Photo: Michael Kelley

nyone involved with rehabilitating former prisoners learns to live with modest accomplishments. In its last study on recidivism, in 2002, the Department of Justice found that over two-thirds of former prisoners were rearrested within three years of being released. There’s no reason to think those numbers have improved. The recidivism rate in California, for instance, has hovered at around 70 percent for the past twenty-five years. On such bleak terrain, gains are measured by a hairsbreadth.

That’s why eyes are turned to Michigan. In 2003, the state launched the Michigan Prisoner Reentry Initiative (MPRI), which amounts to the most comprehensive program in the country for helping parolees transition from prison back into society. The premise of the MPRI is familiar: that it’s both cheaper and safer to invest in preventing ex-cons from reoffending than it is to repeatedly incarcerate them. The methods, however, are new. And preliminary data from the Michigan Department of Corrections is promising: parolees who have been released through the MPRI are returning to prison 27 percent less frequently than similar offender types released without it. Other measures also point to progress. Last year saw the fewest parolees committing new crimes since 2005 and the smallest percentage of total paroles revoked since record keeping began more than twenty years ago.

When parolees are less likely to reoffend, more prisoners can be let go without jeopardizing public safety. Going hand in hand with Michigan’s improved recidivism rates, therefore, has been a correspondent increase in parole approvals. Over 3,000 more prisoners were paroled in 2009 than were paroled in 2006; approvals for violent offenders have gone up by more than half (from 35 to 55 percent), while approvals for sex offenders have more than quadrupled (from 10 to 50 percent). As a result, during the past three years, the number of state inmates in Michigan has shrunk by 12 percent, reversing a sixteen-year trend of steady prison population growth. The turnaround enabled Governor Jennifer Granholm to shut down ten prisons last year, and an additional eight are slated to be closed by the end of 2010.

A few months ago, I sat in on an MPRI employment readiness seminar at the Oakland County parole office. The office is in Pontiac, the former hub of Automation Alley, where over half of the workforce is still employed by General Motors. Some twenty-five parolees—white, black, young, old—crowded into a small room, where they listened to Chaka McDonald, a sharply dressed case specialist, explain things like the difference between a laptop and a desktop.

“Who’s surfed on Yahoo?” asked McDonald midway through the class.

Two men raised their hands.

“Who has e-mail?”

Four hands.

“Who’s familiar with Facebook?”

One hand.

McDonald held aloft a black Dell notebook. “How could it benefit you, knowing what kind of computer this is, if you’re meeting with an employer?”

“The job might be ran on them,” answered a young man in his twenties.

McDonald nodded and moved on to explain how a wireless USB internet adaptor worked. “How could this benefit me, if I go see an employer and I’m not shocked looking at this thing protruding out of this computer?”

A man with a shaved head, who’d been running a comb through his long black goatee, raised his hand. “It shows you have a working knowledge of the technology.”

“There you go. Because at some point we know we gonna have to address the felony. So if I know at some point I’m addressing the felony, if I can let this employer know I’m still computer literate, won’t that diminish that felony a little bit?”

After the class, a reentry coordinator named Sherry Carter pulled the man with the goatee and shaved head into her office. The man’s name was David, and he’d recently been released after doing ten years for having had sex with a thirteen-year-old girl. (David had been seventeen at the time.) He was hoping to get a job at the Silverdome, but he knew his prison tattoos would disqualify him. “SCORPIO” was inked on one side of his neck, “BETTY” on the other, and “D-A-R-K-N-E-S-S” across his knuckles.

“Are you interested in having your tattoos removed?” Carter asked him.

The Pontiac parole office has an agreement with American Pride, a local parlor that performs laser removals. In David’s case, the procedure would likely take two sessions, each costing $100, but the Michigan Department of Corrections, with MPRI funds, was willing to pay for it.

Later, David told me that when he was released from prison the MPRI set him up with an apartment, provided him with bus passes, gave him vouchers for clothes, and helped him create a resume and apply for jobs. MPRI officials also got him enrolled in Oakland Community College, where he will be pursuing a degree in automotive technology.

This sort of assistance is a big change. Prior to the MPRI, Michigan parolees received a little bit of cash and two weeks of transitional housing. After that, one former parole agent says, “They were out on their ass, either sleeping on a park bench or in the shelter.”

Under such circumstances, David told me, he probably wouldn’t have made it. “A guy like me, I did ten years in the joint. You get out and you got nothing. I had one outfit when I got out. That’s the only thing I had in my entire possession. If it wasn’t for the MPRI program, I’d be on the street right now.”

Between 1970 and 2005 the number of incarcerated Americans grew by 700 percent, accounting for one-fourth of the world’s prisoners. A 2008 study by the Pew Center on the States found that more than 1 in 100 adults were behind bars. This isn’t just grim in humanitarian terms; it’s also extremely expensive, costing over $60 billion a year. If states are to cut back on the vast sums spent on prisons, they will need to focus on keeping parolees from reoffending. No state has taken the lead on this more than Michigan. More important, what Michigan’s example suggests, at least for now, is that there’s a way to accelerate releases—and to do it safely. Despite releasing more than 6,000 convicted felons onto its streets, Michigan has seen per capita violent crime decline by over 10 percent since 2005.

nderstanding Michigan’s transition from building prisons to closing them requires looking at how incarceration became such a large part of the state’s budget in the first place. In 1973, Michigan had a prison population of 7,874 inmates, and $38 million, less than 2 percent of the state’s general fund, went to corrections. Crime was rising, however, and prison capacity was insufficient. Like many other state legislatures across the country, Michigan lawmakers attributed their worsening crime rates to what they viewed as overly lenient sentencing guidelines, so they enacted tougher policies and built more prisons. By the end of 1984, Michigan’s inmate population had nearly doubled, reaching 14,658. Five years later, it had more than doubled again, to 31,834.

The increase in incarceration alarmed some state officials, but events conspired to keep the prison population climbing steadily upward. In 1992, a parolee named Leslie Allen Williams confessed to kidnapping and murdering four teenage girls, burying their bodies in shallow graves northwest of Detroit. Williams was a repeat offender who’d been released after serving a minimum sentence for assault. Public outrage was immediate and fierce. “The parole board represents the bureaucratic interests of emptying beds and not one of them represents the public safety interest,” one Michigan prosecutor told the New York Times. At the time, Michigan’s parole board was composed of civil service employees such as criminal justice experts and law enforcement officials. After Williams, however, the legislature granted the governor new powers to appoint and remove board members at will. Republican John Engler soon overhauled the board and encouraged its new members to get tough. In the decade that followed, the parole approval rate dropped by 20 percentage points, and the number of parolees annually returned to prison for technical violations doubled.

By 2003, when Jennifer Granholm became governor, Michigan’s prison population had increased to 50,591, and corrections expenditures had reached $1.6 billion, almost a fifth of the state’s general fund. It was becoming unsustainable. “If something wasn’t done, we’d be building more prisons, and that simply was not in the cards,” remembers Beth Arnovits, executive director of the Michigan Council on Crime and Delinquency.

With the state hitting hard times (Michigan’s recession predates the economic downturn by a decade), Granholm and other officials were finally ready, if warily so, to listen to new ideas. What they heard from prison reformers was that Michigan’s sentencing policies, along with the rest of the country’s, were based on two flawed assumptions: one, that heavier sentences meaningfully deter crime; and two, that fewer grants of parole decrease recidivism. In fact, reformers had long argued, research shows that factors such as police per capita correlate far more closely with crime rates than do prison sentences, and longer lengths of stay do nothing to increase the likelihood of success on parole. A study of 76,000 Michigan convicts, for example, found that prisoners who were paroled later reoffended at the same rate—and, in some categories of offender type, at a higher rate—as those paroled earlier.

As a first step toward reforming the state’s correctional system, Granholm appointed Dennis Schrantz as her deputy director for corrections. Schrantz is a North Carolinian who came to Michigan in 1990 to head the state’s Community Corrections Office, a new agency created to bring down prison admissions by improving probation options. Schrantz’s efforts had reduced the overall commitment rate for convicted felons (that is, the proportion of felons actually being sent to prison) from 32 to 22 percent, the lowest in the country. (The national average is 35 percent.)

Now Schrantz’s attention would be shifted from the front end of the system to the back end: finding safe—and politically feasible—ways to reduce the length of stay for inmates, most of whom were serving more time for their offenses than prisoners in neighboring states. (About sixteen months more, on average.) Schrantz knew this would require a focus on parole. Like thirty-three other states, Michigan uses a form of indeterminate sentencing in which each inmate gets a minimum (determined by a judge) and a maximum (established by the legislature). Often, the gap between the two can be vast—a sentence of three to twenty years, for instance. After serving his minimum, the inmate becomes eligible for parole; if he is never granted parole he eventually “maxes out” and goes free. While indeterminate sentencing has its drawbacks (parole decisions can seem arbitrary), it offered an important opportunity to Schrantz and other prison reformers: they could reduce lengths of stay without seeking authorization from the legislature, where partisan grandstanding has consistently prevented lawmakers from voting for sentencing guidelines that might be portrayed as soft on crime. As long as Schrantz could persuade the board to raise the rate of parole, prison crowding could be brought down.

In order to win over the board, Schrantz first had to make more resources available for released inmates. This didn’t take long. He’d learned from his experience working with probationers that effective programs seldom need to be created; they already exist in the community. “We gave the locals a lot of authority and control, and that proved to be the ticket,” Schrantz says. “The state did what I think the state should do, which is create the policy framework and provide the funding and let the people that are closest to it do the work.” (Nongovernmental partnerships have remained a hallmark of the MPRI. Today, in Detroit alone, the Department of Corrections contracts with more than sixteen nonprofit organizations to work with parolees, offering services that include substance abuse counseling, classes on domestic violence prevention, family reunification, mentoring, and housing and employment assistance.)

Schrantz also opened up communication between parole board members and DOC leadership, something he says had never happened. “States don’t do a very good job of making certain the parole board is part of their process,” he says. “It’s as if the parole board is off on their own, and in many cases they are physically off on their own.” When Schrantz had asked his predecessor’s chief of staff to describe the relationship between the board and the outgoing corrections director, the chief of staff told him that he wouldn’t know a parole board member if she walked into the room.               

Schrantz’s strategy for persuading the board to release more inmates from prison on parole was so straightforward it seems incredible that it hadn’t been tried before. “I asked them, ‘What do you need? Under what conditions would you consider paroling these individuals?’” Schrantz recounts. Members of the board say they were thrilled to be asked this simple question. They first brought up the problem of the mentally ill, who make up roughly a sixth of Michigan’s total prison population (7,100 out of 44,500). Mentally ill parolees reoffend at a much higher rate than other parolees, and the board told Schrantz they rarely felt comfortable releasing them because they knew that the parolees would receive nothing more than sixty days of medication. Schrantz and his team responded by creating the MPRI Mental Health Initiative, which put in place contracts with community providers to offer paroled inmates sustained mental health services and guidance on obtaining other state and federal entitlements. Today, when mentally ill prisoners appear before the board, they bring packets detailing individually tailored, long-term master care plans. Parole approvals have spiked, and now only 28 percent of mentally ill parolees return to prison, compared with 50 percent prior to the initiative.

For regular prisoners, Schrantz worked to equip those going up for parole hearings with plausible plans for what to do once they were released. “What the parole board in Michigan saw was prisoners coming to them for the first time better prepared,” says Le’Ann Duran, director of the Justice Department’s National Reentry Resource Center. “Couple that preparedness with the huge investment that Michigan made in the community-based resources which never before existed, and that’s very attractive to the board.”

Schrantz also introduced into the decisionmaking processes new academic research to which board members and corrections officials had never been exposed. “It was a whole new language,” says Patricia Caruso, Michigan’s director of corrections and a former prison warden. Often, the research had produced remarkably counterintuitive findings, calling into question long-held assumptions. Caruso recalls one especially memorable example of this. “Very early on [Schrantz] said to me that there is not a correlation between misconduct in prison and success on parole,” she says. “When Dennis first said this to me, I was absolutely aghast.” But Caruso says she eventually accepted the logic when Schrantz explained how, for women, unruliness in prison can actually be tied to success on parole. “So many women come to prison because they don’t have a voice. They’re often victims of men abusively guiding their actions,” says Caruso. “So getting a voice, and being able to tell the correction officer to go F himself, may not be the preferred way of conducting yourself in prison. But it does mean that when you get out you’re going to stand on your own two feet … Those types of things—within the first year, I really got it. I really embraced it.”

Meanwhile, Granholm and Caruso gradually replaced every member of the Engler parole board save one. In 2007, after winning reelection and implementing the MPRI statewide, Granholm appointed a former corrections officer named Barbara Sampson as chair of the parole board. “When Barb took over, everything changed,” Schrantz says. Sampson, who previously worked at a nonprofit for battered women, believed strongly in the MPRI and welcomed the new criminal justice research that Schrantz was recommending. When we met in her office in Lansing, her desk was cluttered with folders, books, and binders containing various academic literature on incarceration and recidivism. “We know what we think we know intuitively and with our gut,” Sampson said. “But what does the evidence really say?”

Perhaps the most striking example of how instinct and evidence can diverge concerns sex offenders. Historically, sex offenders have been denied parole far more often than other types of inmates. When Sampson became chairperson, the approval rate for sex offenders was a mere 10 percent. Last year, however, the board approved 50 percent of the sex offenders it considered for release. “For years, the board lumped all sex offenders into one category—evil predators,” Sampson explains. “[But] within the broad category of sex offenders, there are groups. And what we had to do as a board was educate ourselves about the different characteristics of the different groups. For instance, individuals who commit incest, when you look at the data, the re-offense rate is very low.” While the spike in paroled sex offenders has been a particularly hard sell, even with the governor (“She’s no big fan of releasing sex offenders,” Schrantz says. “She had her way, we’d castrate them”), the approach has so far been allowed to continue, without any obviously adverse effects to public safety. Since 2005, reported rapes in Michigan are down 13 percent.

espite the success of the MPRI, there are, unsurprisingly, many opponents of Michigan’s new parole practices. In January, the Prosecuting Attorneys Association of Michigan held a series of coordinated press conferences across the state, enlisting victims and their families to warn against the danger of paroling violent offenders. One attorney likened the spike in paroles to “a fire sale.” Another called it “death for dollars.” Mike Cox, a Republican and former marine who succeeded Granholm as attorney general in 2003 and now hopes to succeed her as governor in November, recently established a Parole Objection Project in his office. “These felons should not see the light of day, and I will do everything I can to ensure they won’t,” Cox promised.

The press has reinforced such sentiments. The New York Times has characterized the Michigan parole surge as a compromise of public safety brought on by the state’s deficit (“Safety Is Issue as Budget Cuts Free Prisoners” was the headline of a front-page story last March), largely glossing over the belief of reformers that the MPRI is an attempt to improve correctional outcomes. Local outlets have often gone much farther in stoking public fears. When I met with Larry Payne, an MPRI community coordinator in Saginaw, he’d just spent the week managing the fallout from recent television coverage of a Bay City house the MPRI had contracted to accommodate sex offender parolees. “They’re scaring the hell out of everybody,” Payne said of the local news station. People in Bay City “were totally under the impression that we were putting pedophiles on the bus stop to snatch their kids.”

While resistance to Michigan’s new parole policy brings out an inevitable quotient of opportunism—no candidate ever lost a political race for being tough on crime—there is no doubt that many of its opponents are informed and sincere. One of these is Mike Thomas, Michigan’s director of the National District Attorneys Association, whom I met at the Saginaw County Courthouse. Every year since 2003, the FBI has ranked Saginaw the most violent city in the nation in per capita crime, and not coincidentally its populace is rapidly fleeing (more than 10 percent have moved elsewhere in the past few years). For Thomas, who grew up in Saginaw and has been its elected prosecutor for more than two decades, Michigan’s crime rate is a deeply personal concern. His views are shaped by victims he considers neighbors and crimes he remembers daily in all their gruesome specificity. And he worries that his city will lose even more residents and businesses if more prisoners hit the streets.

“Just because somebody is in prison for safe cracking doesn’t mean that they can’t commit a violent crime,” Thomas told me when I questioned the effectiveness of longer lengths of stay in prison. “Why do I say that? Because Keith Wood murdered his fifteen-year-old cousin when he got out of prison on his third parole release for the nonviolent offense of safe cracking and burglary. I tried the case a couple years back. He killed her because she had the audacity to have a young black boyfriend from Heritage High School in Saginaw Township. Now, after he was convicted of first-degree murder—’cause he got caught with her nude body in the back of his trunk as he was driving without any lights in the middle of a snowstorm—January comes to mind—then he got religion in jail and told the chaplain there that he’d done a lot of other things. He confessed to another murder that he’d done in our county while he was out on parole for nonviolent safe cracking—a black night manager at a Burger King out in Shields, Thomas Township here—and two or three other murders in Michigan and other states. He also confessed to literally hundreds of other arsons, robberies, burglaries.”

Such stories of parolees who return to crime—and there will always be a proportion of parolees (as well as ordinary ex-prisoners) who return to crime—pose an immense challenge to the MPRI. One incident was horrific enough that it threatened to derail the program altogether. In 2006, a Michigan parolee named Patrick Selepak brutally tortured and murdered three people in New Baltimore. Two of the victims were a young husband and wife. The husband was intravenously injected with bleach while the wife, who was pregnant, was strangled. “The board was distraught,” says Patricia Caruso. “When you see someone you paroled go out and kill innocent people, you question every decision you make.” For a while, the board cut down dramatically on parole approvals, and Michigan’s prison population subsequently increased by more than 2,000.

By the end of 2007, however, the parole board had resumed its earlier pace of approvals. “You had Dennis [Schrantz] and you had the director and everybody doing all of this evidence-based, data-driven work,” Sampson remembers. “And steadily dangling that work, saying, ‘Board, take a look at this information. This is a study that came out of Kansas. Look at the Pew report.’” Caruso, who still has all the Detroit News and Free Press clippings about Selepak in her office, says she was determined not to let “one bad crime become the namesake for everything. It has not stopped us from what we’re doing. It does stop other people. But it has not stopped us.”

Schrantz is adamant about the merit of statistics and the danger of focusing on case studies. When I mentioned some of the stories Mike Thomas had told me, he said, “What these people do is they’re gonna want to talk about cases. They should be forced to talk about trends and large numbers of people. That’s how you run systems. Don’t judge us one case at a time. Judge us by the system.” And trends do bolster Schrantz’s case. Since crime in Michigan is decreasing, it is reasonable to assume that the current allocation of resources for crime control are, despite what can sometimes be terrible failures, making the state safer than it was under the old system.

ltimately, aside from concerns over reoffending parolees, a different problem may pose the greatest threat to the MPRI: Michigan’s severe recession. In reporting across the state, I spent time with numerous MPRI parolees, many of whom had done remarkably well: a former gang member who now works at Habitat for Humanity, a former sex offender who now is a line cook, a former crack addict who now balances two jobs while she attends college, and a former murderer who now takes care of the crocuses and petunias outside his parole office. But what most of them had in common was employment. Absent that, a critical anchor is missing. Michigan’s unemployment rate has reached almost 13 percent, 3 points above the national average. As Mario Dewberry, Wayne County’s reentry coordinator, says, “An individual who’s only been locked up for the last three years, they’re coming out to a completely different world. There’s been so many instances of: The aunt I used to stay with, whether they were renting or paying a mortgage, the house is no longer there. My uncle who had a lawn care business or worked for GM? Gone.” Ironically, the same dire economic straits that spurred Michigan to reduce its prison population in the first place might also endanger the MPRI’s long-term success.

Still, Schrantz and his colleagues in Lansing are optimistic. Perhaps more encouraging, so are many of the parolees I met with. While prison reform advocates complain that opponents and the press focus on frightening anecdotes rather than statistics, I found myself impressed less by Schrantz’s data than by the individual stories of successful parolees.

In Kalamazoo, Michael Brown, a Korean War veteran with a graying flattop, grayer mustache, and several missing teeth, told me he had been in and out of prison six times. Five of those times it was the same parole agent who sent him back on technical violations. This December, he was released through the MPRI. Now Brown has a new parole agent who couldn’t be more different, and Brown has changed, too. “Anything I do, if I’m the slightest bit worried about getting in trouble, I talk to him first,” Brown told me. “In the past, you didn’t have that. They were there to monitor you and they didn’t really care about you. They didn’t care if you had problems. They didn’t care if you wanted to reconcile with your family. The best you could get out of a parole officer was, ‘See you next month.’” I asked what things might have been like if he’d had the MPRI during any of his previous paroles. Brown shook his head. “I probably would’ve never come back. I think things would’ve been a lot different. I’d probably still be married to the same woman and been able to raise my kids instead of not being there for her or them. It’s a world of difference.”