Research on Returning Offender Programs and Promising Practices

Prepared remarks by NIJ Director David B. Muhlhausen given at the Secretaries Innovation Group.

June 13, 2018

It’s a rare event to have so many influential parts of the government convened in one room, and I’m honored to have the opportunity to address you today.

My name is David Muhlhausen and I am the director of the National Institute of Justice, which is the research, development, and evaluation arm of the Department of Justice.

I’ve also been recently appointed by the Attorney General to be the Executive Director of the President’s Federal Interagency Council on Crime Prevention and Improving Reentry.

I’d like to talk about the research on returning offender programs and promising practices in this space. I hope my remarks can provide helpful background on what we know when it comes to adult reentry and recidivism, and then I’d like to take some time to open things up for discussion.

Corrections is an area of criminal justice particularly close to my heart, because I was a manager at a juvenile correctional facility in Baltimore. Before joining NIJ last fall, I spent 18 years with The Heritage Foundation, where my research areas included reentry, recidivism, and program evaluation.

Recidivism in American and the Federal Interagency Council on Crime Prevention and Recidivism

Recidivism is an important issue for the American criminal justice system today. We see offenders cycle in and out of prisons and jails, and this has important implications for public safety and the stability of American communities.

Ninety-five percent of inmates in state and federal prisons will be released eventually, and studies have consistently shown high rates of recidivism.[1] A Bureau of Justice Statistics report published in May found that 83 percent of state prisoners released in 2005 were rearrested at least once within nine years of their release.[2] That means that five out of every six of the prisoners released will be arrested for a new crime.

These high rates of recidivism can have a serious impact both on public safety and on the individual lives of offenders trapped in the revolving doors of America’s prisons.

The silver lining of these shocking statistics is that there’s a lot of room for improvement when it comes to crime prevention and offender reentry in America. There’s room for us to do important work in this space, and this administration is taking decisive steps to help us get there.

In response to these issues, President Trump issued an Executive Order in March 2018 that established the Federal Interagency Council on Crime Prevention and Improving Reentry, also known as the Reentry Council. The Reentry Council is comprised of a dozen agencies across the federal government including the Justice Department, Education, Veterans Affairs, Housing and Urban Development, just to name a few.

Serving as the Executive Director of this Council has been a big honor for me, and an even bigger responsibility.

The Reentry Council is mandated to develop recommendations for evidence-based reforms that will prevent crime, facilitate reentry, and reduce recidivism.

Our job is to produce recommendations based on research that will work to improve public safety, save taxpayer dollars, strengthen public trust in the corrections system, and provide former prisoners with greater opportunities to access services and live productive lives.

We’ve gotten off to a busy start. Over the last few months, we’ve convened meetings with external collaborators and stakeholders to understand their needs around crime prevention and reentry. We’re in the final phases of releasing a 90-day report that will include an overview of the problems of crime prevention and reentry, the work that our partner agencies have done in this space, and what they will do over the next 90 days. The report will also outline what the research and evaluation has told us about the effectiveness of various programs, and appropriate next steps moving forward.

The key in all of the Reentry Council’s work is how we define and use the term “evidence-based.” We’re not developing a list of programs that make us feel good or anecdotally seem to work. We aim to develop recommendations by looking at the empirical evidence and working to understand what programs rigorous research has shown to be effective, promising, or ineffective. When we have an understanding of program effectiveness, we can support and assess what works, continue to research promising initiatives, and discourage programs that don’t work.

Our work at the Reentry Council is ongoing, but I’d like to speak today about some of the research that will help guide and inform our work.

The Need for Rigorous Research Design

All too often, there’s a lack of empirical research to support claims that a program works or doesn’t. As director of the Department of Justice’s research agency, I have no patience for anecdotal or non-rigorous evidence.

Methods matter! This is particularly true for high-stakes issues like crime prevention and reentry. If a state adopts reforms and crime subsequently drops, we can’t consider that crime drop to be rigorous evidence that the reform is effective. If returning members of society report that they found a program helpful, this is only anecdotal evidence at best. It’s not representative of rigorous science. Anecdotal stories, rather than data and evidence, too often drive the narrative in what programs are deemed to be “evidence-based.”

One of the biggest issues we see in determining what works in reentry is the need for more randomized controlled trials.

Randomized controlled trials are the gold standard in research and evaluation. They provide the most rigorous evidence of what works because they include both treatment and control groups, and the samples in each group are evenly matched. Any research method less than a randomized controlled trial can be considered a weaker design.

We often see mixed results in evaluations of crime prevention and reentry programs. As we consider how to translate these results into future funding decisions, it’s important to consider the methods and rigor of the research.

I cannot overemphasize the importance of rigorous research methods when it comes to program evaluation, namely randomized controlled trials. These trials aren’t feasible to examine all aspects of the criminal justice system, but I strongly believe that we need to use them whenever we can, particularly when it comes to programs with the goal of facilitating successful reentry and reducing recidivism.

We have a long history of throwing money at problems without a good understanding of whether the programs are effective. Some of the most important work of the Reentry Council in our first months has been to convene all relevant government actors and conduct research reviews to develop an understanding of what we know and what we don’t know when it comes to crime prevention and reentry.

Defining Reentry Program and Evidence-Based Programs

For the purposes of this review of the scientific literature, prisoner reentry programs are defined as programs that provide services to recently released inmates.

By this definition, corrections programs that provide services only to currently incarcerated individuals are excluded from this review.

However, programs that provided services inside prison and outside correctional institutions are included as prisoner reentry programs.

Evidence-based programs are programs that have been demonstrated as effective by high-quality outcome evaluations that have been replicated and evaluated in at least three sites.

This replication is crucial!

High-quality outcome evaluations are those using rigorous, randomized controlled trials on programs implemented with fidelity.

Do Reentry Program Work?

So, what does the literature say about reentry programs and recidivism? The short answer is that it’s complicated. Due to the lack of RCTs and replications, no reentry programs meet the “evidence-based” definition.

We don’t have a strong understanding of what works and what doesn’t, and there’s a pressing need for additional research to help us better understand the dynamic process of reentry.

Overall, the results of RCTs of reentry programs for reducing recidivism are mixed, at best.

To date, much of the discussion about how to help formerly incarcerated individuals is focused on employment and training assistance. Various types of employment programming are commonly implemented as reentry programs. I want to spend a few minutes talking about what we know regarding the effectiveness of employment programs.

In a nutshell, there are very few employment programs that show promise for reducing recidivism.

In particular, employment-focused reentry programs have had little success in reducing recidivism.

This means that prisoner reentry efforts that rely mainly on job training and subsidized jobs are not likely to succeed.

Despite these findings, employment-focused programs are seen as a panacea. We should not support them expecting that they will serve as a cure-all.

Employment is a big part of an individual’s needs as they return to society, but we know that successful reentry is a lot more complicated than just finding a job.

The Second Chance Act

Table 1: Evaluation of Seven Second Chance Act Adult Demonstration Programs
MeasuresRecidivism Results
 18-Month Follow-Up30-Month Follow-Up
ArrestsNo EffectNo Effect
ConvictionsNo EffectNo Effect
IncarcerationsNo EffectNo Effect

Study Design: Randomized Controlled Trial
N=966 from seven sites
Source: D’Amico and Kim, 2018

A multisite RCT of the Second Chance Act (SCA) reentry programs funded by the U.S. Department of Justice found that the comprehensive services offered failed to reduce recidivism.[3]

The programs included the Kentucky Department of Corrections; Oklahoma Department of Corrections; South Dakota Department of Corrections; Allegany County, Pennsylvania Department of Human Services; Marion County, Oregon Sheriff’s Office; San Francisco Department of Public Health; and San Mateo County, and the California Division of Health and Recovery.

Members of the treatment group were significantly more likely to receive case management, employment assistance, and cognitive behavioral therapy than their counterparts in the control group.

At the 18- and 30-month follow-up periods, SCA programs had no impact on arrests, convictions, and incarcerations.

Randomized Controlled Trials of Employment-Focused Reentry Programs

An RCT of the U.S. Department of Labor’s Reintegration of Ex-Offenders (RExO) Programs assessed the effectiveness of federal grants to 24 local employment-based reentry programs, many of which were operated by faith-based organizations.[4]

At the two-year follow-up, the RExO services failed to affect arrests and incarceration rates.

The Center for Employment Opportunities (CEO) Prisoner Reentry Program is an employment-based program that immediately places recently released prisoners in transitional jobs, usually with New York City or New York government agencies.

At the two-year follow-up period, CEO had no effect on arrests, although CEO did reduce convictions and incarcerations.[5] However, these successful results did not continue during the three-year follow-up.[6]

The Transitional Jobs Reentry Demonstration, a large-scale RCT, tested the effectiveness of providing temporary paid jobs, support services, and job placement assistance to prisoners returning to communities in four large Midwest cities.[7]

At the two-year follow-up, the transitional jobs approach failed to affect all measures of recidivism, including arrests, convictions, prison admissions, and technical violations.

A small-scale RCT of a recognized employment-focused prisoner reentry program operating in Southern California was found to have no statistically measurable effect on recidivism. [8] The program used the STRIVE employment model, which is provided by more than 25 community-based organizations throughout the nation.

The Milwaukee Safe Street Prisoner Release Initiative, operated by the Wisconsin Department of Corrections, provided “reach in” services, including drug abuse counseling, anti-gang counseling, and employment preparation, to inmates six months prior to their release, along with after-release employment-focused services.[9]

After the one-year follow-up, a small-scale RCT found that the program reduced arrests, but did not affect incarceration outcomes.

During the early 1990s, a small-scale evaluation in Seattle randomly assigned 218 eligible prisoners to serve out their sentences or enter work-release facilities. [10]

Participants in the intervention group were required to be involved in gainful employment or job training while participating in the program.

At the one-year follow-up, the program has no effect on arrests and incarceration.

Table 2: Evaluations of Employment-Focused Reentry Programs
Randomized Controlled Trial
Sample Size/SitesMeasuresRecidivism Results
 1-Year Follow-Up2-Year Follow-Up3-Year Follow-Up
Reintegration of Ex-Offenders (RExO) Program
(Wiegand et al., 2015)
24 Sites
 Arrests No Effect  
 Incarceration  No Effect 
Center for Employment Opportunity Prison Reentry Program
(Redcross et al., 2009; 2012)
New York City
Arrests  No Effect No Effect
  Convictions Beneficial EffectNo Effect
  Incarceration Beneficial EffectNo Effect
Transitional Jobs Reentry Demonstration
(Jacobs, 2012)
Chicago, Detroit, Milwaukee, St. Paul
 Arrests No Effect 
  Convictions No Effect 
  Prior Admissions No Effect  
 Technical Violations   No Effect  
Southern California Employment-Focused Reentry Program
(Farabee et al., 2014)
Southern California City
ArrestsNo Effect   
Milwaukee Safe Streets Prisoner Release Initiative
(Cook et al., 2015)
ArrestsBeneficial Effect   
 IncarcerationNo Effect  
Washington State Work Release (Turner and Petersilia, 1996) N=218
ArrestsNo Effect   
 IncarcerationNo Effect

NIJ’s is a clearinghouse for rigorously evaluated criminal justice programs. It includes a number of reentry programs that have been rated “effective,” or show “promise” as being effective in facilitating successful reentry and reducing recidivism. It also can warn against implementing programs that have been shown to have no effects.

However, I do have some concerns with

First, I have serious doubts about the number of programs rated as “effective” based on quasi-experiments.

If you look at the 97 entries for Corrections and Reentry programs that attempt to reduce recidivism, you will find four programs rated as “effective.” Yet, none of these programs was assessed for effectiveness by an RCT.

For 56 programs rated as “promising,” 11 — or 20 percent — were assessed by RCTs and 45 — or 80 percent — were assessed by quasi-experiments.

Alternatively, 37 corrections and reentry programs are rated as “no effects.” More than 56 percent — or 21 of these programs — have been evaluated by at least one RCT.

Table 3: Ratings of Corrections & Reentry Programs (Reentry and Release)
RatingRCTQuasi-ExperimentalNatural ExperimentTotal
Effective0 (0%) 4 (100%)0 (0%)4
Promising11 (19.6%)45 (80.4%)0 (0%)56
No Effects21 (56.8%)15 (40.5%)1 (2.7%)37

This imbalance has led me to be very skeptical of quasi-experimental designed studies that find corrections and reentry programs to be “effective.”

We also need to do more replication studies, particularly on the programs we’ve rated as “effective” and “promising.” is on the right track, but it still has a long way to go.

Promising Alternatives to Employment-Based Programs

Despite lackluster findings in employment-based reentry program evaluations, rigorous evaluations of reentry programs and services have shown us a number of promising alternatives to employment programs. Promising programs include the Comprehensive Offender Reentry Plan, High-Risk Revocation Reduction Program in Minnesota, and the Amity In-Prison Therapeutic Community. However, the results of these RCTs have never been replicated, so we need to be cautious in interpreting these results.

Table 4: Minnesota Comprehensive Offender Reentry Plan
MeasuresRecidivism Results
16-Month Follow-Up
ArrestsBeneficial Effect
ConvictionsBeneficial Effect
IncarcerationBeneficial Effect
Technical ViolationsNo Effect

Study design: Randomized Controlled Trial
N=269 from five Minnesota counties
Source: Duwe, 2012 [11]

The good news is that there are promising programs. The Minnesota Comprehensive Offender Reentry Plan connected caseworkers in prisons with supervision agents in the communities that participants returned to upon their release from prison. Rigorous evaluation has shown that this program significantly reduced recidivism as measured by rearrests, convictions, and incarcerations, but the program did not affect rates of technical violations.

Also out of Minnesota, the High-Risk Revocation Reduction Program shows promise. This is an adult reentry program designed to reduce recidivism among high-risk male offenders who were previously released from a state prison but returned to prison for violating the conditions of their supervised release. One evaluation showed that this program had a statistically significant effect on reducing revocation and reconviction.

Table 5: Minnesota High-Risk Revocation Reduction Program
Recidivism Results
1-2 Year Follow-Up (Phase 1)46-Month (Average) Follow-Up
  Phase 1Phase 2Combined
ArrestsNo EffectNo EffectNo EffectBeneficial Effect
ConvictionsBeneficial EffectNo EffectNo EffectBeneficial Effect
IncarcerationNo EffectNo EffectNo EffectBeneficial Effect
RevocationsBeneficial EffectNo EffectNo EffectBeneficial Effect

Study design: Randomized Controlled Trial
Phase 1 N=239, Phase 2 N=282, from 2 Minnesota prisons
Source: Clark, 2015 [12]; McNeely, 2018 [13]

A final program I’d like to point to as promising is the implementation of Therapeutic Communities, or “TCs.” TCs are residential drug treatment programs in prisons or jails for treating substance-abusing and addicted offenders. While this program is not a reentry program as defined in this presentation, it is worth discussing. There’s a strong emphasis on participation by all members of the program, with the goal of reducing substance abuse and recidivism.

Table 6: Amity In-Prison Therapeutic Community
MeasuresRecidivism Results
 3-Year Follow-Up5-Year Follow-Up
ArrestsNot MeasuredNot Measured
ConvictionsNot MeasuredNot Measured
IncarcerationBeneficial EffectBeneficial Effect
Technical ViolationsNot MeasuredNot Measured

Study design: Randomized Controlled Trial
N=478 from San Diego
Source: Wexler et al., 1999[14]; Prendergast et al., 2004[15]

An evaluation of the Amity In-Prison Therapeutic Community found that participants had statistically significant lower levels of reincarceration rates when compared to the control groups.

However, I do find it very odd that incarceration was the only outcome assessed. Despite the beneficial finding on the only outcome assessed, more randomized controlled trials are necessary to make firmer conclusions about the effectiveness of TCs.

NIJ will continue to support research and evaluation to give practitioners critical information about what works in regard to reentry and crime, and the Reentry Council will continue to develop recommendations based in research and data.

We should pursue and continue to research these and other programs, to better understand what works, what’s promising, and what has been proven ineffective.

Looking Forward

I will wrap up so I can open things up for a discussion and your questions, but I want to conclude by emphasizing that our conversations about what works in crime prevention and reentry need to be grounded in evidence, data, and research.

It’s not enough that a program “seems effective” or makes us feel good. Anecdotal evidence is inadequate. For us to continue to support reentry programs, they must be evaluated and proven effective, and randomized controlled trials are the best and most rigorous way to make this determination.

We have a lot to gain by facilitating successful reentry through evidence-based programs and policies. The high recidivism rate shows that our reentry programs have been unsuccessful in stopping cycles of crime. We’ve failed to provide adequate interventions to help offenders successfully reintegrate into and contribute positively to society.

President Trump articulated this in a January convening to discuss prison reform and reentry. He said: “We’ll be very tough on crime, but we will provide a ladder of opportunity for the future. We can help break this vicious cycle.”

The bottom line is that we simply don’t know enough about what works to help prisoners successfully reintegrate into society. We do know that employment programs are not a panacea, and that there’s a pressing need for additional research to help us better understand what works, what doesn’t work, and what’s promising. This will help us fund effective programs and defund programs that are not proven effective. The Reentry Council aims to break the cycle by truly understanding what we know and don’t know when it comes to reentry.

Everybody wins when released offenders can reintegrate successfully as productive citizens. Thank you.


[note 1] Timothy Hughes and Doris James Wilson, Reentry Trends in the U.S. Bureau of Justice Statistics, June 8, 2018.

[note 2] Mariel Alper, Matthew Durose, and Joshua Markman, “2018 Update on Prisoner Recidivism: A 9-Year Follow-Up Period (2005-2014),” U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, May 2018.

[note 3] Andrew Wiegand, Jesse Sussell, Erin Valentine, Brittany Henderson. 2015. Evaluation of the Re-Integration of Ex-Offenders (RExO) Program: Two-Year Impact Report, Social Policy Research Associates.

[note 4] Ronald D’Amico and Hui Kim. 2018. Evaluation of Seven Second Chance Act Adult Demonstration Programs: Impact Findings at 30 Months. Social Policy Research Associates.

[note 5] Cindy Redcross, Dan Bloom, Gilda Azurdia, Janine Zweig, and Nancy Pindus. 2009. Transitional Jobs for Ex-Prisoners: Implementation, Two-Year Impacts, and Costs of the Center for Employment Opportunities (CEO) Prisoner Reentry Program. MDRC.

[note 6] Cindy Redcross, Megan Millenky, Timothy Rudd, and Valerie Levshin. 2012. More Than A Job: Final Results from the Evaluation of the Center for Employment Opportunities (CEO) Transitional Jobs Program. MDRC.

[note 7] Erin Jacobs. 2012. Returning to Work After Prison: Final Results from the Transitional Jobs Reentry Demonstration. MDRC.

[note 8] David Farabee, Sheldon X. Zhang, and Benjamin Wright. 2014. “An Experimental Evaluation of a Nationally Recognized Employment-Focused Offender Reentry Program,” Journal of Experimental Criminology, Vol. 10, No. 3, pp. 309–322.

[note 9] Philip J. Cook, Songman Kang, Anthony A. Braga, Jens Ludwig, and Mallory E. O’Brien. 2014. “An Experimental Evaluation of a Comprehensive Employment-Oriented Prisoner Reentry Program.” Journal of Quantitative Criminology 31(3):355–82.

[note 10] Susan Turner and Joan Petersilia. 1996. “Work Release in Washington: Effects on Recidivism and Corrections Costs,” The Prison Journal, Vol. 76, No. 2, pp. 138–164.

[note 11] Duwe, Grant. 2013. An Evaluation of the Minnesota Comprehensive Offender Reentry Plan (MCORP) Pilot Project: Final Report. St. Paul, Minn.: Minnesota Department of Corrections.

[note 12] Clark, Valerie A. 2015. “Making the Most of Second Chances: An Evaluation of Minnesota’s High-Risk Revocation Reduction Reentry Programs.” Journal of Experimental Criminology 11:193–215.

[note 13] McNeeley, Susan. (2018). A long-term follow-up evaluation of the Minnesota High Risk Revocation Reduction reentry program. Journal of Experimental Criminology. 10.1007/s11292-018-9330-x.

[note 14] Wexler, Harry K., George De Leon, George Thomas, David Kressel, and Jean Peters. 1999. “The Amity Prison TC Evaluation: Reincarceration Outcomes.” Criminal Justice and Behavior 26(2):147–67.

[note 15] Prendergast, Michael L., Elizabeth A. Hall, Harry K. Wexler, Gerald Melnick, and Yan Cao. 2003. “Amity Prison-Based Therapeutic Community: 5-Year Outcomes.” The Prison Journal 84(1):36–60.

Date Created: June 21, 2018