Final Technical Report: Neighborhoods, Recidivism, and Employment Among Returning Prisoners
Morenoff, Jeffrey, and David Harding. 2011. “Final Technical Report: Neighborhoods, Recidivism, and Employment Among Returning Prisoners.” (Report Submitted to the National Institute of Justice. Grant Award 2008-IJ-CX-0018.)
The rising number of individuals being released from prison has prompted renewed interest among researchers, policy makers, and practitioners in reintegrating former prisoners. Yet relatively little is known about the communities into which former prisoners return and how they affect the likelihood that former prisoners will secure stable employment or return to prison. This research fills an important gap in the literature on prisoner reentry by focusing on the role that community context plays in the labor market outcomes and recidivism of former prisoners. A rich set of longitudinal administrative records were assembled on individuals paroled in Michigan during 2003, including records from corrections, police, and unemployment insurance databases. This report describes the data collected and presents results indicating that neighborhood context predicted both the recidivism and labor market outcomes of former prisoners. The analysis considered the association between baseline neighborhood characteristics (first post-prison neighborhood) and cumulative exposure to neighborhood conditions during one’s time on parole. The analysis of baseline neighborhood characteristics was based on the full population of 11,064 people released on parole in Michigan in 2003, whereas the analysis of time-varying neighborhood characteristics was based on a 1/6 sample (n=1,848). Returning to a more disadvantaged baseline neighborhood was associated with higher risks of absconding and returning to prison for a technical violation, a lower risk of being arrested, and more adverse labor market outcomes, including less employment and lower wages. Cumulative exposure to disadvantaged neighborhoods was associated with lower employment and wages but not related to recidivism. Returning to a more affluent baseline neighborhood was associated with a lower risk of being arrested, absconding, and returning to prison on a technical violation, and more positive labor market outcomes, including greater employment and wages. However, cumulative exposure to affluent neighborhoods was not significantly related to any of the recidivism or labor market outcomes when the full set of controls were added to models. Returning to a more residentially stable baseline neighborhood was associated with a lower risk of absconding and returning to prison for a new conviction, but not with any labor market outcomes; nor was cumulative exposure to residentially stable neighborhoods associated with any recidivism or labor market outcomes. Returning to a baseline neighborhood with a younger age structure was negatively related to the odds of returning to prison on a technical violation, but when measured as cumulative exposure it was associated with an increased risk of being arrested, absconding, and being returned to prison for either a new commitment or technical violation. Being employed substantially reduced the risk of all recidivism outcomes, but there was no evidence that employment mediated the association between neighborhoods and recidivism. Together, these results suggest that the neighborhoods parolees experience during parole were strong predictors of recidivism and labor market outcomes, but there is not a simple answer to the question of what neighborhood characteristics constitute “risky” environments for parolees. Neighborhood socioeconomic composition was a strong predictor of labor market outcomes, as parolees residing in disadvantaged neighborhoods had difficulty securing employment and escaping poverty. For recidivism, the protective effect of living in a residentially stabile neighborhood and the risks posed by spending more time in neighborhoods with higher densities of young people were the most robust predictors. From a policy perspective, these findings suggest that parole outcomes might be improved through more careful evaluation of a parolee’s neighborhood context when approving new residences, placement of institutional housing for former prisoners in more advantaged neighborhoods, inclusion of neighborhood context in risk assessments to better target services to former prisoners in high risk neighborhoods, and place-based parole strategies involving geographically based agent caseloads.