All Commentary
Thursday, August 1, 1985

Competition in the Prison Business

A specter is haunting American corrections—the specter of free enterprise. To be clear at the beginning: no entrepreneurs are challenging the state’s monopoly of penal authority. However, there are those who believe that the actual day-to-day business of imprisonment should not be left entirely in the hands of government employees. It is one thing to believe that only the state has a right to imprison someone. It is another matter entirely to believe that only the state can run a prison in a fair, humane, effective, and economical fashion. The first belief is a matter of political philosophy; the second is an empirical proposition. Given the dismal performance of the state in running its prisons, many people now are willing to entertain, and to test, the proposition that private enterprise can do it better.

The Fruits of Monopoly

What do we normally expect in a business that is protected from competition? More often than not, we find unmet demand, low quality, and high cost. Precisely these conditions now characterize the prison business. Rising crime rates and more punitive public conceptions of justice have increased the demand for imprisonment. Supply, however, has not kept up with demand, so prisons and jails have become seriously overcrowded. The Bureau of Justice Statistics reports that in 1983, state and federal prisons operated, on the average, at about 110 per cent of capacity. In 15 states, 21,420 prisoners were released early because of overcrowding. Over 8,000 other prisoners, in 18 states, were held in local jails because of prison crowding. In a recent survey of corrections officials, police, prosecutors, public defenders, court personnel, and probation and parole officials, a majority of all but the public defenders identified overcrowding as the number one problem of the entire criminal justice system. This was in response to an open- ended question (no prompting or choices offered) and for most groups this problem was far ahead of the nearest contender.

The strain on quantity also affects the quality of imprisonment. Corrections departments have been forced to double-cell and to house the overflow of inmates in tents, gymnasiums, corridors, recreational rooms, and basements. San Quentin, built to hold 2700 but currently housing 3900, must, to insure security, confine most of its prisoners to their cells continuously, allowing them to emerge only for meals and showers. Courts have begun to declare these conditions intolerable. As of 1983, 41 states and the District of Columbia either were under court order to remedy prison conditions or were the subjects of litigation.

However, it is much easier to order reforms than to pay for them. Calculations of the cost of constructing a new prison range as high as $125,000 per bed, but are highly variable. The U.S. Department of Justice, using 1982 dollars, cites average construction costs per bed of $26,000, $46,000, and $58,000 for minimum, medium, and maximum security prisons, respectively. Since they are based on a systematic survey of all states and are specified by prison type, these may be the best figures to use for average costs nationally.

Cost Overrun

Most estimates of construction cost are probably too low, because they ignore such considerations as land purchase, site preparation, overruns, hidden costs, and financing cost (for example, if construction is paid for by a 20-year bond at 10 per cent interest, the real cost will be triple the original figure). Moreover, construction cost is only the beginning; operating cost adds yearly to the burden. As with construction, figures on operating cost vary widely, running from $4,000 to $30,000 per inmate per year. The most common figures are in the mid-teen thousands, as represented by the American Correctional Association’s figure of $15,000 and the federal prison system’s estimate of $13,000.

The cost of constructing and operating prisons has been rising faster than the general level of inflation. This is not surprising; cost is predictably less restrained where the government has a monopoly over imperative services. If current conditions of demand and supply persist, we can expect the cost of imprisonment to continue its upward course. However, two recent developments may deflect this course somewhat. First, taxpayers are beginning to resist. Perhaps the public is aware that it does not get from the penal system full value for its tax dollar. This should put pressure on the government to be more efficient. Second, private entrepreneurs are beginning to offer an alternative. Where demand for a service outstrips supply and where current prices seem unreasonably high, conditions are ripe for competition and the emergence of new sources of supply. We should therefore not be surprised to see this occur.

Despite long-standing and widespread dissatisfaction with the quality, quantity, and cost of imprisonment, it still comes as a shock to most people to learn of the rise, at the beginning of the 1980s, of a new source of supply in response to this unmet, or poorly met, demand. Nonetheless, a number of entrepreneurs, many with prior governmental experience in corrections and other areas of criminal justice, recently have come to believe that they can build and run prisons at least as effectively, safely, and humanely as the state, but with greater efficiency. This would lower the cost to taxpayers while allowing a profit for themselves.

The Private Prison Pioneers

Actually, virtually all aspects of corrections (food services, medical services and counseling, educational and vocational training, recreation, maintenance, security, industrial programs, and so on) are already subject to private contracting on a piecemeal basis. What is new is the idea of running entire, secure institutions completely under private contract. These facilities are usually, but not necessarily, built and owned by the companies that manage them.[1]

The first such facility dates to 1975, when RCA Services Company, a subsidiary of Radio Corporation of America, established the Intensive Treatment Unit, a 20-bed, high security, dormitory style training school for delinquents at Weaversville, Pennsylvania. The second did not appear until 1982, when the State of Florida turned over the operation of the Okeechobee School for Boys to the Eckerd Foundation (created by the founders of a drugstore chain). In 1983, a prison for youthful offenders near San Francisco was contracted by the U.S. Bureau of Prisons to Eclectic Communications, Incorporated.

Most recent of all have been contracts for the private operation of secure institutions for adults. By the end of 1984, the U.S. Immigration and Naturalization Service (INS) had contracts with two private companies for the detention of illegal aliens (including facilities for fatal-lies). One of these, Behavioral Systems Southwest, has been setting up and running detention centers for the INS since 1983. They run facilities totaling 350 beds in Arizona, California, and Colorado and have proposed additional centers in New Mexico. The other INS contractor, Corrections Corporation of America, constructed and opened a $5 million, 350-bed detention center in Houston in 1984 and has been awarded a contract for another facility in Laredo, Texas, to open in 1985 with an expected daily population of 175.

At the local government level, adults are also being held in several privately owned and operated facilities. Corrections Corporation of America owns and runs a 250-bed, medium security correctional facility for Hamilton County, Tennessee. Southwest Detention Facilities owns and runs jails for counties in Texas and Wyoming and is negotiating for similar facilities in Colorado and New Mexico. So far, however, I have learned of only one currently operating commercial prison for convicted adult felons: in 1984, Palo Duro Private Detention Services opened a 575-bed medium security prison for convicted immigration offenders, under contract to the U.S. Bureau of Prisons. These inmates are in the custody of the U.S. Bureau of Prisons because they have been convicted and sentenced to prison terms. The illegal aliens held for the INS are in temporary detention, mostly pending deportation.

States have been slow to follow the federal lead, perhaps partly because of statutory impediments, but private companies are prepared to sign contracts with one state or many. Buckingham Security Limited, a Pennsylvania-based company incorporated in Connecticut, has plans to site and build a 720-bed maximum security prison near Pittsburgh. They have letters of intent from several states to contract for the care of protective custody inmates. These inmates consume extra resources and cause many problems within their original prisons, but are not so troublesome when removed to separate facilities.

Cost Advantages of Private Prisons

What does it cost to build and run prisons in the private sector? It is difficult to give a definite answer to this question even in the public sector, where experience is long and data are plentiful. Early figures on private costs must, therefore, be regarded with caution, but they appear to support an expectation of savings.

At an annual cost per inmate of $40,000, RCA’s Weaversville Intensive Treatment Unit is much more expensive than the other institutions. From independent descriptions of the programs and facilities, it seems likely that much of the cost reflects high quality and small size, combined with the special needs of hard-core juvenile offenders. One expert (James Finkenauer, a criminologist at Rutgers) has described Weaversville as “better staffed, organized, and equipped than any program of its size that I know.”[2] The Florida School for Boys at Okeechobee, at $14,588 annually per inmate, is comparable in operating cost to the American Correctional Association’s estimate of $15,000 for state-run institutions.

Other available cost figures vary considerably. The lowest is $14 a day, per inmate, which Behavioral Systems Southwest charges the INS to detain illegal aliens in a series of converted and lightly secured motels. This is low by any standards, but the $24/day charged to the INS by Corrections Corporation of America for the same service may be an even better deal, since it covers construction of a new facility. The Hamilton County contract with CCA does not include basic construction, but the $21/day fee includes a commitment to spend up to $1 million in renovation during the first year of the four-year contract.

Behavioral Systems Southwest compares their $14/day charge to a current INS cost of $40- $50/day, while Corrections Corporation of America claims their $24/day charge is one-third less than it would otherwise cost the agency (which would compare it to $36/day). Palo Duro Private Detention Services charges the U.S. Bureau of Prisons $45/day for medium security imprisonment, which they say is 25 per cent less than the projected cost if that agency provided the service itself. Southwest Detention Facilities has offered to build and run a regional prison at a fee of $54/day for three counties in New Mexico that currently spend $78/day to house each of their prisoners.

Even the high security prison proposed by Buckingham Security Limited would be competitive at $24,000 per year (about $66/day) per inmate, since that figure includes construction. Buckingham Security’s estimate of $15 million to construct a maximum security prison with 720 beds works out to $20,833 per bed, a steal compared to the U.S. Department of Justice average figure (cited above) of $58,000 per bed for a maximum security prison. Corrections Corporation of America built their 350-bed detention center in Houston for $5 million, or $14,286 per bed.

What these figures suggest is that claims from the private sector of ability to build and run prisons at lower cost are at least plausible and worth putting to the test. The data are too limited and too crude to support any firm or categorical conclusions yet, but they are enough to show that a prima facie case for commercial prisons must be taken seriously. Pending further data, the case for (or against) private prisons rests mainly on a priori arguments, only a few of which can be considered here.[3]

A Priori Arguments for Privatization

Perhaps the strongest presumption in favor of private contracting of imprisonment is to assume that it would be similar in success or failure to the private contracting of other public services. We know that these other contracts are not always successful. However, E. S. Savas, in Privatizing the Public Sector.[4] points out that the most rigorous and comprehensive research (of which there is little so far) supports the general superiority of private over public provision of such diverse services as solid-waste collection, electric power, fire protection, transportation, health care, education, social services, protective services, and a number of others.

Obviously, imprisonment differs from these other activities in important respects, but not necessarily in ways that relate to efficiency and cost. Besides, if a prison system is broken down into its various separate activities, there are only a few aspects that do not have counterparts in both the public and the private sector. Therefore, evidence of successful private delivery of other public services warrants the presumption that it would be feasible for imprisonment also

Personnel is probably the major area where savings could be expected with private contracting. Businesses, for economic reasons, tend to hire and retain approximately the number of employees that they truly need. Governments, for political reasons, frequently era-ploy too many or too few. Retirement programs, sick pay, fringe benefits, and (for blue collar workers) salaries are generally more generous in the public sector, as the Grace Commission has documented. This is because wages and benefits are not subject to the direct discipline of market forces in the public sector, at least not as much as they are in the private.

Profit-and-Loss versus Bureaucratic Incentives

Costs other than personnel are also more likely to be held down by commercial prisons, as a result of the intrinsic economic incentives found in profit-making companies but absent from government agencies. Public service institutions tend to be both inefficient and ineffective, not so much because they are public rather than private, but because of the way they are financed. Profit-and- loss incentives differ fundamentally from bureaucratic incentives.

Entrepreneurs are motivated, under conditions of competition, to provide maximum satisfaction at minimum cost. Bureaucrats, in contrast, are rewarded not so much for efficiency, but in direct proportion to the sizes and budgets of their agencies. Budget-based organizations are thus motivated to grow and to spend. Their money is based on promises, intentions, and efforts, and not strictly on results.

There is another reason why it tends to be inefficient for a governmental unit to attempt to supply all its own services. Economies of scale vary greatly by the type of service and by the size and nature of the area being served, but as Robert W. Poole, Jr. notes: “The one arrangement least likely to be most efficient is for all the services to be provided at the scale defined by the size of the city” (or county, or state).[5] For example, nearly all cities need some jail services, but it may require a multi-city contract to meet efficiently the needs of small cities, while a large city can’ operate more efficiently by using multiple contractors to meet its varied needs (e.g., high and low security, male and female inmates, juveniles and adults, detoxification units, etc). Moreover, economies of scale can change, sometimes rapidly, due to technology or market forces; private vendors can respond to these changes more readily than can political entities.

Other Benefits

So far, I have addressed only the fiscal advantages of commercial prisons, but certain other potential benefits are no less important. One goal of contracting is to inject into the public sector some of the greater flexibility that is often found in the private sector. Flexibility is especially important in the administration of public policy, where concentration of decision-making magnifies the consequences of ignorance, uncertainty, and error. Policies regarding imprisonment, for example, contain implicit or explicit projections about trends, distributions, and patterns of crime and punishment. Even where broad trends are discernible, however, it is beyond the powers of social science to make highly accurate and reliable forecasts.

Because of the scale on which it operates and the ponderous way in which it moves, government is much more dependent than is private enterprise on the long-term accuracy of projections. Of course, the private sector must also be able to predict, if it wishes to make a profit, but it can make better use of short-term (and therefore more accurate) predictions because it can generally respond more quickly to changes in information.

A market in corrections would share in the general advantages of markets over all forms of central planning. The advantage most relevant here is that competition isolates and minimizes the consequences of erroneous predictions, while central planning magnifies them. If a state launches a major prison construction plan and hires an army of civil servants on the basis of a long-term projected trend that does not materialize, or that unexpectedly reverses itself after a few years, the cost will be monumental. If several competing contractors are responding continually to projected needs, it is unlikely that they will all make the same mistakes at once.

Some contractors will predict better than others, or be able to respond more quickly to altered predictions. These companies will survive and prosper by being able to meet the changing needs of the state more effectively. The less successful companies will have to absorb and thereby contain the cost of their inaccurate predictions. In contrast, when the state has a monopoly on the prison business, it can simply pass on to taxpayers the full cost of its errors. Thus, the state has less incentive to avoid errors in the first place.

Speedy Response

This presumption of flexibility and speed of response on the part of private prison contractors is not just a matter of faith; concrete examples can be cited. In 1975, the Attorney General of Pennsylvania ruled that even hard-core delinquents could not be incarcerated in facilities with adult offenders. Faced with the need to immediately relocate all affected juveniles, and having no suitable facilities available, the state turned to RCA, with whom they already had a contract for educational programs for delinquents. In ten days, RCA set up the Weaversville Intensive Treatment Unit, a heavy security facility with 20 beds and 30 staff members. In this case, RCA was able to convert buildings already owned by the state; other contractors have built their own or remodeled existing private structures, such as motels, to make them secure.

While a spokesman for the Federal Bureau of Prisons reports that it takes two or three years to site and build[6] and other sources report that it takes five years or more to build secure facilities, private contractors claim to be able to design, finance, and build a prison in as little as six months. Corrections Corporation of America financed, built, and opened its 350-bed detention facility for the INS in just seven months.

Contracts allow government agencies to experiment with new programs without long-term commitment of funds or of tenured civil service staff. Thus, vested interests in these programs do not accumulate inside the agencies. This avoids the tendency toward bureaucratic self- perpetuation that ordinarily makes public programs difficult even to alter, let alone to eliminate. Of course, advantages of flexibility for the public agency translate into insecurity for the private contractor.

If the vulnerability of a contract is an advantage to one side, it is a disadvantage to the other. However, this problem is intrinsic to all private contracting. To survive and succeed, a contractor must be able to solve this problem in a variety of ways: by providing service that is too good to give up; by accurately anticipating and being ready to meet the shifting needs of different clients; by holding down the administrative cost of hustling from one contract to another; by cultivating multiple clients; or by other techniques that must be the stock-in-trade of any competitive contractor.

Flexibility Promotes Justice

The flexibility of private prison contracts can also enhance justice. Public concerns over justice and punishment have led to reforms in many components of the justice system. Abolishing parole, limiting judicial discretion, banning or restricting plea bargaining, and other such reforms, are supposed to curb abuses and to make punishment more uniform and just. One objection to these reforms is the fear that they will produce further and more dangerous overcrowding of prisons. Therefore, attempts have been made to reintroduce discretion through mechanisms for diversion, sentence reduction, emergency release, and “early community reintegration.”

The search for these new discretionary mechanisms, which are designed to forestall or to relieve the pressures of overcrowding, rests on a faulty assumption: that prison flow can and should be fine-tuned by the state, while prison capacity remains virtually fixed. A penal system primarily concerned with doing justice, however, makes just the opposite assumption: prison flow should respond to crime rates, which are largely beyond the control of the state; therefore, prison capacity must be flexible.

At least at the margins, then, the prison system should be able to expand and contract as the shifting demands of justice require. Flexibility at the margins will tend to maximize the supply and minimize the cost of imprisonment. Commercial prisons, with efficient management, multiple vendors, and renewable, adjustable contracts, offer an increased prospect of achieving this marginal flexibility.

Government Should Foster Competition

Public employee unions will no doubt offer many reasons why government services, including imprisonment of criminal offenders, are best handled by civil servants under direct employment and administrative management by the state. They may be correct, much of the time, in this assertion, but they should be made to uphold their claim in open competition. No longer should they be permitted to foreclose the question through monopolization by fiat.

Competition is as important to the processes of government as it is to those of private enterprise. As a mechanism of evaluation, accountability, and control, it is unmatched. How do we know—and how can we be sure—that the government is running our prisons in the most just, humane, effective, and efficient manner possible? A very good test is to see whether private enterprise can do it better.

Competitive private contracting introduces a market test of price that is absent under governmental monopoly, and it provides a new basis for comparison of all programs, bothpublic and private, on dimensions other than cost. Many of the component activities of the penal system are already subject to competition among private vendors. Why not do the same for the administration of entire prisons? The ultimate legal authority for imprisonment must, by definition, remain in the hands of government, which means that a high degree of regulation will always be necessary. With that proviso, however, rigid legal obstacles to the private administration of prisons should be removed.

1.   Information on the programs described below comes from the following sources: Kevin Krajick, “Punishment for Profit”Across the Board, 21, March 1984, pp. 20-27; Kevin Kra-jick, “Prisons for Profit: the Private Alternative”State Legislatures, 10, April 1984, pp. 914; Philip E. Fixler, Jr., “Behind Bars We Find an Enterprise Zone”The Wall Street Journal, November 29, 1984, p. 34;The Hartford Cour-ant, April 1, 1984;The Philadelphia Inquirer April 16, 1984; and telephone inquiries.


2.   Cited in Krajick, “Punishment for Profit.”


3.   For an extended list ofa priori arguments both for and against privatization of public services generally, see E. S. Savas,Privatizing the Public Sector: How to Shrink Government (Chatham, N.J.: Chatham House Publishers, 1982), pp. 89-91.


4.  Ibid., pp. 93-111.


5.   Robert W. Poole, Jr., “Objections to Privatization”Policy Review, 24, Spring 1984, p. 107.


6.   Cited in Krajick, “Punishment for Profit.”

  • Mr. Logan is Associate Professor of Sociology, University of Connecticut at Storrs.