America releases 600,000 prisoners each year, does little to prepare them for work or improve their habits and, not surprisingly, re-arrests most of them within three years. People who have already spent time in prison or jail move back to some of America's poorest neighborhoods to terrorize neighbors who can ill afford the costs of crime. While quadrupling the number of people in correctional facilities over the past 30 years has played a major role in reducing crime, America's burgeoning prison system has done a poor job insuring that convicts leave the prison gates ready to lead productive lives.

A new Urban Institute study, "From Prison to Home: The Dimensions and Consequences of Prisoner Reentry," provides frightening documentation of America's failure to improve the prospects for released prisoners. The past decade, indeed, has seen nearly every indicator move in the wrong direction: Fewer prisoners get education and drug treatment behind bars while more violate parole terms. Despite tough-on-crime rhetoric, over 100,000 people a year get released without any supervision and per-convict spending has fallen for those who remain monitored. Few released convicts find full-time work and many abandon spouses and children. Some policies reek of negligence: California, for example, releases mentally ill patients with only one day's worth of psychotropic medication.

This situation needs to change. Jeremy Travis, the former head of the National Institute of Justice who conducted the study along with Amy Solomon and Michelle Waul, has often observed that the one thing we know for sure about the people we send to prisons is that nearly all of them will get out. Working to improve prisoners' social integration should find a place at the heart of a conservative, tough-on-crime policy. When hardened gangsters begin terrorizing their neighborhoods as soon as they step off the bus, it's clear that the punishment they received by going to prison didn't do much good. So where does this lead us?

To begin with, we need more research. Prior to the current study, nobody had taken a comprehensive look at prisoner re-entry for 20 years. The Urban Institute plans more studies and one can hope other think tanks and universities will take notice. Existing research, however, suggests that conservatives should support four policies: improved follow-up, better drug treatment, in-prison work programs, and faith-based rehabilitation.

First, states should follow up with nearly every inmate released from prison or jail. The Urban Institute found that most states have no formal way of telling released prisoners how to report for correctional supervision, much less where to find work or get medical care. Parole and probation officers have seen their caseloads swell along with the inmate population: While existing programs in Boston and Orange County have shown that intensive follow-up including home visits can do a lot to protect communities, most areas will not spend money to replicate these programs. Almost nobody, meanwhile, has even tried to replicate intensive supervision for supposedly less dangerous offenders who, for example, "only" break into cars rather than steal them. Long stretches of intensive monitoring for these criminals might provide more effective punishment and better community protection than a few extra months in jail. The police can also play a role and corrections officials can help by telling them where recently released convicts plan to live. Likewise, states should require men and women who have spent more than a few months behind bars to organize their post-incarceration lives. In addition to the traditional bus ticket and cheap suit, released prisoners should walk through the front gate with a plan for their post-prison lives.

Second, we should increase spending on drug treatment for everyone under correctional supervision. The easy availability of drugs behind bars lets prisoners believe that drug use has no real consequences. Since about one-third committed their crimes on drugs, this proves disastrous. Forcing convicts to undergo drug treatment both on probation and while behind bars would improve their behavior upon release. Likewise, states should redouble their efforts to keep drugs out of prisons by improving education and training standards for the guards who are the primary conduits for illegal drugs.

Third, work programs, particularly those linked to commercial enterprises, offer significant hope for rehabilitation. The federal prison system, where nearly all inmates work for pay, presents a model in this regard. The discipline and standards of work can adjust prisoners to social norms while wages offer them a chance to pay restitution, offset the costs of their own imprisonment, and build a nest egg to begin a normal life upon their release.

Finally, we need to allow more experiments with faith-based programs but shouldn't expect them to help everyone. In a 1974 Public Interest article, scholar Robert Martinson famously concluded that "nothing or almost nothing works" in rehabilitating prisoners. While this remains conventional wisdom, Martinson later revised his thinking: Individual programs can work but no broad category or approach presents a foolproof path to prisoner rehabilitation. This indicates that we need to try as many different approaches as possible. A long history of left-wing anti-religious bias has insured that faith remains one of the least tried approaches and, given monotheism's 4,000-plus year track record of success in improving people's lives, it deserves a chance.

Not even the power of faith, however, will successfully reintegrate all of the 600,000 men and women who leave correctional facilities each year. But being tough on crime mandates that we start caring about them.

Eli Lehrer is a former Visiting Fellow, Center for Legal and Judicial Studies at The Heritage Foundation.