Please ensure Javascript is enabled for purposes of website accessibility
Recent News
Home / All News / Tom Roy looks to data in push to cut recidivism
Minnesota Corrections Commissioner Tom Roy sits in his St. Paul office, surrounded by the studies and reports that are the grist of his data-driven approach to the job. (Staff photo: Kevin Featherly)

Tom Roy looks to data in push to cut recidivism

Corrections Commissioner Tom Roy has an abiding faith in data’s power to help Minnesota realize the goal of fewer inmates and less crime. Not everybody does. Therein, Roy said, lies one of his greatest frustrations.

Scanning his desk littered with studies and reports, Roy sees a path to more effective cost containment and reduced recidivism. One of his key problems, he said, is convincing the legislators who hold his purse strings that he is right.

“Clearly legislators have constituencies that are focused on a single topic sometimes,” Roy said. “Especially if those constituents have been victims, that gut-level passion sometimes supersedes — it’s tough for me to say this — good research.”

Roy has no problem punishing offenders, some of whom he knows likely will never change. He just thinks that corrections should be driven by data, not emotion, in its quest to assist those who can change.

Led by DOC Research Director Grant Duwe, the agency has produced reams of peer-reviewed academic studies demonstrating that — if properly staffed and funded — the prison system can help redirect inmates to productive, crime-free lives on the outside.

It can do that, Roy said, using tools like vocational training and counseling both in prison and post-release, through “cognitive restructuring,” and through intensive chemical and alcohol dependency treatment, among other tools — all of which cost money.

“We can demonstrate through the research on what works best, if we can match offenders to the right program and if we have the capacity to give those services,” Roy said.

“We have the science behind it,” he adds. “Providing these services is not soft. It is smart justice in that it prevents future crime.”

National trends

Nationwide, the trend is toward putting fewer people in prison. But that is not what is happening in Minnesota.

Of the 45 states responding to Vera Institute of Justice researchers for a May 2017 study, Minnesota was one of just 15 states that saw both increased prison populations and rising costs between 2010 and 2015. In 10 other states, costs continued to rise despite falling populations. The rest saw both reduced costs and populations.

Minnesota’s population increased 4.1 percent in that five-year period, while costs rose 3.2 percent, to $403.7 million. Yet its cost per resident to maintain prisons was among the nation’s lowest — just $74 per resident here. Only Utah’s costs were lower at $51 per resident. More than half of responding states (28) had costs exceeding $100 per resident.

Roy blames the state’s recent prison population uptick, in part, on creation of new statutes like felony DWI, which has alone filled 600 additional prison beds, he said. Yet incarceration rates remain comparatively low, he said.

Minnesota imprisoned only 178 people per 100,000 residents in 2015, according to the Vera study. Nationally, only Massachusetts did better, with a rate of 159 prisoners per 100,000 residents, though its costs were higher ($88 per resident).

The Vera study pins rising costs on political reactions to high-profile crimes, rising labor costs and health care costs driven by aging prison populations.

Cost-benefit analysis

Numbers like those can be useful in Roy’s legislative messaging—and he is more than willing to trot them out. But DOC digs deeper than many other states by ferreting out cost-benefit ratios on various inmate programs and services.

Some of those programs, the state says, have proven recidivism reduction rates and positive cost benefits per dollar spent, as measured through such factors as lowered medical expenses, decreased property damage and reduced victimization.

In December, Minnesota Management and Budget published a cost-benefit analysis using data supplied by the DOC. It indicates that a number of prison programs are “proven effective” at reducing recidivism, either through the use of randomized control trials or high-quality local evaluation. According to MMB, some of most effective include:

  • EMPLOY: Affiliated with MINNCOR Industries, this voluntary work program enlists participants as they approach their release date. Prison staff helps inmates write resumes, build job-search and interview skills and learn about the employment market outside prison. Follow-up meetings continue with DOC specialists post-release to provide further support and job referrals, Roy said.

    The MMB study says the 15-month program produces an average annual recidivism reduction of 32 percent, and a cost-benefit ratio of $15.90 for every dollar spent.

  • Cognitive behavioral therapy: “COG,” as Roy calls it, essentially is a program to help offenders find ways to change their thinking and is offered under a curriculum called Thinking 4 a Change. Inmates in group meetings run through scenarios where they practice reacting to real-world problems.

    Roy gives the example of an offender who learns his cellmate has looked into his diary. Normally, the inmate would punch the other guy out. With input from others in his group, though, he learns and acts out new reactions. “Generally in the corrections world right now, COG is probably the most promising practice,” Roy said. “Offenders get it, because it’s not rocket science.”

    According to the MMB study, the program, which lasts an average of 25 sessions, reduces recidivism by an estimated 9 percent and brings an estimated $13.40 of benefit for every dollar spent.

  • Chemical dependency: Minnesota’s prison system is also its biggest addiction-treatment organization, Roy said. “At any given time we have 1,000 and we could add another 2,000 or 3,000 if we had the space and staff,” he said.

    He would like that. DOC research shows that 80 percent to 90 percent of inmates are either chemically abusive or addicted, and most probation and parole violations relate to drug and alcohol abuse.

    The MMB report says treatment in prison lasts an average six to nine months — much longer than most programs on the outside. Treatment has a proven recidivism reduction rate of 21 percent and $2.80 of benefit for every dollar spent, the study says.

    Duwe says MMB actually underestimates the cost-benefit ratio of treatment programs. His internal research suggests the benefit is actually $6.32 for every dollar spent. It also leads to $22 million in cost avoidance annually, Duwe says.

    Duwe cautioned that MMB’s results rely more on extrapolations than his own research, generating less precise results. “Their estimates are more abstract, whereas ours are more concrete,” he said.

‘You have to act’

According to Roy, this is the kind of data that he has trouble selling legislators on when he makes his pitch for improving services. To some extent, he said, he understands why.

“Sometimes that data-driven approach appears to people that you are cold to the human interests,” he said. “The real challenge is to weave the human interest into good science and good policy over the long term.”

Rep. Nick Zerwas, R-Elk River, arguably is not one who fails to buy into the scientific approach to corrections. Research and data played a major role in the 2016 push for drug-sentencing reforms, he said.

But Zerwas has some criticisms for Roy. The commissioner tends to see dollar signs attached to every piece of evidence he presents to the Legislature, according to Zerwas.

“In his mind, every problem is because they don’t have enough funding,” Zerwas said. He called that a “flawed logic.”

“At some point as a Legislature, you can’t let perfect be the enemy of the good,” Zerwas said. “You can continue to accumulate and analyze data. But at some point, you have to act.”

Sen. Ron Latz, DFL-St. Louis Park, agrees — to a point. “There is always going to be more information to gather,” Latz said. “But to be unwilling to start the study process is a different question.”

Latz credits Roy with being open-minded and responsive to the direction that data takes him, and the senator thinks that is a good thing.

“He rightly wants decisions to be based on evidence, I think, more than ideology,” Latz said. “I think that government should be more reliant on that kind of analysis in making decisions.”


Leave a Reply