes. The evidence for this integration is compelling; the organizational barriers are real but not insurmountable, as demonstrated by the jurisdictions that have successfully integrated workforce development into reentry programming and incorporated fair chance hiring into procurement requirements for government contractors.
The equity dimension of employment policy as crime prevention is equally important and equally neglected. The criminal record employment barriers that most powerfully undermine the employment pathway to desistance fall disproportionately on Black Americans — who are more likely to have criminal records because of the racial disparities in criminal justice contact that criminological research has thoroughly documented — compounding the labor market disadvantage that racial discrimination independently produces. Employment policy that does not address this racial dimension — that removes formal barriers without addressing the informal discrimination that persists after formal barriers fall — will produce more limited crime reduction benefits than policy that engages the full intersection of criminal record stigma and racial labor market inequality.
References
- Apel, R., & Sweeten, G. (2010). The impact of incarceration on employment during the transition to adulthood. Social Problems, 57(3), 448–479. https://doi.org/10.1525/sp.2010.57.3.448
- Beauchamp, A., & Chan, S. (2014). The minimum wage and crime. B.E. Journal of Economic Analysis & Policy, 14(3), 1213–1235. https://doi.org/10.1515/bejeap-2013-0124
- Blumstein, A., & Wallman, J. (Eds.). (2000). The crime drop in America. Cambridge University Press.
- Gould, E. D., Weinberg, B. A., & Mustard, D. B. (2002). Crime rates and local labor market opportunities in the United States. Review of Economics and Statistics, 84(1), 45–61. https://doi.org/10.1162/003465302317331919
- Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108(5), 937–975. https://doi.org/10.1086/374403
- Raphael, S., & Winter-Ebmer, R. (2001). Identifying the effect of unemployment on crime. Journal of Law and Economics, 44(1), 259–283. https://doi.org/10.1086/320275
- Redcross, C., Millenky, M., Rudd, T., & Levshin, V. (2012). More than a job: Final results from the evaluation of the Center for Employment Opportunities (CEO) transitional jobs program. MDRC. https://www.mdrc.org/publication/more-job
- Sampson, R. J., & Laub, J. H. (1993). Crime in the making: Pathways and turning points through life. Harvard University Press.
- Western, B. (2006). Punishment and inequality in America. Russell Sage Foundation.