Catch and Release

Every day in America some 1,600 people will leave state and federal prisons. Most will start their journey with "gate money" (from $20 to $200), a one-way bus ticket, and little else. Many will be drug abusers who received no treatment for their addiction while on the inside, sex offenders who got no counseling, and illiterate high school dropouts who took no classes and acquired no job skills. A lot of them will be sick: rates of HIV, tuberculosis, and hepatitis C are all considerably higher among prisoners than in the general population. Many of them will be obdurate "churners," who have already been reincarcerated for a new crime or a parole violation and are now being let out again. Only about 13 percent will have participated in any kind of pre-release program to prepare them for life outside. Nearly a quarter of them will be sent home unconditionally and with no supervision. And two thirds (up from one half in 1984), according to the Urban Institute, will return to just a few metropolitan areas in their states, where they will be further concentrated in struggling neighborhoods that can ill afford to accommodate them.

It's not quite fair to say that no one thought about these sorts of things when the rage for incarceration began to dominate American crime policy, in the early 1980s, but it's not far from the truth either. Almost all prisoners get out eventually. What happens when they do, however, is not a topic that held the interest of the legislators who passed mandatory-sentencing laws, abolished parole boards, and eliminated funding for prisoner education. As a result, prison sentences have grown longer while prisons have become places where nothing is done to reprogram criminals for the life outside to which 95 percent of them will return. "Our contemporary prisons basically replicate the social order that produced the offenders to begin with," says Mark A. R. Kleiman, a professor of public policy at the University of California at Los Angeles. "Their signal qualities are violence, idleness, and noise."

Until the early 1980s prison education and rehabilitation programs were deeply embedded in American corrections. But over the past twenty years -- a period in which the U.S. prison population has increased fourfold -- vocational and educational programs for prisoners have dwindled steadily. Funding once earmarked for such programs has gone instead toward constructing new facilities and providing health care for an older and sicker inmate population.

Prison programs lost their funding partly in response to research in the 1970s that implied they had scant success in cutting recidivism. But new studies suggest that certain kinds of programs do work to increase employment and reduce criminality. Adult literacy and GED classes, vocational training with a realistic eye to the job market, cognitive therapy for sex offenders, and drug-abuse counseling that continues after release have all shown modest but cost-effective success. A recent study sponsored by the Virginia Department of Correctional Education, in which ex-inmates were tracked for fifteen years, found that recidivism among those who had pursued an education while in prison was 59 percent lower. More-comprehensive studies on prison educational programs have shown that reincarceration is 20 percent less frequent for participants. Even allowing for the inevitable selection bias (those who enroll in optional prison programs are more motivated to succeed in the first place), these are pretty encouraging results. "It's an ironic story," says Todd Clear, a professor at the John Jay College of Criminal Justice, in New York, "in that just as the evidence for programs that reduce recidivism was growing, the willingness and capacity to fund them shrank."

Meanwhile, the culture of parole has changed too. Resources for supervising parolees have not kept pace with the growing numbers of them, so caseloads are bigger. In the 1970s the average parole officer oversaw forty-five ex-offenders, according to the Urban Institute, whereas today the number is about seventy. One consequence has been an emphasis on surveillance rather than more time-consuming personal relationships between officers and parolees, which has meant that many more parolees are charged with technical violations. Indeed, parole violators make up a rapidly growing class of prisoners: in 1980 they accounted for 18 percent of admissions; today they account for a third, and most of them have been sent back on technical grounds. But many technical violations -- leaving a designated area, not showing up for a meeting with the parole officer -- aren't crimes, and it's not at all clear that improving the capacity to detect technical violations or locking up more parole violators enhances public safety. What we can say is that this approach is expensive: California, which sends more parole violators back to prison than any other state, spends some $900 million a year to house them, for average stays of about five months.

The larger problem, though, is that in many places we have replaced discretionary parole, in which parole boards decide when a prisoner is ready for conditional release, with a regime that eliminates much of the discretion not only from parole but also from sentencing. In so doing we have removed some powerful incentives for prisoners to become the sort of people we would want to send home again. Parole boards came under attack in the 1970s: the right criticized them for being too lenient, the left for being too hard on minorities, and good-government types for being too beholden to the narrow political interests of the governors who appointed them. In 1977 parole-board decisions still accounted for 88 percent of all prisoner releases; by 2000 they accounted for only 24 percent.

In some ways this was an improvement: it removed an element of arbitrariness from the parole process. But the social cost was high. As Joan Petersilia, a criminologist at the University of California at Irvine, explains in a forthcoming book, When Prisoners Come Home,

A majority of inmates being released today have not been required to "earn release" but rather have been "automatically released." Parole boards used to examine a prisoner's "preparation" for release, including whether she or he had a place to live, a potential job, and family support. With determinate sentences [fixed prison terms in which parole boards have no say], these factors are not relevant to release. When offenders have "done their time," they are released no matter what level of support is available to them or how prepared they are for release.

One ramification of this is that in a growing number of cases prisoners have "maxed out" on their sentences and are being released unconditionally. Because of new so-called truth-in-sentencing laws (mandating that a prisoner serve most of his or her sentence behind bars and not on parole), by the time some prisoners get sprung, the penal system no longer has any hold over them. In 1977 only four percent of prisoners maxed out; by 1999, 18 percent did. "We have about 150,000 people getting out scot-free each year now -- no supervision, nothing," Petersilia told me recently. And although some of these are minor offenders who served short sentences, many are what the former assistant attorney general Laurie Robinson calls the "baddest of the bad" -- prisoners who failed to qualify for any early-release credits, or who had committed such violent crimes that they were ineligible for early release. "In Massachusetts," Mark Kleiman says, "you can graduate from a super-max facility at Walpole -- where the lights are on all the time, you never see another human face because all the guards are wearing hockey masks, and you leave your cell one hour a day for exercise -- straight to the street. And that is not atypical. But nobody ought to be able to walk straight from a prison to the street. Inmates need to decompress. That's what halfway houses were supposed to be for."

Over the past year or so some of the dismal facts about "prisoner re-entry" have come under new scrutiny and have begun to generate some creative thinking. For one thing, parole clearly needs to be reformed. It is not working: more than 40 percent of released inmates are back in prison within three years. Part of the problem is the all-or-nothing response to technical violations. "We need a system that does not have as its only sanction ending the experiment of parole for someone entirely," says Kleiman, who advocates a range of prescribed and immediate but less drastic sanctions, such as short confinements in "halfway-back houses," for people who have been caught in technical violations of parole.

A handful of jurisdictions around the country -- including Richland County, Ohio, and Fort Wayne, Indiana -- are now experimenting with a promising institution called re-entry court, which is charged with overseeing a prisoner's reintegration into society. Here "conditions of parole are openly agreed to and openly enforced," as Jeremy Travis and Sarah Lawrence, of the Urban Institute, write. "If a new crime is committed, all bets are off and the parolee is prosecuted for the new crime." The violations of parole that now fill prisons, though, are handled differently -- with "support services, close judicial monitoring, graduated sanctions for failure to meet conditions, and local detention where needed to enforce the orders of the court." Travis has argued for a twist on this approach: making sentencing judges responsible for coming up with a re-entry plan for prisoners. Judges would tell men and women they had just sentenced that they must begin preparing in prison for the return home, and would order drug rehabilitation, job training, or whatever other programs were called for.

At the same time, prison programs have been experiencing a renewal. A few prisons are involving families and community groups in an inmate's release plan before he is let out. The Vera Institute of Justice's Project Greenlight, for example, brings representatives of community organizations into the Queensboro Correctional Facility, in New York, to talk to prisoners about jobs they might seek once they're free. It also provides counseling for prisoners and their families before the prisoners' release. When prisoners serve longer sentences, as they have done in recent years, family ties are likely to be more attenuated, meaning that inmates are likely to require more help to reconstruct them. Of course, not everybody wants to participate in easing a relative's transition from prison; often family members have been the inmate's victims. But when prisoners can go home to families, they seem to fare better, both in the crucial first few months after release and later.

A few prison systems, notably Oregon's, have been trying out more-practical kinds of vocational training, geared to job openings in fields such as telemarketing and computer-aided mapping of water and tax districts. This has cut recidivism. Meanwhile, Missouri's prison system, under the leadership of Dora Schriro, has come up with a more comprehensive approach whose premise is that prison life should actually resemble real life as much as is practicable. Every offender engages "during work and non-work hours in productive activities that parallel those of free society," as Schriro describes the rules in a paper written for the National Institute of Justice. "In work hours offenders go to school and work and, as applicable, to treatment for sex offenses, chronic mental-health problems, and drug and alcohol dependencies. In non-work hours they participate in community service, reparative activities, and recreation."

Schriro says that from 1994 (when the program went into effect) to 1999 the proportion of inmates returned to prison in Missouri for felony offenses fell from 33 percent to 20 percent. That's impressive in itself. Still more promising, however, is the larger idea this approach evokes. It may be that as a society we want to keep our incarceration rates higher than those of other industrialized democracies (though not, surely, as high as they have been, given how many prisoners are parole violators and drug offenders). After all, there is fairly good evidence that the prison boom was responsible for about a quarter of the decline in crime in the 1990s. But if we do want to keep our prisons full, we must endow them with a purpose broader than incapacitation. We will have to take up again, in new form, the goal of remaking prisoners for life beyond bars. We will have to accept that the question before us is not only how stringently we want to punish people in prison but also what kind of people we want to see emerge from it.