How the Release of 6,000 Federal Prisoners Will Test America's Flawed Halfway House System

As I got to the end of my time in prison, I had the chance to live out the final stretch in a "halfway house," or residential reentry facility, as many inmates do. Instead, I chose to stay in prison. That might surprise people who haven't had experience behind bars, but from what I'd heard, halfway houses can be dangerous places. Drugs circulate so freely that in 2009, 73 percent of those sampled in one New Jersey halfway house tested positive for them, according to a New York Times investigation, and a low-risk offender was apparently murdered in another halfway house in that state for the three dollars he had in his pocket.

I was thinking about my decision when it was announced earlier this month that the federal government will release 6,000 nonviolent drug offenders by November 2. About two-thirds of them are destined for halfway houses, and I can't help but wonder how many will have the resources they need to hack it—and how many will be derailed before they even hit the street.

As someone who pined for release from prison for over six years, I want these nonviolent offenders to succeed. But when we talk about releasing inmates, we shouldn't forget that many of them will be dropped into a system of mostly privately-run halfway houses that is begging for reform or replacement by safer alternative.

These facilities have been the targets of scathing investigations for years now. Escapes are common—just last week, the Associated Press reported that 240 federal inmates have escaped en route to federal halfway houses since 2012 alone. According to the New Jersey State Commission of Investigation, in 2009, there was more gang activity in the state's halfway houses than in its prisons, and one study in Pennsylvania even found a higher recidivism rate for those released to halfway houses than for inmates launched directly into the street.

What makes a prison isn't locks, barbed wire, or uniformed authority—it's the people inside. When you take a bunch of institutionalized people and put them in a single dwelling, you essentially create a new prison. You can call it halfway house. transitional living, or the Hotel California, but It's still a form of incarceration. So these 6,000 federal inmates aren't exactly being "freed"—many of them are simply transferring from traditional bar-and-fence facilities to outposts a bit closer to society. The distinction is important, because, as Families Against Mandatory Minimums (FAMM) has noted, virtually all federal halfway houses are operated by private contractors. Given what we know about private-sector prisons—where employees are often paid far less than their counterparts in the public sector, and the facilities enjoy a unique reputation for violence—there's reason to be skeptical of private-sector halfway houses as the way out of our mass incarceration nightmare. None of which is to say that we shouldn't be working to release low-level and nonviolent offenders from prison. But this project should be done in ways that improve the chances that released offenders will be able to thrive in the larger world, rather than reoffending, as so many former inmates do. That would improve the lives of the men and women coming out from behind bars; it would also make the shrinking of prisons a viable political project.

After all, a few high-profile mishaps have the potential to sour the public on reform. I saw this happen firsthand when I entered prison in December 2007; the murders of the Petit family in Cheshire, Connecticut—committed by two ex-offenders who had apparently met in a Hartford halfway house—were only four months old at the time. Parole was soon banned for violent offenders across the state.

Watch the VICE HBO documentary on America's incarceration system, featuring President Barack Obama's first-ever visit to a federal prison:

Electronic monitoring has proven to be an effective alternative to halfway houses, with one Justice Department (DOJ) study finding it accounted for a 31 percent reduction in the failure on part of inmates to comply with terms of supervised release. (A 1993 study by a scholar affiliated with the Bureau of Prisons suggested the two systems are similarly effective, but conceded that "those offenders released via electronic monitoring benefited more by maintaining continuous employment.") So why is it being used rather sparingly? According to the DOJ, when inmates pay the cost themselves, electronic monitoring runs $64 a month, which is nothing in comparison to the $73 a day the BOP says it costs to put inmates up at halfway houses.

A slow trickle out of prisons isn't enough to count as real reform, but a flood that produces a backlash isn't the answer, either. To change things for good, we need a dam—in the form of improved halfway houses or a safe alternative to them—to prevent the prisoners going out from coming back in, and setting back the criminal justice reform project in the process.

Chandra Bozelko served more than six years in York Correctional Institution, a maximum security women's prison in Niantic, Connecticut. She is the author of Up the River: An Anthology (Bleakhouse, 2013) and blogs about her prison experiences at prison-diaries.com.