Tough on Crime,Tough on Families? Criminal Justice and Family Life in America

Mass Incarceration and Family Life

While presidents and presidential candidates may set the tone for broader criminal justice policies pursued by the states, they are restricted in their ability to directly influence the number of people incarcerated in the United States. The federal prison system incarcerates just 13 percent of all prisoners in the United States, the vast majority of whom are incarcerated for drug and public order offenses. In comparison, most inmates in the United States are imprisoned in state facilities and, of these, half are incarcerated for violent offenses (Carson 2015). Jails, housing those serving short sentences as well as those awaiting trial, incarcerate millions of additional citizens (Apel, this volume; Minton 2011). States with historically punitive criminal justice policies—New York, New Jersey, and California, for example—have done away with draconian drug policies and reformed parole revocation procedures, while a host of other states are converting low-level crimes that would have once led to doing time in a jail, such as marijuana possession, to offenses punished with fines, if at all. Yet significant barriers to reform remain, especially given that many proposals designed to reduce incarceration in the United States largely exclude violent offenders, whose long sentences have contributed greatly to the swelling imprisonment rates since the early 1970s. Before talking about whether reforms involving federal prisoners, state prisoners, or local jail inmates hold the most promise—and the most feasibility—we first need a brief reminder of where we are and why it might matter for American families.

So where are we? The legacy of nearly 40 years of punitive criminal justice policies at the state and federal levels can be seen in the expansion of the criminal system and inequalities within it. In 1980, the United States supervised 1.8 million people—520,000 in prison and jail, 1.1 million on probation, and about 220,000 on parole. By 2007, the United States had spent more than 74 billion dollars to do the same for more than 7.3 million people—4.3 million on probation, 826,000 on parole, and 2.3 million people in jails and prisons (Kyckelhahn 2011; Sourcebook of Criminal Justice Statistics 2011). ⁴ Put bluntly, more people have contact with the criminal justice system than ever before (Bonczar 2003; Glaze and Kaebel 2014). ⁵

And why might this matter for American families? On the most basic level, people who have been imprisoned—nearly all of whom are connected to families in some way—are made less able to contribute to their families as a result of their incarceration. Convicted felons may be barred from some jobs, jury service, public services, and voting. This constellation of legal disabilities, piled disproportionately upon the poor and people of color, is overwhelmingly repressive (Alexander 2010; Coates 2015; Wakefield and Uggen 2010; Western 2006) and likely inflicts harm not only on individuals but also on families. The informal consequences of criminal justice contact are more variable but no less severe and likely also spill over to the families of formerly incarcerated individuals. Unlike formal consequences, informal harms are not created by law or debated in legislatures; rather these harms silently ride along with those who have done time. One important such harm—and the harm that our volume specifically focuses on—includes the damage imposed on the families of individuals who have spent time in a local jail, state prison, or federal prison (Western and Wildeman 2009; Wildeman 2009). It is safe to say that incarceration results in global harms for (some) families. For the children of incarcerated parents, parental incarceration increases mental health and behavioral problems, infant mortality, homelessness, grade retention, body mass index, harsh parenting, and material hardship, among many other social problems (Andersen and Wildeman 2014; Cho 2009a, 2009b; Farrington, Coid, and Murray 2009; Geller et al. 2009; Geller et al. 2012; Haskins 2014; Murray and Farrington 2005; Murray, Farrington, and Sekol 2012; Roettger and Boardman 2012; Schwartz-Soicher, Geller, and Garfinkel 2011; Turney 2014a; Turney and Haskins 2014; Wakefield and Wildeman 2011, 2013; Wildeman and Turney 2014), and racial inequality in childhood well-being more broadly (Lee et al. 2015; Wakefield and Wildeman 2011, 2013). The same is true for the romantic partners of inmates, although there is less research on this population. Research describes a number of burdens these partners bear, ranging from mental and physical health problems to material insecurity (Braman 2004; Christian 2005; Comfort 2008; deVuono-Powell et al. 2015; Turney, Schnittker, and Wildeman 2012; Wildeman, Schnittker, and Turney 2012). And these are only the direct effects. Additional research examines the role incarceration plays in marriage markets, which have long been thought to be central to racial inequality in America (e.g., Moynihan 1965; Wilson 1987). This work finds that incarceration is most unambiguously important because it breaks up married or cohabiting families, increasing the likelihood of divorce and separation (Apel et al. 2010; Bacak and Kennedy 2015; Lopoo and Western 2005; Massoglia, Remster, and King 2011).

As the criminal justice system has expanded over the past four decades, few aspects of family life remained untouched. Yet while a great deal is known about the general nature of incarceration and family life effects, far less is known about how the level and character of criminal justice contact moderate such effects. Even less is known about whether the low-level offenders policymakers most often seek to let free early or keep out entirely are indeed the inmates most likely to benefit their families when they return. We have little sense whether picking the low-hanging fruit of criminal justice policy reform will provide the most nourishment for the families affected by the criminal justice system or whether the release of more “difficult” justice-involved populations (such as those in prison for serious violence) might help families even more. As a result, the body of research on incarceration and family life offers little guidance for policymakers, much to the chagrin of those of us who have dedicated our careers to trying to provide such policy-relevant insights. The work described herein is abundantly clear that complication and heterogeneity, not universal harm, are the norms and are the most important challenges to successful policy reform.

A Call for New Questions in Work on Incarceration and Family Life

In light of the complexity described above, we suggest that researchers and policymakers turn their attention to different—and more difficult—questions. Instead of asking whether incarceration is more harmful for some children or partners or families or inmates relative to others, scholars have identified ever more outcomes for which there is evidence of a harmful effect. In a short period of time, researchers have provided substantial evidence on the harmful consequences of incarceration for family life. But largely missing from this literature is an empirical acknowledgement of the variance in the character of criminal justice contact and family life and, as a result, the dramatic variation in incarceration’s impacts for family life.

If the cause of these gaps were simply myopia among researchers, the problem would be easily solved (although it would involve an overhaul of researchers). Yet most researchers in this area are well aware of the limits of our findings for reform. Very few of the data sources on the United States population that are typically used to consider the well-being of American families include information on incarceration, and those that do are extraordinarily limited in scope, even today. The lack of information on criminal justice contact in national surveys made good sense when only one in 1,000 individuals in the general population was imprisoned on any given day. Today, though, this inattention makes significantly less sense, as ex-felons account for a nontrivial share of the voting age population (Bonczar 2003; Manza and Uggen 2006), racial disparities in the cumulative risk of imprisonment are enormous (Pettit and Western 2004), and arrest rates are astonishingly high (Brame et al. 2011, 2014). These numbers are large, by any definition, yet our national surveys, census, and administrative data sources largely (and systematically) ignore them (Pettit 2012).

The consequences of rendering such a large population hidden are not merely academic; the lack of available data on criminal justice contact impedes our ability to design effective policies. Consider, for example, the vast literatures on how family structure (McLanahan and Percheski 2008), school context (Behrends 2015), and neighborhood of origin (Sharkey 2013) affect child adjustment and adult attainment. Studies of this type are only possible with data sources that ask detailed questions about the quality of family life, the conditions of schools, and the context of neighborhoods. And the findings from analyses using these data have led to important policy interventions that have been effective in shaping well-being. The corollary for incarceration would be to interrogate variation related to the correctional context—for example, by examining jail versus prison incarceration, duration and frequency of incarceration stints, or the conditions of confinement. Several qualitative studies have done this (Arditti 2012; Braman 2004; Comfort 2008; Enos 2001; Nurse 2002; Siegel 2011), and they make it abundantly clear that questions such as these are vital. It is difficult, however, to discern clear causal effects, let alone policy reforms, from studies of this sort.

Few large surveys of the incarcerated include more than cursory information on their families, and longitudinal surveys of the general population may have extensive information on family life and child well-being but often have little information on criminal justice contact (National Center for Family and Marriage Research 2012). Administrative data from the United States are similarly limited; studies using them focus on maternal incarceration because mothers are more reliably linked to their children (e.g., Cho 2009a, 2009b; Dworsky, Harden, and George 2011), yet registry data from other countries suggest analyses of these sorts are sorely needed, as they imply more consistent effects of paternal incarceration than maternal incarceration (e.g., Wildeman et al. 2014). By using U.S.-based sources, we cannot properly characterize (let alone understand) the complex nature of the average person’s contact with the criminal justice system, which may involve multiple arrests, pleas, cases, pretrial sentences, and convictions; and imprisonment. That this population is among the most likely to drop out of large-scale surveys (or decline to participate) exacerbates the problem. Using family members as information proxies to document the complex histories of criminal justice contact for this population is similarly difficult (see Geller, Jaeger, and Pace, this volume).

In light of these difficulties, the empirical contributions to this volume answer basic questions of scholarly and practical importance. They leverage international data (L. Andersen, this volume; S. H. Andersen, this volume), qualitative data (Comfort, this volume; Lageson, this volume), administrative sources (Berger et al., this volume), the creative use of existing survey data (Apel, this volume; Wildeman, Turney, and Yi, this volume; Wakefield and Powell, this volume), and provide evidence on the potential (or lack thereof) for linking administrative and survey data (Geller et al., this volume) to privilege variation in those with criminal justice involvement and the character and level of such contact. The findings from these articles echo our call for more data collection and highlight important complexities in the role incarceration of a family member plays in family functioning and well-being. They also call into question the prevailing rhetoric of policy reform proposals.

Challenges to Policy Reform

The limits of current policy reform proposals are large and difficult to ignore. Notably, most reform proposals are restricted to nonviolent (often drug) offenders who are thought to pose little risk to public safety, yet the imprisonment rate in the United States would still be very high—at least 300 per 100,000, and more likely 400 per 100,000—even if all inmates who fit this profile were released. Moreover, concerns about massive racial disparities in imprisonment are unlikely to be allayed if violent offenders are removed from the conversation, as racial disparities in arrests and imprisonment for homicide and violent crime are large. If we preclude prisoners convicted of violent crimes from reform, mass incarceration remains and we do little to reduce racial disparities in the criminal justice system.

As we have described here, our data sources make it difficult for us to adequately describe who is actually a low-level offender, let alone characterize the influence their incarceration may have on their partners and children. For example, in her contribution in this volume, Megan Comfort describes one inmate “realigned” in California (a process meant to be reserved for nonviolent, nonserious offenders) as “bewildered” by his classification as nonviolent. “Nonviolent” offenders, according to the criminal justice system at least, may be surrounded by or even create violence within their families. President Obama has called for the diversion of low-level drug offenders from prisons, but it is important to recognize that low-level drug offenders, however defined, may not be any better for their families (or their communities) than other types of offenders. Moreover, research suggests that, in the absence of compelling evidence to the contrary, “risk” as defined in policy reform proposals is as much a proxy for political costs as threats to public safety when it comes to the formation of criminal justice policy (Beckett 1997; Turner and Gerlinger 2014). If we defined risk in terms of threats to public safety rather than threats to political capital, we might find that releasing a prisoner with a life sentence for one episode of violence in his youth presents much less risk to public safety than diverting an offender addicted to drugs. More broadly, however, society (and families) may be much better served by the more humane treatment of both categories of prisoners.

As folks who mostly do quantitative research, it is easiest for us to visualize the nature of this challenge using a table, even a very crude one. As Table 1 shows, criminal justice policy and family policy align when the individuals who commit low-level offenses are also the most positive contributors to their families—the sorts of partners and fathers so poignantly described in some ethnographic research (e.g., Braman 2004). The same is the case with violent offenders who inflict serious harm on their families. Yet when the criminal infractions are serious but the contribution to family life is also great, family and criminal justice policy do not align. The long prison terms accorded to those who are guilty of one heinous act but remain committed to their families and to avoiding crime in the future thus make little sense on either public safety or family well-being grounds (see Coates 2015 for a number of salient examples). And the same is the case with low-level offenders who do great harm to their loved ones. The cases where “low-level offenders” are less than family-friendly represent a fundamental challenge to family-friendly criminal justice policies—and one that has been largely (if not completely) absent from public discourse surrounding how we might decrease rates of incarceration in our nation.

OPEN IN VIEWER

Table 1

Plausible Overlaps between the Severity of Criminal Offending (High Level, Low Level) and the Quality of Contribution to Family Life (High Quality, Low Quality)

	Low-level criminal offending	High-level criminal offending
High-quality family contribution	Criminal justice and family policy aligned	Criminal justice and family policy not aligned
Low-quality family contribution	Criminal justice and family policy not aligned	Criminal justice and family policy aligned

Second, research on the average effects of incarceration on family life obscures great variation among inmates, the conditions of confinement, and families prior to incarceration. For example, parental incarceration may be moderately harmful for most children or very harmful for a small number of children while relatively inconsequential for everyone else. But few studies of family life and incarceration distinguish between these two very different patterns of effects. While several studies suggest that the latter is the case (Turney 2014b; Turney and Wildeman 2015), the methodological complications of such models are significant (Breen, Choi, and Holm 2015), and the practical implications of such models limit their use for designing effective policy interventions (Lee and McLanahan 2015). Similarly, current research often does not distinguish incarceration spells with respect to their frequency, duration, or the conditions of confinement. While documenting the full array of social, economic, and political disabilities imposed by the prison boom, it is nearly impossible to identify the most promising policy interventions from research of this sort because it ignores contextual variation. The end result of the emphasis on average effects rather than interrogating the “black box” of family effects puts collateral consequences researchers outside of the conversation with those who assess risk within the criminal justice system, especially in an era where risk assessment is increasingly viewed as a tool to help reduce prison and parole populations of “low-risk” individuals. ⁶

Third, reforms to criminal justice ignore the collateral consequences that accompany a criminal conviction and prison term. For example, the commutations (as opposed to pardons) recently offered by President Obama do not remove the civil disabilities associated with a felony conviction, and research shows these effects tend to be rather large (Apel and Sweeten 2010; Manza and Uggen 2006; Pager 2003; Pettit and Lyons 2007). Perhaps more importantly, commutations do not undo the time these individuals have spent away from their families and the subsequent stressors and instability this period of absence can create. As a result, while limiting time served may be desirable for some reasons, it may not be the most fruitful avenue for limiting the collateral consequences of incarceration (Green and Winik 2010; Kling 2006; Loeffler 2013; Ramakers et al. 2014), especially in light of research showing that even short jail stays are often incredibly destabilizing for families (Comfort, this volume).

Finally, research on incarceration effects, partially as a result of data limitations, often uses comparison groups of little use for policymakers. Many studies, again as a result of data limitations, simply compare the incarcerated with the nonincarcerated. Yet these comparisons tell us little about useful alternatives to incarceration, an issue that Signe Hald Andersen (this volume) elegantly lays out in her discussion of a Danish policy reform. It is not reasonable or politically viable to simply do away with criminal justice interventions altogether as a solution to criminal behavior. Instead, research can tell us about useful alternatives to incarceration by comparing the effects of community sentences, treatment programs, or diversion to standard prison sentences on both recidivism and families.

Concluding Thoughts

There is tremendous interest in criminal justice reform and, in our lives at least, there has been no moment with greater potential for change. In our various capacities and networks, each of the editors has struggled with aiding advocates of reform and producing research that is policy-relevant while remaining faithful to our identities as social scientists that do not overstep or overstate what our research can reasonably say. We would be remiss if we did not acknowledge that, at the end of the day, many studies show that even a short jail sentence is enormously destabilizing for inmates and their families—our time might arguably be better spent rethinking not only how we punish but why we do so. Moreover, if the goal is to protect families, we also need to think more critically about how we rehabilitate. Such a debate is worth having, but we also suggest that high quality research can be brought to bear on the question in a complex way. It is to that goal that this volume speaks.

With this volume we hope to contribute to the broader conversation about the harms to families and social life brought on by mass incarceration with an eye toward policy guidance that is honest about the depth of the challenge of reform. We do so with many motivations in mind, but our chief goal is to highlight complexity in the life experiences of prison inmates and their families. Part of the impetus for our collective research agendas is a disheartening but practical recognition that the children and partners of inmates may be more compelling objects of empathy to the broader public than inmates themselves. We worry about whether this is the case; the legacy of public policy in the United States over the past 40 years does not provide good evidence for widespread concern for the welfare of children among the American public or political class and instead suggests broad disinvestment in the life outcomes of vulnerable children and their families, particularly families of color. Mass incarceration must therefore be understood not in isolation but rather as one facet of a constellation of deepening disadvantages, social disinvestments, structural racism, and crushing deprivation. We continue to believe, however, that high quality and thoughtful research capable of clearly privileging one reform proposal over another can improve public safety as well as result in more humane and just outcomes for the imprisoned and their families.

In this volume, we focus on studies that speak to all of the above questions. Our contributions examine variation in sentence length and duration (L. Andersen), facility type (Wildeman, Turney, and Yi; Apel; Geller, Jaeger, and Pace) and alternatives to incarceration (S. H. Andersen); distinguish between characteristics of offenders from characteristics of their criminal history (Comfort; Wakefield and Powell); link the criminal justice system to other social welfare systems (Berger et al.); and detail a largely unrecognized burden of a criminal record in the form of online criminal records (Lageson). We also offer a variety of commentaries, each focused on a different aspect of variation in incarceration experiences (Maruna; Uggen), as well as overarching commentaries that focus on important issues of race and policy applications (Haskins and Lee; Sabol; Rodriguez). In so doing, we hope to add complexity to the study of criminal justice interventions and social effects. Just as Sampson (2011) called for a more concrete “incarceration ledger” to weigh the crime-reduction effects of incarceration against its social harms, we devote this volume to the creation of an incarceration ledger for family life, weighing the instability that incarceration introduces into families against incarceration’s potential to serve as a period of respite and reduce harm.