Should prisons be privatised?

“I believe a big part of our problem is that the very violent inmates, like the three that escaped, [were] sent to private prisons that were just not up to the job”, argued Arizona Attorney General Terry Goddard following the escape of three murderers from Arizona’s Kingman private prison on 30 July 2010. According to the state report released after the events, the private prison faced numerous security breaches that night, bringing the case of private prisons to the forefront. This debate is still alive today.

The long lasting story of private prisons

Prisons have been private, since the medieval age–for almost a thousand years. They were meant to make profit for the owner and the prisoners had to pay for their punishment when the poorest among them depended on bequest and alms. In the eighteenth and nineteenth century, some entrepreneurs specialised in transporting prisoners to the colonies and made large profits. Prisons only became public in the twentieth century following the desire of most countries to get their hands back on incarceration.

This has been the norm for many decades. However, it is progressively questioned in many Anglo-Saxon countries, among which the US. Since the 1990’s the private sector has become increasingly present in American prisons, either through partnerships with public facilities or with total ownership of the prison. The private sector has grown considerably with about 1,200 “private prisoners” in 1985 to almost 50,000 inmates in 1994 and 120,000 in 2017. However, this figure remains less than 10% of the total state and federal prison population.

In general, opponents to prison privatisation consider private prisons as cheap and poor quality facilities, while public prisons are considered to be more expensive but also safer and of better quality. In order to elucidate the debate, one should first investigate whether this popular view is fact or fiction.

Public Vs Private: a tradeoff between cost and safety?

According to D. Perrone et al. (2009), “the empirical evidence regarding whether private prisons are more cost-effective and whether they provide a higher quality of confinement to inmates, however, is inconclusive.”

Nevertheless, in the vast majority of the studies conducted (B. Lundahl et al. (2007)), publicly managed prisons perform better in terms of public safety with clearly better training programs and fewer complaints or grievances.

The Arizona Department of Corrections report of 2011 has shown that if private prisons are cheaper, it is because they host low-cost inmates and leave the sickest and most costly convicts to the public sector. Private prisons also often pay lower wages and hire less-skilled workers. Since labour accounts for two-thirds of the cost of incarceration, this can represent savings of up to 10%. However, the quality of service deteriorates with more violence and deaths due to inexperienced guards.

Consequently, assuming that private prisons are cheaper, a trade-off seems to exist between quality and cost. The basic idea is that, ignoring the effect of competition, the main difference between private or public ownership is the allocation of residual control rights. For instance, if I have my own business rather than working for someone else, I have the residual control rights. This would certainly have consequences! But how and why does it change the way I would work?

Allocation of control rights under incomplete contracting 

Under complete contracting, allocation does not matter. The government should be able to sustain the first-best level of action whether or not he is owning the facility.

However, under incomplete contracting, residual control rights provide incentives to agents in different ways (Grossam and Hart, 1995). An owner of a nonhuman asset possesses residual control rights on this asset; optimal allocation of these control rights can be found. In other words, is it optimal to give the control rights to the manager? How does this affect his incentives to reduce costs or quality? Before answering this question, we must understand why building complete contracts is more or less impossible.

Contracts can never be complete, especially in the case of incarceration services. Even if a broad set of standards is established, many events are still too difficult to describe in a contract, or too difficult to predict, leaving the prison governor some leeway.

The American Correctional Association is a private association that sets prison quality standards and accredits institutions that meet those standards. These hundreds of standards cover a wide range of issues such as administration and management of prisons (personnel policies, staff training, etc.) physical plant, operations, services and inmate programs. These norms are used by the U.S. government to establish the most comprehensive contracts possible.

However, no matter the number and complexity of these standards, they still does not make a fully complete set of contracts feasible. First, they are based on process rather than outcomes. This guarantees that the facilities have rules and staff dedicated to deal with a specific matter; but does not specify the content of the rule. Second, two areas considered crucial for good quality prisons are not properly supervised: the use of force and the quality of personnel. For the use of force, it is only stated that “written policy, procedure, and practice restrict the use of physical force to instances of justifiable self-defence, protection of others, protection of property, and prevention of escapes, and then only as a last resort and in accordance with appropriate statutory authority. […] A written report is prepared following all uses of force and is submitted to administrative staff for review.” It is just as vague for the use of firearms with relatively low restrictions. With regard to the quality of staff, the vacancy rate and the number of hours of training to be provided are specified, but little is said about the quality of the training or the officers that can be employed. Few standards for recruitment requirements are clearly stated and contractors have considerable latitude to save on staff costs. These grey areas leave some room for manoeuvre for the manager, which can have an adverse impact on quality.

The state of New Jersey experienced the unfortunate consequences of incomplete contracting with the case of the ESMOR’s detention centre in Elizabeth. In the 2000s, ESMOR won the contract by underbidding its competitors because it had assumed lower wages for the prison staff in its bid. Due to difficulties in hiring guards at that price, they ended up hiring people who had previously kept goods in warehouses. As a result, the facility was severely understaffed and when a riot broke out, the guards ran away and called the police.

Do private prisons have more incentive to lower quality ?

Both private and public facilities are equally affected by incomplete contracting. However, we do not observe such behaviours in public prisons. Incentives are indeed different depending on, as we said earlier, the allocation of residual control rights. Private contractors seem to have more incentives to lower quality to save costs.

The model of O. Hart et al. (1997) attempts to explain both why private contracts are generally cheaper and why, in some cases, they may offer a lower level of quality than services provided in-house by the government.

In this model, the manager of a given facility can invest his time in two types of investments: cost reduction or quality innovation. Cost reducing investments are detrimental to quality, e.g. because less qualified guards are hired. Consequently, there is something of a trade-off between cost and quality, as introduced in the examples given above. Quality innovations, meanwhile, can, if implemented, increase prisoner rehabilitation or reduce the chances of escape for instance. Neither of these investments is contractible ex-ante.

The government wishes to implement the optimal level of investment in cost reduction and quality improvement in all prisons. Note that the first-best level of cost-cutting is not zero and that it is optimal to damage quality to some extent to save money. This optimal level is determined by the principal taking into account the damages caused to society (here to convicts and citizens living around guardhouses). As these investments are not contractible, discretion, such as the level of investment implemented, is left to the prison director.

If the prison governor is a state employee, he can be replaced by the government. The employee is not a residual claimant on cost savings he implements; the government and manager bargain over the surplus created if the latter is not fired. Consequently, the incentive to do effort either in cost reduction or innovation is too weak. The level of investment for both actions is inefficiently low. If the provider is a private contractor, he is a residual claimant on cost-reducing innovations.

However, if the contractor wishes to implement quality improvements, the surplus generated by this action is split according to the respective bargaining power each party. The contractor would to renegotiate their contract with the government to be rewarded for this innovation. The incentive for the private contractor to exert effort for both innovations is greater than for the government employee. However, it is not high enough for quality innovation since the private manager is still not a residual claimant. Moreover, the social harm of cost reduction is an externality that is not internalised by the private manager. The result is an inefficiently high level of cost reduction which is detrimental to the quality of the service provided.

A compromise appears, either to have an expensive and good quality public prison or to contract with cheap private prisons whose quality of service is degraded. Even if, according to the model, more quality innovations are made in private prisons, they continue to perform poorly.

To conclude, what should we do?

The answer to this question depends on the extent of the social harm deemed acceptable. In the United States, maximum-security prisons are never privately owned to avoid the risk of escape due to lack of security. Which values do we have as a society? A major objection to privatisation is the fear that private providers will hire unqualified guards to save money, thereby compromising the safety and security of the prison. In my view, the fact that high-security prisons are never privately owned suggests that the security and protection of society as a whole are more important than the living conditions of those behind bars. In a sense, this problem is intractable and depends on the philosophy of sentencing and the nation’s sensitivity to the fate of those who have made mistakes.

A more interesting question, for an economist, is to wonder how we could improve the quality of the service provided by private prisons.

First of all, the simplest solution would be to try to build more comprehensive contracts, specifying, for example, recruitment procedures. One could also try to modify the incentives, focusing on outcomes rather than specific tasks. However, this might be difficult to put into practice due to uncertainty. Changing the prisoner assignment process to strengthen ex-post competition and provide additional incentives to improve quality, might be a more effective solution. This could be done by allowing inmates to choose their prison, but this would be a bad idea as they might choose the most lenient prisons. Prisons would have an incentive to “allow gangs, drugs, and perhaps easy escapes” (Hart et al., 1997). A better idea might be to let judges choose the prison for convicts. However, this idea is not perfect either, as some judges could choose bad prisons to impose harsher sentences and others soft prisons to soften the sentence. Moreover, given the shortage of prison capacity in the United States, this idea seems difficult to implement.

The Bureau of Prisons currently “place[s] the prisoner in a facility as close as practicable to the prisoner’s primary residence, subject to bed availability” taking into account such things as “the prisoner’s security designation, the prisoner’s programmatic needs, the prisoner’s mental and medical health needs”.

To conclude on a more personal note, private prisons appear to be a way to save thousands of dollars at the expense of convicts’ human rights. As an economist, it is not always easy to look at the big picture, as it could intractably complicate the analysis. However, violence and deaths due to accidents or abuse should be quantified in monetary terms. Besides, if private prisons are more innovative than public prisons and have better reintegration rates, this should also be taken into account. Nevertheless, given that the value of statistical life is $10 million in the United-States, the death of a guard or prisoner avoided through better management should certainly favour public prisons over private ones. Assuming that the life of a prisoner and an average American citizen are valued in the same way.

References :

Arizona Department Of Corrections (2011), Biennial Comparison Of “Private Versus Public Provision Of Services”.

B. Lundahl, C. Kunz Cyndi Brownell Norma Harris, R. Van Vleet (2007), Prison Privatization: A Meta-Analysis of Cost Effectiveness and Quality of Confinement Indicators. Utah Criminal Justice Center College of Social Work University of Utah

D. Perrone, T. C. Pratt (2009), Comparing the Quality of Confinement and Cost-Effectiveness of Public Versus Private Prisons: What We Know, Why We Do Not Know More, and Where to Go from Here. The Prison Journal, Vol. 83 No. 3, September 2009 301-322 DOI: 10.1177/0032885503256329

 M. Mumford, D. Whitmore Schanzenbach, R. Nunn (2016), The Economics of Private Prisons. The Hamilton Project. Economic analysis.

O. Hart, A. Shleifer, R. W. Vishny (1997), The Proper Scope of Government: Theory and an Application to Prisons Author(s): Contracting for Imprisonment in the Federal Prison System. The Quarterly Journal of Economics, Vol. 112, No. 4 (Nov., 1997), pp. 1127-1161.

O. Hart (2002), Incomplete Contracts and Public Ownership: Remarks, and an Application to Public-Private Partnerships. CMPO Working Paper Series No. 03/061.

U.S. Department of Justice, Federal Bureau of Prisons (2019), Inmate Security Designation and Custody Classification.

 P. Aghion, R. Holden (2011), Incomplete Contracts and the Theory of the Firm: What Have We Learned over the Past 25 Years?. Journal of Economic Perspectives, VOL. 25, NO. 2, SPRING 2011, pp. 181-97.

S. Cabral, S. Saussier (2013), Organizing Prisons through Public-Private Partnerships: a CrossCountry Investigation. BAR, Rio de Janeiro, v. 10, n. 1, art. 6, pp. 100-120, Jan./Mar. 2013.

By Pierre Vannetzel