Toward Understanding Homelessness: The 2007 National Symposium on Homelessness Research. Incarceration and Homelessness. The Research Evidence:  What Do We Know?


Evidence collected so far supports perceptions that there is a tangible link between incarceration and homelessness. However, most of the evidence linking incarceration and homelessness is correlational, and cannot demonstrate that incarceration causes increased risk for homelessness, or vice versa. While there is a need for studies that are capable of assessing causality in this relationship, the associations demonstrated here in the high rates of homelessness among incarcerated populations, and the high rates of incarceration among homeless populations, are consistent with other bodies of research that highlight factors which explain why such high rates would exist. The research reviewed here documents specific and multiple barriers to housing among persons recently released from carceral institutions, and increased vulnerability for arrest and incarceration among homeless persons. This research not only supplies explanations for the high rates just reported, it also implies that addressing these factors could ameliorate the connections between homelessness and incarceration.

Barriers to Housing for Persons Who Have Been Incarcerated

There are structural as well as individual barriers to housing for soon-to-be-released prisoners. These barriers start even before release. For example, one fundamental obstacle to effective discharge planning in prison is that prisons tend to be located in rural areas, whereas most persons released from prison will return to urban areas hundreds of miles from the prison where they were incarcerated. This geographic mismatch renders it difficult to connect returning prisoners to the available housing market or for discharge staff and social workers to even attempt to provide housing assistance, as they are unlikely to have sufficient knowledge of the housing landscape to aid returning prisoners.

Oftentimes, however there will not even be adequate discharge planning and other support services available to incarcerated persons prior to their release. Survey data for state prison inmates from 1997 reveal that only 13 percent of soon-to-be-released inmates reported participating in pre-release programs (Lynch & Sabol, 2001). Most likely, an even smaller percentage receives housing-related assistance (e.g., counseling, search assistance, referrals to local housing providers, vouchers for rent, renter education, etc.) within these programs. An Urban Institute study tracking released prisoners in Illinois found that of clients who responded that they did not have a place to live lined up upon release, only 21 percent participated in pre-release programs. Of those who did participate in pre-release programs, almost half (45 percent) reported that finding a place to live was not covered in the program. In addition, for those who discussed finding housing in their program, only 39 percent received housing referral information (LaVigne et al., 2003). These numbers were similar to the findings of reentry studies in Ohio and Texas (Visher 2006; Visher & Courtney, 2006). The findings suggest that discharge planning involving the provision of housing-related services is rarely a standard part of the pre-release suite of services.

Discharge planning in jails is also frequently inadequate, as exemplified by the case Brad H. vs. City of New York. In this case, the New York State Supreme Court ordered the New York City Department of Corrections to provide comprehensive discharge planning prior to release for discharged individuals diagnosed with a mental illness. Prior to the Brad H. ruling, this class of persons released from jail was treated like other released inmates  they were dropped off in the city in the middle of the night with $1.50 and two subway tokens (Barr, 2003).

Upon release, financial instability can greatly reduce prospects for securing adequate housing. In a survey of housing providers that serve returning prisoners in the District of Columbia, respondents overwhelmingly reported that the inability of returning prisoners to secure jobs to provide income for rent was the greatest client barrier to housing (Roman, Kane, & Giridharadas, 2006). The majority of persons leaving prison have no savings, limited educational attainment and literacy skills, few or no job prospects, and no access to immediate unemployment benefits (Petersilia, 2000). Among those released from prison who are employed, the majority work in unskilled and low-wage jobs that are inadequate for meeting high housing costs, particularly in the urban areas where most prisoners live upon release (Western, 2002). Even among those exiting incarceration who do have skills and experience that would render them employable, having a criminal history poses a substantial barrier to employment, a barrier that is particularly daunting when accompanied by racial discrimination (Pager, 2003). Furthermore, employment opportunities become more restricted when criminal backgrounds, particularly a history of a felony conviction, bar people from being employed in a number of sectors.

After controlling for other factors, incarceration is associated with an earnings loss of anywhere from 10 to 30 percent (Grogger, 1995; Kling, 1999; Lott, 1990). However, economic disadvantages go beyond difficulties in securing employment. Benefits such as Supplemental Security Income, a key income support for persons with disabilities, are stopped while a person is incarcerated and application for reinstatement can only occur after release, often resulting in a wait in excess of several months before benefits resume (Blank, 2006). Persons with criminal histories can be restricted from receiving certain benefits (Legal Action Center, 2004). Many returning prisoners also face the added financial burden of supervision fees, child support, restitution, and other related costs, which often exceed average monthly income (Visher, La Vigne, & Travis, 2004).

Given these financial disadvantages, unsubsidized housing is often cost-prohibitive for persons who are released from incarceration (Scally & Newman, 2003). Many areas, especially urban areas, are witnessing increasingly tight rental markets, with limited numbers of units available for low-income households, particularly in neighborhoods accessible by public transportation. In such a market, individuals with criminal records are at a distinct disadvantage, having to compete with families and others who do not have criminal records and are thus deemed to be more desirable tenants.

These barriers to obtaining unsubsidized housing are coupled with restrictions on government subsidized housing that specifically deny ex-offenders access to public housing and the housing choice voucher program (Legal Action Center, 2004). This categorically puts the largest sources of affordable housing out of reach for many persons released from incarceration. These restrictions also limit the family support available to these persons if their families are living in subsidized housing, as their presence would put all household members at risk for eviction. Current policies pertaining to federal funding for specialized housing have provided little opportunity for housing to be developed to compensate for such restrictions. Ironically, even persons who were homeless prior to incarceration will have increased difficulty in accessing homeless services upon release. This is because persons released from incarceration, even if homeless prior to their incarceration, will not meet the standard criteria for being homeless and will have greater difficulty being eligible for programs targeting the homeless population.

Changes in sentencing structures also create barriers for returning prisoners looking for housing. As crime increased during the 1980s and early 1990s, many states adopted determinate sentencing and truth-in-sentencing laws. Determinate sentencing removed parole board discretion to determine when prisoners were deemed ready to be released. The number of prison releases that were discretionary, meaning that a parole board decided release decisions, dropped from 65 percent in 1976 to 24 percent in 1999 (Travis & Lawrence, 2002). Truth-in-sentencing structures mean that inmates can no longer attain good time toward release for good behavior until they have served at least 85 percent of their sentence. These individuals  those who max out  are not placed on parole supervision, but return to the community unsupervised and absent the support a parole officer might be able to provide (Petersilia, 2003).

Community supervision, most commonly in the form of probation or parole, may facilitate or impede obtaining safe and affordable housing. On the positive side, parole supervision means that, in most states, persons about to be released are required to have an appropriate home plan that demonstrates viable housing arrangements upon release. Receiving community supervision represents a potential source of support, as the supervised person may have access to social services and transitional housing facilities upon release. Even without the mandate to transition through a halfway house or community-based facility, supervision may be particularly helpful if parole and probation officers are knowledgeable about local housing options or housing-related programming. On the other hand, community supervision often includes a variety of restrictions, financial obligations, and conditions that may make it more difficult to maintain housing or employment.

Community opposition creates a number of barriers. Laws restrict the residential options of persons with criminal records, such as barring them from publicly subsidized housing or, particularly in the case of persons convicted of sex crimes, keeping them from living in circumscribed areas. Neighborhoods are often resistant to the siting of any housing that targets persons who have had contact with the criminal justice system. Due to the stigma that accompanies incarceration, the proposals to expand funding and services for reentry programs fail to become a legislative priority.

Finally, indirect and often intangible obstacles arise from the fragmentation that exists across and within service systems that potentially could support the transition from prison or jail to the community. Despite the involvement of corrections and community corrections programs (e.g., probation and parole), housing and homeless assistance providers, and general social services agencies, no single agency or organization is responsible for ensuring that individuals exiting prison are able to find safe and affordable housing. Furthermore, there is little collaboration among systems and little consistency over time. What results is a prisoner reentry system that is disconnected from the housing and homeless assistance services system and from the neighborhoods where released prisoners live. Historically, corrections systems have not provided discharge planning focused on ensuring that releasees have sufficient funds to travel to their destination after release, proper identification, or up-to-date paperwork to apply for public assistance that may have been suspended during incarceration.

Taken together, the barriers reviewed here are legion, and, regardless of whether they involve systemic or individual shortcomings, they can, by themselves or in combination, create conditions that unintentionally facilitate a path toward homelessness or reincarceration for persons upon their reentry into the community.

Incarceration Risks Among the Homeless Population

There are also reasons why persons who are homeless would be at higher risk for incarceration. Homelessness occurs on the economic and social margins of society. In this context researchers have pointed to a related liability for persons who are homeless to incur more arrests and subsequent incarceration for misdemeanors and a range of minor crimes. This is attributed to the public nature of a homeless existence and to attempts at controlling a population that is perceived as unruly, threatening, and offensive (Barak & Bohm, 1989; Snow, Baker & Anderson, 1989). Many aspects of homeless life have become restricted and criminalized, to where acts of subsistence and survival, especially in public places, are illegal and can lead to incarceration (Fischer, 1992; Foscarinis, 1996; Snow, Anderson, & Koegel, 1999, Eberle et al., 2000, Feldman, 2004). While these offenses are often minor, failure to pay fines or follow through with court appearances can lead to incarceration. Solomon and Draine (1995), show how arrest for lifestyle offenses such as trespassing among homeless persons with mental illness often lead to arrests for more serious charges such as burglary, which are likely to result in periods of incarceration.

Arrest rates have been high among homeless populations at least since the Skid Row era of the 1950s and 60s (Metraux, 1999), and there is a line of research that highlights the latent functions of this dynamic. Incarceration has been portrayed as a mechanism for exerting control and distance over the perceived threat to social order represented by the homeless population (e.g., Bittner, 1967; Spradley, 1970), a process Irwin (1985) describes as rabble management. Fischer (1992) points out the role of carceral institutions as sources of, among other things, housing, substance abuse treatment, and mental health care. Taken a step further, shelters, jails, and prisons can be seen as interchangeable waystations on a longer institutional circuit, where a series of institutions provide sequential stints of housing in place of a stable, community-based living situation (Hopper et al., 1997).

View full report


"report.pdf" (pdf, 242.57Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®