Stigma — the social process through which individuals possessing discrediting attributes are reduced from full social persons to devalued, discredited identities in the eyes of those who interact with them — connects the sociology of crime to the broader sociology of social difference, inequality, and marginalization. Erving Goffman’s foundational analysis in Stigma: Notes on the Management of Spoiled Identity (1963) established the conceptual vocabulary through which criminology engages with the social consequences of criminal labeling: the management of spoiled identity through information control and passing, the internalization of stigma through awareness of social devaluation, and the consequences of stigma for social relationships, institutional access, and life chances. Crime and stigma are connected in two directions simultaneously: individuals who commit crimes acquire the stigma of the criminal label through official processing, and individuals who are stigmatized on other grounds — race, poverty, mental illness, homelessness — are disproportionately targeted for criminal justice contact regardless of their actual criminal behavior.
Sociology and Criminology treats stigma as both a consequence of criminal justice processing and a structural condition that elevates criminal justice risk for stigmatized groups — understanding both directions is essential for the comprehensive account of how stigma, crime, and social control interact in ways that reproduce rather than reduce the inequality they address. The stigma of criminal conviction — producing the employment discrimination, housing exclusion, civic disenfranchisement, and social relationship damage that constitute the collateral consequences of conviction — is simultaneously the most direct individual consequence of formal criminal justice processing and one of the most powerful mechanisms through which crime careers are sustained rather than terminated by criminal justice intervention.
Introduction
The sociology of stigma distinguishes between three types of stigma that Goffman identified: abominations of the body (physical differences including disabilities and disfigurements), blemishes of individual character (including criminal records, mental illness, substance use, and other attributed moral failures), and tribal stigma (race, ethnicity, religion, and other group membership characteristics). Criminal conviction falls primarily in the second category — blemishes of individual character — but its interaction with racial and class stigma (the third and first categories) is central to understanding why criminal justice stigma is distributed as it is across the social structure, with individuals already stigmatized by race, class, and mental health status facing dramatically higher rates of criminal justice contact and the criminal stigma that contact produces.
Research by Bruce Link and Jo Phelan extended Goffman’s analysis into a more systematic sociological framework for understanding how stigma affects outcomes. Their modified labeling theory proposed that individuals who are labeled — with mental illness, with criminal conviction, or with other stigmatizing attributes — face not only the immediate social consequences of the label but a restructuring of their social world that affects their relationships, their access to social capital, and their self-concept in ways that compound the initial consequences over time. The expectation of stigma — the anticipation of discrimination and social rejection — itself affects behavior, producing the social withdrawal and identity management strategies that compound the direct consequences of stigma through the avoidance of social situations where disclosure might occur.
The Criminal Record as Stigmatizing Mark
Labor Market Consequences
The employment consequences of criminal records are among the most consequential and most extensively researched dimensions of criminal stigma. Research by Devah Pager — using audit studies that sent matched pairs of applicants (one with and one without a criminal record) to apply for real job openings — documented that a criminal record reduces callback rates by approximately 50% for comparable white applicants and by approximately 64% for comparable Black applicants, establishing both the magnitude of criminal record stigma in hiring and its interaction with racial stigma. The finding that a white applicant with a criminal record was more likely to receive a callback than a Black applicant without one was among the most cited and most consequential findings in the sociology of criminal stigma — establishing that the intersection of criminal and racial stigma produces compound disadvantage that exceeds either stigma in isolation.
Research on the mechanisms through which employer stigma operates has documented both statistical discrimination (employers use criminal records as signals of unobservable risk characteristics and use race as a signal of probable criminal record) and taste-based discrimination (some employers have racial and criminal record preferences that operate independently of productivity concerns). The scale of criminal record discrimination is magnified by the size of the population it affects: research by Shannon and colleagues estimated that approximately 70 million Americans have criminal records, and that one-third of working-age adults have been arrested at least once. At this scale, criminal record discrimination is not a marginal labor market phenomenon affecting a small deviant minority but a structural labor market institution whose consequences for economic participation, tax revenues, and social mobility are macroeconomically significant.
Housing and Social Service Exclusion
Criminal record discrimination in housing — including both the explicit exclusion through “no criminal record” screening criteria and the implicit exclusion through background check requirements that produce disparate impact on Black and Latino applicants — compounds the employment consequences of criminal stigma by adding residential instability to economic marginalization. Research on housing discrimination against formerly incarcerated individuals has found that landlords reject applicants with criminal records at substantially higher rates than comparable applicants without records, and that this discrimination is more severe for violent offense records and drug offense records than for property offense records — establishing a severity gradient in housing stigma that tracks the cultural associations of different crime types.
The systematic exclusion of formerly incarcerated individuals from public housing — embedded in federal regulations that allow and sometimes require public housing authorities to exclude individuals with criminal records — compounds market-driven housing stigma with state-imposed housing exclusion. Research on the consequences of housing instability for recidivism has consistently found that stable housing substantially reduces subsequent criminal justice contact — establishing that housing exclusion is not merely an equitable concern but a recidivism-producing consequence of criminal stigma that costs the criminal justice system far more in subsequent incarceration costs than the housing assistance it denies would require.
Table 1. Consequences of Criminal Record Stigma
| Domain | Consequence | Evidence Type | Magnitude |
|---|---|---|---|
| Employment | Reduced hiring probability; lower wages | Audit studies; administrative data | ~50% reduction in callbacks for white applicants; ~64% for Black |
| Housing | Reduced rental access; public housing exclusion | Survey; audit studies | Significant exclusion, especially for violent/drug records |
| Voting | Disenfranchisement in 48 states for felony; permanent in some | Administrative data | ~5.2 million disenfranchised nationally |
| Education | Federal student loan restrictions for drug offenses | Administrative data | Hundreds of thousands annually affected |
| Professional licensing | Restriction or prohibition for 800+ occupations | Legal analysis | Varies by state and occupation |
| Social relationships | Stigma in family, friendship, and intimate partnerships | Survey; ethnographic | Substantial isolation reported by formerly incarcerated |
| Self-concept and identity | Internalized stigma; diminished agency | Survey; qualitative | Documented across multiple studies |
Social Network Disruption and Stigma
Criminal stigma operates not only through the formal mechanisms of labor market and housing discrimination but through the social network disruption that criminal justice contact produces — the deterioration of existing social relationships and the foreclosure of new relationships that the stigma of criminal record creates. Research by Western and Muller on the social isolation of formerly incarcerated individuals has documented that incarceration substantially reduces the quantity and quality of social network connections — not only through the direct disruption of relationships during confinement but through the stigma-driven avoidance of formerly incarcerated individuals by potential relationship partners who fear the social contamination or practical complications that association with a formerly incarcerated person might produce.
Research by Comfort on the “secondary prisonization” of women whose intimate partners are incarcerated documents a specific form of stigma transfer — the social devaluation that accrues to individuals through their association with the incarcerated — that extends the stigma of criminal conviction beyond the individual offender to their family members and intimate partners. This secondary stigma represents an important dimension of mass incarceration’s social costs that aggregate statistics on individuals with criminal records cannot capture, affecting millions of family members through the social contamination that association with incarcerated or formerly incarcerated individuals produces in social environments where criminal stigma is strong.
Race, Stigma, and Criminal Justice
The Interaction of Race and Criminal Stigma
The interaction between racial stigma and criminal stigma in the United States is among the most consequential dynamics in the sociology of crime — producing compound disadvantage that neither stigma in isolation fully predicts. Research has documented that racial stigma operates in the criminal justice context through multiple mechanisms simultaneously: the implicit racial associations between Blackness and criminality that cognitive psychology documents in police officers, prosecutors, jurors, and judges; the explicit racial discrimination in criminal justice contact that audit and natural experiment studies document; and the structural mechanisms through which concentrated policing in minority communities produces concentrated criminal records in those communities, reproducing racial stigma through the criminal records that racially targeted enforcement creates.
Research by Jennifer Eberhardt and colleagues documenting that police officers more quickly associate Black faces with criminal stimuli, and that more stereotypically Black-featured defendants receive harsher sentences in capital cases, establishes the implicit racial bias dimension of criminal stigma’s racially disparate application. Research by Alexander, Wacquant, and Western documenting the systematic racial concentration of criminal records establishes the structural dimension: not only are Black Americans subjected to criminal stigma more frequently through racially disparate enforcement, but the stigma they acquire reduces their life chances in the same labor market, housing market, and civic space where racial stigma was already reducing them — creating the compound stigmatization that makes criminal justice contact particularly devastating for racial minority individuals.
Intersectional Stigma: Race, Gender, and Crime
Research on the intersection of racial stigma and gender stigma in criminal justice processing has documented how Black women face a distinctive combination of stigmas that produces specific forms of both invisibility (as crime victims whose victimization is less likely to be recognized and prosecuted) and hyper-visibility (as criminal suspects whose behaviors are more likely to be policed and prosecuted than comparable white women’s behaviors). Research by Richie on the criminalization of Black women’s survival strategies — documenting how poverty, domestic violence victimization, and drug use combine to produce criminal justice contact for Black women through pathways that do not apply equally to white women — establishes the intersectional dimension of criminal stigma that single-axis analyses cannot capture.
Research on transgender individuals’ experiences with criminal justice contact — documenting exceptionally high rates of arrest, incarceration, and police violence against transgender women of color — establishes how the intersection of gender identity stigma, racial stigma, and the economic marginalization that discrimination produces creates specific vulnerability to criminal justice contact and the criminal stigma it generates. The specific forms of stigma that intersectional positions produce require intersectional analysis that the single-category frameworks of most stigma research — which examine race or gender or criminal record in isolation — cannot provide.
Stigma, Mental Illness, and Criminal Justice
Research on the relationship between mental illness stigma and criminal justice contact has documented a specific feedback dynamic: mental illness stigma produces social exclusion and inadequate treatment that elevate the probability of criminal justice contact, while criminal justice contact adds criminal stigma to mental illness stigma in ways that further reduce treatment access, employment prospects, and social integration. Research by the Treatment Advocacy Center documenting that individuals with serious mental illness are ten times more likely to be in a jail or prison than a psychiatric hospital — a consequence of deinstitutionalization and inadequate community mental health infrastructure — establishes the specific institutional pathway through which mental illness stigma and criminal stigma intersect in the contemporary American context.
Research on the consequences of dual stigma — the simultaneous possession of mental illness and criminal record stigma — has documented that dually stigmatized individuals face compounded discrimination in employment and housing that neither single stigma alone predicts, and that the combination produces the most severe social exclusion and highest recidivism rates in the reentry literature. Programs that address both stigmas simultaneously — through the integrated treatment, housing, and employment support that the most effective reentry programs provide — show better outcomes than those that address either stigma in isolation, establishing that the intersectional approach to stigma reduction is not only analytically more accurate but practically more effective.
Stigma Management and Identity
Goffman’s Information Management Framework
Goffman’s analysis of how stigmatized individuals manage their social interactions in a world where their discrediting attribute is known or potentially discoverable provides the micro-sociological framework for understanding the social experience of criminal stigma. Individuals with criminal records who are not visibly identifiable as such face the challenge of information management: deciding when, to whom, and how to disclose their record, managing the anxiety of potential discovery, and navigating the social interactions where their record may be relevant.
Research on the strategies through which formerly incarcerated individuals manage criminal stigma — including passing (concealing the record completely), covering (acknowledging the record while minimizing its relevance), and redemptive disclosure (proactively narrating the record as part of a transformation story) — has found that different strategies are effective in different social contexts and with different social audiences. The redemptive narrative — the story of transformation from criminal past to prosocial present that Maruna’s desistance research identified as characteristic of successful desisters — is both a stigma management strategy and an identity transformation that enables genuine behavioral change by reorganizing the self-concept around a prosocial rather than criminal identity.
Reducing Stigma: Policy Approaches
Research on stigma reduction approaches has found evidence supporting several policy mechanisms that reduce the consequences of criminal stigma without denying the fact of prior criminal conduct. Record expungement and sealing — which remove criminal records from public access after a qualifying period — directly address the stigma mechanism by eliminating or limiting the availability of the discrediting information to potential employers, landlords, and others whose discrimination produces the stigma’s consequences. Research by Prescott and Starr documenting that expungement recipients show substantially higher earnings and lower recidivism establishes the effectiveness of this approach.
Fair chance hiring policies — which prohibit asking about criminal records in early stages of the hiring process (the “ban the box” movement) and which implement anti-discrimination standards for record-based employment screening — address employment stigma without eliminating employers’ ability to consider relevant information at appropriate stages. Research on ban-the-box effects has found mixed results that depend substantially on whether the policy is implemented alongside anti-discrimination enforcement or as a standalone procedural reform: research by Agan and Starr found that ban the box in some contexts increased racial discrimination by causing employers without criminal record information to rely more heavily on racial stereotypes as a proxy for criminal record, establishing that procedural reforms without substantive anti-discrimination enforcement may redistribute rather than reduce stigma.
Research on digital stigma — the specific form of criminal stigma produced by the permanent, searchable character of online information about criminal records and criminal justice contact — has documented that digital environments substantially amplify the consequences of criminal stigma by making criminal record information permanently accessible to any potential employer, landlord, or social contact who conducts an online search. Research on the digital permanence of mugshots — which are frequently posted on websites that charge fees for removal, creating a commercial ecosystem of criminal stigma exploitation — documents a specific digital stigma mechanism that operates independently of formal criminal record systems and that existing legal frameworks have only partially addressed. The emerging legal framework for digital criminal record information — including state legislation restricting mugshot websites, privacy laws that establish rights to removal of outdated criminal record information, and the application of consumer protection law to background check industries — represents the developing policy response to a stigma mechanism whose scale and consequences traditional criminal justice scholarship did not anticipate.
The sociology of stigma contributes to criminal justice reform by establishing that the consequences of criminal labeling — the employment discrimination, housing exclusion, social isolation, and self-concept damage that the criminal stigma literature documents — are not incidental side effects of the formal punishment system but are its most consequential practical outcomes for most individuals who experience criminal justice contact. Reforming the consequences of criminal stigma requires not only the formal record reforms that expungement and fair chance hiring represent, but the deeper cultural shifts in how criminal histories are understood — from permanent moral stains to historical events in lives that continue to develop — that can only be achieved through the sustained cultural work that stigma reduction movements have always required.
Research on the stigma of sex offender registration — the public listing of individuals convicted of sexual offenses that subjects them to community notification, residential restrictions, and permanent internet exposure of their identity and offenses — provides a particularly consequential case study in criminal stigma policy. Research consistently finds that sex offender registries produce substantial collateral consequences — in housing (due to residency restrictions), employment, social relationships, and physical safety (due to vigilante violence enabled by registry information) — while providing limited crime prevention benefits: research finds little evidence that registry requirements reduce sex offense rates or recidivism, and some evidence that the housing instability and social exclusion that registries produce actually increase reoffending risk. This evidence establishes sex offender registration as a stigma mechanism whose crime control rationale is empirically weak and whose collateral harm rationale is empirically strong — a specific instance of the general pattern that excessive criminal stigma produces the crime it claims to prevent.
Research on the comparative sociology of criminal stigma across nations — documenting that countries with lower criminal stigma (shorter record retention periods, more accessible expungement, less public record access) show lower recidivism rates and better reintegration outcomes than countries with more extensive permanent criminal record systems — provides the cross-national evidence that stigma reduction is crime reduction. Countries including the United Kingdom, Germany, and France with “rehabilitation periods” after which criminal records are spent and generally unavailable to employers show better employment outcomes for formerly convicted individuals and lower recidivism rates that the comparative evidence suggests are causally connected to the reduced stigma those systems produce.
Research on stigma consciousness — the degree to which individuals are aware of the stigma attached to their group membership and anticipate discrimination in their social interactions — has found that high stigma consciousness produces avoidance behaviors that reduce exposure to discrimination at the cost of reducing access to legitimate opportunities. Research by Link and colleagues on the modified labeling theory of mental illness documented that individuals who anticipated more stigma following psychiatric treatment showed worse employment and social outcomes than those who anticipated less — establishing that the anticipation of stigma, not only its actual experience, produces the social withdrawal and opportunity avoidance that compound the direct consequences of stigmatization. Applied to criminal stigma, this finding suggests that individuals with criminal records who strongly anticipate employment and social discrimination may withdraw from the labor market and social participation even when specific discrimination has not yet occurred, amplifying the consequences of criminal stigma beyond the instances of actual discriminatory behavior that audit studies document.
Research on the role of peer networks in criminal stigma management has found that individuals embedded in networks where criminal records are common — which are precisely the networks that concentrated incarceration creates in disadvantaged communities — face different stigma management challenges than those embedded in networks where criminal records are rare. When criminal records are common in one’s social network, the stigma of a single record is diluted by the normalization of criminal justice contact, reducing the personal stigma consequences while maintaining the community-level stigma consequences of living in a community associated with high incarceration rates. This network normalization of criminal stigma has complex implications: it may reduce the psychological burden of stigma for individuals with records while reducing the deterrent effect of criminal justice contact on future behavior, suggesting that the specific ecology of criminal stigma — its distribution across communities with different levels of criminal justice contact — shapes its consequences in ways that individual-level analysis cannot fully capture.
Research on reintegrative shaming theory — Braithwaite’s distinction between stigmatizing shaming (which marks the person as deviant and excludes them) and reintegrative shaming (which condemns the act while affirming the person’s social membership) — provides the most systematic theoretical framework for understanding how social reactions shape criminal trajectories through stigma. Research testing reintegrative shaming theory across diverse crime types and cultural contexts has found consistent evidence that stigmatizing reactions produce worse recidivism outcomes than reintegrative ones, establishing that the specific quality of the social reaction to deviance — not merely its presence — shapes subsequent behavior through the mechanisms of identity formation and social bond maintenance that the theory specifies.
Research on collateral consequence reform — the growing state and federal legislative movement to catalog, limit, and ultimately eliminate unnecessary barriers to reintegration that criminal convictions impose — provides the most direct policy application of the stigma and crime research, translating the sociology of stigma’s findings about how stigma perpetuates criminal careers into specific legislative proposals that reduce stigma through institutional reform. Research documenting that states with fewer collateral consequences show lower recidivism and better employment outcomes for formerly convicted individuals provides the empirical rationale for continuing this reform trajectory, even as the pace of change remains far below what the evidence supports.
Conclusion
The sociology of stigma and crime has produced empirical findings that are simultaneously among the most consequential and the most policy-resistant in criminological research. The employment, housing, and social relationship consequences of criminal stigma are extraordinarily well-documented; the mechanisms through which stigma sustains criminal careers are theoretically coherent and empirically confirmed; and the policy tools for stigma reduction — expungement, fair chance hiring, record reform, and reduced collateral consequences — are available and evidence-supported. Yet the pace of reform remains far below what the evidence supports, because the political economy of criminal stigma — the employer interests in screening tools, the public interest in information access, and the punitive cultural values that treat permanent stigma as deserved punishment — creates resistance to evidence-based reform that the research evidence alone cannot overcome.
Stigma and crime are connected through bidirectional mechanisms that criminology must engage simultaneously to achieve adequate understanding of either phenomenon. Criminal justice contact produces stigma that reduces legitimate opportunity, elevates recidivism risk, and sustains criminal careers in ways that contradict criminal justice’s formal crime reduction purpose. And stigma — racial stigma, mental illness stigma, poverty stigma — elevates criminal justice contact for stigmatized groups in ways that reproduce the racial and class inequalities of the broader society through the specific institutional mechanisms of the criminal justice apparatus. A criminology adequate to this bidirectionality must study both the causes of crime and the causes of criminal labeling — treating the social reaction to crime as equally important as the behavior that produces it, and treating the consequences of that reaction as equally important for understanding criminal career trajectories as the individual characteristics that conventional criminology has prioritized. The sociology of stigma has established this bidirectionality empirically; translating its policy implications into the institutional reforms that reduce stigma’s crime-perpetuating consequences is the ongoing challenge that requires both the research evidence that criminology continues to produce and the political conditions that make evidence-based reform possible. The integration of the sociology of stigma with the developmental criminology of criminal careers — examining how stigma operates differently across the life course, how early labeling shapes developmental trajectories differently from late-career stigma, and how the timing of stigma reduction interventions affects their effectiveness — represents an important research frontier. Until the broader project of stigma reduction — in employment, housing, civic participation, and social relationship — is advanced with the urgency the evidence demands, the criminal justice system will continue to produce the criminal careers it ostensibly exists to prevent, through the stigmatization mechanisms that the sociology of crime has documented across more than half a century of rigorous empirical research. The scale and urgency of the criminal stigma problem — affecting approximately one-third of adult Americans with criminal records whose life chances are constrained by stigmatizing barriers that research consistently shows serve no crime prevention purpose beyond the initial sanction — establishes that stigma reform is not a marginal policy refinement but a central component of any serious crime prevention strategy. Building the institutional and political will for that reform is the challenge that the sociology of stigma’s research most urgently motivates — and whose successful pursuit would reduce crime more effectively than the enforcement investments that have historically dominated American criminal justice spending.
References
- Agan, A., & Starr, S. (2018). Ban the box, criminal records, and racial discrimination: A field experiment. Quarterly Journal of Economics, 133(1), 191–235. https://doi.org/10.1093/qje/qjx028
- Braithwaite, J. (1989). Crime, shame and reintegration. Cambridge University Press.
- Bushway, S. D., & Sweeten, G. (2007). Abolish the juvenile court? Criminology & Public Policy, 6(4), 727–750. https://doi.org/10.1111/j.1745-9133.2007.00459.x
- Clair, M., Daniel, C., & Lamont, M. (2016). Destigmatization and health: Cultural constructions and the long-term reduction of stigma. Social Science & Medicine, 165, 223–232. https://doi.org/10.1016/j.socscimed.2016.03.021
- Eberhardt, J. L., Davies, P. G., Purdie-Vaughns, V. J., & Johnson, S. L. (2006). Looking deathworthy: Perceived stereotypicality of Black defendants predicts capital-sentencing outcomes. Psychological Science, 17(5), 383–386. https://doi.org/10.1111/j.1467-9280.2006.01716.x
- Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Prentice-Hall.
- Harding, D. J., Morenoff, J. D., Nguyen, A. P., & Bushway, S. D. (2017). Short and long-term effects of imprisonment on future felony convictions and prison admissions. Proceedings of the National Academy of Sciences, 114(42), 11103–11108. https://doi.org/10.1073/pnas.1701544114
- LeBel, T. P. (2012). Invisible stripes? Formerly incarcerated persons’ perceptions of stigma. Deviant Behavior, 33(2), 89–107. https://doi.org/10.1080/01639625.2010.538365
- Lemert, E. M. (1951). Social pathology. McGraw-Hill.
- Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363–385. https://doi.org/10.1146/annurev.soc.27.1.363
- Maruna, S. (2001). Making good: How ex-convicts reform and rebuild their lives. American Psychological Association.
- Pager, D. (2003). The mark of a criminal record. American Journal of Sociology, 108(5), 937–975. https://doi.org/10.1086/374403
- Pager, D. (2007). Marked: Race, crime, and finding work in an era of mass incarceration. University of Chicago Press.
- Pager, D., Western, B., & Bonikowski, B. (2009). Discrimination in a low-wage labor market: A field experiment. American Sociological Review, 74(5), 777–799. https://doi.org/10.1177/000312240907400505
- Prescott, J. J., & Starr, S. B. (2020). Expungement of criminal convictions: An empirical study. Harvard Law Review, 133(8), 2460–2555.
- Richie, B. E. (2012). Arrested justice: Black women, violence, and America’s prison nation. New York University Press.
- Schnittker, J., & John, A. (2007). Enduring stigma: The long-term effects of incarceration on health. Journal of Health and Social Behavior, 48(2), 115–130. https://doi.org/10.1177/002214650704800202
- Shannon, S. K. S., Uggen, C., Schnittker, J., Thompson, M., Wakefield, S., & Massoglia, M. (2017). The growth, scope, and spatial distribution of people with felony records in the United States, 1948–2010. Demography, 54(5), 1795–1818. https://doi.org/10.1007/s13524-017-0611-1
- Taxman, F. S., Perdoni, M. L., & Harrison, L. D. (2007). Drug treatment services for adult offenders: The state of the state. Journal of Substance Abuse Treatment, 32(3), 239–254. https://doi.org/10.1016/j.jsat.2006.12.019
- Tyler, T. R., & Wakslak, C. J. (2004). Profiling and police legitimacy: Procedural justice, attributions of motive, and acceptance of police authority. Criminology, 42(2), 253–282. https://doi.org/10.1111/j.1745-9125.2004.tb00520.x
- Uggen, C., Larson, R., Shannon, S., & Stewart, R. (2022). Locked out 2022: Estimates of people denied voting rights due to a felony conviction. The Sentencing Project.
- Western, B. (2018). Homeward: Life in the year after prison. Russell Sage Foundation.