Factories Behind Bars

Policy Reports | Crime

No. 206
Sunday, September 01, 1996
by Morgan O. Reynolds

Why the Private Sector Needs To Be Involved

States need only so many license plates, only so much furniture for government agencies and only a certain number of workers on prison farms. Thus the private sector must be involved both in creating jobs and in finding markets for prison-made products if prisoner work is ever to employ large numbers of inmates and produce significant income.

Prisoners employed by the private sector, whether by contract or lease, have consistently outperformed those working in prison system industries. For example, in 1923, when the private sector still played a significant role in prisoner employment, a Labor Department survey found that:10

  • Output by prisoners employed by private businesses was $3,173 per worker year.
  • Output by prisoners in state-use and public-account work was $860 per worker year.
  • Productivity was about four times greater under private control than under public control, even if the same industries were compared; for example, furniture manufacturing under private contract had output per prisoner year of $2,740 versus $683 for state use and public account.

Penologist Howard Gill observed that industries operated by the prisons themselves were terribly inefficient and had a lot of idleness compared to prison workshops operated by private industry. Gill wrote, "It appears that idleness increases as public control increases."

Why Prisoners Need to Work

There are both economic and habilitative advantages to having prisoners work. Revenue from prisoner-produced goods and services can offset part or all of the cost of incarceration. Part of the wages paid to prisoners can be used for the same purpose or go to victims for restitution, or to the prisoners' families for upkeep. In addition, working provides other benefits to prisoners.

Projected Gains for Taxpayers

In a survey by the Enterprise Prison Institute, prison industry managers frequently mentioned 25 percent of prisoners as a desired target for employment. Gov. Tommy Thompson of Wisconsin has set a goal of 25 percent of state prisoners privately employed. U.S. Sen. Phil Gramm (R-TX) has proposed that federal prisoners pay 50 percent of their annual support cost through prison work.

What could we reasonably expect under an aggressive expansion of private production by prisoners?

  • A full-time job typically means 2,000 hours of work per year (40 hours per week times 50 weeks).
  • An average wage of $5.00 an hour is feasible, for a total of $10,000 in gross earnings per prisoner each year.
  • If a majority of these earnings were allocated to taxpayer compensation, say 60 percent, then at least $6,000 a year per prison worker would be available to compensate taxpayers.

If one in four prisoners could be put to work for private enterprise over the next five to 10 years, during which time the prison population is projected to increase to 1.6 million, that implies about 400,000 new prison jobs. Their work would reduce taxpayer costs by $2.4 billion per year, or somewhat less than 10 percent of the total cost of prison support

One of the difficulties of creating jobs for prisoners is that many of them are illiterate or semiliterate, or have low IQs, but champions of inmate labor are confident such jobs could be created. The federal system has the best prospects for high rates of payback because many of the prisoners are there for crimes typically committed by more intelligent criminals like counterfeiting, kidnapping and drug smuggling.


Ideally, the money that prisoners earn also could provide restitution to their victims, along with upkeep for the prisoners' families and savings for use after their release. Under the current system, the lack of productive prison jobs has limited efforts to gain restitution for victims. Thus courts rarely order restitution when they send someone to prison.

  • In 1986, 36 percent of felons placed on probation, including 52 percent of burglars, were ordered by state courts to pay restitution.
  • But only 14 percent of felons sent to prison were ordered to pay restitution.

Benefits to Prisoners Putting prisoners to work at productive jobs results in better-behaved prisoners, can help train them in work habits and skills and may increase the likelihood of a productive life when they are released.

Prisoners Overwhelmingly Prefer Work to the Tedium of Prison Life. This is common knowledge among experts. Prisoners especially value opportunities to work for private-sector firms, as is demonstrated by the number who sign up for prison industry jobs wherever formal waiting lists exist.

Prisoners Who Work Behave Better. This is confirmed by prison officials, although hard data from social scientists are less abundant and more controversial. For example, a 1989 survey found mixed results. However, a federal Post-Release Employment Project (PREP) study confirms that employed prisoners do better than those who do not work. Data were collected from 1983 to 1987 on more than 7,000 federal offenders. Study group members worked in prison industries (57 percent), participated in combined work and vocational training (19 percent) or in vocational training, apprenticeship training or a combination of the two (24 percent). Preliminary findings were reported in 1991 when all offenders in the study had been free for at least one year. The investigators used a sophisticated statistical matching procedure to compare study participants with otherwise statistically identical prisoners released in the same quarter. Prison conduct among inmates who participated in work, vocational or apprenticeship programs was better than that of otherwise similar prisoners. Participants were less likely to have a misconduct report during their last year of incarceration, and when they did receive a misconduct report it was less likely to have been serious.

After release to halfway houses, participants in the PREP study were 24 percent more likely to get a full-time or day labor job than those who had not worked in prison. Those who had worked in prison also earned more than those who had not and were more likely to move on to a better-paying job. Only 6.6 percent of those who worked in prison had their parole revoked or were charged with committing a new crime during their first year of supervised release, compared to 10.1 percent of the group who had not worked in prison.

These same findings hold up over a much longer period. Most participants in a follow-up to the PREP study had been released for at least eight years and some for as long as 12 years. Prison work and training programs seem to have been especially effective in reducing the likelihood of recidivism in the long term.

Badger State Industries (BSI), Wisconsin's state prison industries program, employs about 600 of Wisconsin's 10,000 prisoners to produce everything from coffee cups to furniture. A study indicated that the recidivism rate after three years was 15 percent lower for those who worked for BSI than for those who did not. According to BSI's director, Steve Kronzer, "People who worked for prison industries tended to do better than normal."

Sociologists Peter Rossi, Richard Berke and Ken Lenihan concluded from a large experimental study in the 1970s, "Employment for ex-felons is clearly the strongest antidote to reengaging in criminal activities. ...(W)orking clearly lowers the probability of arrest on all sorts of charges."

Origins of Convict Labor

The idea of work by prisoners is hardly new. In 1787 the founding father of criminology in the English-speaking world, Jeremy Bentham (1748-1832), urged replacement of the jails of his day by what he termed "mills for grinding rogues honest and idle men industrious." In the late 18th century, the humane prison reformer Jean Jacques Vilain of the Maison de Force (House of Enforcement) at Ghent, Belgium, opposed life imprisonment and cruel punishment but administered his facility by the Biblical injunction, "If any man will not work, neither let him eat."23
The American colonists, reflecting British traditions in punishment, relied on shaming, maiming or killing rather than imprisonment. Offenders were exposed to public derision in stocks (wooden devices for holding the feet) or the pillory (for the head and wrists) or branded for minor offenses. For more serious offenses, hanging and even burning at the stake were used. Primarily because of the expense, jails were used for temporary confinement only. These Colonial American practices were common not only in Britain but in most other European countries at that time.24

Early Prisons

In April 1790 the Pennsylvania Legislature authorized the first penitentiary in the United States, the Walnut Street Jail in Philadelphia. Inmates were confined in individual cells, where they not only slept but also worked at handicrafts. Quaker activist George Fox, who was influential in bringing about the establishment of the penitentiary, believed that self-reformation would come from long-term isolation under fixed sentences, solitary work and contemplation. The "gaol" at Wymondham in Norfolk, England, erected in 1785, was the immediate inspiration for this first American prison. Wymondham's creator, Thomas Beever, said that the prison was far more effective than whipping, and with hard labor six days a week many prisoners earned more than double their maintenance.25

As other prisons were established toward the end of the 18th century and in the early 19th century, the Pennsylvania system served as the model. But after 1823 the Auburn system, named after the New York state prison in that community, came to dominate U.S. prisons because of its superior economic productivity. The Auburn system confined prisoners at night in solitary cells but brought them together during the day in congregate workshops where they were not allowed to converse.

At first, work by prisoners was done for the state, although prison-made goods were sold on the open market. Almost immediately after the Auburn system was introduced, cabinetmakers in New York protested the competition of prison-made goods.26 Nevertheless, private enterprise quickly began to play a prominent role in the Auburn system through contract prison labor.

Despite - or perhaps because of - the emphasis on handicrafts, Pennsylvania's prisons were never self-supporting. The Eastern Penitentiary, the nation's first, was never able to earn enough to offset the cost of feeding and clothing the convicts. The state's Western Penitentiary finally earned enough from prisoner labor in the 1840s to pay the cost of prisoner maintenance exclusive of prison salaries.

Officials originally intended prison labor to offset the cost of incarceration, or at least a large part of it. Wardens were under substantial pressure to make their prisons financially self-sufficient, and they were "concerned as much about profits as prisoners."28 Newgate prison, opened in New York City in 1797, paid nearly all of its expenses during the first five years from production within the prison. By 1803 a tiny surplus remained after the prison's expenses were paid.29 By 1825 the legislature stipulated that the duty of prison agents was "to cause all the expenses...of any kind, to be supported wholly, or as nearly as shall be practicable, by the labor of the prisoners." [See Table I.]

Contracting with Private Businesses

Under the contract system, prison officials advertised for bids from private employers to hire the labor services of convicts within the walls of the prison, while prison officials retained control over security and sustenance. Contracts usually were awarded to the highest responsible bidder, with the private contractor posting a substantial performance bond to insure the prison against financial default. Contractors typically provided the necessary raw materials and furnished the craftsmen and foremen needed to train and supervise the prisoners. Depending on the terms of the contract, the contractor might also provide tools, machinery and power to operate equipment. The contractor sold the finished products in the open market, and the state received a fixed fee per prisoner per day.

Although the Auburn prison showed a deficit in its early years, it earned a surplus after 1829. The Sing Sing prison, also in New York, posted deficits in its early years, but from 1833 to 1841 earned surpluses totaling $94,000. The prisons of Massachusetts, Ohio and Connecticut, among others, also showed surpluses.

Advocates of the Auburn or congregate labor system argued that it meant lower costs to build and operate prisons, better vocational training and far more revenue for the state than the Pennsylvania system of individual handicrafts in single-cell confinement. Less isolation of the prisoners, gainful employment and a generally favorable experience with the system satisfied legislators. An article in 1839 celebrated the degree of satisfaction this way: "It is surprising how little it costs to do good, if we really set ourselves to work in the right way."

Prisoner Leasing

After the Civil War, convict leases became another way in which prisoners were put to work. Under convict leases, private employers essentially assumed control over nearly all aspects of prison life, including security and living conditions. Prisoner leases usually involved work camps on farms, construction sites (including railroads) and mines outside prison walls. Leases to private employers usually yielded the highest revenues to the state.

The system of leasing prisoners to private businesses for work outside prison walls was first tried in Kentucky in 1825, and during Reconstruction the practice became widespread in Southern and border states whose economies had been devastated. Leasing proved economically successful but politically difficult.

All of the Southern states achieved net profits from prison labor in the limited sense that the taxpayers were relieved of the burden of maintaining prisons. Early lease contracts in the 1860s and 1870s were not always successful, but during the 1880s and 1890s leasing became very profitable and state officials welcomed the additional income from prison labor.
Case Study: Tennessee. A 200-cell prison opened in Nashville in 1831 and prisoners worked at coopering, blacksmithing and manufacturing wagons, harnesses and shoes. Convicts also quarried and cut stone for the State Capitol. Proceeds from these labor services were supposed to defray the expenses of long-term confinement. During the prison's first two decades, legislators praised it as a profit-making organization. Yet some mechanics and tradesmen complained that prison products and prison labor competed unfairly and degraded their respective trades. In 1853 the new Tennessee governor (and subsequent president of the United States), Andrew Johnson, agreed with the protesters and strongly condemned the prison for operating a "State Mechanic Institute" competing with free labor.

At the end of the Civil War, faced with the expensive prospect of refurbishing the old prison or building additional facilities, the poverty-stricken state chose to lease its convict labor for a four-year term to the highest bidder. The prison directors retained the authority to enforce rules "to promote the health, comfort, order and discipline of the convicts, and the proper deportment of the lessee, officers, keepers and guards of the prison." A Nashville furniture manufacturer won the lease, paying 43 cents per day, the highest amount obtained by any state that the prison directors knew of. But a fire gutted one of the prison workshops a year later, destroying much of the manufacturer's equipment, supplies and product. The lessee stopped payments and claimed damages against the state. The legislature disagreed and litigation ensued.

Subsequent labor leases with the railroads produced other problems, including accidents on the job, escapes and administrative and supervisory trouble, along with objections to competition with non-convict labor and to the temporary nature of railroad construction.

But leasing for other types of work was more successful. Branch prisons at Tracy City and Battle Creek provided steady employment in mines. A five-year lease with an agricultural implements manufacturer provided almost half the prisoners at the main prison with jobs building wagons. Prisoners also worked on farms and were subleased to other private businesses. By 1872 Gov. John C. Brown boasted that the businesses employing convicts were meeting their financial obligations promptly, had offered a variety of jobs and had cut down on competitive effects through their diversified production. Tennessee's system, he thought, was superior to those in neighboring states because they spent large amounts of money on their prisons while Tennessee did not.

Case Study: Texas. Between the end of the Civil War and the outbreak of World War I, Texas also routinely hired out prison inmates to private individuals and corporations. The system did not emerge as the result of any "clearly conceived and well-executed master plan. Instead it developed from the plodding and often haphazard efforts of state officials to provide for a growing inmate population at a time when there was little money for state spending and limited public support for innovation in prison policy." Fees paid by private contractors for the labor of prisoners from the 1870s until the end of the leasing system in 1912 offset most of the operating expenses for the Texas penal system. Industries operated by the state inside the prisons "languished...[and] never became profitable" while the private leases outside the prisons "flourished and paid handsome dividends." Railroad contracts were more lucrative for the state than farm labor, but the latter was more common, especially in sugar farming, and yielded net revenue to the state of $3.4 million over the period. Black prisoners predominated on sugarcane farms, and the labor prices charged to companies were only slightly less than the wages of similar free labor. From 1908 to 1912 black prisoners in first-class physical condition cost $31 per month and first-class white prisoners cost $29 per month.

Reformers of the day complained that the contract-lease system did not teach prisoners a trade or reform them to lead productive lives during their terms in prison. As one inmate opined, the prisons were "a nurse and school" to prepare convicts to be career criminals. Yet the meager data show that Texas prisoners overwhelmingly were serving their first term, suggesting a low recidivism rate. During the 1880s, for example, more than 90 percent of Texas convicts were first termers and from 1900 to 1912 first termers ranged from 80 to 85 percent. By contrast, in 1994 only 47 percent of Texas convicts were nominal first termers in the state's prisons, and only one in three was a true first termer, that is, with no prior time served in any prison. Fragmentary national data suggest a similar pattern; recidivism experience has worsened over the last century.

Read Article as PDF