Some states report that their recidivism rates were high but are being reduced now. Others claim that their state recidivism rates were always low. The weekend before President Obama's announcement that his administration's plans to enhance reentry efforts for people leaving prison, reform advocates were passing around a 2014 study from Abt Associates - one that says the recidivism rate is about half of what has been reported by the Bureau of Justice Statistics.
They're all wrong.
My six-plus years at the maximum-security York Correctional Institution gave me a front row seat to the recidivism show. Women would leave and be back within 24 hours sometimes. Recidivism isn't a rate; it's a guarantee.
In my home state of Connecticut alone, there are 35,873 outstanding warrants for arrest, a number twice the size of the state's 16,168 prisoner head count. Because the warrants are outstanding, and therefore not applied to people, they cannot be included in recidivism studies.
However, the likelihood that these warrants apply only to first-time offenders is small. Many are for violations of parole or probation and for absconding and only previous offenders can engage in those offenses.
Connecticut is not alone in these problems. Oregon, the state with the lowest recidivism rate in the country, according to a 2011 study by the Pew Center on the States, and a state that counts only felony convictions within three years of release in its recidivism rate, suffers from undercounting, too. One of its counties has 12,000 outstanding warrants.
The state with the second lowest recidivism rate, Virginia, counts only those released offenders who are sentenced to more than eleven months in prison for a subsequent offense, regardless of the new charge. And of Virginia's 95 counties, one representative county, Roanoke County, has 732 outstanding warrants of which at least 151 are for repeat offenders.
In the Pew Study, Minnesota was reported to have the highest rate, using felony convictions within three years to calculate it. Just one of the state's 87 counties has 4500 outstanding arrest warrants.
When Ferguson, Missouri municipal court judge Donald McCullin wiped out thousands of arrest warrants in August, activists viewed the move as an attempt at racial justice. While the judge never intended it that way, I saw it as a way to make Missouri's reported recidivism rate - 54.4%, the third highest in the nation in 2011 - a bit more accurate.
The best way to combat the number fatigue in the battle for justice reform is to pass the National Criminal Justice Commission Act, introduced by Michigan's junior senator, Gary Peters. The bill proposes a top-down study of what is actually happening in our criminal justice system. Unfortunately, Govtracker.com has given the bill a passage prognosis of 0%.
I would say in not passing the Criminal Justice Commission Act before other reform bills like the Sentence Reform and Corrections Act, we put the cart before the horse but, since we are releasing thousands of people from federal prisons right now, the horse isn't around anymore. We're just waiting for him to come home, recidivate.
For the ways we measure recidivism, simply staying away from crime is success for a released offender. Obeying the law shouldn't be a measure of achievement; it's the bare minimum asked from people who aren't in prison and released offenders - according to reform advocates - are just like everyone else.
Perhaps we need an entirely different way of measuring success in released offenders. Maybe we need to turn from numbers to narrative. The Boston Reentry Study has tracked 122 offenders since their release from incarceration. The Boston Reentry Study has already unearthed unexpected characteristics of prison populations but its effect on actual recidivism is unknown. Following people prospectively rather than counting them retrospectively may lend the support a released offender needs to stay out of prison.
Of course, because of cost, prospective recidivism studies can be done for small groups more effectively than large groups like the thousands being released from federal institutions this fall.
At the beginning of the month, President Obama announced that his administration will be dispensing millions in grant money for programs that aid reentry and rehabilitation to reduce recidivism. At least a portion of that should be dedicated to knowing who's reoffending and who's not. Until we know that, we can't measure anyone's success.