Focus Area: Housing

Securing housing is one of the most immediate challenges individuals leaving prison face upon their release. Research has shown that the types of living arrangements and neighborhoods to which exiting prisoners return are often related to the likelihood that they will recidivate and return to prison. While many of the formerly incarcerated stay with family members - at least early on, others are confronted by limited housing options. This is especially true for those with mental health or substance abuse problems. Obtaining housing is complicated by a host of factors, including the scarcity of affordable and available housing, legal barriers and regulations, landlords’ prejudices against formerly incarcerated individuals, and strict eligibility requirements for federally subsidized housing.

This section provides an overview and examination of key evaluative research investigating the relationship between housing programs and recidivism reduction. Below, researchers highlight the results and conclusions of research that met the criteria for methodological rigor and provide a basis for comparing and discussing effective strategies for overcoming barriers to housing access that have emerged in the reentry field in recent years.

What the Research Says about Housing Programs

From a thorough review of the literature, researchers identified just seven studies of reentry programs involving housing that met the criteria for inclusion. Five studies evaluated halfway house programs, one assessed a more comprehensive supportive housing program and one examined a statewide housing voucher program.

Findings were mixed across the five halfway house program evaluations. Two studies found no evidence that the programs evaluated impacted recidivism rates. In one of the studies, halfway houses were compared to electronic monitoring, and no difference in recidivism was found. The second study evaluated the Ridge House program in Nevada, again finding no effect on recidivism. A third evaluation examined the impact of the length of stay in residential aftercare programs following participation in Pennsylvania’s Motivational Boot Camp Program. Though the finding only approached significance, results suggested that residence in a halfway house for 90 days was associated with lower arrest rates after two years compared to individuals who stayed for 30 days or not at all.

Further, Latessa and Lowenkamp’s study of 37 halfway house programs in Ohio, which met the high standard of rigor, found that halfway houses can have a significant impact on reducing recidivism rates, particularly when targeted to higher risk individuals. The effect of the programs on recidivism was largely determined by a participant’s level of risk to recidivate and by the quality of the particular program. While halfway houses were found to reduce recidivism for individuals assessed at a high risk of recidivism (as compared to high-risk individuals in the comparison group), those at a low risk to recidivate actually experienced higher rates of recidivism than their low-risk control group counterparts if placed in a halfway house.

In addition, recidivism impacts varied based on the particular halfway house’s characteristics, including the types of services it offered, the quality of its staff, and whether it used assessment tools. This variety across halfway houses, coupled with the fact that populations of differing risk levels experienced different impacts, could provide a possible explanation for the fact that the other halfway house studies reviewed did not find effects on recidivism. Further, Latessa and Lowenkamp conducted a subsequent halfway house program evaluation in Pennsylvania which found that residence in one of their Community Corrections Centers or Community Contract Facilities was associated with higher recidivism rates. The researchers noted quality concerns in many facilities as well as the absence of risk-based targeting of halfway houses to those at higher risk.

The two studies of supportive housing programs also demonstrated mixed results. Although an evaluation of Washington state’s three-month housing voucher program showed no evidence of an impact on recidivism outcomes, an evaluation of Returning Home – Ohio, a supportive housing program targeting services to disabled former prisoners returning to Ohio communities, found promising results. Participants in the program were less likely to be rearrested than those in the comparison group and their time to rearrest was longer as well. Notably, however, program participants were rearrested more times than their counterparts in the comparison group.

Future Areas of Research

Given that few studies of reentry housing programs were identified that met the standards for methodological rigor, further high-quality research is needed in this area. Though many studies were identified in the comprehensive literature search, most did not employ strong research designs, necessitating their exclusion. Many other studies were identified that examined housing programs qualitatively but did not examine quantitative reentry outcomes. In particular, studies examining housing programs that adhere closely to a program model (as documented in a process evaluation) would be helpful in determining which types of housing programs show the most promising results, and for which populations they are the most effective.

Summary of Evaluations and Outcomes

Rigor Evaluation Recidivism Employment Substance Abuse
Basic Rigor Fontaine et al, 2012; Fontaine et al, 2009; Markman et al, 2010 (not evaluated) (not evaluated)
Basic Rigor Klein-Saffran 1992 (not evaluated)
Basic Rigor Kurlychek et al., 2011; Kurlychek & Kempinen, 2006 (not evaluated) (not evaluated)
Basic Rigor Latessa, Lowenkamp, & Bechtel, 2009 (not evaluated) (not evaluated)
High Rigor Lowenkamp and Latessa 2002-2007 (not evaluated) (not evaluated)
Basic Rigor Willison et al. 2010 (not evaluated) (not evaluated)