Prison Vouchers

This is the first of several posts in which I’ll be serial-blogging my new article, Prison Vouchers, forthcoming in the Penn Law Review. I’d appreciate comments (especially informed ones, which have a greater chance of making it into the final version, with thanks in the author footnote).

* * *

In this Article, I invite the reader to indulge in a thought experiment. What would the world look like if, instead of assigning prisoners to particular prisons bureaucratically, we gave them vouchers, good for one incarceration, that they were required to redeem at a participating prison?

School vouchers have been debated to death ever since they were proposed by Milton Friedman in 1955 and championed by progressives in the 1960s. Vouchers have also been discussed and used for other government services, like housing and health care.

Vouchers are also no stranger to criminal justice: We see them used for halfway houses, mandatory anti-alcohol programs, and criminal defense lawyers for the indigent. A voucher system was implemented in a few states in the 1970s to allow inmates to buy training and education as part of “Mutual Agreement Programs” (also called “contract parole” programs) that helped inmates work toward parole. This last idea was taken up in the Model Sentencing and Corrections Act in 1978, which suggested that prisoners get vouchers to purchase “specified treatment programs and services directly from either public or private agencies,” though this feature of the Model Act apparently hasn’t been adopted anywhere.

But, as far as I can tell, no one has ever discussed vouchers as a serious possibility for prisons.

This is a shame, because some of the same factors that led early education reformers to suggest school vouchers apply with equal, if not greater, force in the prison context. Both prisons and schools face a similar confluence of three factors:

  1. Both face widespread and serious problems.
  2. The problems in both areas have proven hard to solve through the usual political, administrative, or judicial means.
  3. And allocation of students to schools, and of inmates to prisons, is predominantly done bureaucratically, with limited possibilities for choice.

The prima facie case for considering a market solution, in which the subject population would become consumers and thus drive reform by voting with its feet—essentially, getting rid of (3) to bypass (2) and thereby solve (1)—thus seems strong.

Let me focus on (1) for a bit. Modern American prisons—with their high violence rates, bad medical care, overuse of highly punitive measures like administrative segregation, and the like—are widely believed to be low-quality. (Compare this to the views of early school voucher proponents on the left, who wrote that “[t]he public schools have not been able to teach most black children to read and write or to add and subtract competently” and that the public school system “destroys rather than develops human potential.”)

We should care about prison quality even if we don’t care about prisoners: Bad prison conditions often indirectly hurt the rest of us too. Brutal conditions, as well as excessive use of high-security segregation, make prisoners less useful members of society and more likely to reoffend. The low level of educational, vocational, or rehabilitative programs also contributes to recidivism. And communicable diseases can spill over into the outside world when infected inmates are released. “Every year, more than 1.5 million people are released from jail and prison carrying a life-threatening contagious disease.” The risk in multi-drug resistant tuberculosis in New York in the 1980s and early 1990s may have been linked to poor medical treatment in prisons and jails.

There are thus clear opportunities for gains from prison vouchers—not just to prisoners but also to society at large—as competing prisons seek to attract prisoners by offering better security, medical care, and vocational programs.

But, now focusing on (2), why can’t we “just” fix prisons by other means—by reform legislation, by administrative oversight, or by litigation?

Legislative prison reform is a tough sell. Unlike with schools, where at least all politicians claim to like kids, criminals are widely vilified. Nor are elected officials eager to fund prisons. Some reformers recognize that prison administrators or legislators have little interest in spending money on prisoners or otherwise improving prisoners’ lives, but nonetheless “urge” and “encourage” these same officials to spend the money or implement the reforms. Of course there’s nothing wrong with urging, and some reforms have been implemented even in the face of political pressure to the contrary. But it’s unsurprising that mere urges haven’t gone very far.

Administrative solutions are likewise difficult: Prison officials remain resistant to “scrutiny by ‘outsiders.’” Independent inspection and monitoring, as well as internal oversight mechanisms, like effective grievance systems, are underused.

Judicial solutions are also unpromising. Courts often defer to the judgment of prison administrators, and prisons are exempt from Administrative Procedure Acts in many states. Prisoner litigation, whether on individual claims or in more ambitious prison reform cases, is restricted, for instance by the Prison Litigation Reform Act, and in any case prisoners’ rights are quite limited.

Some have suggested contracting out prison management to the private sector (but holding the method of allocating prisoners constant) as a way of improving prison quality. This is a controversial proposition—others categorically deny that contracting out improves prison quality, and even some of those who are more sympathetic to contracting out grant that the evidence on quality is mixed.

Here, too, the parallels with the public school debate are clear. Whether the blame lies with teachers’ unions or with legislatures unwilling to spend money on schools in politically powerless areas, schools have been hard to reform politically. Litigation hasn’t worked well, and any constitutional rights to a good education are generally weak. Privatization of entire school systems within the context of mandatory government provision has been tried sporadically, and the results haven’t been terribly impressive so far.

The market approach that vouchers represent has an obvious appeal in this context. By empowering the prisoners themselves to reward and punish prisons, it would create powerful incentives for prisons to become better—by the prisoners’ own standards. No longer would one have to “urge” prison administrators or legislatures to reform conditions in the interest of prisoners, or try to convince them that prisoner welfare is aligned with the social interest—a strategy that has not worked terribly well so far. Instead, prison administrators would be moved, as if by an invisible hand, to make their prisons better places.

Moreover, vouchers would also increase prisons’ constitutional flexibility.

First, prisons would be freer to experiment with religiously inspired rehabilitation: faith-based prisons, whose constitutionality under the current regime is dubious, would become fully constitutional. Under vouchers, the prison system would come within Zelman v. Simmons-Harris because prisons would be participating on a neutral basis, independent of religion. As is already the case with halfway houses, residential programs for delinquent children, or alcoholism or drug addiction programs, participating providers could be public or private, religious or secular. The inmate’s ability to choose from a variety of providers, not all of which are religious, would be what makes these programs “voucher-like.”

And vouchers would do more than make the current batch of faith-based prison programs constitutional. When faith-based prisons operate today, they necessarily try to be ecumenical to some extent. But suppose a religious organization isn’t satisfied with such a program, just as it might not be satisfied with ecumenical prayer and other watered-down expressions of religion in public schools. With vouchers, it could become far more ambitious. A religious group could run its own prison and advertise inmate rehabilitation through intensive, and intensely sectarian, exposure to religion.

Vouchers could thus be the best, or perhaps even the only, way to save faith-based prisons.

Second, prisons would be freer to offer inmates packages of features that currently would be considered unconstitutional. Prisoners have dramatically reduced rights, but they still retain some. In general, people—prisoners or not—may benefit from being able to waive their rights in exchange for other benefits; for instance, inmates may agree to waive some part of their due process rights in exchange for better health care. A prison’s ability to offer such a package is limited by the unconstitutional conditions doctrine. I argue that, in a prison system that is more competitive from the inmate’s point of view, the unconstitutional conditions doctrine would (and should) be more permissive, because the inmates’ ability to choose would mitigate the possibility that prisons would be illegitimately taking advantage of inmates.

With all this going for it, what could go wrong?

One possibility is “market failure”—that inmates’ individual decisions won’t succeed in improving overall prison quality. This could be because inmates can’t make themselves better off through their decisions (perhaps they are poorly informed about prison quality), or because some inmates’ decisions will make other inmates worse off (perhaps the better informed inmates will get the best prisons and leave the bad prisons to the uninformed).

Another possibility, one that I think is more serious, is what I call “market success”—that inmates will succeed in improving prison quality by their own standards, and that this is precisely the problem. Prisoners’ preferences aren’t always good; we are, after all, talking about (presumptive) criminals. At worst, satisfying prisoner preferences may directly make the world a worse place (perhaps members of a gang choose to be together and are thus better able to run criminal enterprises on the outside). At best, it may merely make prison a less undesirable place and thereby undermine the deterrent value of prison (perhaps the prison becomes a country club).

I take these counterarguments seriously, and so I make no strong claims about the bottom-line merits of prison vouchers. If the potential downsides (that prisoners, through their voucher spending, drive prisons to adopt undesirable features in ways that can’t be adequately controlled by the political process) are outweighed by the benefits (improvements in prisoner security, health care, and education), vouchers could potentially dramatically improve penal policy. But this paper is meant to spur further research and debate on the question, not to come down on one side or another.

More details next time!