Focus Area: Employment

Individuals returning home from prison often identify employment as the most important factor that helped them stay crime free. While studies have shown that employment can help decrease the likelihood than an individual will reoffend and recidivate, research on the relationship between participation in employment programs and recidivism has yielded mixed results.

This section provides an overview and examination of key evaluative research investigating the relationship between reentry employment programs and recidivism reduction. Below are results and conclusions of research that met the criteria for methodological rigor and provide a basis for comparing and discussing effective employment programs emerging in the reentry field.

From the review of the literature, researchers identified 24 studies meeting the eligibility criteria. These studies fell into one of three broad categories of employment interventions – transitional employment programs, work release, and prison industries and institutional work. For information on vocational training programs, see the summary of research under the Education Focus Area

What the Research Says about Employment Programs

Eight programs were broadly categorized as providing transitional employment services:

Specialized Training and Employment Project (STEP), which provides job training and work assignment within the institution and post-release employment assistance in the community;
EMPLOY and Project RIO, which provide pre-release training and employment planning as well as employment assistance in the community post-release;
the Kintock Group, Inc. Employment Resource Center and STRIVE, both of which offer community-based job retention, employment readiness training, and job placement services;
the Center for Employment Opportunities (CEO) and the Transitional Jobs Reentry Demonstration, which both provide transitional jobs and other employment services; and
the National Supported Work Demonstration Project, which places participants in community-based work crews.

Findings on the effectiveness of these programs were mixed. The National Supported Work Demonstration Project, EMPLOY, and Project RIO did have significant, positive effects on recidivism and the latter two programs also demonstrated strong positive effects on employment. A study of CEO found modest evidence that the program lowered recidivism rates but did not find a program impact on employment. However, evaluations of four programs (STEP, the Kintock Group Employment Resource Center, STRIVE, and the Transitional Jobs Reentry Demonstration) found no evidence of a significant effect on reducing recidivism or helping people find or retain jobs. Clearly, the relationship between employment and recidivism is complex.

Researchers also identified several studies on both prison industry programs and work release programs which presented similarly mixed findings. In total, seven work release studies meeting the criteria were identified, which were roughly evenly split in terms of both recidivism and employment findings (i.e., half of the studies that examined each outcome found an effect, while the other half did not). Similarly, of the eight prison industry studies identified, four found evidence that the programs reduced recidivism, while the remaining four found no evidence of an effect. Finally, one study examined the effects of “prison jobs” - those associated with the operation of correctional facilities (e.g., laundry, cooking) on reentry outcomes, finding no effect on recidivism or employment.

Future Areas of Research

Overall, findings of the impact of employment programs on both recidivism and employment outcomes were mixed. Although several studies of both work release and prison industries programs were identified, and some were rated at a high level of methodological rigor, only one of the evaluations was able to utilize a randomized experiment. Furthermore, findings for both types of interventions were highly mixed. Clearly, more research is needed to examine why these programs do or do not work, and what factors contribute to their effectiveness or ineffectiveness. In addition, researchers have yet to adequately address the question of whether work release enhances employment outcomes for participants. Researchers found only three studies that included this measure, and the findings were conflicting. Moreover, it’s important to note that only one study evaluating each of the transitional employment programs was identified. Therefore, more research is needed on each of these programs, and researchers recommend exercising caution when interpreting the efficacy of these interventions.

Summary of Evaluations and Outcomes

Rigor Evaluation Recidivism Employment Substance Abuse
High Rigor Berk 2007 (not evaluated)
High Rigor Bloom, 2009; Redcross et al, 2010; Jacobs, 2012 (not evaluated)
High Rigor Bohmert, Duwe, and Minnesota Department of Corrections 2010, 2011 (not evaluated)
Basic Rigor Callan & Gardner 2005; 2007 (not evaluated) (not evaluated)
Basic Rigor Cox 2009 (not evaluated)
High Rigor Crouch et al, 1989; Blakely et al, 1991; Menon et al, 1995; Menon et al, 1992; Finn, 1998 (not evaluated)
High Rigor Drake 2003 (not evaluated)
High Rigor Drake 2007 (not evaluated) (not evaluated)
Basic Rigor Duwe, 2011 (not evaluated)
High Rigor Duwe, 2014a; Duwe 2014b (not evaluated)
Basic Rigor Farabee, Zhang & Wright (2014)
High Rigor Flanagan et al. 1988 (not evaluated) (not evaluated)
Basic Rigor Jengeleski & Gordon 2003 (not evaluated) (not evaluated)
Basic Rigor Johnson 1984 (1) (not evaluated)
Basic Rigor Johnson 1984 (2) (not evaluated)
Basic Rigor Johnson 1984 (3) (not evaluated)
Basic Rigor Johnson 1984 (4) (not evaluated)
Basic Rigor Lattimore et al. 1998; 1990 (not evaluated)
Basic Rigor Lee 1983 (not evaluated) (not evaluated)
High Rigor Leonard 2001 (not evaluated) (not evaluated)
High Rigor Redcross et al., 2006- 2012 (not evaluated)
High Rigor Richmond 2009 (not evaluated) (not evaluated)
High Rigor Sabol 2007 (not evaluated) (not evaluated)
High Rigor Saylor & Gaes 1985-1999 (not evaluated)
Basic Rigor Smith, Bechtel, et al. 2005, 2006 (not evaluated)
Basic Rigor Turner & Petersilia 1996 (not evaluated) (not evaluated)
High Rigor Uggen 2000 (not evaluated) (not evaluated)
Basic Rigor Van Stelle, Lidbury, & Moburg 1995 (not evaluated)