By STUART M. BUTLER

By the end of 2013 there were approximately 1.5 million people in state or federal prisons, and the U.S. incarceration rate is the highest in the world. And while there is debate about the relationship between this level of imprisonment and crime rates, there is considerable research to show that a spell of incarceration exacerbates economic and social conditions for families as well as former inmates, especially in low-income neighborhoods. That has led the Obama Administration and some interesting strange-bedfellow groups to call for alternatives to prison for some infractions.

The other side of the prison coin is recidivism. Prisons are often called “correctional facilities” but that is a cruel joke – they do a dismal job in turning lives around. According to the U.S. Department of Justice, about two-thirds of released state prisoners were re-arrested within three years and three-quarters within five. Prison is a revolving door.

It’s true that there can be opportunities in prison for inmates who dropped out of high school to study for a GED to help prepare them for release. However many prisons don’t make this easy. I have helped one inmate studying for his GED by supplying study books – only to discover that the facility places a six-book limit on any inmate’s bookshelf. Prisons may also teach an inmate a new work skill – although many ex-offenders discover they face licensing barriers that makes their new skill worthless.

So why don’t prisons do a better job at helping inmates to re-enter society and workforce after their release? Because they have little incentive to do so. The revolving door is good for business. It means more jobs for guards, more prison construction and usually fatter budgets for the prison system. If prisons did things to cut recidivism it would mean fewer repeat customers. Like Motel 6, their incentive for prisons is to tell people checking out, “We’ll keep the lights on for you.”

What if hospitals had an incentive for you to keep coming back to the emergency room? Well, the irony is that until recently that was the economic incentive. Say a patient was discharged, but became a “recidivist” – reappearing at the emergency room. That could be because they didn’t take their medicine properly, or didn’t change their dressings regularly, or simply did not have the family or social service supports to help them recover. In that case the hospital typically was able to bill insurance, or Medicaid or Medicare, once again for the same diagnosis and treatment. Just like prisons, “repeaters” were good business for hospitals.

But recently, the incentives for hospitals have been altered when they discharge a Medicare patient. Today if that patient is readmitted to a hospital – any hospital – within 30 days for the same diagnosis then the first hospital faces a fine, called a “re-admission penalty.” Hospitals in Maryland face similar penalties for non-Medicare patients, and even if the 30-day readmission is for another diagnosis. The rule went into effect three years ago.

The results? Readmission rates have begun to fall – although it’s still too early to know the full causal effect of the penalty – as hospitals have scrambled to find ways to help their discharged patients take better care of themselves rather than call 911. The new incentive is triggering a change in the hospital business culture. After first grumbling about the rule, hospitals are now taking new steps to make sure patients are more likely to make a full recovery after discharge. For instance, many facilities are hiring new staff to help elderly patients move to an apartment where they are less likely to fall, and helping with follow-up visits or with signing up for support services.

Imagine if prisons faced a readmissions penalty. Let’s say that if an unusually high number of released inmates from a particular prison were convicted and sent back to prison within three years then the prison’s budget would be cut and the bonuses and salary increases of senior prison staff trimmed back. Just as with hospitals, the first reaction would be to complain at the “unfairness” of being held liable for a released inmate’s return to crime. But after that the prison management would start to do a much better job than today in preparing inmates for re-entry into the community. Petty restrictions and surcharges on phone calls to family members would quickly go – the erosion of family ties increases the likelihood of a return to crime. Limits on GED textbooks would certainly vanish.

Prisons would get serious not only about training inmates but also about working with potential employers to help line up jobs. Instead of dumping released prisoners on the street, prison managers, like today’s hospital managers, would become more interested in arranging stable housing for their ex-customers.

The current incentive system for prisons rewards recidivism, just as the Medicare payment system for hospitals used to rewarded readmissions. Changing the incentives for hospitals has encouraged managers to find ways to improve the long-term health of their Medicare patients. It’s now time to change the readmission incentives for prisons.

Stuart M. Butler is a senior fellow in Economic Studies at the Brookings Institute.