Over the past 25 years the United States has seen large increases in both incarceration and homelessness. The jail and prison population went from approximately 500,000 in 1980 to 2.1 million in 2004 (Pastore & MacGuire, 2005), while the homeless population transformed from a small collection of individuals stereotyped as bums and winos to a diverse assortment of families and individuals that, according to best estimates, now include at least 2.3 million who are homeless at some point in a year (Burt et al., 1999). Little is known, however, about the relationship between these two concurrent phenomena. Although service providers have long pointed to anecdotal evidence about the overlap between these two populations, awareness of this nexus from a research perspective is relatively recent and in its nascent stages.
This paper presents evidence for evaluating two assumptions. The first assumption is that persons who are homeless are at increased risk for incarceration and, conversely, release from jail or prison leaves a person particularly vulnerable to an episode of homelessness. Much of the research on the homelessness-incarceration nexus is still documenting parameters. Specifically, this includes the rates at which people cross over from one to the other; the proximate factors associated with an increased probability of such crossovers; and more general explanations of why such a high degree of crossover exists.
The second assumption concerns the centrality of housing, coupled with supports (in whatever form they may take) that help to maintain this housing, in preventing both homelessness and incarceration among persons at risk for both. For those who lack the resources and supports to obtain secure housing upon release, providing such housing stands to mitigate the risks for both homelessness and reincarceration. Evaluating this assumption involves examining the more general experience faced by persons who reenter the community from jails and prisons, as well as examining empirical evidence on outcomes related to interventions that involve housing and supports. Service and housing providers have given limited attention and resources to addressing the needs presented by persons with histories of both incarceration and homelessness. When they have done so, the tendency has been to adapt other models rather than to develop specific interventions focusing on the specific problems presented by this population. Interventions that do exist, whether adapted from other housing models or designed specifically to address the needs of this population, typically are ahead of the research literature on best practices. This paper will review the range of housing approaches, featuring specific housing programs within this continuum, and what, in the absence of extensive evidence in this field, is considered to be best practice.
In describing what will be covered, it is also necessary to outline the limits as to what will be covered here. Community reentry (for persons released from incarceration) and homelessness are both broad topics that touch on a range of other topics. Focusing on the immediate nexus of these two topics necessarily steals attention from areas with a less direct bearing. For example, responses toward this problem do not, by and large, address more general policies regarding incarceration and housing, although this will ultimately be the solution to this problem. Similarly, more general topics such as employment, healthcare, education, and stigma figure into understanding this nexus, but are only touched on in this paper. And some related topics, such as how incarceration of an adult may lead to collateral homelessness for family members, are important but lack research and interventions that provide clearer understanding. This need for future research is addressed in the final section. Opportunities here abound, as knowledge of the nature of the problem and evidence to support current intervention practices still contain numerous gaps with respect to many key questions.
Policy and Institutional Context
This link between incarceration and homelessness can be viewed as a second wave of deinstitutionalization. Deinstitutionalization is a term traditionally used in reference to the exodus of persons treated for mental illness from psychiatric hospitals to the community (i.e., the first wave), and it has parallels to the more recent interactions between carceral institutions and homelessness. The problematic implementation of deinstitutionalization left many persons with mental illness to enter the community unprepared and unsupported, and has been widely thought to be the reason why, from the 1980s on, persons with mental illness figured prominently among the rolls of the homeless population. By comparison, rates of shelter use have been found to be higher among people exiting prison than among people exiting state psychiatric hospitals (Metraux & Culhane, 2004). The number of people exiting prisons and jails to the community was 650,000 and 9 million, respectively, in 2004 alone (Brown, 2006; Harrison & Beck, 2006); just the number of those released from prison alone dwarfs the number deinstitutionalized from psychiatric hospitals (Mechanic & Rochefort, 1990).
Actors involved in this current round of deinstitutionalization involving the criminal justice system can learn from some of the missteps of the previous round of deinstitutionalization involving the mental health system. One key lesson from deinstitutionalization of persons from psychiatric hospitals has been the importance of housing. For the deinstitutionalized mentally ill population, housing had been viewed by community mental health services as a public welfare function, and was largely ignored until homelessness became linked with mental illness. Only as a result of this link did there emerge a consensus in the community mental health field that housing is, in fact, a mental health service and a prerequisite to effectively providing other forms of community-based services (Metraux, 2002). This is an example of where the criminal justice system appears to have learned little from the mistakes made in deinstitutionalizing persons who are mentally ill. The lack of jurisdictional clarity over the problem of post-incarceration homelessness means that people who are homeless at the point of their discharge from incarceration fall under the purview of neither the corrections system, which views its jurisdiction over inmates as ending at discharge, nor the homeless assistance system, as individuals leaving institutions are not considered presently homeless and are therefore ineligible for most forms of homeless assistance. Cho (2004) attributes this jurisdictional gap to a condition of isolationist policymaking, in which sectors of government define their spheres of responsibility too narrowly thus leaving some individuals to become institutional refugees. As Black and Cho (2004) explain, the result is ultimately a scarcity of public funding and resources that target persons who are homeless upon their release from incarceration. As a result, people leaving incarceration enter an uncertain transitional space between institution and community in which services are fragmented at the point where they are most vulnerable (Hopper & Baumohl, 1994).
What Has Changed Since 1998?
Criminal justice involvement among people who are homeless is hardly new: jails and detention facilities have historically served as de facto institutions for persons who were homeless when they were picked up either for violating vagrancy laws or as a benevolent means of quartering (Hopper, 2003). Likewise, shelter operators and other homeless housing providers have long reported seeing high rates of people with recent experiences in correctional settings among their clientele. Some providers of homeless shelters have anecdotally reported rates of formerly incarcerated people as high as 70 percent (Cho, 2004), while a national survey of providers of homeless services conducted in 1996 found that [a]ltogether, 54 percent [of persons receiving homeless services] have some experience of incarceration (Burt et al., 1999).
What is new is a growing level of concern. In terms of reentry, this concern has manifested itself in a changing political climate in which there is a greater receptiveness towards attending to problems related to reentry (Suellentrop, 2006). A recent analysis by Jacobson (2005) suggests that developments in the current political climate may further facilitate efforts to increase programming to address the needs of the formerly incarcerated. In this argument, the huge swell in the prison population, negative public opinions about crime and public safety, and interest in curbing or rethinking public spending practices all create a window of opportunity for policymakers and leaders to create and implement programs that hold the promise of slowing incarceration rates, reducing demand for emergency public services, and ultimately saving or making better use of public dollars. Jacobson argues that evaluation of existing practice for its cost-effectiveness with respect to corrections and other public system utilization is critical, thus furthering the case for supporting evaluation research.
Attending to the needs of persons with histories of incarceration has become a more bipartisan issue, with the Bush Administration first providing $100 million in funds towards reentry initiatives in 2001 under the Severe and Violent Offender Reentry Initiative, and then providing a major impetus for action with Bushs call, in what would become known as the Prisoner Reentry Initiative, for allocating $300 million in funding towards reentry initiatives in his 2004 State of the Union Address. This was followed by the Second Chance Act, a bill that proposed allocating $100 million over two years to help states address reentry issues and that narrowly missed passage by Congress in 2006. This bill represents a start, as considerably more resources would be needed to match the magnitude of the reentry problem. But such beginnings encourage hope that the policy atmosphere will be more open to addressing the needs of those released from jails and prisons now than it has been during the decades-long growth in the incarcerated population.
In the last several years there has also been increased policy emphasis on ending (as opposed to managing) homelessness. More than 200 communities around the country have recently committed themselves to 10-year plans to end homelessness (Interagency Council on Homelessness, 2006; Cunningham et al., 2006). A particular target for many of these plans is the chronic elements of this homeless population. Chronically homeless refers to persons who have been homeless for extended periods, often have one or more disabilities, and disproportionately use other public services and institutions, including jails and prisons. These plans to end homelessness are increasingly seeking to bypass emergency shelters and transitional housing, instead placing persons who are homeless directly into permanent housing with support services, when needed. Insofar as these renewed efforts at addressing homelessness have the capacity and the will to specifically respond to incarceration, this policy focus also promises to be receptive to ameliorating the nexus between homelessness and incarceration.
Both from the reentry and homelessness perspectives, there are grounds to believe that increased attention will be focused on addressing the nexus of incarceration and homelessness. This is, however, still an issue in its infancy. As such, there is a particular need for research that outlines the parameters of this problem and provides evidence for what approaches can effectively address this problem. It is these areas that provide the foci for this paper.