About 40 percent of Minnesota convicts who are let go from prison on supervised release either run from authorities or commit new crimes, a Star Tribune analysis of state corrections data shows.

For some it was fairly predictable that they would have problems. Those who become fugitives or get arrested tend to be veterans of the state prison system who have committed the most serious crimes, such as assaults, aggravated robbery, sex crimes, DWI and drug offenses.

But in Minnesota, that doesn't matter. The length of a prisoner's supervised release, a time when criminals are supposed to adjust to life outside prison under the watchful eyes of corrections officials, is set using the same formula for every offender. Regardless of their likelihood to succeed, most offenders spend two-thirds of their sentence behind bars and one-third in the community.

Minnesota's supervised-release rules are getting a second look.

In 2009, state lawmakers asked the Minnesota Department of Corrections to assess some of its supervised-release practices and make recommendations in a report due in January. The Legislature's directive comes at a time when national studies and experts say that some low-risk offenders should get less supervision so corrections officials can spend more resources on frequent offenders who are considered most dangerous.

Since 2005, nearly 11,000 out of 27,000 offenders failed while on supervised release.

"I think a 40 percent failure rate is too high," said Rep. Debra Hilstrom, DFL-Brooklyn Center, chair of the public safety and policy committee. "We need to be taking steps to make sure that that compliance is higher."

Rewards for compliance

Minnesota's formula for supervised-release time has been around for three decades. In 1980, the state took away the discretion of parole boards and mandated that most prisoners serve the last third of their sentence on supervised release. The change aimed to help officials better plan prison space and reduce disparities based on race, class and an offender's personality.

For supervised release, corrections officials follow many national recommendations such assessing offenders' needs and risks and tailoring supervision conditions to them. For instance, someone who has burglarized homes to feed a gambling habit might be barred from gambling establishments and ordered to attend Gamblers Anonymous meetings.

There are also rewards for those who behave -- compliant offenders may graduate to less supervision, allowing less frequent checks from their officers, for instance.

"We understand and we know that the research tells us that in fact you can probably do more harm than good if you spend too much time with an offender that doesn't need it," said Jill Carlson, field services director for the corrections department. "We are working very hard to target most of our resources to those that are in greatest need and are the greatest risk."

But under state law, offenders can't be rewarded with less supervision time. Two national nonpartisan research organizations recommended such moves in recent years. Calling it "earned compliance credits," the Pew Center on the States recommends ending supervision 15 days early for every month that an offender behaves.

"The research is clear that the carrot and the stick work best together," said Adam Gelb, project director of Pew's Public Safety Performance Project. When offenders are taken off supervision, Gelb noted, resources are freed for agents to concentrate on those who pose problems.

It's one of the ideas state officials will study at the request of the Legislature.

"We want to know what works and what has been proven to keep the public safe," Hilstrom said.

In California, officials seeking to control corrections costs have started using a state law allowing accelerated parole discharge more frequently.

Most offenders there are sentenced to serve three to five years of parole, an official there said, but some low-risk offenders are being assessed for possible release after serving just one year. If everything is going well, they may be let out from supervision in their 13th month.

"It frees up our parole agents to focus on the higher level offender," said Terry Thornton, spokeswoman for California Department of Corrections and Rehabilitation.

Less time behind bars?

Once prisoners are released into the community, surveillance alone doesn't help them succeed, according to a study by the Urban Institute.

The Minnesota Department of Corrections will soon use a $750,000 federal grant to bolster programs for some ex-prisoners, targeting those in the metro area who frequently violate release conditions and are sent back to prison. Increased resources will go to treatment and helping offenders find and maintain jobs and housing, among other items. Participants will also receive mentoring and basic education for obtaining GEDs, the grant application said.

"Going back to prison is a very expensive proposition, so can we become more cost efficient by focusing some resources on that particular high risk population," said Gary Johnson, the corrections department's director of re-entry services.

To save money and change the system, some argue, shortening prison time for some low-risk offenders would have the greatest impact.

Minnesota allows that for only a select few. Mainly non-violent offenders can get out early through work release and the Challenge Incarceration Program, a boot camp-like program with physical labor, chemical dependency treatment and instruction in literacy, critical thinking and life skills.

Retired Appeals Court Judge Jack Davies said he has lobbied for bringing back a parole board. Instead of everyone serving at least two-thirds of his sentence behind bars, Davies believes offenders should be eligible for parole after serving a third of their time. Using ever-improving risk assessment tools, a board could make the decision about who is an appropriate inmate to bring into the community early, he said.

That approach would give prisoners more incentive to improve themselves, experts said. Now, in Minnesota, going to treatment for alcoholism or learning a job skill doesn't affect a prisoner's time behind bars or time on supervised release.

"The corrections department does not have a way to reward or to recognize the people that make the appropriate changes," Davies said.

howatt@startribune.com • 612-673-7192 plouwagie@startribune.com • 612-673-7102