Association for Public Policy Analysis & Management

We estimate the effect of obtaining a General Educational Development (GED) certification on the post-release labor market outcomes of ex-offenders released from state prisons. Our findings correspond to prior research that suggests at most modest effects of obtaining a GED on post-release outcomes. We also observe distinct effects of the GED on white and black prisoners and that the GED effect fades out within a few years of release.

We use regression discontinuity and difference-in-differences research designs to estimate the causal effect of a GED earned in prison. To facilitate our analysis, we construct an analytic data set including a census of all prisoners who have been released from a Midwestern state’s prisons over a 20-year period (N ≈ 150,000) using administrative micro data from two state agencies. These data contain detailed assessments of medical and mental health needs, substance abuse history, and offense types from prison records, as well as wage records from the state UI system. Importantly for our study, the data include detailed records of GED test taking while in prison, including exam scores that we use as the forcing variable in a regression discontinuity design.

This study is motivated by high rates of prison recidivism that suggests rehabilitative expenditures for inmates may be inefficient. Most ex-offenders released from state prisons are re-arrested within just a few years after their release. While ex-offenders face a number of challenges post-release, finding a job can be particularly difficult. A robust set of observational and experimental studies consistently finds that time in prison is a major barrier to ex-offenders when seeking work. In response, there has been a surge in federal, state and privately funded educational programs. These educational and skill programs are popular in prison because they are expected to increase inmates’ post-release employability, which can reduce recidivism by increasing the opportunity cost of future crime and facilitating re-connection to communities.

Over 90% of state prisons and all federal prisons have some sort of educational program for inmates and roughly half of all state inmates will participate in an educational program. One of the most popular educational programs prepares inmates to pass the GED exam, which is an academic certification for those without a high-school diploma. Over 10% of the GED credentials issued each year are from correctional institutions and this proportion rises to 22% for black males. Even with their popularity, however, existing research on the labor market returns to the GED in general are not promising. For ex-offenders in particular, there is limited empirical evidence to demonstrate whether GED completion matters for post-release outcomes. Our study contributes to this knowledge gap by presenting causal estimates of the labor market returns to a GED earned in prison using new and unique data.