« Duke Lacrosse IV: Universities and Student-Athletes | Main | Law Profs as Political Pundits: Should Schools Do Media Training? »

Monday, March 03, 2008

Merit Pay for Prison Wardens?

Tying the salaries of public officials to performance tends to be a policy touted by Republicans as a market-based solution to poor bureaucratic incentives. The most obvious example is merit pay for public school teachers, a prescription urged by Republicans but generally resisted by the National Education Association and the American Federation of Teachers, major constituencies of the Democratic Party. But I suggest a version of the reform that might be more palatable to the Left: Why not have merit pay for prison wardens and tie the bonuses to the employment and recidivism of prisoners following incarceration?

One of the perversities of incarceration is that prison officials have no political or financial incentive to care about what happens to prisoners after release. Few politicians, let alone voters, trace where ex-cons did their time. Whether or not a released prisoner avoids recidivism after release, the warden who supervised the felon gets neither praise nor blame. Small surprise, then, that wardens pay little attention to rehab. As an example, consider the Michigan Department of Corrections’ sanctions for substance abuse, affirmed by the U.S. Supreme Court in Overton v Bazzetta, 539 U.S. 126 (2003). MDOC suspended prisoners’ rights to non-contact visits for two years, as part of a get-tough-on-drugs-and-alcohol policy. Preventing prisoners from speaking to their spouses and children for two years insures that the prisoner will fall back on the only other “family” available – the prison gangs – and, according to every expert who testified in the case, will almost certainly make prevent the re-integration of the prisoner back into any semblance of a normal family life. To avoid a trivial chance that visitors might somehow smuggle contraband past a plexiglass barrier, the MDOC risked imposing a huge costs on the rest of society.

Enter merit pay. The state already has data on ex-cons’ post-incarceration behavior -- parole violations, employment rate, recidivism rate, etc. Why not tie wardens’ compensation to reductions in post-incarceration anti-social behavior of the prisoners released from their prisons? Of course, one would have to control for the usual variables beyond the wardens’ control – SES and offense level of the prisoners, for instance. But these statistical challenges are no greater than those posed by merit pay for public school teachers. Indeed, unlike test scores as a measurement of education, the performance standards for ex-cons – no recidivism, for instance -- is relatively uncontroversial.

So what do you think? Will merit pay for prison wardens become a plank in a new bi-partisan Obama or McCain policy platform? Or are prisons the one sort of Big Government that neither party wants to reinvent?

Posted by Rick Hills on March 3, 2008 at 11:14 AM in Law and Politics | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Merit Pay for Prison Wardens?:


Great question Rick.
I wonder if prison wardens have the same kind of discretion and control that teachers purportedly do in order to benefit from assumptions of merit pay. Also, maybe the wardens are more like principals and corrections officers are better analogized to teachers here?

Posted by: Dan Markel | Mar 3, 2008 11:26:58 AM

This is an interesting question, but I share Dan's concern that individual wardens and guards can't be expected to bear meaningful responsibility for each prisoner's post-release conduct. At the primary school level, each student typically has a single teacher who teaches them for an entire year, at which point the student's progress can be assessed. By contrast, prisoners are subject to the control and influence of many different wardens and guards, and are also affected by other professionals whose job it is to reduce the possibility of future recidivism (psychologists, volunteer educators in some instances). Plus, while we can measure a student's progress at a single moment (after completion of a grade level), the same isn't really true of a prisoner's tendency toward recidivism. We might say that the benchmark is something like number of crimes committed five years after release, but this wouldn't tell us, for example, if recidivism happened because of bad warden-ing or because other unforeseeable intervening factors led to the commission of repeat offenses.

All this having been said, perhaps an incentive system would still have good results--it might encourage guards to act in a way we prefer even if the actual odds of their conduct influencing recidivism is vanishingly small. For this reason, it might be worth introducing the program but only in one direction: give guards credit for good prisoner conduct but no demerits for bad conduct.

Posted by: Dave | Mar 3, 2008 12:11:44 PM

These worries are precisely the same as the worries expressed by the AFT and NEA over merit pay: kids' performance on tests is influenced by many factors outside the school's control. So, at the very least, if you are a skeptic about merit pay for prison officials, you ought to be equally skeptical about merit pay for teachers.

Indeed, it seems to me that Dan's and Dave's objections cut in favor of merit pay for prison officials. Wardens have far more control over their prisoners than principals have over their students: Children go home to households that are largely outside the power of the school entirely. Surely, teachers are right to complain that parents' influence dwarfs that of the school. For prisoners, the prison is home -- a total environment that can be manipulated by the warden for good or ill.

At the very least, merit pay systems would finally give the state an incentive to link recidivism data to prison practices and begin to think seriously about which practices can reduce recidivism (not to mention unemployment, drug dependency, etc). Since James Coleman's 1966 report on educational opportunity in schools, there has been a massive effort to collect data on which aspects of schools might affect educational performance. Some of this data has been famously surprising (per pupil expenditures, for instance, seems to have negligible effects unless one pays close attention to how the money is spent). But no such effort has been made with prisons' practices. Given that our willingness to subject minors to adult sentences seems to be steadily increasing, the analogy between prisons and schools is becoming ever more obvious. Perhaps we just need a Coleman Report for prisons to help us see the light.

Posted by: Rick Hills | Mar 3, 2008 12:36:32 PM

Post a comment