___________________

¹ The risk paradigm is prominent among scholars and practitioners in the correctional field, where risk instruments are commonly used to match returning individuals to particular supervision conditions or to particular treatment services, based on their predicted likelihood of recidivism. This classification process is known as risk assessment (or risk/needs assessment). One popular and well-validated risk assessment tool is the trademarked Level of Service Inventory-Revised and Ohio Risk Assessment System.

as reoffending, rearrest, reconviction, or return to prison; and whether it includes revocations due to violation of technical conditions.

The focus in this chapter is also heavily tilted toward research on return to the community following prison incarceration, which may or may not coincide with a period of parole supervision, although research on criminal offending while on probation is included when it is relevant to the discussion. Finally, much of the extant research cited below combines study designs that are correlational as well as experimental or quasi-experimental, which limits our ability to make strong empirical generalizations about the causal status of the theories and the reported findings. More generally, the evidence for the correlates of recidivism and many “evidence-based” reentry/rehabilitation interventions is based on correlational or quasi-experimental research. To more firmly establish the causal linkages specified by theories linking human well-being and criminal behavior, stronger designs, including randomized controlled trials (RCTs), are needed. And even where experimental evidence is available, as in the case of prison work and vocational programs, such generalizations often depend on the choice of outcomes, treatment heterogeneity and treatment effect heterogeneity, program “stacking” or participation in multiple programs, and ambiguity in defining programs of different types (Nur and Nguyen, 2022).

Personal Risk Factors

Among the so-called “static” risk factors—static because they are either not subject to change or are not amenable to intervention—age and criminal history are among the most salient correlates of post-release criminal behavior (Andrews et al., 1990; Gendreau et al., 1996). Age has a robust correlation with criminal offending in general, so much so that some scholars claim it cannot (and should not) be explained theoretically because it is due merely to the “inexorable aging of the organism” (Gottfredson and Hirschi, 1990, p. 141; see also Hirschi and Gottfredson, 1983). Although many theories challenge this claim and offer their own explanation of the fact that crime declines with age (following late adolescence), there is no dispute about the existence of the age-crime correlation, and post-release offending is no exception.

Past behavior is also a reliable predictor of future behavior, and this continuity has been the subject of research seeking to untangle the degree to which it represents “population heterogeneity,” which emphasizes relatively stable differences across people, “state dependence,” which emphasizes change in response to life events and experiences, or some combination of the two (Nagin and Paternoster, 1991; 2000). Irrespective of the mechanism, this implies that individuals with more extensive prior criminal

records have higher risk of post-release offending.² However, there are two dimensions to criminal history that are not easily distinguished: the number and type of prior offenses at the time of sentencing, and the nature of the commitment offense.

Most studies of returning individuals include both age and criminal history as predictors of post-release criminal offending, and they confirm that these static risk factors reliably predict outcomes. Namely, post-release offending declines monotonically (always decreasing or remaining constant, and never increasing) with age at release, and lengthier criminal history is positively correlated with repeat offending.^3,4

There are also important “dynamic” risk factors—so-called because they are mutable and frequently the target of intervention efforts—or what are sometimes described as “criminogenic needs” (Andrews et al., 1990; Gendreau et al., 1996). Two examples of dynamic risk factors include substance use or abuse and serious mental illness, both of which have been the subject of studies of post-release criminal offending.⁵ Substance use is a major correlate of criminal offending in general (see Tonry and Wilson, 1990, and

___________________

² A complication is that criminal history is also correlated with sentencing—individuals with more extensive or more serious prior records are sentenced to longer lengths of confinement, other things equal. As a result, it is a persistent challenge to untangle the nature of the relationship between criminal history and post-release criminal behavior (see Hester, 2019). There is an additional challenge presented by the length of time that has elapsed within a risk window, as the predictive ability of criminal history appears to degrade with time at risk (Kurlychek, Brame, and Bushway, 2006; 2007).

³ Studies that find no correlation of age with post-release crime tend to measure the former from age at arrest or age at commitment, or else include measures of in-prison behavior (e.g., misconduct) or post-prison statuses (e.g., supervision level) which arguably mediate the relationship between age and repeat offending (Benedict and Huff-Corzine, 1997; Huebner Varano, and Bynum, 2007). On the other hand, studies that measure age of onset (e.g., age at first contact with the criminal legal system) find that age is inversely correlated with post-release criminal offending (Bellair and Kowalski, 2011). However, many of these same studies indicate that when criminal history is measured by the commitment offense, it is frequently uncorrelated with post-release criminal offending, and in fact, some studies find that commitments for violent crime correspond with lower rates of post-release offending than commitments for other types of offenses (see Bales and Mears, 2008). Despite these complicating factors, age and criminal history remain reliable predictors of repeat offending behavior.

⁴ For references see Bales and Mears, 2008; Bellair and Kowalski, 2011; Berg and Huebner, 2011; Berk and Rauma, 1983; Boman and Mowen, 2017; Chamberlain and Wallace, 2016; Cobbina, Huebner, and Berg, 2012; Hester, 2019; Hoffman and Beck, 1984; Huebner and Berg, 2011; Huebner and Cobbina, 2007; Huebner, DeJong, and Cobbina, 2010; Huebner and Pleggenkuhle, 2015; Kubrin and Stewart, 2006; Listwan et al., 2013; Liu and Visher, 2021; Liu et al., 2020; MacKenzie and Li, 2002; Miller, Caplan, and Ostermann, 2016; Mowen and Visher, 2015; Olson and Lurigio, 2000; Zweig et al., 2015.

⁵ Other dynamic factors in the criminogenic needs model relate to peer and family relationships, employment, leisure and recreational activities, and antisocial cognition (Andrews and Bonta, 2010).

the authors therein).⁶ A large share of individuals in prison and jail report being under the influence at the time they committed the offense for which they were incarcerated (Maruschak, Bronson, and Alper, 2021a; Mumola, 1999; Wilson, 2000), and roughly 60 percent of individuals sentenced to prison and jail meet the diagnostic criteria for drug dependence or abuse (Bronson et al., 2017). Studies of released individuals confirm those with a drug use or drug abuse history have a higher likelihood of post-release offending (Benedict and Huff-Corzine, 1997; Boman and Mowen, 2017; Huebner and Cobbina, 2007; Olson and Lurigio, 2000; Zweig et al., 2015), and those identified as drug dependent at the time of release from prison are more likely to be involved in post-release offending (Berk and Rauma, 1983; Huebner and Berg, 2011; Huebner, DeJong, and Cobbina, 2010; Huebner and Pleggenkuhle, 2015; for evidence of a null relationship, see Berg and Huebner, 2011). When investigators are able to measure it, post-release drug use is also positively correlated with post-release offending (Griffin and Armstrong, 2003; Li et al., 2000; Link and Hamilton, 2017; MacKenzie and Li, 2002).

Serious mental illness is another commonly studied dynamic risk factor.⁷ A substantial share of confined individuals meet the criteria for serious mental illness, defined as a mental, behavioral, or emotional disorder that impairs functioning and fits established diagnostic criteria (Bronson et al., 2017; James and Glaze, 2006; Steadman et al., 2009).⁸ This has been linked with a higher risk of post-release criminal behavior (Berg and Huebner, 2011; Cloyes et al., 2010; Listwan et al., 2013; Wallace and Wang, 2020).

In summary, existing research finds that both static and dynamic risk factors are reliable correlates of post-release criminal behavior.

Confinement Experiences

The experience of incarceration itself provides the basis for another set of theories explaining the continuation of criminal offending post-release. Theories of deterrence speak directly to incarceration, emphasizing how

___________________

⁶Goldstein (1985) describes the drugs-crime relationship as stemming from psychopharmacological effects (e.g., intoxication or withdrawal), economic-compulsive effects (e.g., crime to support a drug habit), and systemic effects (e.g., drug market violence).

⁷ Mental illness has been viewed as both a dynamic risk factor and a “responsivity factor” (a response to) incarceration (Andrews and Bonta, 2010).

⁸Bronson and colleagues (2017) estimate that 58 percent of people in state prisons and 63 percent of those serving sentences in local jails meet the diagnostic criteria for drug dependence or abuse. Earlier, James and Glaze (2006) estimated that 56 percent of people in state prisons, 45 percent of those in federal prisons, and 64 percent of those in local jails have had a mental health problem (the corresponding figures for those with a recent history of such problems are 24 percent in state prisons, 14 percent in federal prisons, and 21 percent in local jails). With regard to jail populations, Steadman et al. (2009) estimate the prevalence of current serious mental illness at 14.5 percent for males and 31.0 percent for females.

being sentenced to prison affects the likelihood of post-release criminal behavior (Beccaria, 1764; Bentham, 1789). Known as specific deterrence, a longer length of stay is predicted to lower the expected utility of continued offending (Becker, 1968). This arises because the direct experience of prison contributes to experiential learning about the unpleasant consequences of crime, or to updating of the perceived risk of sanction in the event of a return to crime (Stafford and Warr, 1993).

The assumption that longer prison stays discourage future criminal offending is the basis for a great deal of carceral policy. Yet research on this relationship has produced highly variable findings for reasons that are not yet well understood. There is evidence that longer length of stay is not correlated with recidivism (Walker and Bishop, 2016), is weakly correlated with less recidivism (Cotter, 2020; Estelle and Phillips, 2018; Loughran et al., 2009; Meade et al., 2013; Rhodes et al., 2018), or is weakly correlated with more recidivism (Aizer and Doyle, 2015; Eren and Mocan, 2021; Green and Winik, 2010; Tiedt and Sabol, 2015). In still other studies longer sentences are correlated with both more and less recidivism at different points in the distribution of length of stay or for individuals with different commitment offenses (Mears et al., 2016; Rydberg and Clark, 2016).

It is thus not possible to say with confidence that longer length of stay has a deterrent effect, as longer sentences might actually worsen post-release criminal offending. These findings run contrary to expectations from the specific deterrence doctrine (for reviews of research on length of stay, see Loeffler and Nagin, 2021; Nagin, Cullen, and Jonson, 2009).⁹ A quasi-experimental study found that when Georgia eliminated parole for some incarcerated people, they accumulated more disciplinary infractions, completed fewer prison rehabilitative programs, and returned to prison at higher rates than those whose sentences were unaffected by the reform (Kuziemko, 2013). These results suggest some degree of responsiveness to the incentives of early release.

An opposing theoretical narrative drawn from the labelling tradition suggests that confinement experiences can result in increased likelihood of post-release criminal offending. Theories of secondary deviance seek in part to explain why people who experience prison or longer length of stay might be more rather than less likely to continue offending. According to this view, some post-release criminal offending results from societal reactions to the incarcerated individual. The act of the criminal legal system

___________________

⁹ Evidence related to the confinement experience is also available from studies comparing the use of custodial sanctions versus probation or some other form of diversion on post-release criminal offending (for informative reviews, see Nagin, Cullen, and Jonson, 2009; Petrich et al., 2021). These studies differ from those described in this paragraph because they consider the impact of incarceration on repeat offending along the “extensive” margin (incarceration vs. diversion) as opposed to the “intensive” margin (i.e., longer length of stay among those who are incarcerated).

officially labeling someone as an “offender” and the formal exclusionary processes and stigmas associated with this label can amplify criminal offending (Becker, 1963; Garfinkel, 1956; Goffman, 1963; Lemert, 1951, 1972; Schur, 1971). These labeling processes can contribute to institutional exclusion, identity transformation, subculture formation, or outright defiance by labeled individuals (Braithwaite, 1989; Paternoster and Iovanni, 1989; Sampson and Laub, 1997; Sherman, 1993).

Studies of confinement experiences are also informed by social learning principles (see Bandura, 1977), which often emphasize some degree of prison enculturation that can harden criminal offending after release. Hagan (1993) refers to this as the development of “criminal capital” that might not be easily shed by returning individuals.¹⁰ Research on the prison environment supports the possibility that prisons can be “schools of crime.” For example, there is some suggestion that individuals who have committed the same type of criminal offense experience reinforcing peer effects, thus increasing the chance of post-release criminal offending with that offense type after release (Bayer, Hjalmarsson, and Pozen, 2009; for null peer effects, see Harris et al., 2018). Research on security level also finds that assignment to a more secure prison environment is correlated with more post-release offending (Chen and Shapiro, 2007; Gaes and Camp, 2009), with some indication the effect is at least partly due to peer influences.¹¹ Scholars have only recently begun to probe the network structure of incarcerated individuals, both inside and outside of prison, and to explore the consequences of these networks for in-prison behavior as well as post-release behavior (see Kreager et al., 2016; 2017; Schaefer et al., 2017).

Societal Attachments and Reentry Stressors

Social institutions such as the family, school, and workplace attract a great deal of criminological interest, and studies of post-release criminal offending are no exception. Theories tend to view social ties as sources of informal social control, because they constitute an enduring social bond (Hirschi, 1969), a fount of social capital (Sampson and Laub, 1993), a source of social support (Cullen, 1994), or a set of “local life circumstances” (Horney, Osgood, and Marshall, 1995) that would be jeopardized

___________________

¹⁰ Such explanatory mechanisms include differential associations with patterns and definitions favorable to criminal behavior (Sutherland, 1947), coupled with imitation and differential reinforcements (Akers, 1985; 1998; Burgess and Akers, 1966).

¹¹ There is a positive correlation between gang affiliation and post-release offending (Dooley, Seals, and Skarbek, 2014; Huebner, Varano, and Bynum, 2007; McShane et al., 2003; Pyrooz et al., 2021), although it has been difficult in this literature to isolate the unique influence of prison gang exposure from community gang exposure.

by a return to criminal behavior. Conversely, returning individuals for whom social ties have eroded face a higher likelihood of returning to crime.

Individuals who have been released from prison are uniquely vulnerable to strained relationships, financial hardship, homelessness, and even victimization, particularly during the early period of their return to the community (Binswanger et al., 2007; Comfort et al., 2018; Miller, 2021). The period following release can be thought of as having the potential for crisis (e.g., Western, 2018), especially if individuals return with little more than what they brought with them to prison. Some strain is more or less chronic and a byproduct of the disproportionately lower-class status of incarcerated individuals that reduces access to legitimate opportunities for upward mobility (Cohen, 1955; Cloward and Ohlin, 1960; Merton, 1938). The more acute features of strain can motivate post-release criminal offending by inducing negative emotional states, including anger, frustration, and feelings of injustice, consistent with expectations from general strain theory (Agnew, 1992; 2001; 2006). Individuals who experience these strains are theorized to be vulnerable to using crime as a coping mechanism, especially when they lack the resources and supports for more conventional coping.

Ecological Influences

Ecological frameworks for explaining post-release criminal behavior focus attention on characteristics of communities to which individuals return. Individuals reentering the community from prison concentrate in neighborhoods characterized by higher-than-average levels of economic disadvantage, residential instability, and racial heterogeneity.¹² And just as crime rates are correlated with these indicators of local ecology, criminal offending of formerly incarcerated individuals tends to correlate with neighborhood context. In particular, with exceptions, prior studies document a positive correlation between concentrated disadvantage and post-release offending in a variety of states (Hipp, Petersilia, and Turner, 2010; Huebner and Pleggenkuhle, 2015; Kirk, 2015; Kubrin and Stewart, 2006; McNeeley, 2018; Mears et al., 2008; for exceptions, see Chamberlain and Wallace, 2016; Miller, Caplan, and Ostermann, 2016; Tillyer and Vose, 2011; Wehrman, 2010).¹³

___________________

¹² This pattern has been described as concentrated reentry or mass reentry (Chamberlain and Wallace, 2016), the latter of which unites it with scholarship on mass incarceration (Garland, 2001) and mass probation (Phelps, 2017). Concentrated reentry has a more explicit geographical emphasis, similar to concentrated disadvantage.

¹³ Some evidence suggests the correlation between concentrated disadvantage and post-release offending is strongest for formerly incarcerated African Americans (Mears et al., 2008). There are also unexpected findings from some jurisdictions that concentrated disadvantage is inversely correlated with post-release offending for at least certain groups of returning individuals (Huebner and Berg, 2011; Huebner, DeJong, and Cobbina, 2010; Reisig et al., 2007).

___________________

¹⁴Bellair and Kowalski (2011) found individuals released on parole to areas with more abundant jobs in willing industries have lower likelihood of post-release criminal offending in Ohio. Similar results were found in California (Schnepel, 2018) and a multistate study (Yang, 2017).

true of the enhancement of technical conditions of probation or parole supervision through the use of control- and sanction-oriented technologies (e.g., house arrest, electronic monitoring, curfews) (see MacKenzie, 2006; Sherman et al., 1998; Taxman, 2002; Taxman, Smith, and Rudes, 2020).

The training of community supervision officers also seems to be correlated with post-release reoffending (Andrews and Bonta, 2010; Chadwick, Dewolf, and Serin, 2015; Dowden and Andrews, 2004). The use of incentives to reward positive post-release behavior, as opposed to the use of sanctions to punish negative post-release behavior, also has a growing evidence base in its favor (Mowen et al., 2018; Sloas, Murphy, and Taxman, 2019; Wodahl et al., 2011). Moreover, when individuals on supervision feel they can trust their officer and that the officer is fair, their outcomes include less post-release offending and fewer technical violations (Skeem et al., 2007; Taxman and Ainsworth, 2009; Taxman and Thanner, 2004).

MODELS OF REENTRY SUCCESS

The historical emphasis on recidivism among policy analysts, practitioners, and scholars reflects, in part, a desire by researchers and institutions to establish a common “success rate” indicator. But it is quickly apparent that success and failure are relative concepts—and that crude dichotomies fail to capture the real changes that people returning from incarceration experience. Recidivism is therefore limited as a performance measure or metric for success (King and Elderbroom, 2014). Today, many analysts are calling for “a paradigmatic shift” in criminal justice practices that would better align with contemporary theories of desistance from crime (Bersani and Doherty, 2018; Bushway and Uggen, 2021). For our purposes, we can think of these models in terms of two broad conceptions of “success”: (1) desistance from crime and (2) social integration and well-being.

Desistance from Crime

Desistance refers to why and how people stop committing crime. The key distinction between recidivism and desistance approaches is that the former focuses on a negative outcome (i.e., crime at a discrete point in time), whereas the latter seeks to track positive outcomes that may result in reduced involvement in offending over time, ultimately leading to the complete cessation of criminal behavior. Early models of desistance focused on the relationship between age and crime and the natural process of aging or maturation (Glueck and Glueck, 1940). In recent decades, theories have emerged that explain desistance as the product of social and developmental processes (Bersani and Doherty, 2018; Maruna, 2001; Sampson and Laub, 1993; Uggen and Piliavin, 1998; Weaver, 2019). These models

conceptualize desistance as a process marked by a decline in the rate or severity of offending that is closely linked to other life-course processes, such as work and family transitions (Bushway et al., 2001; Laub and Sampson, 2001). The study of desistance represents an important refinement, as it explicitly recognizes that changes in criminal behavior rarely map neatly onto a recidivist versus non-recidivist dichotomy.

Some desistance researchers explain these shifts in crime over time as a consequence of the development of adult social bonds and informal social controls (e.g., Sampson and Laub, 1993). A long line of research establishes associations between adult employment and recidivism, although the evidence is mixed regarding causality (Uggen, 2000; Visher, Winterfield, and Coggeshall, 2005). Similarly, strong family ties are consistently associated with both employment and reduced recidivism (Berg and Huebner, 2011), although some studies suggest that such effects may depend on the criminal history of the partner (Andersen, Andersen, and Skov, 2015).

In contrast to social control–based models, identity-based models of desistance emphasize the social-psychological processes that link these adult role behaviors to changes in self-concept, identity, and behavior (Giordano, Cernkovich, and Rudolph, 2002; Maruna, 2001; Matsueda and Heimer, 1997; Paternoster and Bushway, 2009). By this logic, desistance is partly a process of “de-labeling” (Maruna, 2001) or nullifying the effects of a criminal label. For Giordano, Cernkovich, and Rudolph (2002), employment and marriage are less predictive of desistance than an individual’s cognitive shifts and development of a “replacement self” (e.g., as the good wife or involved mother). Qualitative research suggests that the process of desistance from crime is a “fragile project,” frequently subject to derailment, setbacks, and recovery (Halsey, Armstrong, and Wright, 2017).

In both control-based and identity-based models, however, desistance theory and research typically link particular roles and statuses of life course development—such as education, employment, housing, family transitions, and civic participation—to subsequent criminal behavior. Shifts in activity patterns, social networks, and underlying identities appear to be key mechanisms in both desistance from crime and recovery from substance dependence (Best and Savic, 2014). Incarceration represents another important life experience that likely shapes desistance patterns (Maruna and Toch, 2005). A recent review of the effects of incarceration on subsequent conviction and reincarceration observed reductions in settings with rehabilitative programming and criminogenic effects in settings without such programming (Loeffler and Nagin, 2021).

In addition to concrete success markers such as employment and housing, formerly incarcerated people who participated in the committee’s information-gathering sessions emphasized social-psychological processes. These include the importance of self-efficacy, feelings of worthiness, healing

from trauma, self-compassion, confidence, and a sense of personal responsibility in the process of successful reintegration into society. Several participants discussed identity transformations that took place in prison or afterwards, often facilitated by participation in therapeutic, educational, and vocational programs. At the same time, they shared accounts of obstacles they encountered during reentry, such as the denial of jobs and housing because they had to “check the box” and other experiences of discrimination related to their criminal record, gender or racial identity, social class, or other characteristics. Regardless of their intrinsic feelings of self-worth, formerly incarcerated persons are burdened by stigma attached to them by society. These experiences are consistent with the conclusions of some scholars that feelings of self-efficacy and the acceptance of a prosocial identity are necessary precursors to successful desistance; however, they may not be sufficient (LeBel et al., 2008).

Social Reintegration and Well-Being

Although facilitating desistance from crime is a key goal of prison reentry programs, people with criminal records are, of course, human beings who are more than the sum of their recidivism risks. It is the view of this committee, particularly informed by those committee members who have been incarcerated, that a myopic focus on recidivism or desistance as the sole indicator of post-release success is problematic. Social integration and reintegration constitute success markers in their own right, irrespective of their effect on crime and desistance (Harding, Morenoff, and Wyse, 2019; Western, 2018).

This conception of success emphasizes flourishing and well-being, often in work, family, and civic roles that benefit families and communities as well as the reentering individual. By this logic, programs that improve post-release education and employability, family functioning, or civic participation may be considered successful and socially beneficial. Research in this tradition often emphasizes economic costs and benefits, as well as reintegration across varied social domains, such as socioeconomic reintegration, familial reintegration, and civic reintegration (Drake et al., 2009; Uggen, Manza, and Behrens, 2004).

In contrast to a recidivism framework organized around a risk paradigm, the committee views post-release success through the lens of flourishing and well-being. Such an approach is consonant with recent scholarship on the social determinants of health, which emphasizes “the environments in which people are born, grow, work, live, learn, play, worship, and age” that affect health and quality of life (U.S. Department of Health and Human Services, 2021). Hastening desistance from crime is an important aspect of post-release success, but social reintegration is a success marker in its own right. This conception of success emphasizes flourishing and healthy adult

development, often in work, family, and civic roles that directly benefit families and communities as well as the reentering individual.

Of course, the experience of criminal punishment and resulting “carceral citizenship” (Lerman and Weaver, 2014; Miller and Stuart, 2017) can directly undermine social reintegration (Brayne, 2014) and well-being. That is, incarceration, monetary sanctions, and other civil disabilities actually impede successful reintegration. Formerly incarcerated people and those with felony-level criminal records face severe restrictions on their work, family, and civic participation due in part to the “invisible punishment” (Travis, 2002) of collateral sanctions. Imposing sanctions such as welfare restrictions or disenfranchisement does little to support victims or improve public safety (see Box 3-1 for further discussion of collateral sanctions).

clearly impact public safety and the well-being of victims. Nevertheless, successful reintegration does not stop at public safety but also benefits communities by contributing to their economic vitality and family stability and by strengthening civil society. Criminal legal institutions, reentry programs, and other agencies that serve individuals returning from prison can thus be evaluated, in part, based on their success in improving participant and victim outcomes in a set of domains central to overall well-being.

The following section reviews existing research on key domains of support and strain in the reintegration process.¹⁵ Effective programming and services support successful reentry in each of these domains by implementing validated assessments, effective cognitive-behavioral therapy and treatment, individually tailored case plans, behavioral incentives, and graduated sanctions. Success in domains such as education and employment is important in its own right, but may also support the cessation of criminal activity. Where that is the case, we note this relationship. However, a domain need not promote desistance from crime in order to be central to successful reintegration. Box 3-2 describes perspectives from correctional leaders on defining successful reentry and the key domains of the reintegration process.

Housing

One significant challenge facing individuals exiting prison is finding stable, affordable housing (Lattimore and Visher, 2021; Miller, 2021; Visher and La Vigne, 2021), which is particularly difficult for people convicted of sex crimes (Dum, 2016). Locating a place to live is one of the immediate concerns individuals exiting prison experience, yet it is often permeated with obstacles. Formerly incarcerated people are 10 times more likely than the general public to be homeless, with homelessness rates higher for women, African Americans, individuals over age 45, individuals who have been incarcerated multiple times, and people recently released from prison (Couloute, 2018). Individuals with sex offense convictions, substance abuse problems, and mental illness are especially susceptible to homelessness (Metraux, Hunt, and Yetvin, 2020). Many formerly incarcerated individuals face financial difficulties securing housing, as many are excluded from housing options (Greenberg and Rosenheck, 2008). Both public housing authorities and private landlords can use screening criteria, including criminal record checks, to exclude reentering individuals (Couloute, 2018). Moreover, a sizeable group of these individuals may have never lived on their own and require substantial support in finding suitable housing after release (Hyde et al., 2021).

___________________

¹⁵ A recent study (Love, 2022), which appeared after the completion of this report, grades the 50 states on their effectiveness in reintegrating persons into civil society after arrest or conviction.

___________________

¹⁶Li and MacKenzie (2003) find that employment is inversely correlated with offending among males on probation, but positively correlated with criminal behavior among females.

prior work credentials pose a number of challenges, as do employment-based reentry programs and parole supervision practices that prioritize placement in any job as opposed to placement in jobs that inspire a high level of satisfaction.

Familial and Social Relationships and Support

Because of what they signify about a returning individual’s societal attachments, familial relationships are central to studying and theorizing the reentry process (Uggen, Wakefield, and Western, 2005). Although it is difficult to disentangle the effects of incarceration from the preexisting characteristics of those who are incarcerated, the criminal legal system itself plays a part in disconnecting formerly incarcerated people from family and other social institutions (see reviews in Kirk and Wakefield, 2018; National Research Council, 2014; Wakefield and Uggen, 2010). Adult social bonds such as relationships with family members have been widely documented to be an important factor in successful transitions away from criminal activity and toward community engagement (Blokland and Nieuwbeerta, 2005; Laub and Sampson, 2006; Naser and Visher, 2006; Sampson and Laub, 2003; Warr, 1998). Family offers an important social network for men and women exiting prison, as many live with their family usually for several months (Visher and La Vigne, 2021). Research indicates that after release from prison people expect a high level of support from their family (La Vigne et al., 2004; Nelson, Deess, and Allen, 1999) and attach importance to commitment to family roles, including as parents (Uggen, Manza, and Behrens, 2004; Zamble and Quinsey, 1997). In addition to acting as a core domain for reintegration, social ties can also support desistance. For example, in a study of interviews with previously incarcerated fathers and mothers, interviewees attributed family connections and parent-child contact as key factors in their post-release success (Charles, Meuntner, and Kjellstrand, 2019).

People who are released from prison generally receive some level of emotional, social, and economic support from family (Martinez and Christian, 2009; Miller, 2021; Visher and Courtney, 2006). Such family support is associated with higher rates of employment, reductions in substance use, and fewer physical, mental, and emotional problems (Harding et al., 2014; Naser and LaVigne, 2006; Naser and Visher, 2006). Further research has suggested that the instrumental support provided by family (e.g., housing, employment, transportation) eclipses emotional and other forms of support in lowering the risk of post-release criminal behavior (Mowen et al., 2019; for a contrary finding, see Taylor, 2016). Familial ties with extended relatives, in particular, may affect repeat offending indirectly by facilitating access to employment for those with a history of employment difficulty (Berg and Huebner, 2011).

___________________

¹⁷ This does not include marital status, which is generally uncorrelated with post-release offending (Bellair and Kowalski, 2011; Listwan et al., 2013). Li and MacKenzie (2003) report that living with a spouse is inversely correlated with offending among males on probation, but is positively correlated with offending among females. For evidence of an inverse relationship with post-release offending, see Boman and Mowen (2017).

behavior have also been found to be factors in desistance trajectories (Lattimore, Dawes, and Berrick, 2018).

However, recent qualitative and quantitative research documents a decline in intimate partnerships, co-residence, and relationship happiness during the reentry period (Comfort et al. 2018). Family members also experience hardships during the reentry process from financial strain, increased anxiety, and trouble with familial and peer relationships (Naser and Visher, 2006; Visher and La Vigne, 2021). These hardships can lead to negative emotional states and maladaptive coping (Liu and Visher, 2019), and family conflict is correlated with higher risk of post-release criminal behavior and substance use (Mowen and Visher, 2015) (See Box 3-3).

Physical Health, Mental Health, and Substance Use

Health challenges facing people exiting prison are often overlooked amid the myriad array of other issues they face (Link et al., 2019; Visher and

Mallik-Kane, 2007; Western, 2018). Individuals report serious mental and physical health problems even at relatively young ages (under 40). In one study, almost half of men reported having a chronic physical health condition; the most commonly reported conditions were asthma, hepatitis, and high blood pressure (Visher and Mallik-Kane, 2007). Research based on the National Comorbidity Survey finds that incarceration is associated with the onset of major depressive disorder, bipolar disorder, and dysthymia (Schnittker, Massoglia, and Uggen, 2012). Among people with a history of incarceration, this study estimates the lifetime prevalence of major depression at 19.8 percent and the lifetime prevalence of post-traumatic stress disorder at 10.8 percent.

In general, incarcerated and formerly incarcerated individuals have an elevated risk of experiencing a diverse set of chronic health conditions compared with the general population (Massoglia and Pridemore, 2015). Consequently, elevated risk of poor health conditions increases mortality both immediately after and years after release from prison (Wang, Pletcher, and Lin, 2009). In general, compared with individuals who have never been imprisoned, incarceration is associated with worse health for formerly incarcerated persons (Massoglia and Pridemore, 2013).

Several studies have also explored mortality patterns following release from prison. One Washington State study found a mortality rate for formerly incarcerated individuals that was 3.5 times higher than for other residents of the state (Binswanger et al., 2007). This excess mortality was not distributed evenly over time, as the risk of death within two weeks of release was 13 times higher than for the general population. Other studies find significant variation in the risk depending on the amount of time the individual served in prison (Patterson, 2013).

In addition, the life domains discussed earlier negatively influence health. Unemployment, poverty, residential instability, and reduced social support are correlated with poor health outcomes (Massoglia and Remster, 2019). While formerly incarcerated people often find it challenging to secure employment, the jobs they do obtain are less likely to offer comprehensive healthcare benefits (Western, 2006). As a result, individuals exiting prison are less likely to have adequate means to treat their health conditions.¹⁸

People exiting prison also have extensive substance use histories. According to data from the Bureau of Justice Statistics, nearly half (47%) meet the criteria for drug dependence (Maruschak, Bronson, and Alper, 2021a). Returning individuals often identify drug use as the primary cause of many of their past and current problems including family, relationship, employment, legal, or financial problems (Lattimore, Dawes, and Berrick, 2018; Visher and

___________________

¹⁸ As of December 1, 2021, the Build Back Better Act contained a provision that would allow people who are incarcerated to receive Medicaid benefits 30 days prior to release (Da Silva, 2021). Such a provision has the potential to improve care and the continuity of care at reentry.

La Vigne, 2021). Yet, many individuals report lower rates of substance use after release compared to pre-prison use. A recent systematic review reports that 18 of 31 studies assessing the effects of drug treatment (the primary modalities included cognitive-behavioral therapy and 12-step programming) found reduced recidivism for the treatment group on at least one indicator, though the pattern of results was somewhat inconsistent (Moore et al., 2020).

Here, as elsewhere, the criminal legal system enmeshes coercion and care, imposing punishment while also providing needed services, including health care (Miller, 2021; Phelps and Ruhland, 2021). The specific conditions of carceral care are likely to play an important part in shaping outcomes. Emerging qualitative studies suggest that reentry services and state-sponsored healthcare have been especially important during the COVID-19 pandemic. For example, a participant in a recent Ohio studied remarked, “I’d probably be in the dirt” without the “good insurance” that accompanied his parole (Vuolo, Schneider, and Laplant 2022, p. 12).

Participation in Peer Support and Help-Giving Roles

Another emerging form of important social support after release is the role of formerly incarcerated people who serve as peer mentors. The employment of formerly incarcerated people to provide peer support, assistance, and advocacy in rehabilitation and reentry programs provides benefits not only to the program participants, but also to those providing the peer support. Applying differential association theory to rehabilitation programming, Cressey (1965, p. 50) proposed that people with criminal histories “can be highly effective agents of change and, further, as they act as agents of change they themselves become the targets of change, thus insuring their own rehabilitation.” The ability of formerly incarcerated program personnel to understand the lived experiences and thought patterns of reentering individuals, and to offer options to manage and overcome circumstantial and psychological obstacles, allows them to be especially effective mentors and guides (Cressey, 1965; Riessman, 1965). These ideas have been the basis for many self-help and mutual assistance groups, including Alcoholics Anonymous and other substance treatment programs (Katz, 1981; Zemore, Kaskutas, and Ammon, 2004).

In turn, individuals serving as peer mentors develop and reinforce new, prosocial identities, hone leadership skills, and often benefit socially and financially from this work while acknowledging their past behaviors (Brown, 1991; Riessman, 1965). Help-giving and mentoring roles can strengthen formerly incarcerated individuals’ active coping strategies, self-esteem, life satisfaction, and sense of self-agency (Aresti, Eatough, and Brooks-Gordon, 2010; LeBel, 2007; LeBel, Richie, and Maruna, 2015; Maruna, 2001). While all of these benefits can help formerly incarcerated people pursue

prosocial lifestyles and reintegrate into society (LeBel, Richie, and Maruna, 2015), Sharp and Hope (2001) note that engagement as a “professional ex-” does not necessarily lead to desistance from all drug use or criminal behavior. Furthermore, work as a “professional ex-” does not necessarily fully mitigate the stigma associated with the label of “ex-offender” (Aresti, Eatough, and Brooks-Gordon, 2010).

Despite these caveats, the value of peer support in reentry was consistently reaffirmed by formerly incarcerated individuals and practitioners who participated in the committee’s information-gathering sessions. They emphasized the value of formerly incarcerated individuals’ perspectives in the creation of appropriate and effective programs grounded in mutual respect, trust, and cultural sensitivity (see Box 3-4).

Voting and Civic Engagement

As noted above, approximately 5.2 million U.S. adults remain disenfranchised due to a felony conviction (Uggen et al., 2020). However, a number of states have recently restored the franchise for this population, wholly or

after a certain period of time. Civic engagement involves more than voting, of course, extending to volunteer experiences, coaching and mentoring, participation in service learning experiences, and attendance at rallies and demonstrations. When they are not formally barred from participating, people with a history of incarceration often lack the resources, ability to mobilize, and in some cases the motivation to be civically engaged. Those who are incarcerated in jails often retain the right to vote, though jail administrators do not often facilitate the voting process for people housed in their facilities (Paikowsky, 2019). Moreover, many incarcerated and formerly incarcerated people may be uncertain about their voting eligibility or fear prosecution for illegal voting (Uggen, Manza, and Thompson, 2006). Although more rigorous studies are needed in this area, some research finds that voting and other forms of civic participation by incarcerated people may enhance public safety (Uggen and Manza, 2004) and willingness to cooperate with law enforcement (Shineman, 2018). When people have resources, motivation, and are able to mobilize and exercise the right to participate in their communities, they may be more strongly tied to these communities and less likely to engage in behaviors that would harm them (Bazemore and Stinchcomb, 2004; Fox, 2010; Miller, 2021; Uggen, Manza, and Behrens, 2004; Weaver and Lerman, 2010).

In addition to holding promise as a marker of broader reintegration and well-being, civic engagement may also support cessation of criminal behavior. Researchers find clear negative correlations between voting and recontact with the criminal legal system, with one study finding approximately 16 percent of nonvoters in 1996 were rearrested during the subsequent three years, relative to five percent of 1996 voters (Manza and Uggen, 2006). A 2012 investigation reports that individuals released in states that permanently disenfranchise are approximately ten percent more likely to reoffend than those released in states that restore the right to vote following release (Hamilton-Smith and Vogel, 2012; for similar findings see Hoover, 2021). This relationship holds when analysis accounts for prior criminal history and when self-report crime data is used in place of official arrest records (Larson and Uggen, 2017). Although some of the association between voting and recontact with the criminal legal system is likely due to preexisting differences between voters and nonvoters, the results suggest a link between political participation and desistance from crime. Other studies show similar patterns for probation and parole (Uggen and Inderbitzin, 2010). It may be the case that voting is tapping a desire to participate as a law-abiding stakeholder in one’s community. Practicing citizenship may, in turn, help to reinforce an identity as a law-abiding citizen.

It stands to reason that the more civically engaged people are, when they have resources, motivation, and the ability to mobilize and exercise the right to vote, the less likely they might be to utilize their power to harm the community by engaging in criminal acts. In particular, symbolic

interactionist and social control theories link desistance from crime to age-graded transitions in work and family life. More broadly, writers such as Alexis de Tocqueville and John Stuart Mill have pointed to the potentially educative and constitutive impact of political participation. Voting may thus engender some degree of identification with the polity and its norms and values. Others point to the expressive impact of voting and its potential to mold “virtuous” citizens. As Winkler (1993, p. 331) put it, “Voting is a meaningful participatory act through which individuals create and affirm their membership in the community and thereby transform their identities both as individuals and as part of a greater collectivity.”

Civic engagement may also take the form of participation in criminal legal system reform and advocacy efforts. While there has been no comprehensive review of the considerable policy and legislative efforts that have been orchestrated and championed by formerly incarcerated individuals and their supporters, people with histories of incarceration have been intimately involved in advocating for criminal legal system reform initiatives, such as police body cameras, alternatives to incarceration, expanding access to health care and therapeutic services for people in correctional custody, and eliminating parole revocations for technical parole violations (Felsenthal, 2018; Rafei, 2021; Sonnenberg, 2017).

This work has often been accomplished within grass-roots organizations or through aligning with broader social justice and legal organizations (Goddard, Myers, and Robison, 2015; Jones and Sayegh, 2021; Ziegelheim, 2018). Formerly incarcerated individuals have established political capital, in spite of the collateral consequences associated with a criminal conviction and wide-ranging attempts to relegate them to the margins (Alexander, 2010; Chesney-Lind and Mauer, 2003; Western, 2018). The voices of individuals who have experienced incarceration have made key contributions to prominent legislative and policy efforts in this area (e.g., the First Step Act, Fair Chance employment laws, the restoration of voting rights, and reductions in the use of solitary confinement). These efforts necessitate an expansion of the framework for what is classified as success for people with criminal legal system involvement to include civic engagement of this type.

Education

Educational opportunities upon reentry (and in prison) can offer potential catalysts for personal transformation and desistance. Education in prison can provide needed credentials, as well as space to achieve personal growth, develop new interests, and increase mutual support, prosocial modeling, and positive socialization (Casey et al., 2013; Waller, 2000). The Bard Prison Initiative, for example, has helped provide college degrees to over 500 incarcerated persons in six New York correctional facilities (Fullilove et al., 2020).

Other models, such as the Inside-Out Prison Exchange Program, offer semester-long academic courses that bring college students together with incarcerated individuals to study as peers in a seminar behind prison walls (Pompa, 2013). Although some students may initially enroll in an Inside-Out course for extrinsic reasons, many who graduate from the Inside-Out class are intrinsically motivated and realize that they are enthusiastic about learning (Wright, 2020). Individuals who have participated in Inside-Out have gone on to create programs, pursue post-secondary education in and out of prison, and engage in scholarship to improve the system and enhance the lives of people involved in the criminal legal system by informing policy makers (Wright, 2020). Recent work by Pelletier and Evans (2019) found that participants in higher education programs within prisons identified numerous positive outcomes beyond avoidance of recidivism, including the development of personal skills, prosocial networks, and bonds with social institutions. In this vein, education can provide a space for people in prison to develop new identities and roles (Søgaard et al., 2016), as they take on the role of “learner.” As Szifris, Fox, and Bradbury (2018, p. 57) state:

Other “hooks” for change can include membership in therapeutic or religious communities or prosocial romantic or family relationships, which allow a reentering individual to embrace a new identity (Giordano, Cernkovich, and Rudolph, 2002).

COMMUNITY AND MACRO-LEVEL IMPACTS ON REENTRY SUCCESS

Much work to support post-release success has been focused on individual-level processes and phenomena, for example by providing vocational training for people in prison. Yet structural contexts, such as the demand for labor, necessarily shape the opportunities for such people or programs to be successful (Hagan, 1997). Similarly, returning from prison with an advanced degree and work skills is likely to provide far more opportunities in a full-employment economy than in the midst of a deep recession or a pandemic. As Farrall (2021) has recently argued, most research on criminal careers and desistance has neglected such macro-level social, economic, and political structures, often as a result of studying single cohorts within limited geographic areas.

A large literature on “neighborhood effects” stresses that residing in a community characterized by poverty, inequality, and socioeconomic

disadvantage can increase the risk of several negative outcomes, including reinvolvement with the criminal legal system. Conversely, living in a neighborhood with ample resources, services, and amenities may mitigate negative outcomes. The immediate environment may help or hinder reintegration after release from prison or successful completion of community supervision (Kubrin and Weitzer, 2003; Simes, 2019; Visher and Travis, 2003).

Neighborhood disadvantage and affluence explained significant variation in arrest for a new crime in a study of individuals in Portland and surrounding areas (Kubrin, Squires, and Stewart, 2007). Other research has expanded on this path-breaking study in a number of ways. Using longitudinal data from the Returning Home project in three states (see Visher and La Vigne, 2021), several studies have identified a variety of neighborhood-level factors that affect reentry outcomes. In one study community cohesion, measured using individual perceptions of networks and ties in community by those returning to the community after incarceration, was protective against returning to prison. However, the impact of community cohesion was dependent on neighborhood-level reentry resources (Liu, 2020). This suggests that communities with low cohesion also suffer from resource depletion and that neighborhood-level deprivation helps to explain why people released from prison fail when they return to impoverished, resource-depleted communities. Another study using the Returning Home data (Liu, 2020) found that parole officers provided more support and spent more time communicating with individuals on parole in more cohesive communities, whereas people returning to disorganized communities received a significantly lower level of support from parole officers and experienced higher rates of return to prison and resumption of drug use. Thus, community and neighborhood factors, specifically community cohesion and material resources, appear to play an outsized role in the success of individuals following release from incarceration.

In addition to surmounting their individual barriers, people leaving prison in recent years have had to contend with widening inequality, declining real wages, lack of access to quality education, rising student debt, a deepening housing crisis, and the enduring effects of structural racism (Bushway and Uggen, 2021). Crime rates fluctuate dramatically over time and across space, which suggests that community context as well as individual factors play a large role in driving criminal behavior and opportunities. While much of the work on reentry generally focuses on changing individual attitudes and behaviors, it is limited by ignoring the constraints imposed by social structure and policies.

A lack of reentry success is not only a criminal justice problem, it is also an employment problem, housing problem, and mental health problem (Wright, Morse, and Sutton, 2021). This has implications for the agencies charged with crime control, but responsibility for reentry success cannot

rest solely upon their shoulders. Such responsibility also extends to institutions charged with higher education, health care, workforce development, housing, public assistance, and the full complement of government and nongovernment organizations charged with public health and well-being. Each organization must be evaluated, in part, based on how well it fulfills its primary mission for persons reentering society after release from incarceration. The research literature on the effects of structural interventions suggests that expanding opportunities and paring back restrictions can make a tangible difference in well-being and socioeconomic attainment. Sampson and Laub (1996), for example, found that the benefits of training provided as part of the GI Bill were significantly larger for veterans stigmatized with an official delinquent past. Moreover, replicating programs that have been successful outside the U.S. context may also help us understand the role of institutional and social context in supporting or undermining the individual drivers of desistance and reintegration (for more on international reentry programs and social contexts see Kazemian, 2021).

Structural Inequality

A new wave of critical reentry research focuses on the systems and practices that contribute to reentry difficulties (Henson, 2020; Middlemass and Smiley, 2019). As a consequence of oppression, marginalization, or isolation associated with characteristics such as poverty, race, ethnicity, gender, and LGBTQ+ status, certain subpopulations enter prison with greater disadvantage, experience disproportionate harm from incarceration, and encounter more obstacles upon release from prison. This section offers several examples of barriers to reentry that are disproportionately experienced by particular segments of the population. These identities are intersectional, and the combined effects of any given individual’s race, ethnicity, gender, class, sexual orientation, and other characteristics can create unique advantages, disadvantages, challenges, and needs. That said, it is important for scholars, practitioners, and policymakers to be cognizant of how patterns of discrimination and disadvantage are perpetuated and sustained against certain groups, and to be responsive to challenges that disproportionately affect certain reentering subpopulations and shape their needs during reentry.

Given the limitations of existing research, the committee acknowledges that this discussion is far from exhaustive. Our objective here is to draw attention to the shortcomings of “one size fits all” approaches to the measurement of success in research, policy, and reentry programing, and to highlight why it is important to recognize and address reentering individuals’ experiences, concerns, and needs in a culturally responsive manner to assess their success. In addition to being more accurate and informative, improved measurement of structural barriers to success has the

potential to produce more effective reentry policies and programs including, potentially, programs targeting structural barriers themselves.

Reentry Challenges Facing People of Color

As discussed in Box 3-5, the development and expansion of the U.S. criminal legal system(s) are inexorably intertwined with the long history of slavery and subsequent civil rights suppression in America (Alexander, 2010; Wilkerson, 2020). Furthermore, racial, ethnic, and sexual minorities have historically experienced oppression and discrimination not only in the criminal

legal system, but throughout numerous interconnected and intersecting policy domains, including education, housing, health care, employment, and political rights. Structural racism and discrimination throughout numerous domains are fundamental sources of disparities in reentry outcomes. Structural racism refers to the normalization and legitimization of historical, cultural, institutional, and interpersonal dynamics that routinely advantage White people and simultaneously produce “cumulative and chronic adverse outcomes for people of color” (Lawrence and Keleher, 2004, p. 1). It is a system of hierarchy and inequity, primarily characterized by the preferential treatment, privilege, and power for White people at the expense of people of color (see Box 3-5).

___________________

¹⁹ After release from prison, White people are likely to live in significantly poorer neighborhoods than they did before prison, whereas Black and Latinx people are more likely to reside in significantly disadvantaged neighborhoods both before and after prison (Massoglia, Firebaugh, and Warner, 2012).

entrepreneurial, using one’s past experience to assist others, and the achievement of heteronormative masculine expectations are factors that many incarcerated men define as being successful post-release (Andersen et al., 2020).

Latino men, as the fastest growing U.S. ethnic group of incarcerated individuals (Carson and Sabol, 2012; Harrison and Beck, 2002), face unique cultural and social needs during the reentry process. For example, they are often characterized by distinct circumstances associated with low educational attainment, limited language ability, and limited access to health care (Harlow, 2003; Schafhalter-Zoppoth, Walther, and Flattery, 2013). These characteristics place formerly incarcerated Latinos at a disadvantage for successful reentry, as many have difficulty securing housing and employment (Schafhalter-Zoppoth, Walther, and Flattery, 2013). Moreover, formerly incarcerated Latinos often return to disadvantaged communities, and remain vulnerable to poor outcomes post-incarceration, as they have complex psychosocial, health, and economic needs (Golembeski and Fullilove, 2005; Lee et al., 2016). However, evidence suggests that family mechanisms of social control and social support can influence the reentry process among Latino men (Lee et al., 2016), and ethnic pride can serve as a source of strength for young men of color transitioning from jail to the community (Upadhyayula et al., 2017).

While the issue of Latino criminalization has often been linked to the issue of immigration, research shows that immigrants reoffend at a much lower rate than native-born residents (Ramos and Wenger, 2019). Evidence suggests that concentrated immigration in a community can serve as a protective factor against reoffending among justice-involved youth (Wolff et al., 2015), including Latina girls (Wright and Rodriguez, 2014). Rather than a “culture of poverty” in which individuals residing in economically distressed neighborhoods adapt in ways that perpetuate their conditions (Lewis, 1965), immigration may contribute to a “culture of resilience” (Wright and Rodriguez, 2014).

Although Mexico is currently the top country of origin among U.S. immigrants today, it is important not to conflate ethnicity and immigration status. In general, recidivism rates are lower for immigrants compared to native-born U.S. residents (Ramos and Wenger, 2019; Light, He, and Robey, 2020), although reoffending rates for individuals who are convicted of illegal entry and reentry in the United States remain high. When incarcerated, deportable non-U.S. citizens housed in federal prisons are provided with fewer educational opportunities and minimal access to substance treatment, and they are ineligible for residential reentry centers (Kimpel, 2018). Moreover, a large number of deportees no longer have strong familial ties in their home countries. Reintegration is made more challenging for them, as they are told to establish roots in a country where they no longer have family (Kimpel, 2018).

Native Americans experience rates of jail and prison incarceration about double that for white Americans, as well as elevated rates of return to prison, mostly for technical violations (Daniel, 2020; Hansen, 2018). While 78 percent

of Native Americans do not reside on reservations, many formerly incarcerated Native Americans return to rural communities, where poverty, drug and alcohol misuse, as well as geographic isolation pose challenges to reentry (Wodahl and Freng, 2017). Despite high rates of poverty, a recent study of fines, fees, and restitution in Minnesota found that Native Americans are subject to higher monetary sanctions than any other racial or ethnic group (Stewart et al., 2021). Kara Nelson from True North Recovery, a participant in the committee’s information-gathering sessions, mentioned that the state of Alaska provides a vast array of services for reentry. But fellow participant Venus Woods from the Cook Inlet Tribal Council pointed out that Alaska natives need culturally competent programming. Conventional instruments for assessing risk and resiliency may not include culturally specific factors relevant to Native American populations (Hansen, 2018). Culturally responsive reentry programs could involve positive elements such as tribal “restorative or reparative” principles, approaches, values, and ceremonies (Melton et al., 2010). The involvement of tribal governments and justice systems, often alongside state and federal court systems, adds another layer of complexity to the coordination of reentry services for Native Americans (Melton et al., 2010).

Gender-specific Reentry Challenges

Although men greatly outnumber women among incarcerated populations, the growth in incarceration of women outpaced that of men from 1978 to 2015 (Sawyer, 2018). From 1990 to 2000, the proportion of women on probation and parole also increased, and in 2004, 85 percent of women under correctional supervision were under community supervision (Bloom, Owen, and Covington, 2004). Nearly two-thirds of women in prison or jail are Black or Hispanic, while nearly two-thirds of women on probation are White (Bloom, Owen, and Covington, 2004).

Criminal legal system (CLS)-involved women are more likely to have histories of childhood and adult abuse and have more serious health problems, higher rates of mental health problems, and substance dependencies than criminal legal system-involved men (Bronson et al., 2017; Harlow, 1999; National Institute on Drug Abuse, 2020; Wolff, Shi, and Siegel, 2009). Many women who suffer from abuse, either as children or as adults, develop mental illnesses (e.g., depression, anxiety, post traumatic stress disorder [PTSD]) (Lynch et al., 2012a; Lynch et al., 2012b). These mental illnesses, if not treated by a professional, may cause women to “self-medicate” through drugs as a coping mechanism to ease the physical, sexual, and psychological pain of abuse (Chesney-Lind, 1997; Daly, 1998; DeHart, 2004). Women are also more susceptible to sexual misconduct and abuse while incarcerated (Bloom, Owen, and Covington, 2004).

Although many incarcerated men also suffer from substance problems, women who enter and exit prison are more likely to have substance abuse

problems (Maruschak, Bronson, and Alper, 2021a). According to one study, not all women had access to prison-based drug treatment programs, and some who did have access found them to be limited in duration and also found that community-based programs after release were scarce (Harm and Phillips, 2001). Certain conditions are associated with women’s drug relapse after release from incarceration, including parole to homes where family members use drugs, reconnecting with drug-using friends, economic strain, crisis events, and negative emotions (Harm and Phillips, 2001). Convictions for drug offenses can have long-term consequences post-release. As noted above, welfare reform–era restrictions on public assistance for people convicted of felony drug crimes had a large effect on African-American and Latina mothers, who are disproportionately poor and need public assistance.

Benda (2005, p. 337) found that “childhood and recent sexual and physical abuse, adverse feelings, living with a criminal partner, and drug use are particularly powerful predictors of women’s recidivism.” For men, job satisfaction and education had particularly strong effects on reducing the risk of recidivism. For women, these factors also reduced recidivism, but were less important than close relationships with family, romantic partners, and friends, along with number of children. Adult social bonds have been found to inhibit criminal offending for both male and female probationers, but the effects are stronger for women (De Li and MacKenzie, 2003). Programming that can help women identify and strengthen prosocial networks during incarceration and during the transition from incarceration to the community can facilitate women’s efforts to successfully reintegrate (Bui and Morash, 2010; McKay et al., 2016).

In addition, compared to men, women place a greater value and rely more heavily on social support following release from prison (Barrick, Lattimore, and Visher, 2014; Clone and DeHart, 2014; Cobbina, Huebner, and Berg, 2012). Once women are released to the community from prison, they are more likely than formerly incarcerated men to reconnect with family and seek out family ties (Cobbina et al., 2012). Positive social support, especially emotional and tangible support, has been shown to be critical to women’s well-being following release from prison (Martinez and Christian, 2009). Most women who return home from prison to economically distressed neighborhoods, however, have small homogeneous networks that result in their marginalization, making successful reintegration more challenging. Since many of these women reside in unsafe, inaccessible, and car-dependent areas, they must often rely heavily on social support to get around (Bohmert, 2016).

In addition to family support, reunification with their children is often a primary goal for women following release from prison (Giordano, Cernkovich, and Rudolph, 2002; Richie, 2001); the possibility of reconnecting often remains a source of hope and inspiration (Bachman et al., 2016). 52 percent of individuals incarcerated in state prisons (51.2% of men and 61.7% of women), and 63 percent of those in federal facilities

(62.9% of men and 55.9% of women) are parents of minor children (Glaze and Maruschak, 2016). More imprisoned mothers than fathers were the primary caregivers for their children before their incarceration. As a result, their family networks are more severely disrupted, and the need to reestablish relations with children and other family members can be especially complex for women upon reentry. The termination of incarcerated mothers’ parental rights can make it difficult or impossible to resume custody of their children (Michalsen, 2011). Moreover, the relationship between family demands and caretaking responsibilities is complicated. Mothers involved with the criminal legal system face particular maternal hardships and additional stigma and discrimination (Mitchell and Davis, 2019; Gurusami, 2019). While childbearing can serve as an impetus for maturation out of crime, children can serve as a source of parental stress, especially when coupled with other life responsibilities. Caring for children while maintaining steady employment may be challenging without reliable and affordable health care (Blitz, 2006). Gender caretaking roles create disproportionate economic constraints for women, which translate into higher rates of women engaging in criminal activity as a means for survival (Belknap, 2015; Wattanaporn and Holtfreter, 2014). When some women feel overwhelmed and unprepared with the obligations of motherhood it can impact their ability to successfully integrate back into the free world and desist from crime.

At the same time, the desire to regain custody can provide motivation to mothers to avoid recidivism. Incarceration of a custodial parent creates serious psychological, emotional, and economic problems for children as well. It also increases responsibilities and sometimes hardships for family members—who are usually women—who must step in as caregivers (Western et al., 2015). Contacts between formerly incarcerated parents and their children are weakly related to monthly income, but strongly associated with stable private housing, as opposed to transitional housing or homeless shelter beds (Western and Smith, 2018).

Due to the importance of relationships and community ties, formerly incarcerated women may be especially drawn toward caregiving both in their personal lives and as vocational options (Chen and Adams, 2019). Unfortunately, the disclosure of criminal records in background checks can eliminate otherwise-qualified candidates from applicant pools for positions in health care, child care, and elder care well after they have served their time in prison and desisted from criminal activity.

Intersecting Effects of Race and Gender on Reentry

Following their release from prison, most women, especially women of color, return to neighborhoods that are characterized by high levels of poverty, unemployment, inequality, segregation, and crime (Massey and

Denton, 1993; Wilson, 1996). Black people tend to be situated in communities with limited political and economic resources (Cobbina et al., 2014; Owen and Bloom, 1995) and reentering Black women report feeling “marginalized within the context of a disenfranchised community,” such that their “needs as women are not a priority” (Richie, 2001, p. 383).

In an effort to avoid criminal activity, Black women on probation and parole often isolate themselves by avoiding everyone, including family members (Cobbina et al., 2014). This is problematic for several reasons. First, women’s networks are smaller than those of men (Campbell and Rosenfeld, 1985), comprising more relatives (Marsden, 1987) and containing fewer ties to non-kin (Marsden, 1987; Moore, 1990), which suggests that women have less ability to use networks to secure employment. Second, compared to White people, Black people have smaller networks that contain a lower proportion of relatives (Marsden, 1987; McPherson, Smith-Lovin, and Brashears, 2006), and they have less frequent contact with others in their networks (Pugliesi and Shook, 1998). Thus, women involved with the criminal legal system who avoid everyone and stay at home to avoid criminal involvement weaken networks that are far from adequate.

Third, when Black women avoid their relatives who have a criminal record, they relinquish a central source of support that has been important to African American families, whose members typically rely on each other. Fourth, structural conditions, including joblessness, persistent poverty, and family disruption, can lead to African Americans traveling in small, isolated social networks that prevent the development of strong social supports (Reisig, Holtfreter, and Morash, 2002). Consequently, their strategies for avoiding offending may intensify this pattern. Overall, women who live in poor neighborhoods, who are disproportionately Black women, generally face unique challenges as they attempt to navigate their communities.

Reentry Challenges for Individuals Experiencing Trauma

High rates of lifetime trauma are common among many incarcerated men and women. There is a high prevalence of victimization and previous trauma among people involved in the criminal legal system, especially among women who bring past trauma into prison settings (Harlow, 1999; Yoon, Slade, and Fazel, 2017). According to the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 2013), trauma is a stressor in an individual’s life that includes exposure to death, the threat of death, actual or threatened serious injury, or sexual violence. Individuals may experience these stresses through direct or indirect exposure. The Substance Abuse and Mental Health Services Administration (SAMHSA) defines trauma more broadly as “an event, series of events, or set of circumstances that is experienced by an individual as physically or emotionally harmful or life threatening and that

has lasting adverse effects on the individual’s functioning and mental, physical, social, emotional, or spiritual well-being” (SAMHSA, 2014). Notably, incarcerated women have histories of victimization and trauma at higher rates than incarcerated men (Swavola, Riley, and Subramanian, 2016). And many of these women have dealt with trauma exposure, interpersonal trauma, victimization, PTSD, and violence before incarceration (Lynch, Fritch, and Heath, 2012; Harner et al., 2015). Such histories are often linked to pathways that lead to women’s imprisonment.

Many individuals involved in the criminal legal system have experienced high levels of violent victimization and undergone significant trauma before, during, and after their time behind bars (Atkinson and Warnken, 2016). Scholars have long underscored the victim-offender overlap, recognizing that “neither victimization nor offending can be understood without full consideration of the other” (Lauritsen and Laub, 2007, p. 56). Nevertheless, most people who have survived victimization do not receive the services they need to heal (Hastings and Kall, 2020). Many victims and survivors face barriers to services, including victim service providers who

do not view them as crime victims (Hastings and Kall, 2020) and see them solely as perpetrators of harm rather than victims of harm, with many unaware that trauma is often a causal factor contributing to criminal activity.

This false dichotomy between criminal offending and victimization has also been formalized through policy and funding structures. For example, in a listening session with the committee, Michelle Garcia of the Washington, D.C., Office of Victims Services and Justice Grants reported that her office had historically been two separate offices referred to in shorthand as “victims” or “perpetrators.” Before the offices were combined, victim service providers did not work with individuals with criminal legal system involvement—Garcia noted that they were not trained or equipped to do so. Similarly, reentry providers did not screen for victim or trauma histories. At the federal level, Victims of Crime Act grantees were for many years prohibited from using grant funds to support rehabilitation and counseling services for criminal legal system-involved individuals (this prohibition was removed in 2016) (Federal Register, 2016). The committee heard additional perspectives from service providers for victims of crime in its information gathering (see Box 3-6).

Spencer, and Wilson, 2021). As noted previously, most survivors who have been previously incarcerated do not receive the services required to heal and experience unaddressed trauma stemming from victimization experienced at some point in their lives (Hastings and Kall, 2020). The failure to recognize individuals involved in the criminal legal system as victims discourages participation in many victim-oriented services. In addition, pressing material needs, including the needs to secure food, housing, employment, transportation, and medical care, create barriers to healing from trauma. Moreover, stigmatization remains a significant barrier to seeking mental health treatment, especially among African Americans and among men in general (Conner et al., 2010; Gary, 2005). Navigating the reentry experience with unaddressed trauma stemming from violence can result in serious negative outcomes for persons released from prison, including criminal involvement and technical violations that send them back to prison (Hastings and Kall, 2020) (see Box 3-7).

Numerous recent studies emphasize that individuals suffer from profound anxiety and feelings of isolation after release from prison (Hyde et al., 2021; Western et al., 2015). Nevertheless, correctional facilities can serve as a venue for people in prison to address the deeper emotional responses tied to trauma through trauma-informed care (Levenson and Willis, 2018). Trauma-informed care in corrections may contribute to successful reentry and reduced recidivism. Some incarcerated people obtain access to vocational, therapeutic, educational, and spiritual programming, as well as physical and mental health care, while in prison. Additionally, prison can serve as a venue where opportunities for self-improvement become available, and some U.S. prisons are actively implementing changes to facilitate these opportunities (Hyatt et al., 2021). More generally, however, formerly incarcerated individuals who participated in the committee’s information-gathering sessions expressed a desire for preparation for reentry to begin at a much earlier stage in prison.

Reentry Challenges Associated with LGBTQ+ Status

Lesbian, gay, and bisexual people are imprisoned at a rate three times that of the general public (Meyer et al., 2017). The incarceration rate for transgender people is twice the general population rate; for Black transgender women, it is 10 times higher (National Center for Transgender Equality, 2018). Moreover, sexual minority adults are twice as likely as the general population to experience homelessness (Wilson et al., 2020).

LGBTQ+ individuals experience extensive social and economic marginalization in employment (Baumle and Poston Jr., 2011; Dilmaghani and Robinson, 2022; Flage, 2019; Mallory, Flores, and Sears, 2021), health care (Brummett and Campo-Engelstein, 2021; Casey et al., 2019), and housing

(Schwegman, 2018), many of which are compounded for LGBTQ+ people of color (Wilson, Bouton, and Mallory, 2022). Given the challenges posed in these domains by incarceration histories, it is likely that formerly incarcerated LGBTQ+ individuals face compounded challenges in these areas. Like other reentering individuals, many LGBTQ+ people have histories of mental illness and substance abuse problems. They are more likely to face employment discrimination, experience bullying and harassment in

educational settings, and to have strained, disrupted, or nonexistent family connections (Movement Advancement Project, 2016).

Barriers to legal employment can lead LGBTQ+ individuals with histories of sex work and drug-related offenses to reengage in those activities (Belenko, Hiller, and Hamilton, 2013; Roe-Sepowitz et al., 2011). Transgender individuals are often housed in jail, prison, and residential reentry facilities that do not match their gender identity, and they may experience harassment and violence in these settings (Movement Advancement Project, 2016). In prison, LGBTQ+ people disproportionately experience sexual victimization by fellow incarcerated persons as well as staff (Beck et al., 2013). They are also at risk of victimization in homeless shelters and halfway houses, where many reentering individuals temporarily reside (Santos, 2021, Movement Advancement Project, 2016). LGBTQ+ people are overrepresented in sex offender registries, in part due to the criminalization of certain sexual acts; this presents additional barriers to housing and employment (Santos, 2021). The need exists for culturally responsive reentry programs and services that are inclusive of diverse gender identities and sexual orientations.

Persons with Disabilities and Reentry

Approximately 40 percent of individuals in state prison and 30 percent of people in federal prison have a disability (Maruschak, Bronson, and Alper, 2021b). That translates to an estimated 760,000 people with disabilities living behind bars, dealing with cognitive, ambulatory (mobility), vision, hearing, or independent-living disabilities (Maruschak, Bronson, and Alper, 2021b). Individuals with disabilities, especially cognitive and intellectual, are at greater risk of serving longer, harder sentences, and they are vulnerable to exploitation, violence, and suicide, especially in solitary confinement (Fazio, Pietz, and Denney, 2012; Shlanger, 2017). Those with physical and cognitive disabilities may face exclusion from programs and services as well as isolation (Cowardin, 1997; Petersilia, 1997; Shlanger, 2017). According to 2011–2012 National Inmate Survey data, disability rates are even higher among incarcerated women and White individuals, compared with men and people of other racial backgrounds (Maruschak, Bronson, and Alper, 2021b). However, an analysis of 2002–2011 Rehabilitation Services Administration data also showed particularly high rates of mental disabilities among Black people in prison (Baloch and Jennings, 2019).

While discussions of mental health in prison generally focus on psychiatric disabilities like bipolar disorder, depression, and schizophrenia, the needs of incarcerated people with intellectual and developmental disabilities have largely been overlooked. As a result, there is little support for people with neurodevelopmental disorders in prison (Young et al., 2018), and there is limited research on this topic.

for their children upon returning home from prison prior to securing employment (Morgenstern and Blanchard, 2006), increasing financial strain on women and children residing in economically distressed neighborhoods (Patterson, 2013; Visher et al., 2004). Moreover, the inability of a large segment of the U.S. population to access housing, employment, education, and public assistance may increase the national poverty rate (Hall, Wooten, and Lundgren, 2016) and in turn increase levels of crime and imprisonment.

While poverty is a risk factor for criminal behavior and incarceration for men and women (Van Voorhis et al., 2010), many individuals returning home from prison will return to families and communities facing similar circumstances that resulted in their incarceration (Belknap, Lynch, and DeHart, 2015). Due to discrimination and low levels of human capital, formerly incarcerated individuals, Black people, and women are disproportionately represented among workers in the secondary labor market characterized by low paying, insecure, dead-end jobs (Martin, 2011). Evidence suggests that the burden of poverty falls heavily on women and children, as most women involved in the criminal legal system are mothers of minor children (Owen and Bloom, 1995). In their examination of 134 women convicted of a felony offense, Holtfreter, Reisig, and Morash (2004) found that poverty status increased the odds of rearrest and supervision violation; yet the odds of recidivism were reduced by 83 percent among those who received state-sponsored support.

Programs and criminal justice interventions providing services to persons released from prison are likely to be located within poor communities (Clear, 2007; Wacquant, 2010). Reuben Miller (2014) has argued that the community-based organizations that serve individuals involved in the criminal legal system manage more poor people and people of color than the correctional system itself. This arrangement provides needed services but contributes to the concentration of social disadvantage within blighted neighborhoods, with disproportionate impacts on the poor, Black, and brown residents in these areas (Miller, 2014). From this perspective, the dominance and proliferation of reentry services in low-income communities of color represents a longstanding collusion between social welfare and criminal justice actors in managing marginalized populations and demonstrates one way that the state has been reconfigured to manage the urban poor (Beckett and Western, 2001; Miller, 2014).

Reentry Barriers Facing Rural Populations

Studies of reentry have focused heavily on the experiences and needs of individuals returning to urban neighborhoods, with less attention to those returning to rural communities, whose circumstances may require different forms of support (Wodahl, 2006; Zajac et al., 2013). People in rural areas

also have more difficulty accessing programs and services such as mental and physical health care, substance abuse treatment, subsidized or transitional housing, and food support (Huebner, Kras, and Pleggenkuhle, 2019; Zajac et al., 2013). The lack of public resources may be due, in part, to the more limited tax base in rural communities (Ethridge, et al., 2014).

Compared to urban areas, rural communities are less geographically dense. People on parole are often required to traverse great distances to check in with parole officers and participate in treatment programs (Huebner, et al., 2021). Reliable transportation is also a necessity for job-seekers and employed persons, but it is often unavailable (Wodahl, 2006; Zajac et al., 2013. Jobs that pay a living wage are also less prevalent in rural areas (Wodahl, 2006). For rural women, traditional notions of “women’s work” may pose additional obstacles to adequately compensated employment (Beichner and Rabe-Hemp, 2014). Compared with people returning to urban areas, individuals reentering rural areas may be less prone to illicit drug use, but more likely to have problems with alcohol or sedatives (Wodahl, 2006). Individuals returning to rural communities may also experience more stigma, due to a relative lack of privacy and anonymity (Huebner, Kras, and Pleggenkuhle, 2019; Zajac et al., 2013).

Structural Reentry Barriers and the Measurement of Success

To summarize this section of the report, the limited research literature has identified reentry and reintegration needs that are particularly acute in particular communities. In many cases, these needs reflect specific manifestations of a general need, such as transportation or housing. In other cases, however, such needs are more localized, such as the need for translation assistance in non-English-speaking communities. This section makes clear that reentry needs and barriers are distributed unevenly in the population of individuals incarcerated and released. Post-release success, whether in terms of cessation of criminal activity or a broader conception of flourishing, is shaped by differential exposure to these barriers. Yet to date, standard methods of measuring success among those released from prison have not accounted for the varied, significant, and systemic differences in barriers to reintegration.

CONCLUSION

In outlining the theoretical rationale for looking beyond recidivism and identifying the barriers to success, the foregoing discussion raises a critical question: What do we know or need to learn about reentry success beyond recidivism? This chapter has illustrated how a singular or primary focus on recidivism ignores scholarly understanding of how the cessation of criminal activity actually occurs, and more broadly ignores the best available

knowledge on the reentry process and needs. Further, an overreliance on recidivism can inhibit the design and delivery of effective, appropriate programs and services for formerly incarcerated individuals. Expanding our attention to the complex, iterative, and multidimensional processes of desistance and reintegration helps us apply and test theories regarding why and how reentering individuals persist in, or desist from, criminal activity.

In orienting to lasting success across a variety of domains, such a focus can also enhance public safety, protect victims, and promote the common good. Correspondingly, an emphasis on recidivism imposes unnecessary and unhelpful constraints on the range of outcomes of interest with regard to the success of people released from prison. This chapter builds on Chapter 2 in concluding that because cessation or reduction in criminal behavior often occurs as a part of a gradual process that may involve setbacks, measures of desistance from crime can offer a more accurate, complete, and nuanced account of an individual’s reduction in criminal activity than do measures of recidivism.

As this chapter has demonstrated, advancing the measurement of success for individuals released from prison is far beyond a methodological issue. Improving our methods for evaluating success also requires theoretical and conceptual work—shifting from a recidivism frame to a desistance frame in measuring criminal legal system outcomes and expanding our understanding of success to encompass the life domains central to successful reintegration and overall well-being. Successful reintegration and post-release success require progress in a number of areas beyond the cessation of criminal activity. This chapter’s analysis of reintegration, reentry barriers, and well-being supports the conclusion that post-release success involves multiple life domains (e.g., health, employment, housing, civic engagement) and not simply involvement in the criminal legal system. Success entails a heightened sense of personal well-being.

Finally, this chapter explores persistent patterns of disparities in reentry supports, and unique reentry needs facing historically marginalized groups. A review of the literature demonstrates that these disparities are both significant and under-studied. The existence of community-level and policy barriers to success is documented in studies that link data on post-release success and local socioeconomic conditions; policies that restrict access to employment, housing, and public benefits; and structural inequalities that disproportionately affect persons of color. Thus, meaningful measurement of success requires attention to systemic disparities in exposure to barriers and access to opportunities that shape post-release outcomes.

A more robust measurement of success could result in policies and programs that better support the needs of those returning from incarceration and more effectively support success. They could also help policy makers, criminal legal system actors, and service providers to recognize and address

structural inequalities shaping post-release outcomes. The following chapter charts possible paths toward improving measures of success, including identifying key indicators of post-release well-being, and considers how we might measure these alternative indicators of success.

REFERENCES

This page intentionally left blank.