The impact of prison labor and education programs and dosage effects on recidivism

As the number of prisoners increases, the cost of their incarceration continue to increase and governments fight to provide services. Any decrease in the tendency of released prisoners to return to crime and prison (recidivism) can lead to significant cost saving. Bonczar notes that “If incarceration rates remain unchanged, 6.6% of U.S. residents born in 2001 will go to prison at some time during their lifetime.” A review of the literature shows that economists have paid little attention to this important topic.

Having prisoners work and/or participate in education programs addresses two possible causes of incarceration and recidivism, poor job skills and poor education.  In the present study, we confine our attention to recidivism and focus on the effects on recidivism of two types of work, prison industry employment (prison labor), and institutional support employment (for example, laundry or kitchen work) (prison jobs), and prisoner involvement in education programs.  A proportional hazards model is used to analyze their impacts over a ten year period using a data set including information on prisoners released from the Ohio prisons in 1992.  The length of post-release time, the inclusion of interaction terms for the three programs, and the estimation of the net savings from the programs make this study a significant contribution to the understanding of recidivism.  Using work from the pharmaceutical industry, the dosage effect (the length of the experience) will be introduced into the propensity score to study the effective length of participation.

Initial findings indicate that internal prison jobs led to savings in the range of $10,400 to $14,600 per prisoner employed.  Adding either prison industry participation or education to having a prison job raises savings to between $11,600 and $17,500.  Finally, with participation in all three programs, Ohio saved an estimated $17,600 to nearly $23,000 on the average male inmate due to these programs’ ability to postpone or eliminate the return to prison.  This initial analysis is being expanded to investigate the impact that length of participation has on both recidivism and savings.  We are expecting an increasing marginal utility of participation in programs up to some dosage level followed by diminishing marginal utility as the dosage increases.