PRIVATE PRISONS

This page intentionally left blank

PRIVATE
PRISONS
CONS AND PROS

CHARLES H. LOGAN

New York Oxford

OXFORD UNIVERSITY PRESS
1990
Oxford University Press
Oxford New York Toronto
Delhi Bombay Calcutta Madras Karachi
Petaling Jaya Singapore Hong Kong Tokyo
Nairobi Dar cs Salaam Cape Town
Melbourne Auckland
and associated companies in
Berlin Ibadan

Copyright © 1990 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.,
200 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without prior permission of Oxford University Press.
Library of Congress Cataloging-in-Publication Data
Logan, Charles H.
Private prisons: cons and pros/ Charles H. Logan,
p. cm. Includes bibliographical references.
ISBN 0-19-506353-8
1. Prisons—United States. 2. Privatization—United States.
3. Corrections—United States—Contracting out. I. Title.
HV9469.L64 1990
365'.973—dc20 89-49028

987654321
Printed in the United States of America
on acid-free paper
For Sharla, Daniel, and Christopher
This page intentionally left blank
Acknowledgments

I am grateful to the National Institute of Justice for the Visiting

Fellowship (86-IJ-CX-0062) that made possible all the research for
this book. The views expressed here are my own and do not nec-
essarily reflect Department of Justice policy. I am especially in-
debted to James K. Stewart, Director of the Institute, for his
enthusiasm, his intellectual support, and his friendship. No other
person has done as much as he to promote nationwide research,
discussion, and action on private-sector involvement in corrections.
Others at the Institute were also helpful in many ways, especially
Joel Garner, Bonnie Gowdy, Dick Linster, Marty Lively, Renee
Trent, Neille Russell, and Ed Zedlewski. Barry Ruback, as a fellow
Visiting Fellow, offered much social support and editorial help
during the early stages of my work. George Cole, also a Visiting
Fellow, was there for me at a later point.
Many government officials involved in prison or jail contracting
were generous with time and information. Officials in Bay County,
Florida, were very helpful, particularly John Hutt and Larry Davis.
In Hamilton County, Tennessee, special thanks go to Floyd (Flop)
Fuller, Superintendent of Corrections, and Bill McGriff, County
Auditor. Bill contributed so much to my chapter on costs that he
is co-author of a separate report derived from it. Bob Schmidt,
now retired from the Immigration and Naturalization Service, pro-
vided a flood of information.
In general, I found that private corrections companies are at
least as open to investigation and research as are public agencies,
viii Acknowledgments

and sometimes more so. While I have talked with people at many
companies, including Corrections Corporation of America (CCA),
Buckingham Security Limited, Wackenhut, Behavioral Systems
Southwest, U.S. Corrections Corp., and Pricor, my closest contacts
have been at CCA and Buckingham. Charles and Joseph Fenton,
the founders of Buckingham Security, were open and candid in all
conversations.
No study of private corrections today can go very far without
access to information on CCA, the industry leader. Fortunately,
I don't see how any company or government agency could be more
generous with its time and information than CCA, or more open
to scrutiny by outsiders, including researchers, journalists, and
even critics and opponents of the industry. So many people at
CCA were helpful to me in supplying materials and providing
access to facilities that I have to apologize to those whose names
I omit. At one time or another, I had special help from Tom
Beasley, Linda Cooper, Richard Crane, Doctor Grants, Don
Hutto, Greg McCullough, and Peggy Wilson. David Myers, Vice
President for Facilities Management, was a crucial link for me in
getting information about, and access to, CCA's facilities.
Other researchers in this area have been helpful in sharing in-
formation and perspectives, including Diane Bast, Bruce Benson,
Jan Brakel, Douglas McDonald, Joan Mullen, Charles Ring, and
Ira Robbins.
Finally, I have saved special thanks for two people most im-
portant to this project and to me. Charles Thomas, Professor of
Criminology and Associate Dean at the University of Florida, is
a leading academic authority on private prisons. He knows more
about this field than anyone else outside the industry, and he was
most generous in sharing his information, materials, and contacts
with me. Last on my list, but first in my life, is Sharla P. Rausch,
my wife and frequent co-author, whose encouragement, assistance,
advice, and support are always deeply appreciated.
 Contents

Introduction, 3
1 The Iron Fist Meets the Invisible Hand, 7
2 Private Prisons Today: A Descriptive Overview, 13
3 Issues and Arguments in the Debate over
Private Prisons, 38
4 The Propriety of Proprietary Prisons, 49
5 Issues of Cost and Efficiency, 76
6 Issues of Quality, 119
7 Issues of Quantity, 149
8 Issues of Flexibility, 162
9 Issues of Security, 171
10 Issues of Liability, 181
11 Issues of Accountability and Monitoring, 194
12 Issues of Corruption, 211
13 Issues of Dependence, 221
14 Private Prisons and the Privatization
of Punishment, 236
Notes, 259
Selected Bibliography, 302
Index, 309
This page intentionally left blank
PRIVATE PRISONS
This page intentionally left blank
Introduction

Americans both love and hate their government; trust and fear it.
Traditionally, their political philosophy has been basically liber-
tarian, though not consistently so. They believe in limiting the
power of government in many ways, yet they also want to use
government to pursue various special goals or interests. This am-
bivalence toward government is particularly evident in the area of
criminal justice.
Those earliest of American reformers, the founders, established
a Constitution premised on a lack of faith in the benevolence and
wisdom of governors. Shortly thereafter, another group of Amer-
ican reformers established a penitentiary system premised on faith
in the redemption of sinners and in the benevolence and wisdom
of at least some authorities whose job it would be to coercively
save those sinners.
This faith in the good intentions of penal authorities and in their
competence to rehabilitate lawbreakers has been strongly chal-
lenged in the past two decades, for a variety of reasons. The reason
most relevant here is a growing disenchantment with government
generally. From the late 1950s to the mid-1970s, self-reported trust
in government declined from almost 80 percent to about 33 per-
cent.' This was part of a general lowering of public confidence in
American institutions and leadership. 2 By the 1980s, taxpayers had
begun to revolt and a presidential candidate with a platform of
"getting the government off our backs" was elected with great
popular support. This was also a time of growing interest in "pri-

3
4 Introduction

vatization": the transfer of assets, and of the production of public

goods and services, from government to the private sector.
The privatization of corrections, or punishment, is an especially
significant part of the broader privatization movement. By chal-
lenging the government's monopoly over one of its ostensibly
"core" functions, this idea directly threatens the assumption that
certain activities are essentially and necessarily governmental.
How does one determine what government functions are nec-
essary or essential? Historical tests are unavailing. No role has
always been the exclusive province of the government. Nor can
current practice be taken as definitive. Ronald Cass, Commissioner
of the U.S. International Trade Commission and Professor of Law
at Boston University, has pointed out the fallacy in using historical
tests to define essential government functions: they produce
a "one-way ratchet" for government expansion. Whatever the
government does or has done becomes an essential government
function, and contraction of the scope of government becomes
impossible.3
Thus, privatization in the area of criminal justice generally, and
of imprisonment particularly, plays an important part in a broad,
ideological debate over the proper scope and size of government.
This book contributes to that broader debate, even though its
topical focus is rather narrow.
Proposals for the privatization of functions such as garbage col-
lection or mail delivery have urged the government to divest itself
partially or even entirely of specific responsibilities. In contrast,
recent proposals for private prisons and jails have suggested only
that it might be more efficient for government to lease these fa-
cilities or to contract for their operation and management by pri-
vate companies. A completely private criminal justice system—or,
indeed, the total dismantlement of the state—may be interesting
to discuss at a theoretical level, but there are no viable policy
proposals along those lines today.
Current proposals for private prisons refer only to private con-
tracting, not to complete privatization in the sense of divestiture.
Nonetheless, penal justice, even in the form of contracting, rep-
resents an extreme test of the limits of privatization. As a result,
all of the arguments against contracting of other public services
have been directed with great intensity against prison and jail
Introduction 5

contracts. If those objections can be answered in this context, it

is doubtful that they could have greater power or validity in other
areas of government activity. So the ideological stakes in the debate
over correctional contracting are high.
My goal in this book is to examine critically all of the objections
to private prisons that have been or could be raised, and to assess
both their validity and their weight. Ten sets of issues will be
explored in separate chapters, covering problems of propriety,
cost, quality, quantity, flexibility, security, liability, accountability,
corruption, and dependence.
In evaluating private prisons on these dimensions, there must
be no double standard between private and public prisons. In
particular, private prisons should not be judged against absolute
or ideal standards without reference to the performance of
government-run prisons. It is too easy to find problems, or to
anticipate potential problems, in either type of prison if we fail to
make relevant comparisons.
Arguments against private prisons vary in soundness and plau-
sibility, but in no area have I found any potential problem with
private prisons that is not at least matched by an identical or a
closely corresponding problem among prisons that are run by the
government. It is primarily because they are prisons, not because
they are contractual, that private operations face challenges of
authority, legitimacy, procedural justice, accountability, liability,
cost, security, safety, corruptibility, and so on. Because they raise
no problems that are both unique and insurmountable, private
prisons should be allowed to compete (and cooperate) with gov-
ernment agencies so that we can discover how best to run prisons
that are safe, secure, humane, efficient, and just.
This page intentionally left blank
1
The Iron Fist Meets
the Invisible Hand

Government production of goods and services is often defended

as being a necessary response to market failure. Where the "in-
visible hand" of the market fails to produce goods and services of
adequate quality, or in sufficient quantity, or at an appropriate
price, or to distribute them properly, the "iron fist" of the state
must take action. Conversely, privatization is often seen as a rem-
edy for government failure in these matters. Does imprisonment
represent an area of "government failure" today?

Imprisonment: Demand and Supply1

Imprisonment, since it serves the public as a whole rather than

individual consumers, can be characterized as a public good. Public
goods are generally financed and arranged for (but not necessarily
produced) by government. Like other goods, they can be analyzed
in terms of supply, demand, quality, and price. In the case of
imprisonment, a clear pattern has emerged, one that is character-
istic of goods or services produced under monopoly conditions.
Quality is low, prices are high, and the supply has not kept up
with demand.
The number of state and federal prisoners totaled 581,609 by
the end of 1987, up 76 percent from 1980. Prison capacity, how-

7
5 PRIVATE PRISONS

ever, has not kept up with the increasing population. Overall, state
prisons in 1987 were rilled to somewhere between 105 percent and
120 percent of capacity, depending on how capacity is measured,
while federal prisons held between 37 percent to 73 percent more
than the maximum number of prisoners that they were meant to
house. Over 12,000 state prisoners had to be held in local jails
because other facilities were overflowing. 2 Due to a virtual mor-
atorium on new construction during the 1970s, these overcrowded
prisons are also deteriorating with age. The average inmate sleeps
in a cell that is nearly 40 years old, and 10 percent are locked in
prisons built before 1875.' Consequently, in 1987 prisons in 42
states and the District of Columbia were under court orders be-
cause of crowding and other conditions ruled unconstitutional. 4

The Cost of Imprisonment

The cost of constructing and operating prisons is enormous, and
usually it is underestimated. Cost estimates for construction vary
according to region, type of prison, program needs, use of prison
labor, recency of data and techniques of adjusting for inflation,
reference to past or planned construction, and definition of what
is included in "cost." 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 se-
curity prisons, respectively. Because they are based on a systematic
survey of all states and are specified by prison type, these figures
probably come closest to reflecting average costs nationally.
Most estimates of construction cost, however, are probably too
low, because they ignore such considerations as land purchase, site
preparation, financing cost, overruns, and hidden cost.5 Taking all
of these factors into account, plus a conservative estimate of
$14,000 per inmate in annual operating cost, one economist has
estimated that a 500-bed prison ostensibly costing $30 million to
build today could end up costing $350 million altogether over a
30-year period. 6
As with construction, figures on operating costs vary widely.
The most common estimates are in the mid-teen thousands, re-
flected by the American Correctional Association's figure of
$15,000,7 and the federal prison system's figure of $13,000.8 Most
The Iron Fist Meets the Invisible Hand 9

estimates of operating costs are too low because they do not take
into account fringe benefits, interagency services, federal grants,
and other off-budget items. Studies of state corrections bud-
gets have shown actual expenditures to be about one-third (Indi-
ana) or 30 percent (New York) higher than the official budgets
reported/'
In short, the current costs of constructing and operating prisons,
while difficult to calculate precisely, are obviously quite high and
are probably much higher than most people, even knowledgeable
ones, assume. Moreover, these costs continue to grow rapidly.
Over the last six years, the budget for care of juveniles in the
custody of the Colorado Division of Youth Services has grown by
76.5 percent, which is over four times as fast as the general rate
of inflation, as measured by an increase of 17.9 percent in the
Consumer Price Index over the same period.10 Nationwide, per
capita spending on corrections at the state and local level has grown
faster in the last 25 years than government spending in most other
categories. From 1960 to 1985, per capita spending in constant
dollars increased by 218 percent for corrections. Spending in other
categories increased less: public welfare, 216 percent; hospitals
and health care, 119 percent; police, 73 percent; education, 56
percent. Spending on highways declined by 21 percent." In Ten-
nessee, the corrections budget has increased 1,790 percent over
the last 20 years, compared to an increase of 648 percent in the
total General Fund and an increase of 810 percent in higher ed-
ucation (the single program with the next largest growth). 12

The Privatization Option

Faced with overflowing and aging facilities, with court orders de-
manding immediate reforms, with already straining budgets and
voter rejections of prison construction bond issues, and with man-
datory sentence laws, toughening public attitudes, and "wars on
drugs" that promise even larger prison populations, government
authorities are ready to consider many different options to help
relieve the strain. Some of these options include: emergency early
release provisions; policies of selective incarceration and release;
community corrections; home confinement with electronic moni-
10 PRIVATE PRISONS

toring; intensive supervision probation; increased use of fines; and

contracting with other jurisdictions for jail and prison space. All
of these options are aimed at either decreasing the prison popu-
lation or using existing prison capacity more efficiently.
Another option is to contract with the private sector to finance,
construct, own, and operate prisons and jails. This option does
not conflict with any of the above options; rather, it supplements
them. Its greatest potential is that of flexibility. It has the capability
to do for corrections what entrepreneurial activity is supposed to
do best: anticipate needs and meet them. In response to the space
crunch in today's prisons, commercial confinement companies offer
an immediate prospect of relatively rapid and efficient increases
in overall confinement capacity. However, contracting need not
always be aimed at increasing the number of available prison cells.
If the need for secure confinement should decline, or if a viable
alternative is developed, it should be easier to alter contracts or
change contractors than to restructure entrenched public bu-
reaucracies.
Private prisons and jails are not meant to be used in lieu of
existing facilities, but to widen and diversify the prison system.
One purpose of competition, in addition to cost containment, is
to maximize choice. The prison crisis described above has caused
government managers to see the urgent need for alternatives. In
the early 1980s, the private sector responded to this need in the
form of proprietary (i.e., privately owned) companies specializing
in the management of correctional and confinement facilities.
These companies, and the facilities they have been running, will
be described in the next chapter.

Sources of Opposition

Opposition to contracted prisons comes from several sources. Or-

ganizations that have either opposed or called for a moratorium
on private prisons include the American Federation of State,
County, and Municipal Employees (AFSCME), the National Sher-
iffs' Association, the American Civil Liberties Union (ACLU),
and the American Bar Association (ABA).
It is interesting to note that contracting provoked relatively little
The Iron Fist Meets the Invisible Hand 11

controversy during the many years when the only people being
held against their will in private facilities were mental patients and
juveniles. It may also be significant that most of the adult con-
finement facilities under private contract in the early 1980s were
for the detention of illegal aliens. Finally, in this vein, the first
privately contracted state prison holding adults at all levels of
security is a facility for females.
It would be a mistake, however, to infer from this pattern that
those who object to private prisons care less about the rights and
welfare of mental patients, juveniles, aliens, or women than they
do about those of healthy adult male citizens. Rather, the issues
loom larger now because greater numbers of people, and therefore
greater interests, are at stake. Neither governmental monopoly
over the confinement function nor government employee monop-
oly over the relevant jobs was seriously threatened until privati-
zation began to encroach on the vast heartland of corrections.
That's when the real turf battles began and opposition was orga-
nized by the AFSCME, the National Sheriffs' Association, and
other groups.
Employee unions and sheriffs' associations are opposed to pri-
vate prisons largely for personal reasons: contracting threatens
their jobs and, more significantly, their power. 13 Some correctional
officials, too, view contracting as a challenge to their professional
status and as a means of lessening their control. Others, however,
believe that contracting will extend their capabilities as managers
of contracts.
Probably the most powerful opponent to private prisons is the
AFSCME, which is the sixth largest of all the AFL-CIO interna-
tional unions. 14 It represents over 50,000 corrections workers na-
tionally. 15 As of 1981, correctional employees were unionized in
29 of 52 jurisdictions (state, federal, and District of Columbia). 16
Union strength is weakest in the southern tier of the United States,
which is where the private prison industry has concentrated most
of its efforts. The AFSCME and other public employee unions are
opposed to contracting out virtually all public services, but their
opposition to prison privatization seems especially vehement.
Unions were the major force behind legislation in Pennsylvania
imposing a one-year moratorium on new privatization of prisons
or jails. They also forced the nonrenewal of a jail management
12 PRIVATE PRISONS

contract in that state when the AFSCME threw its support behind
two candidates for county commission running on a "take back
the jail" platform. In San Diego County, California, the probation
and correctional officers union in 1982 prevented Radio Corpo-
ration of America (RCA) from receiving a contract to run a 100-
bed juvenile detention facility, by invoking an old county bylaw
that prohibits contractors from running public facilities.17
The ACLU, a longtime champion of the rights of individuals
against encroachment by the state, apparently fears private enter-
prise even more than government. ACLU officials are concerned
that prisoners' due process rights are more likely to be jeopardized
in private than in government-run prisons. Indeed, they believe
that private prisons are intrinsically threatening to civil liberties.18
A substantial part of the ACLU's opposition to private prisons,
however, stems from its belief that there is already too much in-
carceration and privatization would only lead to more.
In February 1986, the House of Delegates of the American Bar
Association passed a resolution recommending that "jurisdictions
that are considering the privatization of prisons and jails not pro-
ceed . . . until the complex constitutional, statutory, and contrac-
tual issues are satisfactorily developed and resolved."19 The
president-elect of the ABA was quoted at that time as declaring:
"I am personally hostile to the notion [of private prisons]."2" Ac-
cording to Thomas Beasley, Chairman of Corrections Corporation
of America and a member of the ABA, the resolution was brought
forward with three speakers in favor but without any opportunity
for opposing debate.21 The report that accompanied the ABA
resolution was written by a vigorous opponent of private prisons.
In later chapters, I will examine in detail the issues raised by
these and other critics of privatization. First, however, the follow-
ing chapter gives a descriptive overview of current or recent con-
tracts for the private operation of secure confinement facilities—
private prisons and jails.
2
Private Prisons Today:
A Descriptive Overview

"Private prison" is not a very precise term, though its general

meaning is fairly clear. It refers to a place of confinement that
is managed by a private company. "Prison" will be used broadly
in this book to include any place of involuntary confinement
within the justice system. This would include facilities of the
Immigration and Naturalization Service (which is part of the De-
partment of Justice) and juvenile justice facilities, but not
places for involuntary civil confinement of the mentally ill.
Within the current context, "prison" will be more or less
synonymous with a "detention," "confinement," "correc-
tional," or "penal" facility. 1 These are not primarily residential
facilities, like foster homes, group homes, halfway houses,
or community treatment centers. Prisons, detention centers,
jails, reformatories, and similar institutions are places of incar-
ceration.
The term "private" (or, sometimes, "proprietary") prison im-
plies private ownership, at least of the management company,
and sometimes of the facility's buildings and grounds. Private
prisons are those that are privately owned, operated, or man-
aged,2 under contract to government. No prison today is com-
pletely private, in the sense of being independent of government
authority, control, and revenue. Private prisons now operate
only under contract to government. Hence, they may often be

13
14 PRIVATE PRISONS

referred to simply as "contractors," "vendors," or "service

providers."
Reference to "proprietary" or "commercial" prisons will reflect
the fact that these are businesses. However, they may be organized
on either a for-profit or a not-for-profit basis. They may be closely
held, publicly traded, or employee owned (profit sharing). What
they all have in common is that they are private entities operat-
ing under contract to government. They perform government
functions and may be regarded for some purposes as "quasi-
governmental," but they are not governmental in organizational
form.
Correctional facilities can be divided into four general types: (1)
juvenile open facilities; (2) juvenile institutional facilities; (3) adult
community facilities; and (4) adult confinement facilities. The Bu-
reau of Justice Statistics (BJS) classifies juvenile facilities as "open"
or "institutional" based on degree of access to community re-
sources and on the degree of security provided by the personnel
and the physical plant. 3 Most open facilities are shelters, halfway
houses or group homes, a few are set up as ranches. Institutional
facilities include detention or diagnostic centers, training schools,
and ranches. Adult facilities are classified by BJS as either "com-
munity" or "confinement," depending on whether the proportion
of residents who are regularly allowed to depart unaccompanied
(e.g., for work or study) is greater than half (community) or less
than half (confinement).
Proprietary facilities are found among all four correctional
situations. Since they are part of the criminal and juvenile
justice systems, they all involve the exercise of state authority
and at least some deprivation of liberty. Thus, although the con-
cept of "proprietary prisons" is mainly associated with confine-
ment or institutionalization, it could be taken to include open
environment and community facilities as well. However, the
private community-based and open facilities are so numerous,
so widespread, and so broadly accepted that they will be de-
scribed here only briefly, in statistical outline. The more recent,
unfamiliar, and controversial examples of institutional and con-
finement facilities will be described in greater detail, including
case histories.
Private Prisons Today: A Descriptive Overview 15

Juvenile Open Facilities

Private, low-security juvenile facilities have been part of the cor-

rections system in America since the nineteenth century. There
has been relatively little controversy about these facilities, whether
run by nonprofit or for-profit organizations. Indeed, there has been
much praise, even from individuals and organizations opposed to
private prisons, for the substitution of private community-based
programs in place of state institutions of juvenile justice. For ex-
ample, Jerome Miller has been widely applauded for changing the
juvenile justice system of Massachusetts from a network of state
institutions in 1969 to mostly community-based programs by 1974.4
The Massachusetts Division of Youth Services now operates all of
its community-based facilities (and half of its secure programs)
through private contractors.5
The Bureau of Justice Statistics counted 82,272 persons held in
2,900 juvenile detention and correctional facilities on February 1,
1983.b The 1,877 private facilities, found in every state but Dela-
ware, outnumbered the 1,023 public facilities by a large margin.
While the privates comprised 65 percent of all facilities, they held
just 38 percent of the juveniles in custody, since they were generally
smaller units and provided more long-term care.7
In juvenile facilities with open environments, private contracting
is now standard practice. Eighty-one percent of such facilities were
privately operated as of 1983 and they housed about three-quarters
(72 percent) of the juveniles held in that type of custody.8
One reason for the wide acceptance of private facilities with
open, or noninstitutional, environments may be that they are not
perceived as examples of the private exercise of state power. How-
ever, even open facilities are basically coercive in nature when
they are used by the juvenile justice system. That is, they exercise
authority backed up by actual or potential use of state force. Any
court placement, even in I he most open environment (like a group
home), carries with it the possibility of subsequent judicial or other
official intervention. Only 18 percent of the 25,855 admissions to
private open facilities during 1982 were voluntary without adju-
dication. '' Because the vast majority of residents in facilities with
16 PRIVATE PRISONS

"open" environments were placed there involuntarily, the phrase

"open facility" does not imply complete freedom to come and go.
Unpublished data from the 1982-83 Children in Custody census
show that 97 percent of the private open facilities were nonsecure
in terms of hardware or guarded exits. In terms of custody, how-
ever, about 18 percent of the private, open facilities were classified
as medium or strict. 10

Adult Community Facilities

As of June 30, 1986, according to the Directory of Juvenile and

Adult Correctional Departments, Institutions, Agencies and Parol-
ing Authorities, published by the American Correctional Associ-
ation (ACA), there were 4,758 adult state correctional inmates in
community homes, of whom 61 percent were in contracted facil-
ities. One survey found adult community-based facilities under
contract in 32 states.11 The International Halfway House Associ-
ation estimates that about 1,000 of its members are private vendors
providing community programs for correctional agencies.12 At the
federal level, the Bureau of Prisons relies exclusively on contracts
to place offenders in community treatment centers. Of 310 separate
contracts, 218 (or 70 percent) are with private providers; the rest
are with local public agencies. However, many of the public con-
tracts are low volume, so that nearly all of the actual placements
are in privately operated centers. Thus, in June of 1989, 94 percent
of the 3,287 federal offenders placed in community treatment were
placed in private facilities.13
I will have little more to say in this book about correctional
facilities that are community-based and residential. For nonsecure
facilities, both juvenile and adult, the practice by government of
contracting with private parties for operation and management is
widely accepted. That level of acceptance is significant, because
even nonsecure facilities involve the exercise of authority by the
criminal justice system, albeit at an ostensibly low level of coercion.
This suggests that the private exercise of authority usually reserved
for the government is not categorically objectionable; rather, it is
controversial primarily in proportion to how overtly coercive it
appears. Thus, to set the stage for the debate over "private pris-
Private Prisons Today: A Descriptive Overview 17

ons," I will give more detailed descriptions here only of juvenile

and adult facilities designed for relatively secure confinement.

Juvenile Institutional Facilities

Juvenile correctional facilities having an institutional environment

are most often run by government employees. Still, there are many
that operate under private contracts. In the 1982-83 census, there
were 808 institutional facilities nationwide, of which 187 (23 per-
cent) were private. 14 While almost all (91 percent) of the public
institutional facilities were secure in terms of hardware or guarded
exits, only 39 percent (73) of the private institutional facilities had
that sort of security. What made these facilities "institutional" in
BJS terminology was primarily their self-description as having a
strict or medium level of custody (true of 92 percent of this
group). 15
This category—private juvenile detention and correctional fa-
cilities identified by BJS as having an institutional environment—
may contain several examples of what 1 am calling "proprietary
prisons." Unfortunately, it is impossible to trace the identities of
those facilities directly, because of confidentiality guarantees pro-
vided by the Bureau of the Census, which gathers these data, and
by the Bureau of Justice Statistics, which analyzes them. However,
several such facilities have been identified by journalists and re-
searchers in recent years.

Weaversville Intensive Treatment Unit

Of the currently operating institutions for secure confinement of
juveniles, perhaps the longest-running for-profit enterprise is the
Weaversville Intensive Treatment Unit, which opened in North
Hampton, Pennsylvania, in 1975. When the Attorney General of
Pennsylvania ruled that even hard-core delinquents could not be
confined with adult offenders, officials turned for help to RCA
Services, a division of the Radio Corporation of America that had
previously run treatment programs for the state. In ten days, RCA
renovated a state-owned building to establish Weaversville and
was given a contract to run it. 16
18 PRIVATE PRISONS

High security and control are maintained at Weaversville by a

fence, locked internal and external doors, intensive supervision
(35 staff members for 22 youths), and room restriction when nec-
essary. 17 The security is necessitated by the nature of the popu-
lation: boys aged 15-18 who have failed in or run from other
programs and who have committed violent or otherwise serious
crimes, such as burglary, robbery, assault, sex offenses, weapons
offenses, arson, vandalism, and theft. 18 In spite of the far-reaching
security, however, the atmosphere is relaxed. Each resident has
his own carpeted room, with his own key. 19 Length of stay averages
six months. The program emphasizes behavior modification, ed-
ucation, and vocational training. Contract renewal has been com-
petitive, originally on a yearly basis, later changed to every three
years.20

Eckerd Youth Development Center

While Weaversville was a first for the private sector in terms of
its high-security custody for chronic and serious offenders, it re-
mained in the tradition, for private facilities, of very small size.
By the end of the 1970s, the private sector had not yet been en-
trusted with a large and secure correctional institution.
That threshold was crossed in the summer and fall of 1982, when
the Florida School for Boys at Okeechobee, one of Florida's three
large juvenile institutions (with 400 to 450 inmates), became the
Jack and Ruth Eckerd Youth Development Center. The state re-
tained ownership, but issued separate contracts for operating the
program and for managing the facility. Both contracts were
awarded (through open competition, but with only one responding
bidder) to the Eckerd Foundation, a nonprofit arm of the Eckerd
Corporation, a major drug manufacturer and drugstore chain. 21
Prior to 1982, the Eckerd Foundation had for many years run
programs in Florida for emotionally disturbed and delinquent
youths.
The state had planned to close the Okeechobee school but found
that it could not afford to do so. For years it allowed the facility
to deteriorate, with no money budgeted for physical improve-
ments. Conditions were bad, and led to a lawsuit filed by the ACLU
and other groups, charging "cruel and abusive conditions of con-
Private Prisons Today: A Descriptive Overview 19

finement." The suit named Okeechobee and two other state-run

institutions, but not the Eckerd Foundation, which the critics agree
simply inherited the conditions. The Foundation donated $280,000
of its own money to upgrade the staff and worked to improve the
physical plant and equipment. 22
The population, aged 14 to 18, is predominantly hard-core, se-
rious felony delinquents, with a complicating mix of severely dis-
turbed and first-time offenders. Fligh security is provided by close
supervision, a fence with electronic sensors, and the isolated lo-
cation of the training school. The average stay is about six months.

Florida Environmental Institute

The Florida Environmental Institute, a program of the nonprofit
Associated Marine Institutes, is at least partially institutional and
secure. Run since 1983 under contract to the Florida Department
of Health and Rehabilitative Services, this is a three-phase program
for Dade and Broward counties. The third phase is an open, non-
residential program of participation in one of the marine biology
programs operated by the Associated Marine Institutes. However,
phases one and two, which consist of work, vocational, educa-
tional, and values training, occur in tents and trailers in an isolated
swamp.23 These phases are institutional in the sense of having no
interface with an outside community, and they have at least a
minimal level of security in the form of isolation. The program
handles serious juvenile offenders.

Shelby Training Center

Another institutional juvenile facility is managed by Corrections
Corporation of America, the leading proprietary prison company
in the country (CCA will be described in more detail in the next
section). Opened in May 1986, the Shelby Training Center in Mem-
phis, Tennessee, was built for Shelby County by CCA—in ten
months with an investment of $6.5 million of company funds—to
hold male delinquents, primarily property offenders. The Training
Center is a secure lockup, with close supervision, locked windows
and doors, and a fence, though it looks like a college campus from
the street. The complex consists of six housing units with 150 single
20 PRIVATE PRISONS

rooms; six classrooms; library; music and art room; mechanical,

woodworking, and electrical shop; dining area; and kitchen. It has
84 employees. The facility is accredited as being in accordance
with American Correctional Association standards, as is its sister
facility, Tall Trees, a 50-bed, nonsecure, community-based juve-
nile facility.
Under a state law effective in 1982, the state pays the county a
per diem fee to hold locally those offenders who would otherwise
be placed in state institutions. The county passes this per diem on
to CCA, minus 2 percent for administration costs. While the facility
has a capacity of 150, the population has been held to 110, which
has been the limit that the state will reimburse, though that may
change.24

Artesian Oaks
Artesian Oaks, in Saugus, California, is a fenced-in campus for
juvenile parole violators. This 100-bed facility is operated by Man-
agement and Training Corporation of Ogden, Utah, a company
that manages federal Job Corps training centers.25

Adult Confinement Facilities

The most recent and most controversial form of proprietary prison

is the adult confinement facility, in which unchaperoned access to
the community is not permitted for most of the population. Private
versions of these facilities began to appear with increasing fre-
quency during the 1980s.
By mid-1989, about a dozen private companies were running
about two dozen adult confinement institutions totaling some 7,000
beds in about a dozen states. If that last sentence seems obsessed
with qualifiers, it is because it is not always clear whether to classify
a particular institution as a "confinement facility" or as a "resi-
dential program," and because contractual prisons and jails can
spring so rapidly into or out of existence (as this book goes to
press in late 1989, there are now about 9,000 of these beds). Thus
no definitive census of these facilities exists.
The types of adult confinement facilities currently under contract
include: jails; regular state and county prisons; facilities for of-
Private Prisons Today: A Descriptive Overview 21

fenders nearing release to parole; lockups for return-to-custody

parole violators; and detention centers for the Immigration and
Naturalization Service. They can be divided into three groups:
federally contracted, state-contracted, and locally contracted.

Federal Contracts

Hidden Valley Ranch

In 1984, the federal Youth Corrections Act expired, leaving the
federal Bureau of Prisons with a number of young adult offenders,
aged 18-26, with remaining sentences to serve for such acts as
bank robbery and mail fraud. The BOP signed a three-year con-
tract with Eclectic Communications, Inc., a for-profit company,
to house about 60 of these offenders at Hidden Valley Ranch in
La Honda, California. The ranch, located in the mountains on 3.5
acres leased from San Francisco County, has the air of a suburban
school, but is surrounded by a 12-foot chain link and razor-wire
fence. 26 The staff of 30 is headed by Tom Keohane, a 27-year
veteran of the Bureau of Prisons.27
The Bureau of Prisons was not the first federal agency to contract
for confinement, nor the one to make the most use of that ar-
rangement. That distinction belongs to the U.S. Immigration and
Naturalization Service. The INS has contracted at least half a dozen
facilities for the detention of illegal aliens awaiting deportation.
These facilities are generally minimum to medium security and
often include arrangements for families. Contracting provided the
INS with a rapid, flexible, and cost-effective response to dramatic
increases in illegal immigration, particularly in the Southwest. Jails
and other facilities operated by local governments could no longer
handle the overflow. Even if overcrowding was not a problem,
existing facilities were unable to accommodate families, and they
were expensive to operate, having been built for a higher level of
security than the INS needed.

Behavioral Systems Southwest, INS Contracts

In 1980, the INS awarded its first facility management contract via
competitive bidding to a for-profit company, Behavioral Systems
22 PRIVATE PRISONS

Southwest. BSS was founded as an offshoot of a behavioral mod-

ification training program by its current President, Theodore R.
Nissen, and Vice President, Tamara S. Lindholm. Both Nissen and
Lindholm had long careers in the California Department of Cor-
rections. BSS now manages several holding facilities for the INS,
community treatment centers for the federal Bureau of Prisons,
and re-entry programs for the corrections departments of Arizona
and California. 28 With 130 employees, many of them ex-convicts,
BSS in 1986 supervised 465 inmates. zy
For its first facility—the Pasadena Immigration Holding Facil-
ity—BSS converted a former convalescent home to house 125 men,
women, and children. 30 Over 3,000 aliens a year are processed by
the Pasadena facility.31 BSS later won contracts for other INS
holding facilities in San Diego and Aurora, Colorado, near Den-
ver. The buildings have locked doors and razor wire on the roof;
the staff is not armed.32 BSS leases its buildings for ten years on
the private market and operates them for the INS on annual con-
tracts, thus exposing itself to some risk if a contract is not re-
newed.33 The company did, in fact, give up its contract to run the
Aurora facility in 1987, when the INS contracted with the Wack-
enhut Corporation to build a new facility there. However, they
were able to convert the facility to another use for the remaining
two years of their lease. They had originally converted a warehouse
into the 85-bed facility at a cost of $150,000.34
BSS has been reported to carry insurance of $5 million and
its profits have been variously reported as 8 percent on a gross
of $6 million annually, 35 or 3.5 percent on revenues of $4.5
million. 36

Corrections Corporation of America, INS Contracts

Another INS contractor is Corrections Corporation of Amer-
ica (CCA), which is now the leading proprietary manager of
correctional facilities. As of summer 1989, CCA operated 16 cor-
rectional facilities in 5 states with 4,238 beds.37 Originally fi-
nanced with $10 million in venture capital from the Massey
Burch Investment Group, the same firm that started Hospital
Corporation of America,38 CCA is now a publicly traded cor-
poration with assets of over $60 million in 1989.39 CCA is self-
Private Prisons Today: A Descriptive Overview 23

insured for $5 million in general liability coverage, which in-

cludes personal coverage of company officers while acting in
their official capacities.40
The company, headquartered in Nashville, was founded in 1983
by Thomas Beasley, a Tennessee businessman, attorney, and
former Tennessee Republican Party chairman, 41 and by T. Don
Hutto, a former commissioner of corrections in Arkansas and Vir-
ginia and President of the American Correctional Association from
1984 to 1986.
In April 1984, CCA opened the Houston Processing Center, a
350-bed dual-purpose facility holding adult illegal aliens awaiting
deportation by the INS and convicted alien offenders in the custody
of the Bureau of Prisons. The company owns the facility, and
operates it under annual contracts with the INS and the BOP.
CCA took just seven months to locate a site and finance, design,
and construct the $5 million facility. 42 During part of the construc-
tion period, CCA leased a former motel to house 140 aliens
temporarily. 43
On the outside, the Houston facility resembles an office building,
including landscaping, while inside it is like a dormitory. Locked
doors provide perimeter security and no weapons are carried in-
side.44 CCA maintains a staff of 55 full-time employees45 and pro-
vides space for several INS officials, whose duties include on-site
monitoring of the CCA contract. 46 The Houston Processing Center
is fully accredited by the Commission on Accreditation for Cor-
rections,47 and operates under the regulations and standards of
the INS.48
In a second contract that began in March 1985, CCA oper-
ates the Laredo Processing Center for the INS in Laredo,
Texas. CCA constructed this 200-bed facility, at a cost of $3
million, in just 145 days. 49 Unlike the Houston facility, which is
restricted to adults, the Laredo facility is designed for men,
women, and children.

Wackenhut Corrections Corporation, INS Contract

The Wackenhut Corrections Corporation (WCC) is a subsidiary
of the $300-million Wackenhut Corporation of Coral Gables, Flor-
ida, one of the world's largest providers of security and related
24 PRIVATE PRISONS

services. Wackenhut, which operates in 40 different countries on

6 continents, has offices in over 100 U.S. cities and 35,000 em-
ployees worldwide. 50 On May 4,1987, just five months after signing
a contract with the INS, Wackenhut opened a new 150-bed facility
in Aurora, Colorado, near Denver, to hold illegal entrants and
immigration offenders who have finished serving sentences in BOP
prisons and arc awaiting deportation. 51 Actual construction time
was just three months. 52 The facility is owned outright by Wack-
enhut; it was not financed through third parties. The facility ad-
ministrator is Craig Dobson, a 20-year veteran of the Bureau of
Prisons and the founder and first director of the National Institute
of Corrections Jail Center in Denver. In August 1989, the facility
was accredited as meeting the standards of the American Correc-
tional Association.

State Contracts

Marion Adjustment Center

The first facility to have the full combination of private ownership,
private operation and management, incarceration of adult felons
sentenced by a state, and a classification level (on paper) of at
least "minimum security,"53 was the United States Corrections
Corporation's Marion Adjustment Center in St. Mary's, Kentucky,
which began receiving inmates on January 6, 1986.
Without detracting from this distinction, it should in fairness
to history be noted that other vendors had previously achieved
all of these elements and more, though not in full combination
at one facility. Most notably, on October 1, 1985, Corrections
Corporation of America assumed exclusive possession (but not
title) and full operation of the Bay County (Florida) Jail. With
a level VI (highest) security wing holding serious offenders such
as capital murderers and rapists, with postconviction as well as
pretrial inmates, with federal and state in addition to county
commitments, and with cells for men, women, and juveniles,
this facility had it a l l . . . except for private ownership simultane-
ous with operation. That element was added later with CCA's
Private Prisons Today: A Descriptive Overview 25

construction and ownership of the Jail Annex, which opened on

April 30, 1986.M
Still, the opening of Marion Adjustment Center was a milestone
in the history of proprietary prisons. Whether the label of "prison"
is accurate or not, the mere fact that it is widely referred to as the
first contracted state prison in modern times makes Marion a po-
litical test of public perceptions of private prisons.
On legal, as opposed to political grounds, however, perhaps the
U.S. Corrections Corporation (USCC) should not be too eager to
identify itself as the first overseer of a privately owned and operated
state "prison." Under a 1972 statute, the Kentucky Corrections
Cabinet is authorized to establish "community residential centers"
to hold convicted felons. The state attorney general, however,
defines the Marion Adjustment Center as a minimum security
prison, for which no specific contract-authorizing legislation exists.
A citizens' group has brought a lawsuit against the state based on
that opinion."
USCC was incorporated in January 1983 by co-owners Milton
Thompson, an architect, and J. Clifford Todd, a builder and de-
veloper, with an initial investment of $1.9 million.56 To support
their bid for the two-year contract, they purchased an old seminary,
St. Mary's College in St. Mary, Kentucky, for $695,000. They were
able to open the facility without any remodeling at first; residents
were employed at that task later. 57
The population at Marion is drawn from state prisoners nearing
their parole eligibility dates.58 The state selects cases for Marion
on the basis of low risk, and USCC can reject any they regard as
unsuitable. This selection process may help explain why 63 percent
of Marion residents are in educational programs, compared to 20
percent of the inmates in other state institutions. 59 During the first
7 months of operation, there were 4 walk-aways, 3 of whom were
recaptured within 24 hours. In a comparison state facility, walk-
aways averaged 1.5 per week. 60 The staff in 1986 numbered 46,
with 33 of them in security. An on-site state employee monitors
the contract.
The facility initially had a physical capacity of 400 and an au-
thorized capacity of 200.61 Kentucky guaranteed USCC a minimum
daily population of 175. Such an arrangement has the advantage,
26 PRIVATE PRISONS

over a minimum payment clause, of ensuring that the state gets

what it pays for, while protecting the contractor.62 In late 1988,
under a second contract from the Kentucky Corrections Cabinet,
a new unit was opened with 250 beds, bringing the authorized
capacity to 450.
USCC carries insurance that will cover up to $1.5 million per
incident. 63

New Mexico Women's Correctional Facility

On July 1, 1988, the state of New Mexico signed a contract with
Corrections Corporation of America calling for a new 200-bed
prison to be designed, financed, constructed, and then operated
by CCA commencing April 1, 1989. Because the state was tardy
in appropriating funds, the facility did not actually open until June
1, 1989. Its purpose is to hold all of the state's female felons,
anticipated to number about 160 (any excess beds, up to the ca-
pacity of 200, may be filled by CCA with inmates from federal or
other state jurisdictions). Located on 40 acres in Grants, New
Mexico, this prison has high external security and internal security
at all levels of custody. This makes it the first minimum through
maximum security contemporary state prison to be run by a private
contractor.
Prior to mid-1989, New Mexico held its female prisoners in
another penitentiary in Grants that doubled as a reception and
diagnostic center. When the CCA facility opened, all current em-
ployees had the option of either becoming part of CCA's staff of
85 full-time employees, or remaining with the state. The state is
responsible for preservice training of all new CCA employees.
The contract has a 4-year initial term, with an option for suc-
cessive 2-year extensions, up to 20 years. The extension option
must be exercised after the first 2 years and every 2 years thereafter,
thus occurring always in the middle of 4-year terms. If the contract
is terminated, the state will buy the facility and all associated
property from CCA, or it may, at its option, make this purchase
anytime during the contract, at a decreasing percentage of the
property's initially appraised value. If the contract runs the full 20
years, the state may buy the property for one dollar.
CCA, which is just beginning its operation at Grants as this is
Private Prisons Today: A Descriptive Overview 27

being written, will offer a program of services to include education,

recreation, counseling, and medical and dental care. The contrac-
tor will run an inmate work program, a work- and school-release
program, and a furlough program. In addition, space will be pro-
vided for an industry program to be run by the state.
CCA will classify inmates, using the state's classification manual,
and will impose discipline, with grievance procedures, in accor-
dance with New Mexico Corrections Department (NMCD) policy
and procedures and state law. However, the NMCD retains final
review and authority over all major decisions affecting inmates:
classification as it affects custody level; discipline; grievances; al-
location or revocation of good time; computation of parole eligi-
bility or discharge dates; and approval of work, medical, or
temporary furloughs. CCA may take no action on these without
a prior written decision by NMCD. The NMCD will provide a
contract monitor, who will have office space in and full access to
the CCA facility.
CCA must maintain property, general liability, and civil
rights liability insurance, naming the state as an additional in-
sured, and it has pledged to defend the state in any lawsuit. In
addition, CCA has indemnified the state against any claims, and
all costs of defending against claims, arising out of the opera-
tion of the prison.
The contract obliges CCA to seek and obtain ACA accreditation
within 24 months of operation.

Texas Preparole Facilities

As of this writing (summer 1989), the Texas Department of
Corrections is in the process of bringing on line four major con-
tracts to add 2,000 beds to their system. Wackenhut and CCA
are each building, and will operate, two 500-bed minimum se-
curity prisons to hold state felons who are within two years of
their release dates. The first of these facilities was opened by
Wackenhut in Kyle, Texas, on June 28, 1989. Scheduled to
open later in the summer are the second Wackenhut facility, at
Bridgeport, Texas, and the two CCA facilities, at Cleveland,
Texas, and Venus, Texas.
The Texas Board of Pardons and Paroles also has three other
28 PRIVATE PRISONS

prerelease facilities under private contract: Pricor Incorporated

(described later, under local jail contracts) is providing 210 beds
at the Sweetwater Reintegration Center and 223 beds at the Hous-
ton Reintegration Center; and CCA provides beds for 200 pre-
parole cases at its Houston Processing Center (where it also holds
INS detainees).
On the other side of the parole process, Wackenhut Correc-
tions Corporation operates the 619-bed Central Texas Parole
Violator Facility for offenders from all over the state who have
violated their paroles.64 This kind of lockup is often referred to
as a Return-to-Custody (RTC) facility and the State of Texas is
loooking to contract for several more. At least one other state,
California, also views the RTC as a prime candidate for privat-
ization.

California Return-to-Custody Facilities

As of the summer of 1989, the California Department of Correc-
tions had six RTC facilities under contract, with plans to add more
later. Although minimum security, these are real prisons, with
razor wire and other security measures. Without going into indi-
vidual details, the six prisons and their contractors are: (1) a 200-
bed RTC at McFarland contracted to Wackenhut; (2) a 200-bed
RTC at Mesa Verde (Bakersfield) contracted to Gary White and
Associates; (3) the 200-bed Eagle Mountain RTC contracted to
Management and Training Corp.; (4) a 200-bed RTC at Baker
contracted to Eclectic Communications Inc. (ECI); (5) Hidden
Valley Ranch, an 88-bed RTC at La Honda contracted to ECI65;
and the Leo Chessley Center at Live Oak, a 100-bed RTC for
females, also contracted to ECI.
In addition to the contractors' own staff at these RTCs, the
state provides California corrections officers to supervise secu-
rity. Despite reservations about privatization in the Department
of Corrections, and opposition from correctional employees'
unions in the legislature, the department plans to contract
more facilities like these.66 Such contracts allow the state to re-
spond quickly to rapid increases in cases, without the long-term
commitment entailed in construction and civil service em-
ployment.
Private Prisons Today- A Descriptive Overview 29

Local Contracts

While no state has yet contracted for the operation of a medium

or maximum security prison for males, something like that has
already occurred on a small scale at the local level, in contracts
for the management of jails. By their nature, jails must be equipped
to hold and process all types of offenders as they enter the criminal
justice system. In addition, jails often serve as temporary catch-
basins for the runoff and spillover from all sorts of other institu-
tions. Thus, overcrowding anywhere aggravates the already serious
overcrowding in local lockups. Traditionally the poorest level of
government, counties have been under the greatest pressure to
find alternative solutions to crises in the demand for and supply
of prisons. Despite organized opposition from the National Sher-
iffs' Association, one solution being explored by counties is private
contracting.

Bay County Jail and Annex

A few years ago, Bay County, Florida, was devoting 65 percent
of its budget to its jail, yet it faced overcrowding, medical prob-
lems, violations of state regulations, and lawsuits, including one
inmate rape case in which corrections officers testified that they
had not been on the floor for over two hours.67 At the suggestion
of a local newsman, the county commission considered the option
of privatization. After an open and competitive process of Re-
quests for Qualifications and subsequent bids, the county awarded
a contract to Corrections Corporation of America.
Under the contract, CCA assumed full management of the Bay
County Jail starting October 1, 1985, and agreed to build (and
own) a new jail annex in addition to renovating the jail. The annex
opened on April 30, 1986. Its facilities for 200 men, women, and
(separately housed) juveniles supplemented the 204 beds for adult
men at the main jail. The Bay County/CCA facilities hold every
type of offender, from misdemeanants to capital murderers. Pre-
trial detention cases and convicted offenders awaiting transfer or
serving sentences of one year or less are also placed here. Most
are county inmates, but some are state or federal prisoners.
30 PRIVATE PRISONS

The Bay County Jail is secured by means of locked exits, locked

internal gates, and locked cells. The annex has a chain-link and
razor wire fence, internal gates, and some locked cells in addition
to the more open, dormitory-style housing areas. Both facilities
have remote monitors and lock-control stations. Consistent with
Florida policy followed in all county jails, no arms are carried
inside either facility, but some CCA security guards arc certified
to carry arms outside.
A county employee monitors the contract and has final authority
over all recommendations regarding "gain time" made by CCA
personnel in disciplinary actions. Rules for inmates and procedures
for discipline and allocation of gain time were established in con-
formity to Florida statutes and Department of Corrections policies.
Where not in conflict with Florida law or policy, CCA abides by
the often more exacting standards of the American Correctional
Association. Under contract, CCA is required to provide insurance
and to indemnify the county against all legal damages resulting
from the operation of the jail. At first, CCA carried a purchased
policy; later they instituted a self-insurance plan.
Bay County Sheriff Lavelle Pitts was strongly opposed to the
contract. He retained all of his salary and duties as sheriff except
for administration of the jail, but 70 of the 75 deputies who pre-
viously worked in his jail became CCA employees on completing
40 hours of training by CCA.68 CCA paid them for accrued vacation
time and gave them raises ranging from 10 percent to 20 percent.
After one year with the company, employees are eligible to par-
ticipate in an employee stock ownership program. 69
The 20-year contract with CCA specifies conditions for termi-
nation by either party. The fee charged by CCA may increase with
inflation, but by no more than 5 percent a year. There is a pro-
vision, however, for negotiating adjustments every three years in
the event of unforseen circumstances. A budget limit is set each
year based on projected population size and projected inflation.
The fee varies by population level, with lower per diem charges
above specified population breakpoints. The contract specifies a
minimum payment by the county of an amount equal to the charge
for 285 prisoners per month. CCA may rent unused space to other
jurisdictions, at a price at least equal to that charged to Bay
County. Any excess is shared equally by the county and CCA.
Private Prisons Today: A Descriptive Overview 31

Silverdale Detention Center

On October 15,1984—one year before it took over the Bay County
Jail—CCA assumed management of the Silverdale Detention Cen-
ter, a minimum to medium security work farm for adults under
the jurisdiction of Hamilton County, Tennessee, at Chattanooga.
Silverdale has a capacity of 400, including 100 beds for females.
Since the Hamilton County jail has no arrangements for females,
some of the women at Silverdale are pretrial detainees; the rest
of the prisoners are convicted offenders. As reported in the New
York Times, Silverdale "houses state prisoners serving long terms
for felonies, including murder, county prisoners serving less than
a year for misdemeanors, and a growing number of prisoners serv-
ing mandatory 48-hour sentences for driving while intoxicated.
Also growing in number are second offenders serving a minimum
of 45 days and third offenders serving a minimum of 120 days
under Tennessee's tough laws on driving while drunk." 70 The max-
imum sentence is 6 years, and about one-third serve more than
one year, but the average stay is short, about 45 days. 71
As part of the contract, Hamilton County leases the Silverdale
facility to CCA. All buildings and grounds remain the property of
the county, including the $1.6 million in renovations and additions
invested by CCA during its first year of operation. If the county
terminates the contract, however, it must reimburse CCA for any
remaining unamortized portion of this investment. The contract
runs for 32 years and is automatically renewable at 4-year intervals,
though conditions for termination are included. Payment rates,
however, are renegotiated every year.
All county employees at Silverdale were retained by CCA, but
there was 50 percent turnover within two years. 72 The warden of
Silverdale under its operation by the county is now the Hamilton
County Director of Corrections and the monitor for the CCA
contract. Unlike Sheriff Pitts, the displaced warden in Bay County,
the Hamilton County corrections director is favorably impressed
with the new management of his old facility, in spite of his initial
skepticism toward the idea of a proprietary operation. 73
CCA expanded bed space and introduced many changes at Sil-
verdale. A list prepared by the warden identifies 74 innovations
effected by CCA during its first year. 74 Public officials doubt that
32 PRIVATE PRISONS

the county could have afforded to make these improvements on

its own.75
In contractually mandated pursuit of accreditation, corrections
officers receive more training than was previously provided by the
county, and many operating procedures are tighter than before.76
As in all CCA facilities, the officers inside are unarmed; after
experiencing a minor disturbance, Silverdale obtained riot equip-
ment, including mace, shields, and riot guns, to be issued if
needed.77 Procedures for the discipline of inmates and the allo-
cation of gain time are specified in CCA's contract and are similar
to those for the Bay County Jail. 78

Santa Fe County Jail

Santa Fe County, New Mexico, is the site of a third CCA contract
at the local level of government. After running its new jail for a
year, the county decided to consider the private alternative. As a
result of competitive bidding among five contenders, the county
awarded a three-year contract to CCA, commencing August 1,
1986.79 The jail—the Santa Fe County Detention Facility—has 147
beds, plus a section for juveniles with 20 housing cells and 4 book-
ing cells.80
Through supplemental contracts, CCA also rents some of its
beds at the Santa Fe County Detention Facility to other jurisdic-
tions, including the city of Santa Fe, the Federal Bureau of Prisons,
the U.S. Marshall's Service, San Miguel County, and Pecos and
Rio Arriba, New Mexico.81 CCA has posted a $325,000 certificate
of deposit as collateral on a performance bond for this contract.82
CCA assured the staff of 58 that they would retain their jobs
and salaries but that there would be retraining and reorganization
required. The company "also promised to assume employees' ac-
crued vacation; [to] offer 10 paid holidays [plus optional unpaid
holidays], 12 days of annual leave and 12 sick days; [and to] provide
comprehensive medical and life insurance and a stock ownership
plan." 83
A 1984 New Mexico statute authorizes contracting for two
county jails as pilot projects.84 The law grants peace-officer powers
to private jailers but denies them the power to award or take away
Private Prisons Today: A Descriptive Overview 33

gain time; that decision is left to sheriffs. It limits contracts to three

years, and requires the contractor to assume all liability and to
buy enough insurance to cover itself, as determined by the Risk
Management Division of the General Services Department. Coun-
ties are required to inspect both private and public jails at least
two times a year. If these inspections uncover apparent violations
of statute, the district attorney must sue the sheriff, jail adminis-
trator, or contractor. Finally, the New Mexico law provides for
termination of contracts on 90 days notice, for cause—which may
include failure to meet minimum standards or other deficiencies
that seriously affect the operation of the jail. 85

Volunteers of America Regional Corrections Center

Corrections has a long history of volunteers who come into insti-
tutions or who work with offenders and ex-offenders in the com-
munity. Volunteers of America (VOA) has been involved in
corrections since its founding in 1896.86 It was not until September
1984, however, that VOA contracted to assume total operation of
a jail.
The Volunteers of America Regional Corrections Center, in
Roseville, Minnesota, a suburb of St. Paul, has 40 beds, of which
about 25 at a time were occupied in 1985.87 It has a staff of 17 full-
time, 3 part-time, and 12 volunteer workers. All the inmates are
women serving sentences of up to one year.88 The VOA leases its
facility—a former juvenile detention center—from Ramsey
County, and receives prisoners from county, state, and federal
courts.89 Two other counties also place prisoners at the VOA fa-
cility, under separate contracts,90 and some prerelease prisoners
are received from the federal Bureau of Prisons.

Pricor Incorporated: Two Jails

Pricor Incorporated is a private corrections firm headquartered in
Nashville, Tennessee, where it spun off from CCA early in the
latter's development. Like CCA and Wackenhut, Pricor issues
public shares of common stock and operates on a national level.
34 PRIVATE PRISONS

While Pricor lately has been foeusing on juvenile facilities, it also

operates adult preparole facilities (described above) and jails. One
jail contract is for the Tuscaloosa Metropolitan Minimum Security
Detention Facility, a 150-bed jail, where Pricor books all arrestecs
and holds just the minimum security cases for the city and county
of Tuscaloosa, Alabama, and for the city of Northport. A second
contract is for the Greene County Jail, in Greeneville, Tennessee,
a 160-bed jail for all county arrestees and detainees.

Butler County Prison

On October 1, 1985, Butler County, Pennsylvania, turned over
management of its county prison (and jail) to Buckingham Security
Limited, under a two-year contract. It thus became one of the first
two jails (along with CCA/Bay County) to transfer from public to
private management. The contract specified that Buckingham was
to be responsible for management and operating costs, while Butler
County was responsible for capital improvements, medical ex-
penses, and jail employee labor costs.91 The contract was later
renewed for one more year, then allowed to expire without renewal
when a new county commission was elected following an intense
campaign against the prison contract. More on that development
shortly.
Buckingham Security was incorporated in Connecticut by Peter
Savin, a Hartford builder, and Joseph and Charles Fenton.92 Buck-
ingham had previously planned to build a 720-bed prison in Penn-
sylvania to hold protective custody cases from several states. These
prisoners require special resources inside a normal prison, where
they must be segregated from the general prison population. In a
regional facility, however, they can be served more efficiently, due
to economies of scale and greater homogeneity of population. That
plan, however, foundered on the shoals of siting problems, liability
issues, and a one-year moratorium on prison privatization imposed
by the Pennsylvania legislature. The moratorium did not affect the
preexisting contract with Butler County.
Charles Fenton, who had served as the warden of maximum
security penitentiaries in Marion, Illinois, and Lewisburg, Penn-
sylvania, during a 23-year career with the federal prison system,
became warden of the Butler County Prison under the contract.
Private Prisons Today: A Descriptive Overview 35

The prison, or jail, is a high-security facility holding both pretrial

and postconviction county prisoners, with some state and federal
offenders as well. All are adults, both male and female.
The capacity of the prison at the start of the contract was about
100.93 Buckingham increased the total capacity by over 10 percent
in the first year, without additional staff, and another 50 percent
increase was anticipated soon thereafter. 94 The facility was reha-
bilitated with inmate labor under Buckingham's supervision,
thereby reducing the county's capital improvement costs. Previ-
ously, inmates performed no work of any sort.
The contract was monitored by a county employee. Disciplinary
protocol was outlined in the contract. However, since the jail of-
ficers were still county employees, there was less of an issue of
private delegation than is the case under some other contracts.95
In contrast to prior Butler County custom, under the contractor's
management police no longer were allowed to bring their firearms
with them into the jail. 96
A professional system of key control, head counts, security sys-
tem checks, cell inspections, and classification, admission, and re-
lease procedures was established at the jail for the first time.97
Buckingham ended the county practice whereby officers and other
staff did not venture into living areas, passageways, and recreation
yards while inmates were there.98
An interesting feature of this contract is that up until the day
before it was to go into effect, Buckingham had intended to fire
the entire staff of unionized county employees and replace them
with workers of its own choosing and in its own employ. However,
the union obtained an eleventh-hour court ruling that required the
company to retain the county workers. The resulting contract ad-
justment left Buckingham in the strange if not unique position of
being a private management company supervising public employ-
ees who remained on the county payroll.
In view of the rocky start, the ensuing hostile atmosphere, and
the long-standing and vehement opposition to private contracting
on the part of the union leadership, one might have expected all-
out internal warfare. Surprisingly, however, labor relations at But-
ler Prison were in some ways better after the contract than they
were before. Every year since the prison workers first unionized
under AFSCME, negotiations with the county had ended in im-
36 PRIVATE PRISONS

passe followed by arbitration. Under Buckingham's management,

the workers achieved their first labor contract with the county.
The employees gained higher pay, while management was allowed
to redefine job responsibilities and to eliminate much costly over-
time and part-time work." Although union opposition remained
vehement, employee morale was high. During the first few years
after Buckingham assumed management, no employee left, in con-
trast to a turnover of three or four per year previously.
In a first-year report, Buckingham summarized its accomplish-
ments as follows:
The County Commissioners have saved money and are confident
that the prison for the first time is under competent, professional
management. The union for the first time in history has a signed
contract with the county. Employees have better working condi-
tions, higher pay and greater pride. The sheriff has fewer hassles
and less expense. The prison board is confident that they have a
smoothly running prison, functioning in accord with local, state and
federal laws. Inmates have brighter, cleaner, safer and more peace-
ful living conditions. New programs have been instituted that have
positively impacted on work release, health, education, cleanliness,
physical fitness, work and recreation.1""

While Buckingham might be expected to toot its own horn, this

assessment was confirmed by county officials. In a letter to Warden
Fenton dated November 13, 1986, the Chairman of the County
Commission, Richard M. Patterson, said:
Less than one year ago, we had a great deal of concern about the
Butler County Prison. It occupied our time almost daily. Control
was in question. Both the employees and the prisoners were in a
serious state of turmoil. Court action was involved, and the public
was agitated by negative media comment.
Within three months, due only to the professionalism of Buck-
ingham Security], the whole matter has made a one-hundred-eighty
degree turn, and all is quiet and all is under control, including the
cost. 101
Not everyone was happy with the contract, however. The union
(AFSCME) was opposed from the beginning and was already or-
ganized throughout the state in support of a legislative moratorium
on prison privatization. At the next election, they gave their full
Private Prisons Today: A Descriptive Overview 37

support, in money and campaign labor, to two candidates for

County Commission who promised, if elected, to put the jail once
again under direct county management. The union disagreed with
the contractor, and County Commission members and candidates
disagreed among themselves, as to whether the contract was saving
the county money. According to accounts from all sides, however,
it was labor and county politics even more than economics that
drove the conflict. 102 This became the major issue of the campaign
and the two union-backed candidates were elected, defeating and
replacing two incumbents who supported the contract with
Buckingham.
I have given extra attention to this contract because it reflects
the intensity of the debate surrounding private prisons. Most of
the rest of this book will be devoted to identifying and examin-
ing the issues and arguments involved in that debate.
3
Issues and Arguments in the
Debate over Private Prisons

This chapter gives a concise overview of arguments on both sides

of the controversy over private prisons. It identifies the issues and
lists in summary form a series of claims and counterclaims that
will be examined in detail in the remainder of the book. The
chapter is intended as a reference tool; many readers may pre-
fer to skim through it and proceed quickly to the more comp-
lete discussions of each issue that begin with the next chapter.
However, readers who want to sec a conceptual framework
defining and organizing the issues, and a comprehensive but
condensed guide to arguments on both sides of each issue, will
find that here.

Private Prison Issues'

Privately contracted prisons raise many philosophical, empirical,

and policy questions. These cover at least the following range of
issues:
1. propriety 6. security
2. cost 7. liability
3. quality 8. accountability
4. quantity 9. corruption
5. flexibility 10. dependence

38
The Debate over Private Prisons 39

1. Questions of propriety may be philosophical, political, or

legal. Is it proper for imprisonment to be administered by anyone
other than government officials and employees? How might private
delegation of authority affect the legitimation of prisons in the eyes
of inmates or the public? Is the "profit motive" more or less com-
patible with doing justice than are the motives to be found within
state bureaucracies, employee unions, or nonprofit agencies?
Should prison contracts permit the private exercise of quasi-judicial
authority (i.e., classification, discipline, allocation of gain time)?
2. Is cost as likely to be reduced with the privatization of cor-
rections as it has been with some other public services? Or does
experience with privatization in other areas suggest that the net
costs may actually be higher in the long run, as a result of "low-
balling" or due to the added costs of supervision and of the con-
tracting process itself? Can the process of contracting help clarify
the true costs of both public and private service delivery in
corrections?
3. Will privatization increase the quality of imprisonment due
to innovations by private companies? Or will commercial com-
panies cut corners to save costs and thereby lower quality? What
are the advantages and disadvantages of government control versus
competition as a quality control mechanism? Can the advantages
of competition be obtained without involvement of the private
sector? How can the contracting process be used to specify and
clarify standards?
4. How might privatization affect the quantity of imprisonment?
Will it merely help meet an independently determined demand,
or will commercial companies lobby to increase the demand?
5. Will private contracts bring with them the greater flexibility
characteristic of small businesses and entrepreneurs? Do they re-
duce red tape and avoid the perpetuation of agencies and programs
commonly found in government? Can the private sector more ac-
curately anticipate and more rapidly respond to the correctional
needs of government? Or will contracts bring with them their own
form of rigidity, such as restrictions on what can be expected or
demanded? Do contracts encourage short-term, over long-term,
planning?
6. Can security be ensured in private prisons? What are the legal
limits to the delegation of authority to use deadly force? How does
40 PRIVATE PRISONS

the training of government correctional personnel compare to that

of the staff of private companies? What steps can be taken to
prevent, ensure against, or deal with a possible disruption of pri-
vate prison operations due to strikes or bankruptcy?
7. Does a private prison contract simply extend and add to the
liability of government, or does it defray and reduce liability costs
through insurance and increased incentives to avoid expensive
lawsuits?
8. Is accountability decreased because private prisons are less
accessible to public scrutiny, or increased because the private sec-
tor is more vulnerable than the state to legal controls? Do contracts
diffuse responsibility, or do they increase it by providing anoth-
er mechanism of control over prison managers? How account-
able are correctional institutions and personnel under current
arrangements?
9. Would the potential for corruption in running prisons be
higher, lower, or merely different in form under contractual ar-
rangements? Can close monitoring, along with competition and
market processes, keep the bidding for and the granting of con-
tracts honest, or is collusion inevitable? How do the possible forms
of corruption under public-private management differ from those
under purely public systems? Which forms are easier to control?
10. How can government protect itself from merely substituting
a private monopoly for a public monopoly? Should the government
retain some correctional capacity in its own hands in order to
forestall dependence on a private provider? Should it contract only
to multiple providers? Or does the possibility of competition at
some time in the future limit the potential for abuse by a solitary
contractor?

Arguments for and against Private Prison Contracting

In the chapters that follow, each of the issues defined above will
be examined closely. Generally, the format of those chapters will
be one in which an issue is raised as a negative or critical question
posed by opponents of private prisons, followed by a response on
the positive side. Here, that order is reversed, with arguments
favorable to contracting for management of prisons followed by a
The Debate over Private Prisons 41

separate set of arguments in opposition. The arguments have been

abstracted and adapted from general literature on privatization as
well as from discussions specific to prisons.2

Arguments for Contracting

1. Propriety
a. Contracting enhances justice by making prison supply more
responsive to changes in demand, both upward and
downward.
b. Contractual wardens have an incentive to govern inmates
fairly in order to enhance their legitimation, induce co-
operation, lower costs, and ensure renewal of contracts.
c. Contracting does not jeopardize due process; private and
public wardens are equally subject to the rule of law and
accountable to the same constitutional standards.
d. Contracting, in conjunction with governmental monitoring,
adds a new layer of independent review of correctional
decisions and actions, thus improving due process.
e. Contracting can help clarify the purposes of imprisonment
and the rules and procedures that define due process.
f. Contracting for operating prisons is compatible with federal
law and the laws of many states; specific enabling legislation
has been passed in some states.

2. Cost
a. Contracting allows prisons to be financed, sited, and con-
structed more quickly and cheaply than government pris-
ons; also, private firms are more apt to design for efficient
operation.
b. Contracting across jurisdictions permits economies of scale.
c. Contracting may reduce overly generous public employee
pensions and benefits.
d. Contracting typically indexes fee increases to the Consumer
Price Index, while government costs have been shown to
rise faster than the general level of inflation.
e. Contracting discourages waste because prodigality cuts into
profits.
42 PRIVATE PRISONS

f. Contracting counteracts the motivation of budget-based

government agencies to continually grow in size and to
maximize their budgets.
g. Contracting makes true costs highly visible, allowing them
to be analyzed, compared, and adjusted.
h. Contracting avoids cumbersome and rigid government pro-
curement procedures; vendors can purchase more quickly,
maintain lower inventories, and negotiate better prices and
values.
i. Contracting, through more effective personnel manage-
ment, better working conditions, and less overcrowding,
may increase employee morale and productivity while low-
ering absenteeism and turnover.

3. Quality
a. Contracting provides an alternative yardstick against which
to measure government service; it allows for comparisons.
b. Contracting motivates both governmental and private pris-
ons to compete on quality as well as cost.
c. Contracting, by creating an alternative, raises standards for
the government as well as for private vendors.
d. Contracting adds new expertise and specialized skills.
e. Contracting promotes creativity and enthusiasm by bring-
ing in "new blood" and new ideas more often than is pos-
sible under civil service.
f. Contracting promotes quality and high standards by forcing
officials and the public to evaluate expenditures carefully,
rather than masking costs through overcrowding and sub-
standard conditions.
g. Contracting will expand the political constituency con-
cerned about legislative reforms of the correctional system.
h. Contracting could hardly do worse than some current (pub-
lic) prisons, in terms of quality.

4. Quantity
a. Contractors can help alleviate today's capacity crisis by
building new prisons faster than the government can.
The Debate over Private Prisons 43

b. Contracting will allow quicker response in the future to

meet changing needs or to correct mistakes resulting from
inaccurate predictions or faulty policies.
c. Contracting facilitates the distribution of inmates across
agencies or jurisdictions, thereby maintaining occupancy
rates at an efficient level (i.e., near capacity but not
overcrowded).
d. Contracting helps limit the size of government.
5. Flexibility
a. Contracting allows greater flexibility, which promotes in-
novation, experimentation, and other changes in programs,
including expansion, contraction, and termination.
b. Contracting can avoid capital budget limits through leasing,
or spread capital costs over time through lease-purchasing.
c. Contracting reduces the levels of bureaucracy (red tape)
involved in management decisions.
d. Contracting reduces some of the political pressures that
interfere with good management.
e. Contracting avoids civil service and other government (and
sometimes union) restrictions that interfere with efficient
personnel management (i.e., hiring, firing, promotion, and
salary setting; assignment of duties, work schedules, va-
cations, and leaves; adequate staffing to avoid excessive
overtime).
f. Contracting reduces the tendency toward bureaucratic self-
perpetuation.
g. Contracting promotes specialization to deal with special-
needs prisoners (protective custody, AIDS patients, and so
forth).
h. Contracting relieves public administrators of daily hassles,
allowing them to plan, set policy, and supervise.
6. Security
a. Contracting may enhance public and inmate safety through
increased staff training and professionalism.
b. Contracted corrections officers are less likely to go on strike
because they are more vulnerable to termination.
44 PRIVATE PRISONS

7. Liability
a. Contracting may decrease the risks for which government
remains liable, through higher quality performance and
through indemnification and insurance.

8. Accountability
a. Contracting increases accountability because market mech-
anisms of control are added to those of the political process.
b. Contracting increases accountability because it is easier for
the government to monitor and control a contractor than
to monitor and control itself.
c. Contracting promotes the development and use of objective
performance measures.
d. Contracting can help enforce adherence to procedures and
limit or control discretion in the discipline of inmates.
e. Contracted prisons will be highly visible and accountable,
in contrast to state prisons which, at least historically, have
been ignored by the public and given (until recently)
"hands-off" treatment by the courts.
f. Contractors are forced to be more responsive to the atti-
tudes and needs of local communities when siting a prison.
g. Contracting can require prisons to be certified as meeting
the standards of the American Correctional Association.
h. Contracting motivates vendors to serve as watchdogs over
their competitors.
i. Contracting will encourage much broader interest, involve-
ment, and participation in corrections by people outside of
government.
j. Contracting provides a surgical solution when bad man-
agement has become entrenched and resistant to reform.

9. Corruption
a. Contracting gives managers more of a vested interest in the
reputation of their institution.
b. Contracting pits the profit motive against other, less benign
motives that can operate among those whose job it is to
punish criminals.
The Debate over Private Prisons 45

10. Dependence
a. Contracting can increase the number of suppliers, thus re-
ducing dependence and vulnerability to strikes, slowdowns,
or bad management.

Arguments against Contracting

1. Propriety
a. Contracting for imprisonment involves an improper dele-
gation to private hands of coercive power and authority.
b. Contracting may put profit motives ahead of the public
interest, inmate interests, or the purposes of imprisonment.
c. Contracting prisons raises legal questions about the poten-
tial use of deadly force.
d. Contracting creates conflicts of interest that can interfere
with due process for inmates.
e. Contracting may face legal obstacles in some jurisdictions.
f. Contracting threatens the jobs and benefits of public em-
ployees; it is antilabor.
g. Contracting may threaten corrections officers' sense of au-
thority and status, both inside and outside the prison.
2. Cost
a. Contracting is more expensive because it adds a profit mar-
gin to all other costs.
b. Contracting creates the special costs of contracting: initi-
ating, negotiating, and managing contracts, and monitoring
contractor performance.
c. Contracting may cost more in the long run as a result of
"lowballing"—initial low bids followed by unjustifiable
price raises in subsequent contracts.
d. Contracting may cost more in the long run if high capital
costs inhibit market entry and restrict competition.
e. Contracting lacks effective competition in "follow-on" con-
tracts, which are commonplace.
f. Contracting costs the government extra for the termination,
unemployment, and retraining of displaced government
workers.
46 PRIVATE PRISONS

g. Contracts with cost-plus-fixed-fee provisions provide no in-

centive for efficiency.
h. Contracting may have a higher initial marginal cost than
would expanding government services.

3. Quality
a. Contracting may reduce quality through the pressure to cut
corners economically.
b. Contracting may "skim the cream" by removing the "best"
prisoners and leaving the government prisons with the
"worst," which will spuriously make the private prisons
look better by comparison.
c. Contracting will decrease the professionalism of rank and
file prison employees because they will be underpaid and
insecure and thus not able to develop a career orientation.

4. Quantity
a. Contracting creates incentives to lobby for laws and public
policies that serve special interests rather than the public
interest; in particular, private prison companies may lobby
for more imprisonment.
b. Contracting, simply by expanding capacity and making im-
prisonment more feasible and efficient, may unduly expand
the use of imprisonment and weaken the search for
alternatives.
c. Contracting on a per prisoner, per diem basis gives private
wardens an incentive to hold prisoners as long as possible.
d. Contracting creates a kind of underground government,
thus adding to total government size.

5. Flexibility
a. Contracting may limit flexibility by refusal to go beyond
the terms of contract without renegotiation.
b. Contracting may be stopped in advance, or suddenly re-
versed in midstream, by adverse public reaction, legal chal-
lenges, partisan politics, or organized opposition by interest
groups, including public employee unions.
The Debate over Private Prisons 47

c. Contracting reduces ability to coordinate with other public

agencies (police, sheriff, probation, parole, transportation,
maintenance, and the like).

6. Security
a. Contracting may jeopardize public and inmate safety
through inadequate staff levels or training.
b. Contracting may limit the ability of the government to re-
spond to emergencies, such as strikes, riots, fires, or
escapes.
c. Contracting increases the risk of strikes, which may not be
illegal for contractor personnel.
d. Contracting may cause high employee turnover at
transition.

7. Liability
a. Contracting will not allow the government to escape
liability.
b. Contracting may cost the government more by increasing
its liability exposure.
c. Contracting shifts risk away from the government, which
is the party best able to bear it.

8. Accountability
a. Contracting reduces accountability because private actors
are insulated from the public and not subject to the same
political controls as are government actors.
b. Contracting diffuses responsibility; government and private
actors can each blame the other.
c. Contracting may encourage the government to neglect or
avoid its ultimate responsibility for prisons; supervision
may slacken.
d. Contracting reduces accountability because contracts are
difficult to write and enforce.

9. Corruption
a. Contracting brings new opportunities tor corruption (i.e.,
political spoils, conflict of interest, bribes, kickbacks).
48 PRIVATE PRISONS

10. Dependence
a. Contracting lowers the government's own capacity to pro-
vide services, which makes it dependent on contractors.
b. Contracting carries the risk of bankruptcy by the vendor.
c. Contracting may involve exclusive franchises that simply
replace public monopolies with private monopolies.
4
The Propriety of Proprietary Prisons

The most strongly expressed, and least critically examined, objec-

tions to private prisons are those that are presented as statements
of "principle." Some of the government's most vehement critics,
especially when it comes to running prisons, nonetheless—for rea-
sons of principle—are championing the government as the only
suitable manager of these institutions. The ACLU, for example,
regards imprisonment as among the "functions which rightfully
belong to government." ' The officially recorded policy of the
ACLU states:
The delegation of control and custody of prisoners to private enti-
ties, in and of itself, raises serious constitutional concerns. Because
the deprivation of physical freedom is one of the most severe in-
terferences with liberty that the State can impose, and because of
civil liberties concern created by private management. . . the power
to deprive another of his/her freedom cannot be delegated to private
entities. 2
Sandy Rabinowitz, Director of the Houston office of the ACLU,
declares that "the whole concept [of private prisons] is really
frightening." 3
Mark Cuniff, Executive Director of the National Association of
Criminal Justice Planners, says: "We're talking about taking away
people's liberty, and I have questions about the propriety of anyone
but the state doing that." 4 These organizations, like many critics
of private prisons, raise a troublesome question but make no effort
to explore possible answers. It is simply taken for granted that if

49
50 PRIVATE PRISONS

the power in question is strong enough, only the state may legit-
imately apply it.
John Dilulio, Jr., a political scientist at Princeton and a brilliant
analyst of prisons, is one critic who has carried the propriety ob-
jection beyond the level of intuitive reaction and into the realm
of more serious thought. Writing in Public Interest and elsewhere,
Dilulio identifies the central question as a normative one: Who
should govern prisons?5 In a thoughtful answer, Dilulio begins by
accepting for the sake of argument that private prisons could do
everything that prisons are supposed to do, and do so better and
cheaper than government prisons. Even then, Dilulio insists, pri-
vate prisons would be undesirable, as a matter of principle. With
admirable clarity and consistency, Dilulio points out that the issue
of motive (whether profit or other) is irrelevant, as is the issue of
scope. The private administration of even one halfway house, he
says, would raise the matter of principle just as sharply as the
private ownership and operation of every prison in the country. 6
It is not clear why Dilulio stops there. The final extension of this
argument would seem to be that even specific aspects of the care
and custody of prisoners—such as food service, health care, and
treatment programs—must also be provided directly by gov-
ernment.
While Dilulio does not take his argument to its limits, he does
insist that proponents of private prisons should be prepared to
extend their position to encompass private police, judges, juries,
and executioners. 7 Indeed, Dilulio asks them to go so far as to
explain why we should not turn over to private contractors the
process of selecting a president, supposing for the sake of argument
that professionals could do a much more competent job of it than
can the general electorate. 8 These might be fair questions to ask
of those who advocate total privatization, divestiture of all gov-
ernment authority, and dismantlement of the state. However,
where the issue is contracting part of one particular government
function (and delegation of relevant authority), it is not necessary
to defend the privatization of all functions or delegation of every
authority.
It must be made clear that contractually managed prisons are
still government prisons. They do not exist on their own authority.
The Propriety of Proprietary Prisons 51

A case might be made for truly private prisons independent of

government authority, but no one arguing for prison contracting
is attempting to make that case. In the current argument, the choice
is only between (a) direct governmental provision through salaried
employees versus (b) governmental procurement through contract.
Those who govern prisons at the facility level (administrators
and correctional officers) derive their authority and legitimacy only
indirectly from the electoral process. Unlike Dilulio's presidential
analogy, none of these actors is elected; they are all hired. Their
legitimacy depends on an uncorrupted chain of accountability lead-
ing back to those who are elected and who set policies and issue
orders in the name of the people. Our elected leaders exercise
very little direct power; rather, they issue instructions and direc-
tives that are carried out by subordinates. The bigger and more
centralized the state or government becomes, the longer the chains
down which authority is transmitted and the greater the power
that is exercised by those at the ends of the chains. The integrity
of each link in a particular chain is therefore very important. How-
ever, it is false to assume that the integrity of a chain of civil
servants is necessarily superior to a contractual chain.
Compare the chains of authority in the following two examples:
(1) an elected county commission chooses a private jail vendor in
an open, competitive process at a public meeting, using publicly
announced criteria developed by the commission; and (2) an
elected mayor appoints a friend and financial supporter to run the
jail as a public employee. These examples, which are not unreal-
istic, make it clear that public employee jailers are not necessarily
chosen through processes that are more open, democratic, or in-
herently honest than the methods used to select contractual jailers.
Dilulio argues that, all other things held equal, something which
he calls "the public interest" or "the common good" requires that
prisons be run directly by government employees.'' He does not
identify any specific public interest that can be shown to be ill-
served by contracting. Rather, he presents an apparently analytic
truth—a tautology. Seemingly as a matter of definition, only gov-
ernment employees, not contractors, can be public servants—i.e.,
can serve "the public interest." Dilulio does not use the phrase
"government employees." He simply refers to "government."
52 PRIVATE PRISONS

However, government cannot act without recruiting individual ac-

tors; the relevant distinction here is between two kinds of govern-
mental recruitment: employment versus contract.
Dilulio asks, rhetorically, whether the government's responsi-
bility to govern "ends at the prison gate," as if a contract could
cause it to do so.10 He concludes that "we are most likely to
improve our country's prisons and jails if we approach them not
as a private enterprise to be administered in the pursuit of profit,
but as a public trust to be administered on behalf of the community
and in the name of civility and justice." " This thesis, however,
is based on an erroneous assumption. Contracting does not con-
stitute abdication of responsibility, and it is not necessary to choose
between contracting, on the one hand, and civility, justice, and
fulfillment of the public trust, on the other.

The Derivation and Delegation of Authority

to Imprison12

How can it be proper for anyone other than the state to imprison
criminals? Perhaps the place to start is by asking what makes the
punishment of lawbreakers proper for the state itself. By what
right does the state imprison?
In the classical liberal (or in modern terms, libertarian) tradition
on which the American system of government is founded, all rights
are individual, not collective. The state is artificial and has no
authority, legitimate power, or rights of its own other than those
transferred to it by individuals.
Why does this transfer take place? In the instance of criminal
justice, John Locke argued that individuals in the state of nature
have the right to punish those who aggress against them. However,
there will always be disagreement over interpretations and appli-
cations of natural law; people cannot be unbiased in judging their
own cases; and the victims of crimes may lack the power to punish.
For these reasons, said Locke, people contract to form a state and
completely give over to it the authority to bring malefactors to
justice. Locke's contractarian view of government can be used
either to justify or to challenge governmental monopoly of the
legitimate use of force. On the one hand, Locke gives pragmatic
The Propriety of Proprietary Prisons 53

reasons for limiting to government the power to judge and punish.

On the other hand, Locke insists that this authority does not orig-
inate with the state, but is granted to it. Moreover, that grant is
a conditional one. Citizens reserve the right to revoke any of the
powers of the state, or indeed, the entire charter of the state, if
necessary.
Robert Nozick, like Locke, sees the right to punish as one held
by individuals in a state of nature. He also insists that no collective
rights or entitlements emerge beyond those held by individuals.
Thus, the right to punish is not exclusive or unique to the state.
Is it, however, special to the state in some way? Is there an ar-
gument for individuals turning over their punishment power to a
state rather than to some private agency?
In Anarchy, State and Utopia, Nozick answers as follows. Pun-
ishment, to be just, can be administered only once (or up to the
amount deserved); anyone who punishes will thus preempt others
in their exercise of this right. When persons authorize an agent to
act for them, they confer their own entitlements on that agent.
The more clients on whose behalf a protection agency acts, the
fewer others whose exercise of the right to punish has been
preempted or displaced. Therefore, a dominant protection agency
(a state) has a higher degree of entitlement to punish, in the sense
that it preempts the fewest others.
Whatever reasons may exist for placing the power to punish in
the hands of the state, the major point is that it must be transferred;
it does not originate with the state. The power and authority of
the state to imprison, like all its powers and authority, are derived
from the consent of the governed and may with similar consent be
delegated further. Because the authority does not originate with
the state, it does not attach inherently or uniquely to it, and can
be passed along to private agencies.
Anarchists go further. They argue that people may delegate their
rights, including the right to punish offenders, directly to private
agents acting on their behalf. Here, I defend only the weaker
(libertarian but not anarchist) claim: any legitimate governmental
authority may be further delegated, through the government, to
private agents. This assumes the existence of a legitimate and
representative government so that the chain of authority is unbro-
ken from its original source—the people.
54 PRIVATE PRISONS

In short, the state does not own the right to punish. It merely
administers it in trust, on behalf of the people and under the rule
of law. There is no reason why subsidiary trustees cannot be des-
ignated, as long as they, too, are ultimately accountable to the
people and subject to the same provisions of law that direct the
state.

Legitimation of Authority

In any prison, someone will need authority to use force, including

potentially deadly force in emergencies. Questions of legitimacy
in the use of that force, however, cannot be resolved simply by
declaring that for state employees some use of force is legitimate,
while for contracted agents none is.
In a system characteri/ed by rule of law, state and private agen-
cies alike are bound by the law. For actors within either type of
agency, it is the law, not the civil status of the actor, that determines
whether any particular exercise of force is legitimate. The law may
specify that those authorized to use force in certain situations
should be licensed or deputized and adequately trained for this
purpose, but they need not be state employees.
The distinction between a contractual relation and salaried state
employment, in terms of the derivation of authority, may be more
apparent than real. The authority of a corrections officer, for ex-
ample, derives from the fact that he or she is functioning not just
on behalf of the state, but within the scope of the law. Consider
the case of state-employed prison guards who engage in clear-cut,
extreme brutality. We do not say that their acts are authorized or
legitimate, or even that they are acting at that moment as agents
of the state. In fact, we deny it, in spite of uniforms and all the
other trappings of their positions. We say that they have over-
stepped their jurisdiction and behaved in an unauthorized and
unlawful fashion. The state may or may not accept some account-
ability or liability for their acts, but that is a separate issue. The
point here is that the authority or legitimacy of a position does
not automatically transfer to the actions of the incumbents.
There is, in effect, an implicit contract between a state and its
agents that makes the authority of the latter conditional on the
The Propriety of Proprietary Prisons 55

proper performance of their jobs. This conditional authority can

be bestowed on contractual agents of the state just as it is on civil
servants. Where contractually employed agents, such as correc-
tions officers, have identifiable counterparts among state-salaried
agents, there is no reason why their authority should not be re-
garded as equivalent. Thus, the boundaries of authority for con-
tracted state agents should be no less clear than those for state
employees; they could be even clearer if they are spelled out in
the contract.
What about authority inside the prison itself? Would private
prisons lack authority and legitimacy in the eyes of inmates?
Legitimation constitutes one of the most effective methods of
cutting the cost of exercising power in a wide range of social
organization 13 ; prisons are no exception. Since legitimation is gen-
erally granted in exchange for the fair exercise of power, a profit-
seeking prison has a vested interest in being perceived by inmates
as just and impartial in the application of rules. Commercial pris-
ons, unlike the state, cannot indefinitely absorb or pass along to
taxpayers the cost of riots, high insurance rates, extensive litigation
by maltreated prisoners, cancellations of poorly performed or con-
troversial contracts, or even a wave of adverse publicity. Thus,
the self-interest of a for-profit prison company is more likely to
increase, than to decrease, its concern with fairness. Moreover,
the state is more likely to renew a contract with an organization
that has a good record of governance than with a contractor that
generates numerous complaints and appeals from inmates. In
short, economic self-interest can motivate good governance as well
as good management.

Symbols vs. Substance

Many critics of private prisons are extremely concerned about
matters of symbolism. For example, Ira Robbins, Professor of Law
at American University, asks:
When it enters a judgment of conviction and imposes a sentence,
a court exercises its authority, both actually and symbolically. Does
it weaken that authority, however—as well as the integrity of a
system of justice—when an inmate looks at his keeper's uniform
and, instead of encountering an emblem that reads, "Federal Bu-
56 PRIVATE PRISONS

reau of Prisons" or "State Department of Corrections," he faces

one that says "Acme Corrections Company"?'11

I suspect that prisoners care more about practical than philo-

sophical distinctions. They care more about how officers treat them
than about what insignia grace their uniforms. To the extent that
they are treated with fairness and justice, inmates will be more
inclined to legitimate their keepers' authority and to cooperate
with them.
Dilulio, too, shares this concern with the symbols of authority:
"The badge of the arresting policeman, the robes of the judge,
and the state patch on the uniform of the corrections officer are
symbols of the inherently public nature of crime and punish-
ment." 15 It is true that symbols are important to perceptions of
legitimacy. However, perception is affected by substance as well
as by symbol. In fact, substance is ultimately the biggest part of
what creates, sustains, or undermines a symbol. "Pinkerton" and
"Brinks" are private trademarks that symbolize the public virtues
of honesty, reliability, and accountability as well as security, law,
and order. "City Hall," while it obviously refers to a public insti-
tution, is a phrase that, for many people, symbolizes corrupt but
overwhelming power in the service of private ends (as in, "You
can't fight City Hall").
There is no government monopoly on probity and integrity. The
symbols of government and the symbols of business enterprises
are functionally similar in this respect: they each represent the
reputations of the organizations behind them. Since reputations
are ultimately founded on deeds, the degree to which any symbol—
whether of government or of business—is honorific or legitimizing
will be determined, over time, by the quality of what stands behind
the symbol.
If it were symbols, rather than substance, that we were worried
about, we could find plenty of them in a contractual situation. A
contract itself is a powerful symbol of legally enforceable obliga-
tions and responsibilities. For private prisons, a license could be
required, and hung on the gate if that would make people feel
better. The vendor's employees could be ceremoniously deputized
and sworn. They could be given official-looking uniforms and
badges, for that matter.
The Propriety of Proprietary Prisons 57

Michael Walzer is another who argues that administering crim-

inal justice is an enterprise that is largely symbolic and therefore
requires not just technical expertise, but the application of social
values. For this reason, says Wal/er, criminal justice should always
be in the hands of "representatives of the people": "Police and
prison guards are our representatives, whose activities we have
authorized. The policeman's uniform symbolizes his representative
character." 16 Most public employees, however, including police
and corrections officers, are neither politically appointed nor dem-
ocratically elected. They are hired, and cannot be said to "rep-
resent" the public by virtue of their selection. Rather, they
represent by virtue of their function; that's their job. How they
do their job determines how well they represent the public. More-
over, the values that criminal justice workers arc supposed to
administer are codified in law. They are not the personal value
preferences of either public officials or employees, whether
elected, appointed, or hired through civil service procedures. The
important question, then, is whether relevant legal values will be
served more faithfully by public employees or by contractual
agents. This is not a question of principle; it is an empirical
question.
The great concern with symbolism on the part of those who
question the propriety of private prisons indicates that their ar-
gument is not substantive. Essentially, it is theological,17 or rather,
theocratic. Substantively, however, what we are ultimately trying
to symbolize is legal authority, not government employment. Em-
ployment is merely one method of conveying or delegating legal
authority; contract is another.

Contracting and Sovereignty

Are there any theoretical limits as to which functions and powers

(or as to how much of any one of them) government can delegate?
Operating prisons, which is an exercise of executive, and perhaps
quasi-judicial, power, is widely recognized as one of the basic
functions of government. However, it is only one among many
that have been carried out through the use of private agents. David
M. Lawrence, Professor of Public Law and Government at the
58 PRIVATE PRISONS

University of North Carolina, notes that many delegations of ex-

ecutive and judicial powers are well established:
Important judicial and executive powers have been delegated, in
some cases for decades or even centuries, without the validity of
the delegation being questioned. The power of arrest has been
delegated to railway police, to humane society agents, and to bail
bondsmen. The power to seize and sell property has been delegated
to certain lienholders. The power to destroy buildings, without per-
sonaljiability, in order to stop the spread of fire has been delegated
to anyone at the scene of a fire. The power to adjudicate grievances
between employees and employers has been delegated to private
arbitrators. And the authority to determine which law schools' grad-
uates may sit for the bar examination has been delegated to the
American Bar Association. Only the last of these has been chal-
lenged on delegation grounds, and the challenges consistently have
been refuted. 18

Laurin Wollan points out that the perception that criminal justice
is inherently and exclusively a function of the state is strongest
when it is least closely examined. Moreover, it makes a big dif-
ference whether the question is asked broadly ("who should be
responsible for the welfare of prisoners?") or narrowly ("can a
private company provide prisoners with better food at lower
cost?").19
When criminal justice is broken down into specific activities,
or into functions and subfunctions, it is no longer so clear that
any of these tasks must be the exclusive province of the state.
Wollan presents a typology in which criminal justice is divided
into 22 functions across 6 categories, and he gives examples of
privatization for each of these functions. In many cases, private
performance of discrete functions has been going on for some
time with little or no controversy. Specific aspects of criminal
justice that already have experienced varying degrees of privati-
zation include:
Community security and prevention
Initial detection and accusation
Investigation and evaluation of evidence
Victim services
The Propriety of Proprietary Prisons 59

Bail services
Bounty hunting
Legal aid and representation
Prosecution
Presentence investigation and sentencing recommendations
Transport of prisoners
Incarceration
Prison services (food, medical, education, and so forth)
Reintegration (i.e., halfway houses)
Community corrections programs
Alternative sentencing supervision
Probation services
Other functions widely regarded as "governmental" have also
been privately delegated or contracted. National defense is clearly
an essential governmental function. Yet not only does private en-
terprise produce the full range of materiel for the armed services,
it participates directly in our defense. The Distant Early Warning
System, which warns of attack by missiles or aircraft over the
Arctic, is manned and operated by a private contractor.20 Police
protection, fire protection, even the entire management and
administration of some cities have been provided under contract.
Private courts adjudicate many civil cases. Private prosecutors have
operated to some extent in California and to a considerable extent
in England. Even capital punishment is sometimes administered
by private contractors in the United States today. It is hard to find
any specific governmental function or power, the administration
of which has not been delegated at least in some part at some time
to private agents.
If there is no qualitative limit to delegation—no type of function
or power that can never be delegated in any degree—is there
perhaps some quantitative limit? The question here is not what
function can be delegated, but how much of it.
The concept of sovereignty places a theoretical limit on the
delegation of state power. While there is no limit to the number
or type of functions that the state may delegate to private parties,
there is a limit to the amount of power that can be ceded. The
60 PRIVATE PRISONS

power to coerce is part of what defines the state. It can give up

any amount of that power, but if it gives up too much, it ceases
to be sovereign; if it gives up all of it, it ceases to be a state. The
"far end" of that limit is the point at which the state loses actual
sovereignty, at least over a particular function; at the "near end,"
the state begins to lose effective control.
Correctional law expert William Collins, drawing from a leading
treatise on municipal corporation law, concludes that delegation
becomes "excessive" when the government gives away so much
power as to compromise its ability to act in the public interest in
a certain area.21 In operational terms, this would mean loss of
control over basic policy or ultimate supervisory authority, rather
than over ministerial matters. "The more the contract makes the
private provider simply the administrative extension of the. . .
government, and leaves ultimate authority in the hands of govern-
ment officials, the more likely the contract will satisfy judicial
scrutiny." 22
With respect to the operation of prisons and jails, Collins be-
lieves that it would be excessive delegation to relinquish control
over admission or release (including granting or denial of good
time). He is less certain about questions of classification (i.e.,
custody level), which affect the conditions, but not the fact, of
confinement. Note, however, that to pass the sovereignty test it
is not necessary that private prison officials be excluded entirely
from decisions and actions relating to discipline or to release
dates; only that they not be given final control in those areas.
So long as the government retains final authority and the power
of review over disciplinary actions or good time—decisions that
affect the liberty of prisoners—delegation of responsibility for
these functions would not be "excessive" because sovereignty
will not have been lost.
In addition to preventing erosion of its sovereignty, there is
another, and much more important, reason for the state to retain
final (not exclusive or total) authority over decisions affecting pris-
oners' liberty. In the political philosophy of liberal, or libertarian,
constitutional government, the rights of individuals are more im-
portant even than the sovereignty of the state. From this perspec-
tive, the strongest argument for placing in private hands something
less than final authority over prisoners is not that this will preserve
The Propriety of Proprietary Prisons 61

state sovereignty, but that such a restriction will help to preserve

due process.

Issues of Due Process

Certain aspects of prison administration have a quasi-judicial char-

acter. Examples would include imposing solitary confinement or
other disciplinary actions, making or contributing to parole deci-
sions, allocating "good time" sanctions that affect the date of re-
lease, and classification procedures that significantly affect the
conditions of confinement. Discretionary decision-making in these
areas, whether done by public or by private prison staff, clearly
requires some elements of due process. Moreover, even where
prison decision-making is administrative rather than judicial, the
coercive environment in which it occurs can still make the question
of due process relevant.
The best approach to the question of delegation is to view it as
just one aspect of the broader issue of due process, within the
general problem of power. The central issue in due process is not
who exercises power or how much power is delegated, but how it
is exercised and what safeguards exist to prevent it from being
abused.

A Due Process Theory of Constitutional Delegation

In his comprehensive review of delegation issues, David Lawrence
has noted that the ability of the federal government to delegate
power is both broad and clear. "Since Carter v. Carter Coal Co.,
decided a half-century ago, the federal courts have consistently
allowed delegations of federal power to private actors. . . . Private
exercise of federally delegated power is no longer a federal con-
stitutional issue. Nor is the private exercise of governmental power
delegated by state or local governments a federal constitutional
issue. . . . " 23
Among court decisions at the state level, Professor Lawrence
finds no clear or consistent doctrine to distinguish constitutional
from unconstitutional delegation. As a step toward producing such
a doctrine, he suggests that the closest approach to a coherent
62 PRIVATE PRISONS

principle or standard against which to judge a potential delegation

of governmental power is the test of due process:
In summary, a due process basis for reviewing private delegations
permits a court to approach and resolve the problem in terms of
the essential danger that such delegations present: that govern-
mental power may be used to further private rather than public
interests. A court can address the danger directly to determine
whether it exists in a particular instance and then test the mecha-
nisms available to protect against the danger. This approach, well
within the traditions of due process, not only permits handling the
basic dangers raised by private delegations, it has the further ad-
vantage of being more likely to force a court to address those con-
cerns directly and to articulate the considerations behind its
decision.24
Judicial constraint on the delegation of power to financially in-
terested private parties is not essentially different from judicial
constraint on the exercise of power by financially interested public
officials.25 Both rest on due process concerns, not on a theory of
delegation that distinguishes private actors as a class from public
actors as a class. An objection to a private delegation cannot be
based solely on the assumption that a private actor will, by defi-
nition, have conflicting interests, financial or otherwise. Nor can
one assume that a comparable public actor will have fewer con-
flicting interests. Due process procedures are designed to prevent
conflicting interests or other biases from interfering in either case.
Due process is grounded on a justifiable lack of trust toward
those who exercise power. Neither business nor government is
inherently deserving of trust; it has to be earned. Moreover, trust
is not enough; power must be limited by procedural safeguards
even when it is in the hands of those whom (at the moment) we
do happen to trust. Differences between government and business,
insofar as they relate to questions of due process, are not just
exaggerated by opponents of privatization, but distorted so as to
portray business as inherently less trustworthy than government.
In my view, government and business are equally untrustworthy,
but given comparable procedural safeguards there is no aspect of
the administration of a prison that can be entrusted to government
employees but not to private contractors. By comparable proce-
dural safeguards, I do not mean that the specific mechanisms of
The Propriety of Proprietary Prisons 63

accountability must be exactly the same for government and private

operations, only that the degree and standards of accountability
should be comparable between the two. There is nothing intrinsic
to the natures of business and government that requires higher
standards of accountability for one than for the other.

Contracting Contributes to Due Process

Contracting can make positive contributions to due process. One
of the strengths of contracting is that it forces us to make visible
and to treat as problematic some important issues of authority and
due process that we might otherwise ignore or take for granted.
Due process requires preset rules and rigorous adherence to them.
It is universalistic, not individualistic: discretion, individualization,
and "creativity" in punishment are detrimental to due process.
Contractual arrangements offer an excellent means of limiting and
controlling discretion, of clarifying rules, and of enforcing adher-
ence to procedures.
In a recent note in the Yale Law Journal, David Wecht argues
that private prisons may increase prisoners' due process protections
by forcing courts to move away from their historical pattern of
deference to prison administrators.26 While courts actually began
to abandon their "hands-off" approach to prison administration
long before the recent emergence of contracted prisons, Wecht's
note argues that court supervision will now be accelerated as a
result of the longstanding suspicion of courts toward delegation of
power to private, for-profit entities. In order to ensure that con-
flicting private interests do not interfere with either prisoners'
rights or the public interest, courts will insist on strong procedural
safeguards. These might include strict legislative and contractual
standards; training and certification requirements for prison staff;
independent, state review of policies and of discretionary or ad-
judicatory decision-making; and greater liability to lawsuits.
Unfortunately, Wecht dismisses too lightly a warning that courts,
on equal-protection grounds, must apply the same standards of
due process to government as to private prisons.27 Instead, he
argues that private prisons should be held to higher standards and
scrutinized more closely. Public actors, in Wecht's view, presum-
ably are not greatly influenced by private incentives that might
64 PRIVATE PRISONS

conflict with public interests. But government employees no less

than others have self-interests that can conflict with the rights of
prisoners. It will not do to say or imply that public prisons are run
by civil servants who have no profit motives and therefore can be
trusted not to compromise prisoners' rights, health, or safety.

Equal Protection of Due Process

Being suspicious of authority in the hands of commercial prison
managers is an example of having the right attitude for the wrong
reasons. It is not because these managers pursue profit that we
should be vigilant, but because they wield power. A constructive
response to this suspicion would be to require as part of a contract
that commercial prisons codify the rules that they will enforce,
specify the criteria and procedures by which they will make dis-
ciplinary decisions, and submit to review by a supervisory state
agency. In short, the requirements of due process should be built
into the conditions of the contract. But this is no different from
the attitude we should have toward the state itself, and toward its
employees.
Since due process is an important problem for all prisons, we
should not design solutions that apply only to private prisons. To
do so distracts attention from the problem in public institutions.
Our focus should be on the procedures that will best protect the
due process rights of inmates regardless of whether they are applied
by government employees or by contracted agents. The procedures
that will do this best will probably be the same or similar in either
case. It should not be assumed a priori that one system or the
other requires more stringent safeguards.
It is also no solution to propose, as some have, that only those
decisions having implications for due process must be left in gov-
ernment hands. 28 The whole point of having procedures is to reduce
our reliance on being in "the right hands." And the whole point
of constitutional guarantees of due process is that decisions af-
fecting life, liberty, or property cannot simply be entrusted to
agents of the government. Whether prisons are run directly by the
"iron fist" or respond also to the "invisible hand" is less important
than a rule that, either way, those hands should be bound by the
same requirements of due process.
The Propriety of Proprietary Prisons 65

Evaluating, sanctioning, and controlling inmate behavior are

integral parts of every aspect of a prison program; they cannot be
handled by a separate and distant staff. Trying to contract for the
overall management of a prison by a private company, while re-
stricting exclusively to government all decisions and actions that
require due process protections, would be futile. Worse yet, for-
mally defining "administration" as the business of the private com-
pany and "rights protection" as the business of the state would
discourage contractors from maintaining an attitude of full re-
sponsibility. While protection of due process is ultimately guar-
anteed by the state, it should be made a responsibility of
contractors as well. Each contract should establish a system of
supervision whereby the state can monitor the discretionary de-
cisions of the contractor, and whereby inmates can appeal what
they view as unfair treatment.
In a comprehensive review of the delegation doctrine as it may
affect private prisons, Ira Robbins observes: "State courts gen-
erally invalidate statutes and administrative regulations that del-
egate adjudicative power to private parties when there is no
provision for judicial review of the private adjudications. When
there is provision for such review, however, the delegation is up-
held." 29 Thus, courts probably will accept arrangements in which
classification decisions, disciplinary sanctions, good time deter-
mination, and other quasi-judicial decisions are first made by pri-
vate prison officials and then reviewed by or made subject to appeal
before government authorities and courts.30
One essential element of due process is the provision of mech-
anisms for independent review and appeal. It is important that the
reviewing agent be disinterested, or at least not influenced by the
same interests as the agent whose initial decisions are being re-
viewed. A warden who reviews the judgments and decisions of his
officers, has some independence, but not a great deal. A Disci-
plinary Hearing Officer from a central office is somewhat more
independent, but still remains part of a common administrative
structure. In a contract situation, initial decisions made by the
contractor may be reviewable and reversible by a government mon-
itor as well as subject to challenge in court. In this way, contracting
adds another layer of independent review, which enhances due
process.
DO PRIVATE PRISONS

Recognizing the importance of independent review to the pro-

tection of due process, some corrections systems provide for an
ombudsman to hear and act on inmate grievances. Ideal charac-
teristics for an ombudsman are independence, impartiality, ex-
pertise in government, universal accessibility, and power only to
recommend and to publicize.31 In Connecticut, a private research
institute provides the ombudsman for the state's correctional sys-
tem. The Connecticut ombudsman's function is to serve as an
informal check against the power of the state.
If it is acceptable for Connecticut to have initial decisions made
by state employees, subject to informal but influential review by
a private agency, it should be at least as acceptable for a contracted
prison to have initial decisions made by people who are not on the
government payroll, subject to formal review by the state and in
accordance with rules and procedures that are also subject to state
review. It should not be necessary to require that the whole pro-
cess, including initial decisions, be left in the hands of state em-
ployees. Indeed, it would seem that impartiality is increased when
a government agency reviews decisions that have been made by a
private agent rather than by another government agency or, worse,
by a subdivision of the agency itself.
If due process is satisfied in a government prison by the existence
of channels of independent external review of decisions, then due
process will also be satisfied if initial decisions by the staff of a
private prison are subject to independent external review. In nei-
ther case does due process require that the prison's staff be ex-
cluded from the decision-making process.

Inmate Discipline and Good Time Decisions

Critics of private prisons fear that due process, particularly as it
affects discipline and allocation of good time, will be trampled in
the pursuit of profit. Most of all, they fear that private wardens
will try to retain custody of their charges as long as possible to
maximize their per diem revenues. The Legal Director of the In-
diana Civil Liberties Union predicts:
Private prison operators would not only want to reduce costs, but
to enhance revenues by maintaining their facilities at capacity and
The Propriety of Proprietary Prisons 67

by creating new demands for their services. Inevitably, private

prison officials would have a role, even the key role, in making
classification decisions, parole recommendations, awarding good
time credits and meting out disciplinary sanctions. It will be only
too easy for them to abuse these powers in order to increase the
length of incarceration and their own income.32
Three observations are needed to put this fear in its proper
perspective. First, a private penitentiary has a financial incentive
to hang on to prisoners only if nonreplacement is assumed. Under
conditions existing today and projected well into the future, that
is not a rational supposition. As regards good time policies and
decisions, overcrowding actually has an unbiasing effect on a pri-
vate prison, but a biasing effect on a government prison. In a
private prison, there is no economic interest either for or against
denial of good time if systemwide overcrowding guarantees re-
placement of inmates. In contrast, overcrowding gives the state
system a material bias toward early release through liberal use of
good time, in order to avoid the possibility of expensive court-
ordered reforms or massive fines. Second, the argument that pri-
vate prisons will have an incentive toward unfair and improper
denial of good time assumes that such decisions will confer only
benefits and carry no costs; no commercial company with a com-
petent legal staff would make so unrealistic an assumption. Third,
it should be noted that only a small proportion of disciplinary cases
(often less than 10 percent) result in revocation of a prisoner's
good time credits; the majority involve other sanctions.33 In those
cases where significant prisoners' rights are at issue, due process
can be protected by provisions for appeal to higher authority.
In his Model Contract and Model Statute proposals to the Amer-
ican Bar Association, Professor Robbins argues that officers in
private prisons should have only the most limited possible role in
the disciplinary process: the role of a complaining witness before
a state judicial officer.34 Robbins believes that all discretionary
decisions potentially affecting a liberty interest must be reserved,
both by statute and by contract, to state officials. Such decisions
would include classification; transfers; formulating rules of con-
duct; any disciplinary action; allocation of sentence credits; parole
recommendations; calculating sentence credits and dates of release
68 PRIVATE PRISONS

or parole eligibility; assigning, supervising, or rewarding inmate

work; and determining eligibility for furlough and work release.'5
The basic premise behind Robbins' stance is that private em-
ployees are inherently biased in ways that government employees
are not. In brief, no private corrections officer should be involved
in discretionary decision-making that affects an inmate's liberty,
because the officer's employer has a material interest in the out-
come of those decisions. Commenting on discipline, Robbins de-
scribes this interest as both monetary and institutional. The
monetary interest is an alleged "pecuniary interest in increasing—
or at least, not decreasing—each inmate's stay," while the insti-
tutional interest is an alleged "institutional bias in favor of disci-
plining prisoners." 36 To support the latter charge, Robbins simply
asserts that: "A decision to deny certain privileges or services, for
example, would reduce the operating costs of the company and
would promote its administrative convenience. . . . Moreover, any
exercise of nonreviewable discretion by the company would be
cheaper than complying with due process constraints." 37
Aside from the fact that no one has ever proposed that private
prison contractors should be allowed to exercise nonreviewable
discretion, Robbins' assertions can be countered along several
lines. First, the daily actions and decisions of employees every-
where are guided largely by personal motives, interests, and in-
centives that are often quite different from, and may even conflict
with, those of the employing organization. This is equally true of
both private and public employees. Employees are supposed to
follow official organizational policies and rules, which can be ex-
amined for potential biases, but departures from official policy are
more likely to be influenced by personal incentives, such as con-
venience, than by some general consideration of the interests of
the company (or state).
Second, private organizations are not unique in their potential
for conflicts of interest. Public organizations also have self-interests
that may bias and distort the decisions of workers and adminis-
trators. We cannot avoid such biases by allowing only public em-
ployees to make important decisions. As mentioned earlier,
overcrowded government facilities operating under limited budgets
are under great pressure to release offenders early regardless of
the merits of individual cases. This is a type of pecuniary bias, and
The Propriety of Proprietary Prisons 69

it is clear that government officials such as sheriffs, wardens, and

parole board members respond to it. In addition, nonpecuniary as
well as pecuniary biases are found in all organizations, whether
public or private. For example, saving face and maintaining per-
sonal authority can influence both private and public corrections
officers in disciplinary matters.
Third, in neither type of organization do potentially biasing in-
centives operate in only one direction. In the case of private pris-
ons, the alleged pecuniary incentive toward delaying release in
order to maximize revenue is counteracted by a second pecuniary
bias toward early release in order to avoid prisoner litigation. A
third pecuniary "bias" favors neither early nor late release in order
to maintain a reputation for fairness and integrity, which helps to
secure and renew contracts and cuts the costs of conflict with in-
mates. Thus, even a private warden who fit Robbins' stereotypic
caricature of someone who cares only about the bottom line would
not inherently be biased in any one consistent direction. Will the
cost-calculating warden revoke a prisoner's good time to gain a
little "extra" per diem revenue? Or will he bribe inmates with
liberal grants of good time credit in order to buy their cooperation
and to avoid the expense of the extra paperwork required by dis-
ciplinary proceedings? Or will he decide that it is least costly in
the long run to govern firmly but fairly and consistently?
The most fundamental fallacy in Robbins' argument is that it
rests on the false assumption that due process requires purity of
heart on the part of those who make decisions. If due process
awaits the appearance of mortals who are certain to be impartial
and objective, then due process will wait forever. The whole point
of due process is to substitute reliance on procedures for reliance
on motives or character to guarantee fair treatment. We must insist
on procedures that will enhance due process regardless of whether
they are administered by public or by private employees, each with
their varying mix of occupational, personal, and institutional
biases.
The single most important element of such procedures is the
provision of a mechanism for independent review so that deci-
sions can be evaluated by someone whose biases will probably
be different from those of the original decision maker. An inde-
pendent layer of review is preferable to concentrating decision-
70 PRIVATE PRISONS

making power among persons whom we presume to be impartial.

Therefore, rather than requiring government officials to make
all discretionary decisions in a contracted prison, the initial
decision-making authority should be ceded to the private officials
responsible for running the prison. These decisions, in turn, should
be subject to independent review by the government.
Take the case of the Hamilton County prison run by Corrections
Corporation of America. There, the County Superintendent of
Corrections serves as an onsite monitor and also makes all de-
cisions regarding gain time, discipline, and work assignments.
Professor Robbins, presumably, would be satisfied with this ar-
rangement because these decisions are left entirely in government
hands. I am not satisfied, because what I see is a situation in which
the Superintendent acts as Disciplinary Hearing Officer, Warden,
and Secretary of Corrections all rolled into one. He is the first,
last, and only discretionary decision-making official. There is no
internal or external independent review, unless a case goes to
court. Due process, and therefore justice, would be enhanced in
this prison if initial decisions were made by CCA employees, sub-
ject to review by the Superintendent. Professor Robbins' Model
Contract and Statute would preclude this potential enhancement
of justice through privatization.

Inmate Classification
Classification of inmates according to the degree of security or
supervision they require, which determines where and with whom
a prisoner will be housed within a facility, can have a major effect
on the conditions of confinement. Classification decisions must
be made not only at the time of entry, but throughout an inmate's
sentence as behavior or circumstances change. Logically, the closer
the decision makers are to the actual operations of the institution,
the more information they will have on which to base their clas-
sifications. This suggests that classification decisions are best made
by those who actually run an institution.
Critics worry that private prisons will make classification deci-
sions in accordance with their own agendas, not in the best interests
of prisoners or of the public. It is not at all clear, however, that
these separate concerns are truly in conflict.
The Propriety of Proprietary Prisons 71

From any standpoint, classification decisions should be based on

accurate predictions. Accuracy is not something that serves one
set of interests better than another. The causes of justice, order,
economy, and inmate welfare are all served best when prisoners
are subjected to the minimum degree of control sufficient to ensure
security. In the public system, the highest status, largest budgets,
and biggest staffs attach to maximum security institutions. There
is thus an incentive to overclassify inmates (i.e., a bias toward
higher security designations). In addition, most states overclassify
inmates because higher security space is often all that is available.38
The economic cost of overclassification is not borne directly by
those who do the classifying, but passed along to taxpayers. More-
over, while oversecurity increases economic costs, it decreases po-
litical costs by minimizing the risk of escapes or loss of control.
Public officials are more vulnerable and sensitive to political costs
than to economic costs and thus tend to overclassify.
In a private prison, on the other hand, there is a financial in-
centive to treat cases at the lowest possible level of security, so
that more prisoners can be held under lighter supervision or in
facilities less costly to construct and administer. Nonetheless, there
are restraints against carrying the downward classification too far.
Escapes, violence, lawsuits, and other consequences of inadequate
security carry both direct financial costs and indirect public rela-
tions costs in the form of lost contracts due to a poor reputation.
Thus, while private vendors have some incentives to engage in
underclassification, they also have countervailing disincentives.
It is not necessary here to decide whether classification biases
are more likely to occur in private or in governmental prisons. The
major point is that they can occur in either type. The implication
of this for due process is the same as in the case of discipline.
Review procedures and other checks and balances are neither more
nor less needed in private prisons than they are in prisons run by
government employees.

The Profit Motive vs. Other Motives

Before we look at motives, we should note one point of logic at

the outset. Strictly speaking, the motivation of those who apply a
72 PRIVATE PRISONS

punishment is not relevant either to the justice or to the effec-

tiveness of the punishment. Of course, for punishment to be a
moral enterprise, it must be meted out for the right reasons. This,
however, is a stricture that applies more to those who determine
and decree the punishment than to those who carry it out—to
legislative and judicial more than to executive agents. The im-
mediate agents of punishment may be humans with motives vir-
tuous or venal, or robots with no motives at all; that does not
affect the requirements of justice. Still, the matter of motives—or
rather, one particular motive—seems to be of such great impor-
tance to so many opponents of proprietary prisons that it must be
dealt with. These critics believe that "criminal justice and profits
don't mix." The ACLU especially has complained repeatedly that
"the profit motive is incompatible with doing justice."
If it is legitimate to examine the motives of interested parties,
then to be consistent we ought to examine the motives of all parties,
including state agencies, public employee unions, prison reform
groups, and "public interest" groups.39 All these parties, like pri-
vate vendors, have motives that reflect self-interest as well as al-
truism, and agendas that are hidden as well as overt. For example,
the ACLU's National Prison Project, by making the point that
more efficient prisons will mean more imprisonment, may be as
much opposed to prisons per se as to running them like a business.
The ACLU does not object to the profits that are made from the
private administration of community correctional programs that
serve as alternatives to prison.
A consistent objection to the existence of vested interests in
punishment would have to scrutinize the public sector as well as
the private. Is it wrong for state employees to have a financial
stake in the existence of a prison system? Is it wrong for their
unions to "profit" by extracting compulsory dues from those em-
ployees? Is it wrong for a state prison bureaucracy to seek growth
(more personnel, bigger budgets, new investment in human and
physical capital) through seizing the profits of others (taxation)
rather than through reinvestment of its own profits? Are the sanc-
tions of the state diminished or tainted when they are administered
by public employees organized to maximize their personal bene-
fits? If not, would those sanctions be compromised if they were
administered by professionals who make an honest profit? Ad-
The Propriety of Proprietary Prisons 73

mittedly, I have posed these questions in prejudicial language, but

I have done so to make a point. The notion that any activity
becomes suspect if it is carried out for profit, as compared to salary
and other benefits, is simply an expression of prejudice.
Of various possible motivations for serving as an agent of pun-
ishment, the profit motive is among the most benign. Compare,
for example, some alternative motives: enjoyment of power, ar-
rogance, self-righteousness, zealotry, machismo, malice, cruelty,
sadism, vengefulness, adventurism, aggression, hostility, hatred,
anger, resentment, prejudice, or displacement. No one has pro-
posed that all criminal sanctions be administered by unpaid vol-
unteers motivated by pure love of justice. If someone does propose
it, watch out! Great injustices are often done in the name of noble-
sounding values. The history of corrections, from the penitentiary
to the juvenile court, is a chronicle of good intentions that pro-
duced bad results.40 The clear lesson from this history, drawn by
criminologists of all political persuasions, is that criminal justice
policies and practices must be judged by their consequences, not
by their motives. In particular, declarations of "public service"
should not be taken at face value. 41 Instead, public service should
be judged as an outcome, regardless of whether the motivating
force behind it is probity, power, or profit.
Replacing "public servants" with "profit seekers" in the man-
agement of prisons will not trade those whose motives are noble
for those whose motives are base. Rather, it will replace actors
whose motives we suspect too little with actors whose motives we
might suspect too much. Still, that is a step in the right direction
when we consider the high cost of relying on good intentions in
the past.

Constraining (Everyone's) Self-interest

But won't a commercial institution be "driven by profit" and, as

a result, be tempted to put its own welfare ahead of the welfare
of inmates, the needs of the state, or the interests of justice? This
concern is legitimate, but it is at least partially misplaced when it
is portrayed as a problem unique to commercial enterprises. Ac-
tually, the problem exists for public as well as private, for nonprofit
74 PRIVATE PRISONS

as well as profit-making organizations. If justice and the profit

motive really were incompatible, then justice would be doomed,
because in one form or another the profit motive is universal. Like
the rest of society, politicians, government bureaucrats, and other
agents of the state are motivated by self-interest. The field of public
choice, a hybrid of economics and political science, is founded on
this insight, and one of its founders, James Buchanan, recently
received a Nobel Prize for his extensive research and theory in
this area.
Consider the case of prosecutors. One does not have to be cyn-
ical to understand that prosecutors, like other people, are moti-
vated by self-interest and not purely by love of justice. Job security,
prestige, and power are among the incentives of prosecutors. Many
have political ambitions. They are rewarded along these lines ac-
cording to their conviction rate. Although justice is served best
through careful review of cases and the prosecution of lawbreakers
on the most deserved charge, conviction rates are maximized by
generous plea bargaining and by pursuing the most and the easiest
cases. It is this structure of incentives, rather than caseload, that
explains the prevalence of plea bargaining.42
Suppose now that prosecution were privately contracted. The
profit motive of a commercial prosecutor would be at least as
defensible, morally, as the mixed motives of a public prosecutor.
More importantly, however, the profit motive has the advantage
of being more controllable. We cannot easily structure and ma-
nipulate the complex incentives of public prosecutors. In contrast,
if we wish to encourage a commercial prosecutor to concentrate
on making the charge fit the crime, rather than on the quantity or
rate of convictions, we simply structure this incentive into the fee.
The fee can be set by an agency authorized to reflect society's
concerns (and accountable for doing so) and can vary according
to charge of conviction. To discourage wrongful prosecution or
improper charges, fines could be imposed on commercial prose-
cutors whose convictions are overturned on review or appeal.
This proposal is sketchy at best, but it illustrates the point that
the profit motive is subject to creative public management in a
simpler fashion and to a greater degree than are the motives and
incentives operating in public bureaucracies.43
Among the most universal of motives is one that could be called
the "convenience motive"—the desire to behave in ways that max-
The Propriety of Proprietary Prisons 75

imize one's own convenience. Compared to the profit motive, the

convenience motive has few societal benefits; it is much more
asocial and self-interested. Not only individuals, but institutions,
too—from hospitals and universities to courts and prisons44—tend
to operate according to their own convenience unless they are
motivated to do otherwise. For public or nonprofit institutions,
this motivation takes the form of political pressure. For private,
profit-making institutions, the motivation can take economic as
well as political forms, because market mechanisms of discipline
and supervision are added to those of the state apparatus. Shaped
by market mechanisms, the profit motive can act as a powerful
constraint on the convenience motive. Businesses must often sac-
rifice their own convenience if that will increase their profit. Busi-
nessmen understand that to sustain any competitive enterprise it
is generally necessary to satisfy some needs other than one's own.
Competition does not just contain costs; it advances other goals
as well. When it is possible for a commercial company to take
business away from a competitor (including the state) by showing
that it can do a better job, then that company becomes a self-
motivated watchdog over other companies (and over the state).
Such a company will have an interest in critically evaluating the
quality of its competitors' services and an interest in improving
its own.
In the case of prisons, the existence of competition, even po-
tential competition, will make the public less tolerant of facilities
that are crowded, costly, dirty, dangerous, inhumane, ineffective,
and prone to riots and lawsuits. Indeed, the fact that these con-
ditions have existed for so long in monopolistic state prisons is a
big part of what makes private prisons seem attractive. The pos-
sibility of an alternative will make the public, the courts, and the
government justifiably more demanding in their expectations.
Without competition, the state has had a monopoly over both
service and supervision, over both doing justice and seeing that it
is done properly. With competition, there will be a proliferation
of agencies having a direct stake in both, without detracting from
the state's role as the final arbiter of justice.
For these reasons, among others, the profit motive is not nec-
essarily in conflict with the pursuit of justice; it can, in fact, be
conducive to it.
5
Issues of Cost and Efficiency

Among claims made for the superiority of proprietary prisons, the

most frequent and most salient—but not necessarily the strong-
est—is that they will be less expensive, or at least more efficient.
So far, very little systematic empirical evidence exists either for or
against this premise. As a result, both advocates and critics of
proprietary prisons base their contentions on theoretical (or often
ideological) grounds and on analogies to privatization in other
areas. Those arguments will be reviewed here, along with all the
empirical evidence available at this time.
The relative cost and efficiency merits of public and proprietary
prisons can be examined within three areas: finance, construction,
and operation. Greatest attention here will be given to the area
of operation.

Private Financing and Lease-Purchase Arrangements

State and local governments are faced with the need to finance $5
to $10 billion in new prison construction. The traditional method
of financing is through general obligation bonds. These bonds are
secure, and government can generally borrow at slightly lower
interest rates than private enterprise. This, however, is assuming
that it can borrow at all. Some jurisdictions have borrowed to the
point that they have very low credit ratings. Others have legal debt
ceilings. And increasing numbers of voters are reluctant to au-
thorize further borrowing.

76
Issues of Cost and Efficiency 77

Lease-purchase arrangements and private financing offer gov-

ernment the opportunity to pay for new prison space out of an-
nually appropriated operating budgets. While lease-purchase
arrangements are long-term in outlook, they include nonappro-
priation clauses allowing the government to terminate the lease.
The main purpose of these clauses is to avoid classifying the ex-
penditures as debt. But governments are not likely to terminate
their leases, especially as their equity in the property increases.
Private financing, even when it is through a nonprofit corpora-
tion or authority that can sell tax-exempt revenue bonds, generally
carries a slightly higher interest rate. However, it avoids the cost
of a referendum, has lower transaction costs, and is much faster
than the issuance of general obligation bonds. This greater speed
can save many months worth of inflation in the cost of the eventual
construction. California, for example, has reported needing about
$1.3 billion for new prison construction. Assuming a 5 percent
inflation rate, an 8-month to 10-month delay associated with a
general obligation bond would result in a $43 to $54 million in-
crease in total construction costs.'
Private financing and lease-purchase deals are options for gov-
ernment to consider, but they are not by any means a simple
solution to the high cost of financing new jails and prisons. The
issues involved are very complex and the relative advantages of
different financing schemes depend on a jurisdiction's credit rating
and can shift rapidly with changes in interest rates and tax laws.2
Moreover, governments will want to consider other factors besides
cost. For example, a straight lease—the most expensive option—
might be preferred because it offers flexibility in meeting a tem-
porary need.
Recent tax reforms have made private financing less attractive
to investors by eliminating tax write-offs for accelerated deprecia-
tion. However, depreciation is still allowed if the facility is both
owned and operated privately. Thus, there may be a trend toward
this form and away from private financing by itself.3

The "Bypassing" of Voters

Avoiding debt limitations and capital-budget restrictions—the
reputed advantages of lease-purchasing—raise several political
78 PRIVATE PRISONS

questions. Where voters have turned down bonds or passed prop-

ositions restricting debt, private financing may be seen as no more
than a scheme to spend taxpayer's money without direct voter
approval. Similarly, an off-the-balance-sheet technique such as
lease-purchasing, may enable a municipality to acquire a new jail
or prison without the appearance of long-term obligation or capital
expenditures. Under this arrangement, some people claim that
"it's the brokers, architects, builders and banks—not the taxpay-
ers—who will make out like bandits." 4
On the issue of imprisonment, many government officials view
the public as inconsistent. In opinion polls and letters to the editor,
citizens demand stiffer sentences for lawbreakers, yet bond ref-
erenda to pay for new jails often fail. Rather than inconsistency,
however, this may reflect dissatisfaction with current government
performance in constructing and operating prisons. The public may
sense that it does not get full value for its correctional dollar. It
wants more and better imprisonment but does not want to pay a
lot more for it. This is not necessarily irrational. Indeed, it is a
reasonable reaction to dissatisfaction under conditions of monop-
oly. Only where competition and comparison demonstrate the true
price of a product does it become irrational to demand more with-
out being willing to pay what it takes.
Even if a majority of the public did not want more prisons, that
would not settle the question. We live in a representative republic,
not a direct democracy, and under constitutional rule of law, not
direct majority rule. Where the crime rate is high and our laws
call for imprisonment of serious offenders, we must provide space
for them. Where courts declare overcrowding and other conditions
to be unconstitutional, we must expand jail space and make the
conditions acceptable. Thus, a fiscally strained government in a
high-crime area or facing court orders may be forced to "find ways
around" a prison-expansion referendum. Such a government is not
necessarily behaving in a high-handed fashion; it may simply be
living up to its responsibilities.
In any case, while a referendum may be the last word on one
particular type of financing, it does not preclude other forms, nor
is it the last word on the issue of construction. Indeed, it is not
always clear just what it means when voters defeat referenda on
the issuance of new bonds to finance prison construction. The
Issues of Cost and Efficiency 79

results may simply reflect public apathy and confusion. Relatively

few voters bother to cast ballots on these issues, often allowing a
well organized and financed opposition campaign to prevail.
James Jacobs studied one prison bond referendum that was nar-
rowly defeated in New York State.5 He found strong and organized
opposition among a small number of groups, combined with gen-
eral public apathy and low voter participation. According to Ja-
cobs, voting in favor of the bond, when examined by area,
correlated positively with the crime rate; thus the bond was sup-
ported strongly in politically liberal but high crime New York City,
and rejected strongly in politically conservative but low crime up-
state New York. To some extent, the split probably reflected the
general pattern of upstaters not wanting to pay for New York City's
problems, but Jacobs was not wholly satisfied with that explana-
tion.6 He concluded that the vote was the result of confusion and
apathy, not conscious rejection of prison expansion as a policy.
Opinion polls consistently show strong public support for more
spending on criminal justice. An ABC News poll in 1982 found
that 69 percent of a national sample would approve of building
more prisons and 62 percent said they still would approve even if
this meant their taxes would go up as a result. In a Gallup poll
the same year, these figures were 57 percent and 49 percent, re-
spectively.7 Thus, a failed prison construction referendum may
reflect objections that are specific to the issuance of bonds and the
expansion of public debt, not to the expansion of the prison system.
In any case, with lease-purchasing, voters still have the option of
objecting, if they wish to, through their elected representatives.

Construction

Private companies have demonstrated repeatedly that they can

locate, finance, design, and construct prisons more rapidly than
the government can. Corrections Corporation of America reports
its construction costs to be about 80 percent of what the govern-
ment pays for construction.8 CCA notes that it can build not only
faster, thereby saving inflation costs, but also at a lower immediate
cost, since construction contractors charge the government more.9
A few examples illustrate these points clearly.
80 PRIVATE PRISONS

In 1975, the Attorney General of Pennsylvania ruled that even

hard-core delinquents must not be incarcerated in facilities with
adult offenders. Faced with the need to relocate all affected ju-
veniles immediately, and lacking suitable facilities, the state turned
to RCA Services Corporation, with which they already had a con-
tract for educational programs for delinquents. In 10 days, RCA
set up the Weaversville Intensive Treatment Unit, a heavy security
facility with 20 beds and 30 staff members.
In the Pennsylvania 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.10 While a spokesman for the Federal Bureau of Pris-
ons states that it takes two or three years to site and build their
prisons11 and other sources report that it takes five years or more
to build secure facilities, 12 some private contractors have been able
to design, finance, and build prisons in six months.13
In Houston, Texas, a privately owned, 68,000 square-foot de-
tention facility for 350 illegal aliens was financed and built in just
seven months, to be operated under contract to the Immigration
and Naturalization Service. T. Don Hutto and G. E. Vick describe
the process:
INS's request for proposal required a response within 30 days from
the date of advertisement, complete with plans of the proposed
facility. Work on programming and site acquisition began imme-
diately. Site criteria were established and feasibility studies con-
ducted in a matter of days. . . . Preliminary drawings and specs were
completed in two weeks and ready for contractor review and pricing.
. . . After the contract was awarded, the project was fast-tracked
with construction drawings completed in the sequence required for
each building system.' 4
In Laredo, a 150-bed detention facility for adult and juvenile
males and females was built and opened by a private company
within eight months of signing a contract with the INS.15
When Wackenhut signed the contract to provide a new detention
facility for the INS in Aurora, Colorado, it paid for the facility up
front. Design and construction, from the signing of the contract
to the opening of the institution, took five months. 16 In another
contract, Wackenhut took just 90 days to perform a $1.4 million
renovation of an eight-story building, hire and train 150 local staff,
Issues of Cost and Efficiency 81

and open what at that time was the largest (619-bed) private de-
tention facility in the country. 17
Time savings such as these are almost impossible to achieve in
the public sector if normal bureaucratic procedures are followed.

Operation

Salaries account for about 80 percent of most prisons' operating

expenses.18 Starting salaries for corrections officers are generally
low, however, so cutting salaries may not be the best method of
reducing labor costs. Although there are occasional stories of lower
wages,19 contractors generally pay salaries comparable to govern-
ment, especially when they take over the operation of an existing
facility.
According to CCA, more effective personnel management can
cut costs without cutting salaries.20 Adequate and appropriate staff-
ing, better working conditions, and more efficient procedures
improve productivity and morale, decrease absenteeism and turn-
over, and reduce expensive reliance on overtime. For example,
police chiefs around Bay County say that after CCA took over
management of the jail there, booking time was cut in half as a
result of streamlined procedures.21
CCA reports that it achieves savings in the key area of security
personnel through efficient scheduling and facility design, and
through strategic use of electronic surveillance systems. These
management and capital investments have enabled CCA to reduce
labor costs to about 60 percent of operating costs.22 A building
design or work schedule that eliminates one post can save over
$100,000 a year in salaries and fringe benefits, since it requires
more than 5 staff positions to fill a post 24 hours a day. 23
Red tape and other restrictions endemic to the civil service sys-
tem impede efficient personnel management in the public sector.
E. S. Savas gives an extreme example from New York City:
Inefficient staffing was legitimized by a state law that called for an
equal number of police officers on duty on each shift, despite the
fact that crime statistics showed few criminals working in the small
hours of the morning. Because of this legislated inefficiency, if more
police were needed for assignment to evening duty, when most street
82 PRIVATE PRISONS

crimes occur, more would also have to be hired and assigned when
there was little work for them to do.24

When the Eckerd Foundation took over operation of the Okee-

chobee School for Boys, it eliminated what an independent study
referred to as "a convoluted personnel bureaucracy." 25 Under the
state personnel system, facilities "cannot add, delete, reclassify,
exceed 10% above the minimum pay range, or carry out other
personnel functions on a local level." State administrators found
the personnel process time consuming and frustrating. In contrast,
under Eckerd, the Superintendent at Okeechobee "has the latitude
to hire, fire, and exceed the minimum salary, as long as he stays
within the budget." Positions can be added or reclassified with a
telephone call to get concurrence from Foundation headquarters.
Dismissals are formalized, with an appeal mechanism, but there
is no union.
The existence of a union of county employees at the Butler
County Prison in Pennsylvania did not prevent Buckingham, a
private management company, from instituting personnel reforms.
All prior full-time employees retained their jobs (as the result of
a court order) and received pay hikes, but management was able
to redefine job responsibilities and work schedules in order to
reduce overtime and eliminate part-time, nonunion positions.26 At
first the union insisted, like the police in New York City, that every
shift had to have an equal number of staff, regardless of need; in
a later agreement, however, differential staffing was instituted. 27
Other new personnel efficiencies covered work assignments, ver-
ification of worker presence, and more systematic provisions for
sick and vacation relief. Turnover fell from three or four employees
annually under the county, to zero during Buckingham's first
year.28

Skepticism Regarding Cost Savings

Critics of privatization challenge the assertion that prisons can be

run more efficiently and less expensively by private companies.
For example, John Hanrahan, a leading opponent of private con-
tracting of public services, asserts: "One basic item that should
Issues of Cost and Efficiency 83

make contracting-out more expensive, all other things being equal,

is that contractors exist to make a profit, while governmental units
have no such motivation." 29 By specifying "all other things being
equal," Hanrahan makes his argument tautological. To escape the
tautology, he would have to show that the profit motive adds more
to expenses than it subtracts and also that it does more to increase
expenses than do comparable incentives characteristic of nonprofit
organizations. In a competitive environment, profit margins must
be kept as low as possible, as part of the process of keeping total
costs down. Among current prison contractors, RCA makes a
profit of 5 percent at Weaversville, while Behavioral Systems
Southwest runs its detention centers with a profit margin of 8
percent.30 Naturally, if "all other things" were equal, these two
contractors would have costs that were excessive by 5 percent and
8 percent, respectively. By the same token, however, other costs
need only be lower by 5 percent or 8 percent for these contractors
to compensate for their profits.
Another critic, John Donahue, argues that the scope for cost
savings in running prisons is very limited. Imprisonment is such a
simple, basic arrangement, he says, that there is little room for
improvement in efficiency. "Prisoners must be sheltered, fed,
cared for when sick, protected from each other, and prevented
from escaping. These do not appear to be the type of tasks that
allow for major innovations in technique."31 John Dilulio also
argues that so much of running a prison is either a fixed cost or
judicially mandated that it is hard to imagine that there is any room
for greater efficiency by the private sector.32
If this argument were correct, there would also be little room
for variation in public prisons, either in performance or in cost.
Yet, as Dilulio himself points out, prisons vary greatly in both
performance and cost, with no simple relationship between the
two.33 Since there is variation, it must be true that there is room
for variation.

Reasons to Expect Cost Savings

Peter Drucker suggests that public service institutions tend to

be both inefficient and ineffective, not simply because they are
84 PRIVATE PRISONS

public rather than private, but because of the way they are fi-
nanced.34 Profit-and-loss incentives differ fundamentally from
budget-driven bureaucratic incentives. Entrepreneurs are com-
petitively motivated to provide maximum satisfaction at mini-
mum cost. In contrast, bureaucrats are rewarded not so much
for efficiency, but in direct proportion to the size and total
budget of their agencies.35
Bureaucratic organizations increase in size and budgetary re-
sources on the basis of promises, intentions, and efforts, not
strictly on results. Because they depend for support on their
ability to appeal to a broad constituency, they must be all
things to all people and alienate no one.36 This compromises
their effectiveness because they cannot concentrate their efforts
successfully.

Economies of Scale
Economies of scale vary greatly by the type of service and by the
size and nature of the area being served. However, as Robert W.
Poole, Jr., points out, "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 [political jurisdiction]." 37 For example,
nearly all cities need jail facilities, but a multi-city contract might
meet 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).
Private contractors should be able to realize significant econo-
mies of scale by contracting across jurisdictions. When a private
prison rents to a secondary contracting jurisdiction space that is
unused by the primary contracting jurisdiction, all parties con-
cerned benefit. The company can keep its unit costs low by running
closer to capacity. The secondary contracting jurisdiction benefits
because it can rent space rather than build it. The primary con-
tracting jurisdiction benefits because the company can charge the
primary jurisdiction a lower fee and share with it the higher fee
charged to the secondary jurisdiction.
Economies of scale could also be achieved by governmental units
Issues of Cost and Efficiency 85

contracting directly with each other, rather than through private

contractors. Interj urisdictional prisons and j ails have been widely
advocated, but cooperation between governments has been hard
to achieve. Private contracting may help overcome some of the
political, fiscal, and administrative obstacles to establishing re-
gional, interjurisdictional facilities.38 Such facilities offer a poten-
tially more efficient means of accommodating low prevalence
cases, exceptional needs cases, and other special categories of
detainees.

Wastefulness in Public and Private Sectors

Waste and extravagance can be found in both the public and
the private sectors, though they tend to take different forms in
each. Business is better known for padded expense accounts,
expensive perquisites like company aircraft, and lavish offices
(though these are found at the upper levels of government bur-
eaucracies also). Private use of company or agency cars is com-
mon in both sectors.
Dereliction of duty, tardiness, absenteeism, and abuse of sick
leave are probably less prevalent in private industry, where man-
agement controls are stronger.
Employee theft occurs in both sectors, but only in very large
corporations (which are occasionally the targets of elaborate em-
bezzlement schemes) is it as common as it is in government.39
Personal use of telephones and mail privileges and other types of
petty pilferage are so common in government that they are hardly
even perceived as theft, except for occasional abuses so flagrant
as to be scandalous.
The form of waste that is most characteristically governmental
(though it would be more accurate to identify it as "budget-based"
rather than "governmental") is the deliberate splurging that occurs
toward the end of each fiscal year. It is universally understood
among bureaucracies that to end the year with a surplus is to invite
a reduction in the next year's budget. Most certainly, it weakens
the case for an increase.
In sum, while various forms of waste can be found in either
sector, prodigality is probably much greater in the public sector.
86 PRIVATE PRISONS

Public vs. Private Retirement Benefits

In 1983, the President's Private Sector Survey of Cost Control (the
Grace Commission) reported the results of its efforts to identify
sources of waste and inefficiency in the federal government. They
found that the largest single source was retirement programs. His-
torically, government had to compensate its employees for lower
salaries by offering them greater security, both before and after
retirement. However, when the Federal Salary Reform Act of 1962
mandated "comparability" of salaries, this trade-off was under-
mined, to the benefit of government workers and the detriment
of taxpayers. Today's federal blue-collar workers are better paid
than their private counterparts and their retirement arrangements
are two to three times as generous.40 Federal employees also take
two-thirds more sick pay than do those in the private sector.41
Assuming that the bulk of correctional workers are blue-collar and
that state and local (like federal) retirement systems and other
benefits are also relatively generous, the findings of the Grace
Commission suggest that there is room for savings in the overhead
and indirect costs of prisons.

Inflation and Contracting

Another potential cost advantage of contracting is in controlling
the effects of inflation. Prison contracts, if they are not renegotiated
annually, typically include a provision constraining the vendor to
limit fee increases according to inflation, as measured by the Con-
sumer Price Index (CPI). In contrast, growth in government costs
usually outpaces the general level of inflation, and government
costs in the corrections area have been rising particularly fast.42
Economist Harry W. Miley, Jr., found that from 1974 to 1984, the
per inmate costs of the South Carolina Department of Correction
increased by an annual average of 11.4 percent while the CPI
increased an average of 7.7 percent a year. 43

Purchasing
Government agencies require bureaucratic controls (red tape) to
regulate their purchasing procedures because they lack the more
Issues of Cost and Efficiency 87

automatic restraints of a profit-oriented firm. With their greater

latitude, private prisons can shop more effectively and obtain bet-
ter prices. Because they can purchase more quickly, they can main-
tain lower inventories.
An American Correctional Association report on the private
operation of the Okeechobee School for Boys illustrates the dif-
ference between public and private purchasing arrangements:
Purchasing practices constitute an additional bureaucracy at the
State level. In addition to Department of Health and Rehabilitative
Services policies and procedures, there is a Department of General
Services—Division of Purchasing—Bureau of Institutions which is
responsible for training school purchasing. This Bureau has a sched-
ule for purchasing commodities needed in an institution. Conse-
quently, the facility's Business Manager and Purchasing Agent have
to plan ahead for at least six months and forecast the quantity needed
for every item in those categories. Once a purchase order is issued
by the Department of General Services it cannot be canceled by an
institution for any reason without going back to General Services
with complete written justification.
The Foundation, in contrast and with minimum interference,
gave complete latitude to Okeechobee to purchase needed com-
modities. There is little delay in receiving goods, and no six-month
lead time. 44
In Hamilton County, Tennessee, the county had to go through
a formal bidding process every week for prison kitchen supplies.
After a private operator took over all prison purchasing, the county
was able to eliminate two buyer positions.45

Property Management
The Grace Commission's comparison of public versus private
property management is also instructive and relevant to the is-
sue of prisons. The Government Services Administration em-
ploys 5,000 people to manage 8,600 buildings worth $9 billion,
at a cost of $125 million. In contrast, a large life insurance com-
pany requires only 300 managers for 10,000 buildings worth $8
billion, at a cost of only $9 million.46 In other words, the pri-
vate company manages more buildings (of nearly comparable
value), with far fewer managers, at a small fraction of the cost
88 PRIVATE PRISONS

of government management. It therefore seems reasonable to

suppose that the management of prison property would also be
more efficient in private hands.

Experience with Contracting for Specific Prison Services

Virtually every aspect of operating a prison—such as food ser-
vice, medical service and counseling, educational and vocational
training, recreation, maintenance, security, industrial programs,
and so on—is already subject to outside contracting. In addi-
tion, many states and the federal government now contract out
the majority of their community corrections programs. If cor-
rectional administrators say that it is cheaper for them to farm
out these various aspects of corrections separately, that may not
prove, but it certainly does suggest, that it could also be
cheaper to administer an entire subsystem, such as a prison, un-
der private contract.
Camille and George Camp's comprehensive national survey of
correctional agencies identified 52 agencies that had contracts with
the private sector in 38 states plus the District of Columbia.47 Those
contracts encompassed 32 varieties of service, covering "literally
every aspect of institutional operations."48 Fifty agencies reported
a total of about $200 million in contracts.49 Looking only at their
largest contracts, 22 agencies reported saving $9.5 million in all,
that is, they spent 26 percent less than they would have if they
had provided those services themselves. On the other hand, six
agencies spent 17 percent more on their largest contracts than they
would have if they had provided the services themselves, for a
total loss of $800,000.50
Since the aforementioned figures pertain only to the largest
contracts (excluding construction or architect fees), the total
value of savings due to private contracts is not known. How-
ever, we do know that three-quarters of the agencies reported
some—at times considerable—savings. Moreover, at least some
of the money-losing contracts were a response to court orders
or some other motivation to upgrade services51 and thus not pri-
marily designed to save money. 52 Reviewing the cost implica-
tions of their survey in a later article, the Camps concluded:
Issues of Cost and Efficiency 89

"Contracting with the private sector has proven to be cost-effec-

tive most of the time." 53

Cost Inferences from Privatization of Other Public Services

All public services—not just corrections—either have direct coun-
terparts in the private sector, or can be broken down into com-
ponents, each of which has a commercial analogue. Not counting
the postal service, the jobs of one-quarter of all federal employees
are identical to jobs in private business and industry. The Office
of Management and Budget has estimated that $1.7 billion a year
could be saved by contracting this work to the private sector.54
Therefore, the literature on privatization of other public services
is relevant to the question of whether we should anticipate cost
savings in the privatization of corrections.
The contracting process has been shown to save the govern-
ment money even when the government, after study, decides
not to contract out. Circular A-76 of the Office of Manage-
ment and Budget specifies the procedures for federal agencies
to follow in determining whether some of the goods and ser-
vices they produce could be more efficiently procured under
contract. The first step is to write a clear Performance Work
Statement defining the activity in question. The agency then
analyzes its current method of carrying out that activity in-
house and determines what changes would be necessary to
make it as efficient as possible. The estimated cost of a reorga-
nized in-house operation is then compared with the estimated
cost of contracting the activity to a commercial firm. About 45
percent of these comparisons favor the in-house operation, but
the government agency wins either way by achieving greater ef-
ficiency. In 1984, OMB reported that almost 1,700 cost studies
conducted since 1979 showed an average savings of 20 percent
over previous costs, "regardless of whether Federal employees
or contractors won the competition." 55
Other empirical research has shown economic benefits in the
privatization of such diverse services as solid-waste collection, elec-
tric power, fire protection, transportation, postal service, health
care, education, social services, protective services, and a number
90 PRIVATE PRISONS

TABLE 5.1. Operating Costs, Per Capita, at Okeechobee and Dozier

Schools for Boys
1981-82 1982-83 1983-84
(Before) (Transition) (After)
Dozier $12,155 $13,604 ( + 10.7%) $17,215 ( + 41.6%)
Okeechobee $10,853 $11,310 (+ 4.2%) $14,617 ( + 34.7%)
Differential 6.5% 6.9%
Source: Adapted from American Correctional Association, Private Sector Operation of a
Correctional Institution (Washington, D.C.: U.S. Dcpt. of Justice, National Institute of Cor-
rections, April 1985), p. 69.

of others.56 Obviously, imprisonment differs from other public

services in important respects, but not necessarily in ways that
relate to efficiency and cost. Evidence of successful private delivery
of other services is cause enough to anticipate that it also would
be feasible for corrections.

Okeechobee—An Early Cost Comparison

In a report sponsored by the National Institute of Corrections, the

American Correctional Association compared Florida's privately
run Okeechobee School for Boys with the state-run Arthur G.
Dozier School for Boys. Table 5.1 is adapted from that report.57
In the AC A report, the dollar figures given in Table 5.1 were
interpreted as showing that Okeechobee did not become more
efficient under private management. While Okeechobee continued
to operate at a lower per capita cost under the Eckerd Foundation,
just as it did under state management, it did not show any decrease.
At Okeechobee, the per capita operating cost increased from
$10,853 to $14,617, while at Dozier it went from $12,155 to
$17,215.
A very different interpretation, however, is actually more con-
sistent with Table 5.1 than is the ACA study's conclusion that
savings were not achieved because costs were not reduced. The
three fiscal years shown in the table are the years before, during,
and after the takeover of Okeechobee by Eckerd. The "promise"
Issues of Cost and Efficiency 91

of Jack Eckerd (which was more like a boast or a bet) was not
that his foundation could decrease costs over time, but that it could
run the school for less than the state could.58 An early claim of 10
percent in achievable savings was reduced to a prediction of 5
percent before the agency solicited bids, so the lower target of 5
percent seems fair to use as Eckerd's test.
Costs are ordinarily expected to increase through time as a result
of inflation, so the proper design for a before-and-after study is to
compare the rates of increase under the two different types of
management. A lower rate of increase found in a before-and-after
comparison is most correctly interpreted as a cost saving. To then
attribute such a saving to private management would require an
assumption that Okeechobee in state hands would have increased
in costs at the same rate that Dozier did in state hands.
This is a logical assumption, and there is some empirical support
for it. Dozier's major cost increase was for additional staff, hired
in response to the recommendations of a 1981 legislative study of
the state's training schools.""9 In 1981, Okeechobee's client/staff
ratio (1.7) was higher than Dozier's (1.6).6" It is thus reasonable
to assume that the state would have upgraded the staff at Okee-
chobee at least as much as it did Dozier, at a comparable per-
centage increase in cost.
As shown in Table 5.1, costs at Okeechobee during Eckerd's
first year increased 4.2 percent—6.5 percentage points below Do-
zier's increase of 10.7 percent. The difference, 6.5 percent, is an
estimate of the increase in Okeechobee costs that would have
occurred under state management but was avoided by the change
to private management. This compares quite favorably to Eckerd's
claim to be able to run the school for "5 percent less" (than it
would cost the state to run it during the same year). In other words,
Jack Eckerd's boast was fulfilled, at least in the short run.
On the other hand, Okeechobee's cost control advantage across
two years (6.9 percentage points) was not very much greater than
its advantage across one year (6.5 percentage points). Additional
data would be needed to determine whether Eckerd's efficiencies
were temporary or continuing. Both Okeechobee and Dozier
changed in important ways across these three years.61 Such changes
make comparisons from one year to the next very difficult.
92 PRIVATE PRISONS

Despite these problems, many simple cost comparisons between

public and private facilities can be cited. Most of the compari-
sons are favorable to private management, while a few favor
government.

Simple Cost Comparisons Favorable to Private

Management

In a 1982 census of juvenile facilities, the average cost to house

one resident for one year in a private facility was $21,256; in a
public facility it was $22,009. However, there was great variation
by state in costs and in the relative cost advantages of private
versus public facilities. Public facility costs were higher in 30 states
and private facility costs were higher in 17 states. Much of the cost
advantage of private facilities was probably due to their open en-
vironments. While 90 percent of the private facilities had open
environments, this was true for only 61 percent of the public fa-
cilities. The average daily cost per inmate for all public facilities
was $60, a little higher than the cost of $58 for private facilities.
However, the cost for "open" public facilities was $53 and for
those with an "institutional" environment it was $62. Thus, cost
was associated more with type of environment than with type of
management.
One privately run juvenile facility that has an institutional rather
than an open environment is RCA's Intensive Treatment Unit at
Weaversville, Pennsylvania. While expensive, this program is com-
parable or a little lower in cost than equivalent state programs. Its
per diem of $130 is about 11 percent less than the $141 and $152
costs at two comparison state facilities.62 RCA's staff salaries at
Weaversville are often lower than equivalent state positions and
their medical and pension benefits are more modest.63 In 1983,
Pennsylvania allocated $912,819 for Weaversville. RCA came in
under budget, with expenses for that year of $868,449, of which
$59,761 was their proprietary fee.64
At the Marion Adjustment Center, the Kentucky Corrections
Cabinet reports that U.S. Corrections Corporation's cost is about
25 percent higher than the cost of state-operated minimum security
facilities, but comparable to the cost of contracts for community
Issues of Cost and Efficiency 93

corrections, which is how the Cabinet defines the facility.65 An-

other source compares the USCC fee of $25 per day with a state
cost of $21. This would be 19 percent higher, but it includes $1
million for remodeling.66 The state cost probably does not include
financing, construction, and other capitalization. A third source,
however, compares the USCC per diem fee of $25 in 1986 with
1983-84 costs of $22.74 and $26.83 at two similar state-operated
institutions. 67 Allowing for inflation from 1983-84 to 1986, the
USCC fee would be lower than state costs.
Facilities of the Immigration and Naturalization Service (INS)
are generally cited as less expensive when in the hands of private
operators. In 1984, Behavioral Systems Southwest (BSS) charged
the INS, for its Pasadena facility, about half of what the INS was
paying the Los Angeles County Jail two years before.68 Another
source compared a BSS per diem of $14 to INS costs of $40 to $50
(producing savings of 65 to 72 percent).69 CCA's per diem charge
of $23.84 in 1984 compared to an INS cost of $34.85 a (32 percent
savings).70 While these comparisons can be countered with others
that favor government operators (see below), the INS has reported
an overall savings, across contracts, of about 6 percent.71
In addition to its INS contracts, Behavioral Systems Southwest
also operates six minimum security reentry facilities. As reported
in Money magazine: "The firm charges $14 to $33.50 a day per
detainee, or 15% less than it costs California to run similar pro-
grams. The difference is mostly labor costs. While starting cor-
rectional officers at a state-run prison usually earn slightly more
than $2,000 a month, Behavioral Systems' monitors—they don't
carry firearms or wear uniforms—make about half that." 72 The
comparison between prison guards and halfway house monitors,
however, may not be appropriate. Also, the implication that cost
savings require salary cuts for similar work does not hold up when
training and duties are held constant. For example, when CCA
takes over a facility previously run by the government, where
comparison of function is more appropriate, it generally raises the
salaries of employees who stay on.
Santa Fe County, New Mexico, spent $94 a day to run its jail
before CCA took over in 1986. CCA costs in 1987 were $45 a day
(52 percent lower). New Mexico Attorney General Paul Bardacke
attributed the lower cost to better use of the facility.73 The jail was
94 PRIVATE PRISONS

originally overbuilt and rarely more than half full, but CCA has
subcontracted space not used by the county to other jurisdictions.
While Santa Fe County's budget for the jail under its own operation
was $1.5 million a year, its payment to CCA came to $858,678 a
year for the population size at the time of takeover. Assuming an
arbitrary 12 percent increase in population, that would rise to about
$975,771 in a year. The contract with CCA sets a ceiling price of
$1.3 million a year during the three-year life of the contract.74 At
the end of the first year, Dr. Patricio Larragoite, Chairman of the
Santa Fe County Commission, reported savings of $400,000 under
the contract and projected a total of $1.5 million in savings over
the three-year period.75
In Florida, CCA's final bid of $24.50 per diem to operate the
Bay County Jail was 12 percent lower than the sheriffs final bid
of $27.80.76 As an added benefit, CCA's figure (unlike the sheriffs)
included $700,000 in renovations. When CCA and the sheriff each
projected their total jail costs for the following year, CCA's esti-
mate of $2.5 million was 22 percent lower than the sheriff's estimate
of $3.2 million.77 The sheriff's true costs were probably understated
in his final bid. According to County Commissioner John Hutt,
the sheriffs proposed budget prior to county negotiations with
CCA worked out to about $48 per inmate day. During county
negotiations with CCA, however, the sheriff presented about six
different revised budgets, in which he shifted costs from the cor-
rections budget to the law enforcement budget.78 His final figure,
however, was still higher than CCA's. In addition, CCA initiated
a work program that provided the county with $660,000 worth of
labor in one year.79
Volunteers of America in 1985 charged Ramsey County, Min-
nesota $57 a day for fully confined inmates80 in its women's de-
tention center. This was 29 to 37 percent less than the $80 to $90
estimated cost it'Hennepin County (Ramsey County's prior, public
contractor) were to build new space for them. 81
In Ohio, the Corinthian Corporation proposed to build a jail
for Summit County or Cuyahoga County. The 100-bed, $3.5 mil-
lion facility would hold minimum security cases only: misdemean-
ants or nonviolent first offenders. The corporation's proposed
charge of $50 a day would be lower than the $57 per diem cost of
the current Cuyahoga County jail or the $70 charged by Lake
Issues of Cost and Efficiency 95

County to hold prisoners for other jurisdictions. Those facilities,

however, probably handle a broader range of offender than the
Corinthian Corporation expected to house. The $44 to $50 sliding
scale at the Warrenville Workhouse might reflect the public cost
for a comparable population.82
A three-county group in New Mexico negotiated with a private
company, Southwest Detention Facilities, to build and operate a
regional prison. The company offered to do this for $54 per inmate
day, 31 percent less than the counties' costs of $78 a day.83
During its first year of operating the Butler County Prison, Buck-
ingham Security saved the county $100,000 in overhead by elim-
inating 15 part-time, nonunion jobs, while retaining 21 full-time,
unionized county workers and reorganizing work schedules. The
cost of running the jail was reduced from $700,000 to $600,000.
Of that, $270,000 was for costs the county pays directly—capital
improvements, medical expenses, and jail (county) employee sal-
aries. The remaining $330,000 was Buckingham's management fee,
which covers food service, utilities, maintenance, management sa-
laries, and a profit of 5 to 10 percent.84

Simple Cost Comparisons Favorable to Government

Management

Just as some INS facilities, like those cited above, did better fi-
nancially under private management, similar facilities showed
lower costs under the aegis of the government. In 1985, the per
diem cost at privately operated INS facilities was $37.26 as an
unweighted average across five sites; the cost at publicly operated
INS facilities was $31.89 as an unweighted average across six sites
for which there were data.85 However, as noted by the Pennsyl-
vania Legislative Budget and Finance Committee, which reported
these figures, there is so much variation among INS facilities by
location, size, security, and services that comparing average ex-
penditures is not very meaningful. The range among the govern-
ment-run facilities was $17.65 to $68.14. Among the privates, the
range was $17.76 to $88.69. A Massachusetts legislative report
cited these comparative figures to emphasize the point that such
comparisons are practically useless.86
96 PRIVATE PRISONS

Norman Carlson, Director of the Federal Bureau of Prisons from

1970 to 1987, informed Congress that the privately operated Hid-
den Valley Ranch charged $92 per diem to hold Youth Corrections
Act offenders. At the Bureau's three other facilities holding YCA
offenders, he said, the cost was $55 per inmate day (40 percent
less). Even so, Carlson regarded Hidden Valley as "cost-effective"
because it provided flexibility during a period of transition as the
Youth Corrections Act expired.87 Data on Hidden Valley are con-
flicting, however. Another source reports that Bureau of Prisons
records showed payments to the contractor of about $76 per inmate
day. This was described by the contractor as "about what it would
cost the government to do the job itself."88
California plans to contract for several minimum security facil-
ities to hold parole violators. The estimated cost of $16,000 to
$20,000 per year per inmate would be one-third to two-thirds
higher than the $12,000 figure for low-security state prisons.89 State
officials caution, however, that these figures may not be properly
comparable.9"
The State of Alabama in 1985 decided not to contract after its
own Stanton Correctional Facility for juveniles was compared with
Florida's privately managed Eckerd Youth Development Center
at Okeechobee. The study showed that "privatization of correc-
tional facilities in Alabama would significantly raise costs, not re-
duce them." 91 The annual cost difference was $2,694 per inmate
(or $7.38 per diem).

Problems with Simple Cost Comparisons

Simple comparisons like those presented above can be used to

support opposite conclusions: that proprietary prisons are less
expensive—or more expensive—than their governmental coun-
terparts. The key word here, and a source of confusion, is "coun-
terpart." Researchers who compare institutions must face the fact
that facilities vary widely on a great many factors that affect costs;
so much so that most simple comparisons of per diem rates are
not very meaningful.
Region or location of a facility affects wage rates, property val-
Issues of Cost and Efficiency 97

ues, construction costs, and the price of food, fuel, utilities and
many other costs.
The age of a facility affects maintenance, depreciation, and costs
related to efficiency of design. If buildings are still being financed,
the speed at which the debt is being retired has a substantial impact
on per diem costs, just as housing costs vary by length of mortgage.
Construction costs and the purchase or rental of land may be
included in some budgets or per diem figures and not in others.
Population size, homogeneity of inmates, and security level and
custody needs of inmates all affect costs. So, too, does the match
between the physical design of a facility and the nature of its
population. For example, a facility that must be built, staffed, and
programmed to accommodate a mixture of security levels may not
be as efficient as one that is designed for a more narrow population
and purpose.
Processing costs vary by sentence length and turnover. Treat-
ment costs are a function of the range of services and programs
offered.
The foregoing are a few of the many factors that make one
correctional facility inherently more expensive than another. Un-
less these items are explicitly taken into account, it is not very
useful to compare the per diem cost of a private facility to its
governmental "counterpart."92
Simple comparisons also suffer from being insufficiently thor-
ough. A complete analysis of the cost of a prison should include
construction, depreciation, debt servicing, rent or rent equiva-
lence, taxes paid or foregone, overhead, indirect costs, and many
other complexities. The official budgets of government-run prisons
generally do not include all of these components.
One of the inherent difficulties in comparing costs of government
and private services is that the usual mechanism for reporting
government expenditures—the budget—is not as thorough or as
accurate as the cost-accounting mechanisms that are used by con-
tractors, who want to be sure that they recover all of their outlay.
As a result, government is often unaware of the true costs of its
own services. In a 1971 study of refuse collection in New York
City, E. S. Savas showed that the full cost was 48 percent greater
than what was shown in the city's budget. 93 In a later analysis
98 PRIVATE PRISONS

across cities nationwide, Savas found that the true cost of municipal
refuse collection averaged 30 percent higher than what was shown
in the city budget.94 An independent study of 18 Connecticut cities
also found actual costs of collection to exceed the budget figures
by an average of 30 percent (with a range up to 256 percent).95
In contrast, Savas found that cities that contracted for their
refuse collection were able to set user fees fairly close to actual
costs, while cities with municipal collection charged user fees that
had no relation to real expenditures. His conclusion: "Cities with
contract collection know how much the service costs; cities with
municipal collection do not." 96
There is reason to believe that government is often just as ig-
norant of the true costs of corrections as it is of the costs of refuse
collection.

Hidden Costs of Corrections

Generally, reports of government correctional costs are taken from
a single budget, either of a facility or of the agency in charge of
the facility. These budgets vary a great deal in terms of what
components they include. It is probably fair to say, however, that
no agency or facility budget shows all of the direct and indirect
costs of corrections.
Costs that do not appear in an agency's budget can be referred
to, for convenience, as "hidden costs." This does not imply that
they are deliberately concealed; only that they are not readily
apparent or easily discernable. Most will come from the budgets
of other government agencies, where they will probably not be
identified as expenditures on corrections.97 For example, litigation
and liability costs are generally taken from the budget of the state
or county attorney; fringe benefits and pensions often come out
of some general fund rather than the budgets of particular agencies
or facilities. Services provided by other agencies should be (but
rarely are) prorated into the budgets of correctional agencies, and
services provided centrally by a correctional agency should be pro-
rated into the budgets of individual facilities. Facility budgets
commonly list only operating costs, omitting land purchase, con-
struction, financing, depreciation, and other capital outlays.
The problem with hidden costs is not merely that they are under-
Issues of Cost and Efficiency 99

estimated in comparisons; they are also harder to control. When

costs are out of sight, they are often out of the minds of the people
who incur them and of the people who might want to curtail them.
Legislative budget review committees will lack the information
necessary to evaluate competently the reasonableness of requests
for funds. Outsiders (and even insiders) will find it hard if not
impossible to do realistic cost-effectiveness analyses. Program
managers lose at least some incentive to hold down costs when
they do not get direct and accurate feedback. For example, if fringe
benefits come from a separate budget, correctional managers and
workers are not encouraged to view pensions, health insurance,
and so forth as a trade-off against salary, or against other expenses
(like equipment) that affect working conditions.
A list of cost components frequently missing from the budget
of a correctional agency or facility would include at least the
following:
1. Capital costs: land purchases, construction, major equip-
ment, depreciation or amortization.
2. Finance costs: service and interest on bonds.
3. Employment benefits: longevity bonuses, pensions, in-
surance.
4. External administrative overhead: prorated share of the ex-
penses of centralized executive offices (governor, mayor,
and so forth) or administrative offices (e.g., personnel ser-
vices, central purchasing, data processing, general services
administration).
5. External oversight costs: inspections, program monitoring,
administrative or judicial reviews and appeals of decisions,
auditing and other comptroller services.
6. Legal service costs: counsel, litigation, and other legal ser-
vices occasioned by the activities of the correctional agency
or facility in question but charged to other budgets (includes
publicly funded litigation costs of inmate plaintiffs or de-
fendants as well as defense of the institution or agency and
its political jurisdiction).
7. General liability costs: successful legal claims, punitive dam-
ages, fines, court costs, general liability insurance premiums
or costs of administering a self-insurance plan.
100 PRIVATE PRISONS

8. Property insurance costs: premiums or self-insurance costs

for fire, theft, and casualty protection (or risk-cost of un-
insured losses).
9. Staff training costs: when provided at cost or subsidized by
another agency.
10. Transportation costs: transportation services, vehicles, ve-
hicle maintenance, fuel, parts, and related costs may be
provided by other departments.
11. Food costs: other government agencies may provide surplus
food or subsidies.
12. Interagency personnel costs: personnel may be borrowed
from other agencies for either routine purposes or emer-
gencies.
13. Treatment or program costs: other agencies may provide
hospitalization, medical and mental health care, education
services or programs (including vocational education or job
training), recreation, counseling, or other treatment pro-
grams and services.
14. Opportunity costs: taxes or rent foregone from alternative
uses of land or buildings.
15. Unemployment and workmen's compensation costs.

When asked to provide a per prisoner figure for the cost of

running their facilities, most corrections agencies simply divide the
operating budget by the average daily population. This, as noted
above, may leave out many hidden expenses. In the absence of
information that indicates a more thorough accounting, it would
be reasonable to add 35 percent to most figures reported by public
agencies as an estimate of their real costs.98
In 1985, George and Camille Camp asked state correctional
agencies to report their average daily cost per prisoner." The 42
states that responded reported an average cost of $38.87 (and a
range of $20.27 to $84.72). They were then asked to give an es-
timate of the total cost of correctional confinement and care taking
into account expenditures by other agencies. These estimates
ranged from $22.02 to $100, with an average of $44.11. The average
estimated total cost was 13.5 percent higher than the average re-
ported agency cost. The estimates of other agencies' costs may
have been anything from informed appraisals to outright guesses.
Issues of Cost and Efficiency 101

Six states did not answer the second part of the question, one
reported a total cost six cents higher than the agency cost, implying
an uncanny degree of accuracy, and eleven states indicated that
there were no outside costs of corrections, which seems very
unlikely.
Even the estimated total costs in this study did not include con-
struction and financing. The Camps suggest that a "more accurate,
yet conservative estimate is a 20 percent addition in expenditures
above those in the correctional agency's budget. In some systems,
other than correctional agency expenditures for corrections may
account for up to 35 percent more in prisoner expenditures."100
The Camps' data, however, show a 45 percent addition for Col-
orado, a 104 percent addition for Ohio, and a 128 percent addition
for New Hampshire. A study by the Correctional Association of
New York found that the real cost of housing inmates in New York
City jails was 54 percent higher than the figure used by the Cor-
rection Department. 101
Agencies or facilities that do not pay pensions and fringe ben-
efits out of their own budgets are probably underestimating
their expenses by 25 to 30 percent. The Camps found that sal-
ary accounts for an average of 80 percent of correctional budg-
ets and that retirement and other fringe benefits "frequently
amount to one third of salary expenses."102 A broader study by
the U.S. Chamber of Commerce reported that government
spends an average of 31 percent of payrolls for pensions (com-
pared to 13 percent for private enterprise).103 Adding current
retirement contributions to reported cost figures may not be suf-
ficient, however. Government pension systems are often under-
funded, because they are subject to less stringent (and less
fiscally sound) funding requirements than private pension sys-
tems must meet.104 Since pension obligations eventually must be
honored, part of their cost has simply been delayed, and thus
hidden in another way as well.

Hidden Costs of Contracting

A private contractor's charges will also underestimate the total
cost of corrections, albeit to a lesser degree. While we can have
some confidence that a contractor's per diem will reflect all of the
102 PRIVATE PRISONS

costs to the private vendor, plus a margin for profit, we cannot be

sure that it will contain all of the costs to the government purchaser.
A contractor's fee will include (1) items that are visible in both
public and private operations; (2) costs that are not visible in a
public agency's budget but are explicitly factored into a contractor's
bill; and (3) costs that are special to contracting or to the private
sector, but which can be included in the fee to government. Beyond
those costs that can be explicitly incorporated in a contractor's fee,
there will be some costs to government that cannot be passed to
the contractor, to be charged back as part of the fee. An example
would be the cost of preparing Requests for Proposals (RFPs) and
evaluating the responding proposals. In addition, some costs may
remain with the government because it is more efficient for the
government to perform the service itself than to purchase it from
a contractor.
Those costs to the government that are not included in the con-
tract are the "hidden costs" of contractual corrections. That is,
they are hidden in the sense that they do not appear in the per
diem costs reported by the contractor. Thus, while contracting
makes costs more visible, it does not by itself reveal all costs.
Expenditures that the government retains, however, will be more
easily identified as a result of the contracting process. For example,
when asked how much it costs to run their jail, most county offi-
cials, looking at their jail budget, will not think to include any part
of the cost of running the county hospital. If, however, they have
a contract specifying that the county is responsible for the bills of
hospitalized inmates, they are more likely to include this compo-
nent in their cost computations.
Other government costs that are generally ignored also may be
uncovered through examination of contractors' costs. For example,
the ACA study of Okeechobee notes that Eckerd must pay over
$175,000 for insurance and spend $250,000 of allocated funds on
overhead, which the report incorrectly refers to as "costs that
Dozier does not have." 105 But the state of Florida does have over-
head and self-insurance costs in running Dozier, Okeechobee's
sister school. These costs should (but do not) appear in the budget
for that facility.
One paper critical of private prisons contends that among the
hidden expenses of proprietary jails are the costs of emergency
Issues of Cost and Efficiency 103

situations, such as escapes, riots, fires, natural disasters, public

health problems, employee strikes, or bankruptcy. 106 These emer-
gencies also occur in public prisons107; we do not know yet whether
private jails will be more or less likely than their public counter-
parts to incur catastrophic costs. However, the risk-cost will prob-
ably be more rather than less visible in private prisons. Contracts
generally specify the liability of contractors for costs like these and
require that the private company insure itself against them and
indemnify the government. Contractors therefore must take these
costs into account explicitly when they bid for contracts or
renewals.
The contracting process itself—including the cost of soliciting,
evaluating, drawing up, monitoring, renegotiating, and terminat-
ing contracts—is said by critics of contracting to add to the total
cost of services. The federal government uses 4 percent of the total
contract cost as the incurred expense for contract administration,
though the American Federation of State, County, and Municipal
Employees feels this figure is "often too low."108
Monitoring costs may or may not be hidden, depending on the
provisions of the contract. In any case, these costs will accrue to
the government, since if they are charged by contract to the private
provider they will then be calculated into the provider's fee. This
procedure forces the government to identify and make explicit the
cost of an activity that exists in one form or another with or without
a contract.
Most discussions of monitoring treat it as a new cost, attributable
only to proprietary prisons. However, all public correctional fa-
cilities have, or should have, at least some provisions for moni-
toring, supervision, or inspection. Monitoring is no less important
for public facilities than for private ones. The Council of State
Governments and the Urban Institute, in their study of private
prisons, focused on this subject:
It can also be argued that states should monitor their own state-
operated facilities as carefully as they do a contracted institution
and, therefore, monitoring expenses should be about the same for
both modes of operation.1"9

Thus, only differential costs—not all of the cost of monitoring a

private prison—should be attributable to the contracting process.
104 PRIVATE PRISONS

Assuming that it costs money to monitor both public and private

correctional facilities, then the question is whether the degree of
monitoring required by a private facility is more or less expensive
than the degree of monitoring needed in a public j ail. For example,
it would be more expensive if a private facility required the hiring
of new or specially trained monitors, different from those that
monitor public facilities. It also would be more expensive if mon-
itoring a private facility required the keeping of duplicate records
whereas a public facility maintained only one set of centralized
records. On the other hand, monitoring a private contractor might
be less expensive if the contractor kept better records, or had a
more sophisticated and efficient information management system.
It also would be less expensive if private facilities passed their
inspections more often the first time around, thereby eliminating
repeat inspections.
Another kind of indirect or hidden cost of proprietary correc-
tions may occur if contractors pay some of their workers wages
that are below the level of subsistence. In this case, an apparent
savings in lower per diem fees may have to be balanced by a
"hidden" government cost in the form of added welfare payments.
Similarly, some of the apparent savings from lower fringe benefits
may be spurious. Contractor contributions to social security and
company retirement plans will be included in the fee charged to
the government. However, if those contributions are not adequate,
they may eventually be added to government costs in the form of
post-retirement welfare benefits.110

"Hidden Rebates" from Contracting

Some of the costs of running a private prison, although they are
charged to the government as part of the company's fee, eventually
return to the government in some form of tax. These could be
referred to as "hidden rebates." Hidden rebates are expenses paid
through a private contractor that return to the government as
revenue.
For example, business taxes, property taxes, sales taxes, PICA
(social security) and workmen's compensation contributions, un-
employment taxes, telephone and utility taxes, fees for water,
sewage, and waste disposal, inspection fees, and license fees are
Issues of Cost and Efficiency 105

all costs of business incorporated in a contractor's fee, but they all

return to the government as revenue. So do taxes on the profits
of a private prison.
If the government contracts for the operation of a prison, the
hidden rebates described above would have to be subtracted from
the contractor's fee to calculate the true net cost to the government.

Hamilton County: A Relatively Thorough Cost

Analysis

Correctional officials will find it very difficult to identify and es-

timate interagency costs. A county auditor, however, is in a good
position to do so. The analysis that follows is based on the work
of Hamilton County (Tennessee) Auditor Bill McGriff. 111
On October 15, 1984, Corrections Corporation of America as-
sumed management of the Hamilton County Penal Farm, a 350-
bed minimum to medium security county prison holding convicted
county misdemeanants, state felons, and some pretrial detainees
under the jurisdiction of Hamilton County at Chattanooga. The
cost of the contract is renegotiated by the county every year. For
that purpose, Bill McGriff prepares an annual analysis estimating
and comparing the total cost to the county of: (a) reassuming direct
county operation of the prison, versus (b) continuing to contract
with CCA for the operation of the facility.

Costs under County Management

The question facing the county each year is whether they could
run their prison for less than the fee that the contractor is re-
questing for the upcoming fiscal year. Since fiscal year 1983-84
was the last year in which the county managed the prison itself,
and many things have changed, it is necessary to identify and
annually reestimate component costs, rather than relying on out-
dated budget figures.
McGriff based his analysis on several assumptions:
1. Staffing would remain the same as CCA's, with certain ad-
justments, if the county took back the facility.
106 PRIVATE PRISONS

2. Prison employee salaries would have increased since fiscal

year 1983-84 by the same amount as the salaries of other
county employees.
3. Nonsalary expenses would have increased at a rate equal to
inflation as measured by the Consumer Price Index (CPI)
plus, where appropriate, a rate equal to the increase in the
prisoner population.
4. The county would have incurred no extraordinary expenses,
such as a lawsuit settlement beyond the level of insurance
coverage.
5. The county would issue $1.6 million in bonds at 7 percent
for 15 years to pay CCA for the new facilities they built at
the Penal Farm.
These assumptions were designed to be conservative, i.e., to un-
derestimate costs to the county if it had retained, or if it took back,
management of the prison.
The assumption that county staffing would be the same as the
contractor's is realistic for purposes of pricing a resumption of
control. However, it could well underestimate what the staff size
might have grown to under continued county management, since
one of the goals of contracting is to achieve greater staffing
efficiency.
Assumption 2 had to be modified in fiscal year 1986-87 because
a county wage study indicated that prison employees, among oth-
ers, had been underpaid by the county. If the county took back
the prison, it would have to pay salaries responsive to the wage
study, whether higher or lower than CCA's. Furthermore, the
personnel department indicated that since corrections officers at
the prison must have the same certification and receive the same
training as the sheriff's jail officers, they should be paid accord-
ingly. Jail officer trainees with six months' experience were paid
at grade 8, while the entry level for prison officers prior to the
contract was at grade 4. On the other hand, the county had a policy
of not upgrading a position more than two grades in one year,
which might or might not have applied to the prison officers, who
had already been out of the county system for a year. To be
conservative, McGriff upgraded their positions only two grades.
This is a significant underestimate, since it is reasonable to suppose
Issues of Cost and Efficiency 107

that corrections officers returning to the county system (especially

those with experience) would be paid no less than entry-level jail
employees. Note that underestimation of salaries also implies
underestimation of fringe benefits. Together, these two categories
constitute close to half of total costs.
Assumption 3, by using the CPI, probably underestimates infla-
tion in other county costs. Since World War II, the cost of services
provided by the government has tended to rise substantially faster
than the CPI,"2 and correctional costs have recently been rising
even faster than other government costs.
Assumption 4, that no extraordinary (i.e., unforeseeable or in-
calculable) expenses would occur, is a necessary supposition almost
by definition. However, since sooner or later such expenses are
bound to arise, the assumption has the effect of underestimating
the potential costs of county management. Under contractual man-
agement, the contractor serves as a buffer for many such potential
costs.
In computing the total costs of the prison under county man-
agement, McGriff included the following components:

Hamilton County Penal Farm Component Costs

Salaries & wages Fringe benefits
Consumable maintenance supplies Maintenance & garbage
Utilities Insurance
Medicine & personal care County hospital care
Food & kitchen supplies Depreciation
Uniforms Interest expense
Capital outlay (equipment) Other direct costs
Other operating expenses Other indirect costs

The costs listed in the column on the left are those that would
ordinarily be found in any prison budget. The costs listed on the
right are often taken from other budgets and not accounted for in
the budget of the facility being examined. 113 As Hamilton County
Auditor, however, McGriff was in a position to be unusually thor-
ough in identifying the indirect and interagency costs as well as
the regular budget line-item costs of operating the Penal Farm.
Several of the items in the right-hand column need some
elaboration.
108 PRIVATE PRISONS

For maintenance and garbage, McGriff at first used a figure

provided by the public works administrator. However, when he
had his own people check out actual costs, counting the number
of pickups and identifying costs not billed back to the facility, he
found that the real cost was about twice that amount.114
Insurance includes both property and liability insurance."5
McGriff made a conservative estimate of the cost of insurance for
the prison, based on county insurance covering both its jail and
its sheriffs department. The cost of the (estimated) jail portion of
that coverage was used to estimate the cost of covering the Penal
Farm.
The Workhouse Records Clerk keeps records on time served by
prisoners. Her salary is separate from the Penal Farm's budget.
County hospital care would not come from the Penal Farm's
budget in any case. Since the county defines all prisoners as in-
digents, any hospitalization is paid for out of the $3 million annual
contribution the county makes for indigent care to Erlanger Med-
ical Center, the joint city/county hospital. McGriff pulled Erlanger
hospital costs for Penal Farm prisoners from this total, separate
from the costs for jail inmates.
Depreciation and interest costs are calculated for all construction
at the prison prior to the contract. In addition, if the county ter-
minates the contract, it must reimburse CCA for the $1.6 million
in renovations and additions invested by the contractor during its
first year of operation. To estimate the cost of this reimbursement,
McGriff assumed a bond rate of 7 percent and a depreciation
period of 40 years.
The categories labeled "other direct costs" and "other indirect
costs" are listed below.

Other Direct Costs Other Indirect Costs

Personnel County Commission
Accounting County Executive
Financial management County Auditor
Data processing County Attorney
Purchasing Finance Administrator
County Physician
Human Services Administrator
Issues of Cost and Efficiency 109

"Other direct costs" include activities of those central offices

that routinely perform services for all county agencies: personnel,
accounting, financial management, data processing, and purchas-
ing. Some portion of the activities of these offices would be directed
toward the Penal Farm. The County Physician was (at that time)
a doctor who worked part-time for the county. All he did for
Hamilton County was to treat prisoners at the jail and the prison.' lf>
The Human Services Administrator is the head of the county's
Human Services division. 117 Since the prison falls under this di-
vision, the services of the Administrator and her secretary are
"direct" rather than "indirect" costs.
"Other indirect costs" are those incurred by the activities of
other county officials at the executive level. These officials, and
their staffs, must spend some portion of their time dealing with
matters pertaining to the Penal Farm. The matters requiring their
attention may be occasional, periodic, or seemingly constant, but
they are distinct from the direct, routine dispensations included in
"other direct services."
All these interagency costs come from budgets other than those
of the Penal Farm. Thus, each item must be prorated to calculate
what proportion of these monies goes toward the existence and
operation of the prison. As described below, the prorated amounts
were based on county expenditures (obligations) and other data
for fiscal year 1985-86.
For personnel, McGriff assumed that time and costs would dis-
tribute in a manner equal to the number of Penal Farm employees
expressed as a percentage of total county employees. This is cer-
tainly conservative. If correctional workers have higher turnover
than other county employees (which is often the case) or if their
recruitment requires a more extensive background investigation
(which is less often the case than it should be), then the assumption
of equal effort per worker from the personnel department would
be false. The prorated personnel costs would be underestimated.
The County Physician (at that time) attended only to prisoners,
so his salary and fringes were split between the Jail and the Penal
Farm. When total Penal Farm prisoner days for fiscal year 1985-
86 were added to jail prisoner days, the prison accounted for 54
percent of the total. Under the assumption that jail and prison
inmates have equal daily needs for a doctor's services, the auditor
110 PRIVATE PRISONS

prorated 54 percent of the physician's salary and fringes to the

Penal Farm.
The attentions (and salary and fringes) of the Human Services
Administrator and her secretary were prorated to the Penal Farm
at 34 percent. This is equal to the county's total Penal Farm ob-
ligations, expressed as a percentage of the county's total human
services obligations.118
All other direct and indirect costs were prorated at a rate equal
to total Penal Farm obligations as a percentage of the county's
total general fund obligations. This technique assumes that the
ratio of external costs to internal costs was no greater at the prison
than in the average county operation. As in the case of personnel,
such an assumption is quite conservative. For example, auditing
and purchasing for the prison were more difficult than for other
county operations, and thus are underestimated by the prorating
technique. As a case in point, the county had to go through a
formal bidding process every week for prison kitchen supplies.
After the contractor took over all prison purchasing, the county
was able to eliminate two buyer positions. Also, it is a good bet
that when the prison was under direct county administration, it
caused more headaches per dollar of internal spending, and re-
quired more time from some county executives, than did the av-
erage county operation. The County Attorney, for example,
probably spent more time on prison matters than on many other
county matters, prior to contracting. County commissioners every-
where cite the county jail or prison as a disproportionate source
of their problems, particularly when they are uninsured against
personal liability in the case of lawsuits.
Based on the calculations and estimation procedures described
above, Table 5.2 presents McGriff s estimated costs to the county
if it were to resume management of the prison for fiscal year 1986-
87. Table 5.2 shows a conservatively estimated total cost of
$3,413,741 for fiscal year 1986-87, if the prison were under direct
county management.
Items 1-9 in Table 5.2 are those costs that appeared as line items
in the Penal Farm budget when it was under county opera-
tion . Items 10-18 are expenses that would not appear in the Penal
Farm budget under county administration; rather, they would be
charged to other budgets. Note that the total cost shown in Table
Issues of Cost and Efficiency 111
TABLE 5.2. Hamilton County Penal Farm Estimated Total Cost if
Operated by County, FY 86-87
1. Salaries & wages $1,239,380
2. Fringe benefits 320,491
3. Food & kitchen supplies 404,966
4. Medicine & personal care 28,694
5. Utilities 198,587
6. Consumable maintenance supplies 56,532
7. Uniforms 61,237
8. Equipment 45,506
9. Other operating expenses 108,045
Subtotal: operating budget items $2,463,438
10. Maintenance & garage 70,195
11. Insurance 41,885
12. Clerk of workhouse records 16,238
13. County hospital care 238,886
14. Depreciation on pre-CCA construction 57,500
15. Interest on pre-CCA construction 74,878
16. Amortized purchase of CCA addition 152,000
17. Other direct costs" 204,888
18. Other indirect costs" 93,833

Total cost for year $3,413,741

Prisoner days (avg. daily pop. = 364) 132,788
Cost per prisoner day $25.71
"Other direct costs: personnel, accounting, financial management, data processing, purchasing,
County Physician, Human Services Administrator.
''Other indirect costs: County Commission, County Executive, County Auditor, County At-
torney, Finance Administrator.

5.2 is 38.6 percent higher than the subtotal, which includes only
prison budget line items. In my earlier discussion of hidden costs,
based on other studies, I suggested adding 35 percent to most
prison budgets as a conservative estimate of indirect costs. If Ham-
ilton County, like many other jurisdictions, had charged "fringe
benefits" to a general budget, the total would have been 74 percent
higher than the subtotal.
At an estimated total cost per prisoner day of $25.71, Hamilton
County would be fairly frugal. A 1986 survey showed a reported
cost of $30.26 per prisoner day for 10 jails in the East South Central
region (Alabama, Kentucky, Mississippi, Tennessee)."9 In 1983-
1 12 PRIVATE PRISONS

TABLE 5.3. Hamilton County Penal Farm Estimated Cost of

Operation Under Contract to Corrections Corporation of
America, FY 86-87
Total payments to CCA ($21.82 avg. per diem) $2,897,685
Superintendent's budget (monitoring) 67,783
Continuing, noncontracted county costs" 387,502
"Hidden costs"" 23,458
Subtotal $3,376,428
Less "hidden rebates'" ($64,000)

Total $3,312,428
Total prisoner days (avg. daily pop. = 364) 132,788
Cost per prisoner day $24.95
"Clerk of Workhouse Records, county hospital care, depreciation on pre-CCA construction,
and interest on pre-CCA construction (itens 12-15 from Table 5.2).
''Other indirect costs of county management (item 18 from Table 5.2), reduced by 75 percent
(auditor's estimate) under CCA management.
c
Local sales, property, and business taxes.

84, the reported cost per prisoner day across all of Tennessee's
state adult confinement facilities was $30.17.120 Although this last
figure includes capital as well as operating expenditures, it is prob-
ably not as inclusive of other costs as is the Hamilton County
estimate, and it was calculated three years earlier.
Clearly, then, Hamilton County's estimated costs were low rel-
ative to those of other government-run facilities in the region.
Therefore, Hamilton County provides a fairly severe test of a
private contractor's ability to lower government costs.

Costs under Contracting

After estimating costs under county management, total prison and
related costs to the county under contractual management can be
calculated rather clearly, simply, and thoroughly. The fee per pris-
oner day is fixed by contract and the number of prisoners, while
not predictable in advance, is known precisely for any past or
current period. Table 5.3 calculates what it cost the county to run
the Penal Farm under contract during fiscal year 1986-87.
Payments to CCA averaged $21.82 per diem, rather than the
contracted $22, because of an agreement by which the contractor
Issues of Cost and Efficiency 113

charged a lower fee for offenders convicted of drunk driving. These

offenders serve their time on weekends. Beyond its payments to
CCA, the county incurs other costs, which are included in Table
5.3 and described below.
Tennessee law calls for a Director of Corrections (also referred
to as superintendent) in counties with correctional facilities. Prior
to the contract, the warden of the Penal Farm doubled as super-
intendent. The former warden now serves only as superintendent,
while an officer of CCA serves as warden. The superintendent acts
as contract monitor for the county. He has an office at the facility
plus an office downtown. He spends about two-thirds of his day
at the facility and the other third downtown. He has a secretary
and some other expenses. The superintendent has final say on
disciplinary matters, "good time" allocation, and release decisions.
At least part of the superintendent's budget represents new ex-
pense to the county under contracting, since a monitoring function
has been added and a previously combined role has been split into
two paid positions. On the other hand, it is not just new cost; a
new function (independent monitoring) has been added, which the
county did not have before. In keeping with McGriffs conservative
methodology (underestimating costs under county operation and
overestimating costs under contracting), the superintendent's en-
tire budget is included in Table 5.3, even though some of his time
is spent on correctional matters other than the prison, such as
electronic monitoring, community corrections, and community ser-
vice programs.
In addition to the superintendent, some other correctional ex-
penses continue to be paid by the county directly, rather than
through the contract. In Table 5.3, these also have been added to
the cost of the contract, in the line labeled "Continuing, noncon-
tracted county costs." They all relate to the prison and were dis-
cussed in connection with Table 5.2. Some of these costs would
be the same under county operation (Table 5.2) or under
contracting (Table 5.3)—specifically: the salary of the Clerk of
Workhouse Records, county hospital care for prisoners, and de-
preciation and interest on construction prior to the contract (items
12-15 in Table 5.2).
Some other costs, referred to in Table 5.3 as "hidden costs" of
contracting, are basically the same as the "other indirect costs"
114 PRIVATE PRISONS

TABLE 5.4. Hamilton County Penal Farm Costs under County

vs. Contract to Corrections Corporation of America, FY 85-86,
86-87, 87-88
1985-86 1986-87 1987-88
County operation $2,853,513 $3,413,741 $3,642,464
(per diem) ($25.05) ($25.71) ($27.49)
CCA contract $2,746,073 $3,312,428 $3,346,300
(per diem) ($24.10) ($24.95) ($25.25)
Savings $ 107,440 $ 101,313 $ 296,164
(as %) (3.8%) (3.0%) (8.1%)
Prisoner days 113,928 132,788 132,514
(avg. pop.) (312) (364) (363)

(item 18) in Table 5.2, but at a lower level. The prison continues
to demand some attention by county executives. However,
McGriff estimated that county officials now spend at most one
quarter of the time they used to on prison matters, so this cost
estimate was reduced accordingly.
The "hidden costs" under contracting are offset by what are
referred to in Table 5.3 as "hidden rebates" from contracting:
every year CCA pays about $64,000 back into the community in
local sales, property, and business taxes that would not have existed
without the contract.

Comparison of Costs
Table 5.4 compares, for three fiscal years, the total costs to Ham-
ilton County when the prison was managed under contract, with
the total costs (as estimated by the methods described above) that
would have occurred if the county had resumed management itself.
Fiscal year 1985-86 was the first full fiscal year under the con-
tract. CCA's fee that year was $21 per diem. The higher per diem
in the table is based on total county costs under the contract, not
just the payments to the contractor. In fiscal year 1986-87, the fee
was raised to $22, where it remained the following year.
Table 5.4 shows savings to the county of at least 3.8 percent the
first year, 3.0 percent the second year, and 8.1 percent the third
year. The savings dipped a little when the contractor first raised
Issues of Cost and Efficiency 115

its fee; however, a considerable increase was evident the following

year, when CCA held (or was held) to the same fee while the
county's estimated costs increased.
Recalling the conservative nature of the county cost estimates,
the savings should be described as certainly more than those iden-
tified in Table 5.4. For example, consider the effects of McGriff's
assumption 2 (regarding county salaries) on the figures for fiscal
year 1986-87, where savings were lowest. If McGriff had estimated
the average pay of prison guards as equal to that of novice jail
guards, rather than two grades lower, this would still have been a
low estimate. However, it would have added $148,676 to estimated
county costs, and the estimate of savings for that year would have
been 7 percent, rather than 3 percent. This adjustment would not
affect the estimate for the previous fiscal year, 1985-86 (see earlier
discussion of assumption 2), but the $148,676 increase in county
salaries would continue during fiscal year 1987-88, so the estimated
savings for that year would be 12 percent rather than 8 percent.121
There still remain several other downward biases in the estimates
of county costs and therefore in the estimates of savings. These
biases, however, such as the underestimation of governmental in-
flation, the assumption of no unforeseen expenses, and the con-
servative prorating techniques for "other direct costs" and "other
indirect costs," are known only as to their direction; it is too
difficult to estimate their magnitude. Making a subjective allow-
ance for their existence, however, a reasonable and still cautious
estimate of real savings over the three years would range from 5
to 15 percent per year.
In discussions before the County Commission, McGriff has re-
peatedly emphasized the very conservative nature of his estimates
of the costs of county operation. Where he could not get figures
in which he had confidence, he either left costs out or used as-
sumptions that he thought would err on the low side. Because of
McGriff's consistently understated assessments of the costs of
county operation, we can have confidence at least in the direction
(if not the absolute size) of his findings. It is clear that Hamilton
County saves money by operating its prison under a private con-
tract. The savings are at least in the range of 3 to 8 percent, and
more probably in the range of 5 to 15 percent.
When the County Commission reevaluates the contract every
116 PRIVATE PRISONS

year, it takes into consideration some costs and benefits that

McGriff was not able to quantify. For example, CCA carries $5
million in liability insurance. In the event of a successful lawsuit,
an indemnification clause in the contract could save the county
(and perhaps the commissioners personally) a considerable, but
unpredictable, amount of money. Also, the commission believes
that CCA provides better management and more professional
training than previously existed and spares county officials many
of the daily hassles involved in running a prison. The additional
staff training, new inmate classification system, computer records
management system, and other improvements provided by the
contractor would have cost the county money to have achieved on
its own. Grand jury reports have all been positive since the contract
began, thus eliminating the time and expense required of the
county to correct the sorts of problems identified by earlier grand
juries.
Two benefits in particular make the facility and its operation
under contract not truly comparable to the alternative county ver-
sion. These are: the physical improvements made by the contrac-
tor, and the added service gained by splitting the superintendent
function from the warden function.
The county benefited from $1.6 million in new construction that
was handled by the contractor. The cost for this is factored into
both sides of the analysis. However, if the county had overseen
the construction itself (i.e., had there been no contract), it almost
certainly would have cost more. Inmate housing constructed by
the county in 1981, for example, cost approximately $65 per square
foot. The contractor's cost to construct inmate housing in 1985 was
$48.62 per square foot. Thus, for the amount of money that was
put toward construction, the county would not have been able to
add as much space to the prison without the contract.
In addition to the new construction, the contractor invested
capital and labor in repair and preventive maintenance of every
aspect of the physical plant, including plumbing, heating, and elec-
trical systems that the county had allowed to deteriorate. The cost
for this work was included in the contractor's fee, but the analysis
does not include any estimate of what the extra repair and main-
tenance would have cost under county administration.
The contract added human as well as physical capital. Under
Issues of Cost and Efficiency 117

the contract, the county has two full-time managers (each with a
secretary) performing three functions: warden, superintendent,
and monitor. Without the contract, the county would have only
one person (with one secretary) to perform as both warden and
superintendent, and it would have no monitor. It should be em-
phasized that monitoring is not just an added cost; it is an added
benefit as well. 122
The county has also added quality as well as quantity to its human
capital. The warden under CCA is a man with much more expe-
rience than the county would have been able to attract on its own.
Moreover, each CCA facility has behind it the quite considerable
experience and expertise of the top corporate officers in Nashville.
Thus, the prison operation that Hamilton County has under its
contract is not the same as what it would have if it took the op-
eration back, or if it had never contracted. It gets more (and
better) 123 for less money by contracting.

Conclusion

Private prisons will not necessarily be less expensive than those

owned and run directly by the government. A very safe general-
ization from the broader literature on contracting for public ser-
vices is that often it saves money, but sometimes it does not. It is
too soon to say much more than that for prisons, but there are
many theoretical reasons, and the beginning of some empirical
evidence, to support the proposition that private prisons can offer
to government at least the potential for gains in efficiency.
Whether or not proprietary prisons are less expensive than those
run by the government, their greatest economic benefit may be
that they make more visible the true costs of correctional facilities.
As stated in a report to the National Institute of Justice:
Government accounting systems are generally incapable of isolating
the full costs of a public activity or service. For a specific function
such as prison security or standards compliance, the direct costs are
usually buried in the expenditure records of several agencies, and
the indirect costs are particularly elusive. One of the advantages
typically ascribed to contracting in other fields is its ability to reveal
the true cost of public service. Corrections is no exception. Under
118 PRIVATE PRISONS

a contract system, the costs of confining particular numbers of clients

under specified conditions will be clearly visible and more difficult
to avoid through crowding and substandard conditions. While cor-
rections authorities might welcome the opportunity to demonstrate
clearly that more prisoners require more resources, it remains un-
clear whether legislators and voters will be prepared to accept the
real costs of confinement practices that meet professional
standards. 124
Correctional authorities should welcome the chance to reveal
the true costs of uncrowded, properly run prisons and jails. Voters
and legislators can then make realistic choices. To get this infor-
mation, however, as well as to provide the maximum range of
choices, there must be competition and information from the pri-
vate market.
6
Issues of Quality

Critics' Predictions of Poor Quality

Even if private prisons do prove to be less expensive, critics charge

that this can come only at the cost of sacrifices in quality. Indeed,
responsiveness to economic incentives, which proponents see as
essential to competition on both cost and quality, is regarded by
many opponents as the natural enemy of value. Achieving econ-
omy at the expense of quality is commonly referred to as "corner
cutting."

Corner Cutting
Critics contend that corner cutting is an almost inevitable conse-
quence of the pursuit of profits. "It's impossible to make a profit
and not cut those corners," declares Stefan Presser, a Houston
ACLU attorney. 1 Corner cutting by private prisons, it is charged,
will mean poor food and less of it, fewer services, and cheaper
labor with lower professionalism and less training. John Donahue
insists that "private firms will be unable to reduce labor costs
without debasing the quality of the work force and, with it, the
conditions of confinement for prisoners."2 Barry Steinhardt, Ex-
ecutive Director of the ACLU of Pennsylvania, predicts: "Since
the object of private prison operators will be to maximize profit,
companies will inevitably look to reduced services and unaccept-
ably low standards."3

119
120 PRIVATE PRISONS

Those who would be affected most adversely by reduced labor

costs—unionized public employees—maintain that cutting back on
wages, benefits, or number of workers will yield dire consequences:

[Private prisons will have] fewer correctional officers . . . more es-

capes, more inmate attacks. . . more riots. [Staff will be forced to
work] longer correctional careers [with] more heart attacks, more
alcoholism, more nervous breakdowns—in short, more death.
Lower salaries [will] mean greater turnover; less qualified personnel;
less job commitment; and in many cases, exploited workers. . . .
[A]s the companies cut corners to bolster the bottom line, law and
morality will fall by the wayside. .. . 4

That may be an extreme expression of the thesis, but it captures

the essence of the argument: private companies, by their very
nature, must put cost before quality and therefore quality will
suffer.
Concern with quality is certainly called for in the field of im-
prisonment, and it is true that excessive concern with costs can
jeopardize quality. However, whether competition and profit-
seeking lead to corner cutting or to improved quality will depend
less on the intrinsic nature of private business than on the nature
of government's oversight and regulation of the contracting pro-
cess. If government becomes caught up in the lowest bidder syn-
drome, competition for business and the need for profit may indeed
cause a reduction in standards. For this reason, concern with cost
savings should not outweigh considerations of quality when eval-
uating programs or proposals.
Experience with competitive contracting in other contexts has
shown that market forces can be used to assure quality as well as
cost containment. However, to enhance this effect it is generally
necessary to pay a premium, in the form of a somewhat higher
than perfectly competitive price. Thus, prison contract proposals
should be compared on a cost-benefit basis, and not on cost alone.
For example, when the Immigration and Naturalization Service
evaluates proposals and bids for its contracted facilities, it uses a
weighting scheme that gives greater weight to quality than to cost.5
"You get what you pay for" is a message that both government
planners and taxpayers need to hear. But that message can be a
promise as well as a threat. The real question is not whether private
Issues of Quality 121

enterprise will be motivated to produce whatever is demanded (and

paid for), but whether it is able to meet demands for higher quality.

Cream Skimming
A frequent objection to contracting for any public service, includ-
ing prison management, is that contractors, instead of producing
quality, will simply take it for themselves. "Skimming the cream"
refers to the possibility that private prisons may be able to decide
which cases they will accept, or that the government will place
only less problematic cases with a contractor, retaining the more
difficult cases itself. A 1986 textbook portrays skimming as char-
acteristic of prison privatization:
Most privatization plans call for skimming off the best of the worst—
the nonserious offenders who can be efficiently processed. Thus,
the correctional enterprise faces the possibility of having to manage
only the most costly, most intractable offenders on a reduced
budget, and with the worsened fiscal and personnel situations that
would result from such a development. 6

There has indeed been selectivity in some of the contracts for

secure confinement facilities. The INS reserves its own facilities
for more difficult cases and screens for escape risks those aliens
placed in the privately contracted Pasadena and San Diego facil-
ities, 7 For the Marion Adjustment Center in Kentucky, a con-
tractual prerelease center, the state allegedly selects only its best
prisoners out of a desire for the contract to succeed.8 Tom Keo-
hane, warden at Eclectic Communications' juvenile facility, re-
ported that their contract with the U.S. Bureau of Prisons (now
expired) effectively allowed them to reject inmates they considered
likely to cause trouble. 9 In a proposal that never became reality,
the Corinthian Corporation of Beria, Ohio, offered to build a 100-
bed minimum security jail to receive only misdemeanants or non-
violent first offenders. A local county official complained that Cor-
inthian's charges, while less than those of other jails in the area,
were still high for that type of prisoner.'"
Not all contracts are oriented toward minor offenders. Most of
the existing county jail contracts obligate the vendor to take what-
ever offenders the sheriff brings in for detention, from the mildest
122 PRIVATE PRISONS

misdemeanants to the most dangerous felons. When CCA officials

offered to take over the entire Tennessee prison system, they did
not want just the cream. In Pennsylvania, the Weaversville Inten-
sive Treatment Unit receives many of the state's most serious
juvenile offenders and has little control over its referrals. 11 At
Okeechobee, in Florida, the Eckerd Foundation incarcerates pre-
dominantly hard-core, serious felony delinquents, with a compli-
cating mix of severely disturbed and first-time offenders. When
cream skimming does occur in private prisons, it will usually be
because the contracting agency or jurisdiction wants it that way,
not because companies are unwilling to provide a full range of
custody and security.
In the government's own prison system, "cream-skimming" goes
on all the time, except that it is called "offender classification."
When it occurs entirely within a single system, it is seen as a
legitimate management tool designed to enhance security, effi-
ciency, effectiveness, and other ends. Ernest van den Haag has
argued persuasively that prisons could be built and run much less
expensively if security classifications were more sharply defined. 12
Though they might balk at some of van den Haag's proposals for
"no-security" prisons, most correctional officials would agree in
principle that it is wasteful to confine inmates in prisons that have
much higher security than they need.
In practice, offender classification operates with uncertain suc-
cess. Predictions of offender behavior are unreliable, there are few
clearly established standards for classification, and the classification
of particular inmates can change at any time. Despite extensive
classification efforts in prisons today, facilities at all levels, from
the most to the least secure, are not homogenous in their popu-
lations. They each have the full range of offender "types."13
Hence, "cream skimming" may not be as easy to demonstrate in
practice as it is to discuss in principle.
The crucial flaw in the charge that privatization will "skim the
profitable cream and leave the losses to the government" is that
it ignores the fact that all imprisonment is paid for by the govern-
ment. Assuming an informed governmental purchaser, the only
way that a contracted prison can keep its contract, let alone make
a profit, is by operating at a lower cost than the government could
operate that same prison, with the same prisoners. This is true
Issues of Quality 123

whether the facility and its inmates are "cream" or "milk." Even
if a prison also generates revenue through a prison industry, work
crews, work-release programs, or businesslike management of the
commissary, this will bring a profit to the contractor only to the
extent that the contractor can do these things more efficiently than
could the government. Otherwise, the government has the option
of running the prison itself and keeping the "profit"—i.e., the
excess payment that it would have made to the contractor. The
only way that "cream skimming" could constitute a relative loss
to the government would be if the government failed to maintain
effective competition between contractors, or between itself and
contractors. If this occurs, the government will waste money
whether contractors take the cream, the milk, or a homogenized
mixture. So long as profit is the result of a gain in efficiency, rather
than the result of inattention, corruption, or other dereliction of
duty on the part of government in the contracting process, it will
be a gain to taxpayers as well as to the contractor.
To illustrate how different this positive view of "cream skim-
ming" is from that of the critics of privatization, consider the
following statement by John Dilulio, to which I have added com-
ments in brackets that offer the positive interpretation:
[I]f extensive privatization does occur, it is likely to create a two-
tiered correctional system [that's a gain of one tier more than there
is now] in which managers on the public tier have an unwelcome
monopoly on [or rather, competitive superiority in] the worst fa-
cilities and the most hard-to-handle inmates. Public prison managers
will then govern facilities where costs, staff turnover, and violence
run high [but not as high as if they were contracted] while productive
inmate activity and staff morale run low [but not as low as if they
were contracted].' 4
As this illustrates, the important question is not who gets to run
the easy facilities or who has to run the tough ones; the question
is who can run which ones better.
Why should it be supposed that private enterprise will be inter-
ested only in the "cream" of corrections? This would, of course,
be true in a tautological sense, if "cream" were defined as whatever
part of corrections is profitable to a contractor. Suppose, however,
that cream and milk are defined on other grounds, such as difficulty
of management. The profit motive attaches a premium to hard
124 PRIVATE PRISONS

work and risk-taking, but that premium is double-edged. Prisons

and prisoners that are hard to manage will be expensive whether
they are in contractors' hands or in the government's. What counts
is the differential. If it is profitable for private companies to handle
troublesome inmates, they will be motivated to do so. If this pop-
ulation can be handled more efficiently by the government, outside
contracting would not be feasible. That is a rational solution and
should not be referred to, pejoratively, as "cream skimming."
Assuming that, indeed, government is better at running the dif-
ficult and expensive facilities while the private sector is better at
running the easy and less expensive ones, the critics of privatization
have one legitimate point concerning "cream skimming": direct
comparison of such apples and oranges makes the private sector
look good and the government look bad. As Dilulio points out:
Where facilities are modern, staff is abundant, populations are
small, and offenders are not hard-core, both public and private
agencies have run safe, clean, cost-effective institutions. Most in-
mates, however, are confined [by government managers] in places
where facilities are huge, the physical plant is in disrepair, trained
personnel are lacking or hard to retain, and populations are large,
racially polarized, and dangerous. 15

This, however, is a basic methodological point about proper com-

parisons, not a valid argument against contracting per se.
"Dumping," the opposite of skimming, may occur if a state tries
to use a contractual facility to selectively rid itself of its most
difficult cases.16 Both skimming and dumping will cause problems
only to the extent that they are noncontractual manipulations of
one party by the other. Such practices can be avoided by identifying
as clearly as possible in the contract the nature of the intended
population and the respective powers of selection on each side.
Related to skimming, but less controversial, is the prospect that
contractors could be used to handle specific categories of prisoners.
Protective custody cases, medical cases, the aged or handicapped,
youthful offenders, and females require separation from the rest
of the adult, male prison population and often have special treat-
ment needs. Protective custody cases, for example, total about 7
percent of all prisoners, or around 35,000 cases nationally, but
require a disproportionate share of resources. Courts often rule
Issues of Quality 125

that the conditions under which they are held, involving solitary
confinement purely for purposes of segregation rather than pun-
ishment, constitute cruel and unusual punishment. 17 Since these
cases are few as a proportion of any one public facility, greater
economies of scale could he achieved if they were contracted to a
private facility, particularly one that served multiple jurisdictions.
Many prison officials view this particular form of prison privati-
zation more favorably than other forms. With specialized pop-
ulations, it is easier to see that a contract can be mutually
advantageous to the government and the contractor.
Hence, these proposals are not characterized as either "skim-
ming" or "dumping."

Mud Slinging
Speaking of "dumping," this seems as good a place as any to
examine a couple of the more unsavory charges that critics of
private prisons have directed at certain members of the industry.
These accusations are important not so much because of the sub-
stantive issues they raise as because they reveal the ideological
intensity to which the debate over private prisons sometimes rises
(or sinks).
The ACLU and some other critics frequently repeat a story
about an entrepreneur who planned to "build a prison on a toxic
dump purchased for $l." 1!i Apparently the ACLU, if it cannot
defeat private prisons by other means, is prepared to smear them
with toxic waste—or at least, with toxic waste stories that mislead
through innuendo.
The truth is that for $1, Beaver County, Pennsylvania, trans-
ferred to Buckingham Security Limited 60 acres of land as part of
an agreement in which, at a cost to itself of $350,000, Buckingham
would clean up, completely remove, and safely dispose of some
toxic industrial chemicals contained in a shallow, monitored stor-
age pit in the woods at the back of the land, and build a prison
on the front of the land (not "on" or even significantly near the
former dump site). 19
The clearly intended point of the story, as presented by the
ACLU, is that only a greedy private company could have such
dirty hands. However, if the ACLU and others want to make the
126 PRIVATE PRISONS

general point that penology and pollution don't mix, they could
find a better target than a private company that wanted to clean
up a toxic waste site in order to replace it with a prison.20
Another story uncritically repeated for its shock value, and to
imply that private wardens will not run humane institutions, is the
story of "axe-handle Charlie." Charles Fenton, co-founder of
Buckingham Security Limited, had a long career in the Federal
Bureau of Prisons, including terms as warden at three major pris-
ons. Several critical sources have referred to an incident in his
career when Fenton was Warden of the penitentiary at Lewisburg,
Pennsylvania.
The facts of the case (Picariello v. Fenton) are readily available.21
Warned that he was about to receive two busloads of unusually
dangerous offenders from the Atlanta penitentiary and that most
of the prisoners had succeeded in removing their shackles during
their trip in the enclosed rear of the bus, Warden Fenton had good
reason to be prepared for the possibility of a serious disturbance.
When he and his men met the bus, they were equipped with riot
batons and pick handles, these being among the riot gear approved
by the Federal Bureau of Prisons. The pick handles were used
only as a show of force and to constrain, not hit, certain offenders.
Some who refused to cooperate during processing were pointed at
or touched lightly with the handles to designate them for removal
to a separate room where they had to lie on the floor. If they
attempted to rise, they were held down with a pick handle.
Private prison critics have repeated many times a distorted ver-
sion of this story in which a jury is said to have found Fenton liable
in a "brutality case" in which inmates were allegedly "beaten with
axe handles while shackled and handcuffed." 22 In reality, the court
and jury found that no inmates were beaten, with axe handles or
anything else, while in restraint or otherwise. In fact, the plaintiffs
did not even claim that they had been "beaten" in the usual sense
of the word; rather, that they were "terrorized" and subjected to
"assault and battery" in the legal sense of "offensive touching."
The U.S. District Court, in its review of the case, found that no
one was beaten or terrorized or offensively touched, that escorting
and restraining the prisoners had involved only "privileged con-
tact," and that all of Warden Fcnton's actions had been reasonable,
except for one. The action found by the court to be "not reason-
Issues of Quality 127

able" (but also "not extreme and outrageous") was the decision
to keep one plaintiffs handcuffs on while he was in his cell, for
three days. The court awarded this plaintiff $200 as full and fair
compensation for his discomfort.
These two stories represent extreme examples of attempts to
discredit private prison companies and to imply that they cannot
be trusted to run decent, humane institutions. I have discussed
them at some length here because of the frequency with which
they are repeated and because of the danger that dramatic stories
like these might foreclose a fair test of whether private companies
can, in fact, run high-quality prisons. These particular stories hap-
pen to be misleading. Sooner or later, however, some scandalous
or horrifying stories about private prisons probably will be proven
true. This will happen not because the prisons are private, but
because corrections is a dangerous and difficult business where it
is easy to make mistakes. And when things go wrong in a prison
the results can be calamitous.
Much of the rest of this chapter will, of necessity, rely on an-
ecdotal evidence regarding the quality of performance by private
prisons. As illustrated above, this type of evidence needs to be
viewed with caution.

General Goals and Standards

Most people, including legislators, judges, and criminal justice

administrators as well as the general public, are ambivalent and
inconsistent in their views of criminal justice. They mix together
utilitarian and nonutilitarian considerations. They want prisons to
be places of rehabilitation, deterrence, incapacitation, and retri-
bution all at once. They offer little consensus on priorities, to
resolve the inevitable conflicts among their many goals and ex-
pectations. Consequently, it will not be any easier to evaluate
private prisons than it has been to evaluate their public counter-
parts, let alone to compare the two.
These observations about conflicting goals, however, indicate
one of the most important contributions that private prisons can
make to the enhancement of quality. Without a contracting pro-
cess, an agency may never face up to the question of just what its
128 PRIVATE PRISONS

purposes and goals are. It may even remain deliberately vague or

ambiguous about its activities in order to satisfy conflicting de-
mands placed upon it. In contrast, jurisdictions concerned about
their contractors' fulfillment of contractual obligations will need
to clarify and specify their goals and performance measures. Every
jurisdiction has laws, regulations, and policies that apply to the
administration of prisons. These can be incorporated into contracts
simply by reference (though the laws would probably apply even
without reference in the contract). In addition, contracts can be
used to spell out standards, regulations, and performance measures
beyond those that apply to all prisons in that jurisdiction.

ACA Standards
In a study for the National Institute of Justice (NIJ), the Council
of State Governments and the Urban Institute report that private
prison contracts often include a requirement to adhere to standards
developed by the American Correctional Association (ACA) and
to seek accreditation by the Commission on Accreditation for Cor-
rections, which uses the ACA standards.23 In some cases, the ad-
herence to ACA standards was proposed by the contractor, rather
than requested by the government agency. Generally, ACA stan-
dards go beyond what is required locally; where they are incom-
patible, or less stringent, the local regulations take precedence.
As noted in the report to NIJ, 24 ACA standards cover such areas
as:

Security and control Work programs

Food service Educational programs
Sanitation and hygiene Recreational activities
Medical and health care Library services
Inmate rules and discipline Records
Inmate rights Personnel issues

Missing from this list, but very important to the ACA, are stan-
dards that relate to crowding, such as the size of cells and recreation
areas or the total floor space per inmate.
The ACA standards all refer to internal conditions, such as
Issues of Quality 129

standards of security and decency, rather than to external results,

such as rehabilitation or crime control. They are concerned pri-
marily with process rather than with outcome and they emphasize
nonutilitarian, rather than utilitarian, criteria. In short, they relate
to the "quality of confinement" and the "quality of life" in prisons,
in terms of criteria important to prisoners, their keepers, and the
public: security, order, safety, space, sanitation, food, recreation,
work, discipline, and programming.
As of early 1987, Corrections Corporation of America had se-
cured accreditation of two of its nine facilities and was preparing
to apply for accreditation of the remaining seven. As CCA pointed
out, this ratio compared favorably to the one-fifth of state and
federal facilities and less than 1 percent of local jails that are
accredited.25 In fairness to the Federal Bureau of Prisons, it should
be noted that 38 of their 47 institutions were accredited for three-
year terms as of 1986.26 However, within the category of state and
local facilities, accreditation is clearly an unusual event.
By mid-1989, CCA had achieved accreditation at six facilities
and was pursuing accreditation at others.27 When their Bay County
jail facility was accredited, the ACA's regional administrator for
standards and accreditation reported that CCA had received one
of the highest ratings of any such facility. 28 Wackenhut Corrections
Corporation received accreditation of its INS detention facility in
Aurora, Colorado, in August 1989.29
The private sector already plays a major role in the supply and
accreditation of health services in prisons. Doyle H. Moore, found-
er of Prison Health Services, which supplies the services of nurses,
doctors, psychiatrists, and dentists, points out that "[o]f the na-
tion's 3,900 jails, only 160 are accredited by the National Com-
mission on Correctional Healthcare and we provide the health
services at 31 of them." 30 The company guarantees to get its clients
accredited within one year or pay a penalty. In 1985, Prison Health
Services did not have to pay any of the $1 million it faced in
potential penalty fees.31
Accreditation, while a useful measure of quality, is not always
reliable. As John Dilulio32 and Stephen Gettinger33 point out, even
accredited public prisons are sometimes crowded, dirty, violent,
or deficient in work or other programs, so there is no guarantee
that private prisons will be free of these problems even when they
130 PRIVATE PRISONS

are accredited. Still, it is likely that accredited institutions, as a

group, will tend to be of higher quality than most others, so ac-
creditation status is still useful as a single (albeit not definitive)
indicator of quality. 34 Moreover, accreditation must be renewed
every three years, so the longer a facility's history of accreditation,
the more reliable that is as an indicator.
American Correctional Association standards generally exceed
constitutional minima, and many are more stringent than those
specified in state statutes. In turn, the standards of private prison
companies sometimes exceed those of the ACA. Staff training is
an example.

Training

The ACA calls for a minimum of 120 hours training for new staff.
CCA requires "at least 160 hours of training for new correctional
officers, a minimum of 40 hours of additional training each year,
and at least 24 hours of training per year for management per-
sonnel."35
As a comparison, new federal Bureau of Prison staff also
undergo 4 weeks (160 hours) of formal training and a minimum
of 40 hours training each year.36 The Immigration and Naturali-
zation Service trains its officers for 6 weeks (240 hours) at the
Federal Law Enforcement Training Center.37 The intensive train-
ing of federal correctional officers, however, may not be typical
of the larger number of state and local correctional workers.38
Florida is unusual in the amount of training it requires of cor-
rectional officers: 360 hours. CCA testified in support of changing
the law to apply to private as well as to public corrections officers.39
On the other hand, perhaps reflecting different standards for ju-
venile as opposed to adult corrections, Florida's contract for pri-
vate operation of the Okeechobee School for Boys requires only
a minimum of 40 hours preservice training, and 40 hours per year
in-service training for all direct child-care employees.40
Corrections officers at the privately owned Marion Adjustment
Center "receive the same training as their state-employed coun-
terparts." 41 At Silverdalc Detention Center, however, because
CCA is seeking accreditation, the private guards "have received
more training than did the county guards." 42
Issues of Quality 131

The most important training, however, may not be the formal,

classroom type. Nor do most prison jobs require higher education,
and therefore a more expensive labor pool. What an effective
officer needs is on-the-job training in the "simple, paramilitary
routine of numbering, counting, checking, looking, monitoring
inmate movement, frisking convicts, searching cells, and so on." 43
In short, officers must be trained in how to run things "by the
book."
To support their warnings about lack of training in the private
sector, some critics have pointed to the case of Danner, Inc., a
one-time private provider of detention space to the INS. Danner
did not train its employees in security and emergency procedures
or in the use of firearms. During an escape attempt, one prisoner
was accidentally shot and killed by a Danner guard. While this
case indicates the importance of training, it is not a fair represen-
tation of correctional contractors. Danner's primary function was
to provide transportation for spare parts and shipping crews in the
port of Houston. As a secondary activity, it supplied security
watchmen for ships.44 Lacking training and proper facilities, Dan-
ner should not have been involved at all in providing detention,
even on a one-time emergency basis. There is a lesson here, but
it does not necessarily generalize to companies whose primary
business is incarceration. Danner, not incidentally, is no longer in
business. The same cannot be said for the many public institutions
of confinement where even worse things have happened.
Training is important to the staff of any prison, public or private.
Indeed, it is too important to be taken for granted in either type
of facility. Contracts for private prisons certainly should require
that corrections officers be at least as well trained as their public
counterparts. But if contracts call for higher training levels, it
merely invites the question, "Why don't we expect this of our
public employees also?" Raising expectations is, after all, a major
purpose of competition, and either side may have something to
learn from the other. 45

Turnover

Even the best training program will be diminished in its contri-

bution to quality if staff turnover is very high. Not only is training
132 PRIVATE PRISONS

an ongoing process, but there is no substitute for experience, which

accumulates over time in an efficient, stable work force.
Unfortunately, high turnover is endemic to corrections, from
top to bottom. Among sheriffs and top administrators in depart-
ments of corrections, there is potential for discontinuity at every
election. In 1987, the average length of service among directors
of adult correctional agencies was 3.6 years.46 John Dilulio, an
insightful student of prison governance, believes that this problem
of fluctuating executive leadership "is largely responsible for the
fact that prisons have been ill-managed, under-managed, or not
managed at all."47
Among the line staff, turnover in corrections has historically
been very high. The figure most commonly cited for turnover
among correctional officers is a national average of about 30 per-
cent. A recent textbook, for example, says: "Turnover is high,
about 28 percent nationally, and absenteeism runs as much as 15
percent in some prisons." 48 The National Manpower Survey of
state correctional institutions found that workers resigned volun-
tarily at an average annual rate of 19 percent and were hired at a
rate of 32 percent. At institutions with 25 to 74 employees, the
resignation rate was the highest—an average of 28 percent—and
the hiring rate averaged 47 percent.49 A more recent source cites
a national average turnover rate of 18 percent a year among cor-
rectional officers in adult state systems, but does not indicate
whether this is a quit or hiring rate. 50
It is not clear yet whether private prisons will be able to reduce
staff turnover significantly or reliably. In an occupation with low
pay and extreme stress, the turnover problem may be intractable.
One of the early secure facility contracts showed at least a short-
term negative impact on turnover. Severe problems of transition
caused a turnover rate that was already exceptionally high among
the state employees to almost double during the first year after
the private takeover. This contract, at the Okeechobee School for
Boys, will be discussed in some detail later in this chapter.
Corrections Corporation of America has reported a facility staff
turnover rate of about 15 percent overall, though it is sometimes
higher during the first year of a contract.51 CCA had its highest
turnover at Silverdale, the prison in Hamilton County, Tennessee.
Three years after that contract went into effect, less than half of
the original staff remained. 52
Issues of Quality 133

Staff turnover is likely to be a problem for private prisons, just

as it is for others. It is too soon to predict whether it will be higher
or lower, and in any case it will probably vary so much over time
and across facilities as to make generalization difficult even when
more data are available. Rates of turnover will be useful in future
research to comparatively evaluate governmental and private pris-
ons. Of course, the meaning of a turnover rate is not self-evident;
it must be interpreted according to information about its cause.
However, turnover is at least relatively easy to measure and it is
one of several factors that can be used to evaluate prison quality.

Credentials

Private companies can bring to corrections experience and exper-

tise that would not otherwise be available to government, partic-
ularly at lower levels. Many of the officers of proprietary prison
companies are veterans of state and federal corrections systems;
others have expertise in management, business, finance, construc-
tion, and the law. As Richard Crane, then a Vice President at
CCA, put it:
If anyone should be using the "no experience" argument it should
be us. Our over 160 years of correctional experience among top
management is surely better than any you will find in most state
corrections systems.53

Among the current or recent executive officers at CCA are two

who were former state directors or commissioners of corrections,
one of whom is a past president of the American Correctional
Association and recipient of the E. R. Cass Correctional Achieve-
ment Award. Three are former state prison wardens. Three hold
doctorates in law; one of these, a specialist in correctional law, is
the former Chief Counsel for the Louisiana Department of Cor-
rections and a legal issues trainer for the National Institute of
Corrections. Several other officers hold degrees in criminal justice
or business. One is an architect, another a Certified Public Ac-
countant. The company's board includes a former Chairman of
the U.S. Board of Parole.
CCA's competitors, large and small, also have personnel with
training and experience. These would include: Wackenhut, the
134 PRIVATE PRISONS

largest provider of private security in the country; Pricor, whose

officers include one who worked 15 years as a state prison budget
director54 and another who is a former Commissioner of Finance
and Administration and Commissioner of General Services for the
State of Tennessee; Buckingham Security Limited, one of whose
co-founders was once warden at the tough Lewisburg and Marion
federal prisons during a 23-year federal corrections career; Be-
havioral Systems Southwest, whose two chief officers had long
careers in the California Department of Corrections; Eclectic Com-
munications, Inc., whose warden at Hidden Valley Ranch worked
27 years in the federal prison system, including stints as Warden
at maximum security institutions.55
Private corrections companies will not necessarily have strong
correctional experience and training among their initial executive
officers. U.S. Corrections Corporation, for example, was founded
by two men with no corrections experience (they hired that ex-
pertise when they got their first contract). However, the contract-
ing process does give a jurisdiction a higher degree of control over
selecting correctional authorities on the basis of experience than
often occurs in the political processes of election and appointment.
Sheriffs, for example, are often elected without prior experience
in law enforcement, let alone in running a jail. A small city
or county has little chance of hiring directly the high level of
competence, training, and experience it can obtain through
contracting.
Ironically, the extensive prior experience of private corrections
officials is sometimes used against them. Critics ask, rhetorically,
what these people can do in the private sector that they couldn't
do in the public sector. Or critics search through the records to
find some problem faced by private corrections officials when they
were in government service. In both cases, critics make the mistake
of attributing to individuals something that is characteristic of a
system. It is hard to serve for long at the level of Warden or high-
er without being involved, directly or indirectly, in a lawsuit,
and critics of private prisons have drawn some ammunition from
this fact.
The case of "axe-handle Charlie" has already been discussed.
Another example is an article in the Nation, which cites a Supreme
Court ruling that the Arkansas prison system in 1978 was operating
Issues of Quality 135

under conditions of cruel and unusual punishment. 56 The Director

of Corrections in Arkansas from 1971 to 1976 was T. Don Hutto,
who later became Vice President of Corrections Corporation of
America, then President of CCA International and a member of
the Board of Directors of CCA. Since most states in recent years
have had jails, prisons, and prison systems under court orders for
conditions of confinement, it is very misleading to imply that this
particular case is indicative of malfeasance on the part of either
CCA or Hutto. To the contrary, Hutto is widely regarded as having
had a distinguished career in corrections. While Director in Ar-
kansas, he helped dismantle that state's armed trusty guard system
by replacing inmate trusties with qualified staff. He has been
elected President of the American Correctional Association and
in 1987 received that Association's E. R. Cass Correctional
Achievement Award. His prior experience, including his record
of reform in a troubled state prison system, is an asset, not a
liability, that he brings to his service in the private sector.

Inspections

In January 1983, the Tennessee Corrections Institute inspected

Hamilton County's prison and found violations of 62 out of 177
regulations. Earlier reports were similarly poor. In February 1985,
four months after the prison was taken over by CCA, the same
inspector found 21 violations, most of which were corrected within
a few months. Subsequent inspections found even fewer faults. 57
A Hamilton County grand jury inspects the prison four times a
year. According to County Auditor Bill McGriff, the grand jury
reports since CCA took over "have all been glowing." 58 In con-
trast, many grand jury reports were critical of the facility under
the county.
On January 30, 1987, a grand jury inspected the New Mexico
State Penitentiary and the privately run Santa Fe County Detention
Facility. After comparing the two institutions, several grand jury
members recommended that private management of the state
prison be considered. They cited economy, efficiency, and clean-
liness as potential benefits.
In Bay County, the State Correctional Internal Inspector found
136 PRIVATE PRISONS

76 violations of Florida codes or standards at the county jail prior

to its takeover by CCA. Four and a half months into the contract,
a subsequent inspection found only nine violations, all of which
were rectified by the time of a follow-up one month later. 59

Private Parts Reflect Public Wholes

Whatever effect contracting may have on quality, it cannot out-

weigh completely the effect of the purchasing agency's own com-
mitment. Quality supplied is determined most strongly by quality
demanded. Private prisons will thus largely reflect the public sys-
tems of which they are a part. This principle was noted by Kevin
Krajick, one of the first observers of the private prison scene, in
a brief comparison of two private juvenile facilities: RCA's Weav-
ersville Intensive Treatment Unit, and the Eckerd Foundation's
Okeechobee School for Boys.60
That comparison is worth elaborating here. It illustrates not so
much a difference between RCA and the Eckerd Foundation—
both had successful track records running small-scale programs for
juveniles—as a difference between the missions assigned to each
by the respective states. Pennsylvania asked RCA to quickly create
a small, intensive treatment facility for hard-core delinquents. The
state was more concerned with speed and quality than with cost.
Florida asked the Eckerd Foundation to take over a large, run-
down, troublesome, underfunded facility, and to run it for even
less money.
The results are described in the two sections that follow.

Pennsylvania, RCA, and Weaversville

Weaversville, as Krajick noted, "is part of a relatively progressive
juvenile system; Pennsylvania has made a commitment to running
small, well-funded institutions." Krajick quotes the manager of
Weaversville as saying, "We're able to do what we do because
somebody up there in the state really cares about these kids. They
give us whatever we need." The state pays RCA $40,000 per year
for each resident, 5 percent of which is RCA's profit. 61
Though many of the residents are serious offenders, the atmo-
Issues of Quality 137

sphere is relaxed. Staff outnumber residents and are well educated

and trained; they include teachers, psychologists, and caseworkers.
There are daily group therapy sessions. Krajick quotes James Fin-
kenauer, a Rutgers University expert on delinquency programs
nationwide, as saying, "Weaversville is better staffed, organized,
and equipped than any program of its size that I know." 62
The employees at Weaversville are enthusiastic about their jobs
in spite of pay that is lower than they might get in a comparable
state job. Indeed, some of those employees gave up state positions
to do the same work in the private sector. They cite as a major
compensation the opportunity to do quality work. "There's far
more creativity in working for a corporation," says Henry Gursky,
project manager at Weaversville. "There is a direction towards
excellence and innovation here that seems to be unique to private
industry," adds Arthur Eisenbuch, Weaversville's clinical
psychologist.63
Pennsylvania state officials, who might be seen as competitors
as well as sponsors and monitors of the Weaversville facility, also
praise the program. Robert H. Sobolevitch, Director of the state
welfare department's Bureau of Group Residential Services, says
that Weaversville is "the best example of a private operation. This
is going to be the national model. It's the hottest thing in correc-
tions." While the program is expensive, Sobolevitch reports that
costs at Weaversville run slightly under spending at Pennsylvania's
comparable state-run juvenile institutions. 64
At least at Weaversville, the fear of critics that companies "dri-
ven by profit" must necessarily sacrifice quality to cut costs has
not materialized.

Florida, Eckerd, and Okeechobee

The private takeover of the Okeechobee School for Boys occurred
within a totally different context from that which produced the
private Weaversville Intensive Treatment Unit. Both Weaversville
and Okeechobee were intended to hold hard-core, serious juvenile
offenders, but there the similarity ends. Weaversville was produced
as a new facility in response to a legal necessity, and it was sup-
ported by a jurisdiction already spending relatively high amounts
on its own facilities. Okeechobee, in contrast, was perhaps the
138 PRIVATE PRISONS

most neglected and deteriorated facility within a system of large

state training schools that had been allowed to run down in antic-
ipation of deinstitutionalization. Prior to contracting, Okeechobee
and two other state training schools were facing lawsuits alleging
cruel and abusive conditions of confinement.
Krajick underscored the contrast between Okeechobee and
Weaversville:
Florida gives Eckerd less than one-half the money per resident at
Okeechobee that Pennsylvania gives to RCA for Weaversville. Ju-
venile-justice experts say that Florida imprisons too many delin-
quents in training schools to begin with, and that all the schools are
too big for their own good. Okeechobee alone is more than twice
the size of the Pennsylvania institutional system. . . . 6 5

With funding from the National Institute of Corrections, a re-

search team for the American Correctional Association undertook
a study to evaluate the effects of privatization on Florida's Okee-
chobee School for Boys.66 Because of late timing, they were unable
to use a before-and-after design for most of their analyses. Instead,
they compared Okeechobee (after the takeover by the Eckerd
Foundation) with the Arthur G. Dozier School for Boys, another
training school of about the same size that the state continued to
run. They gathered and examined data on clients, staff, and man-
agement and administration. In addition, because the school
seemed to be suffering from transition problems at the time of
their original study, they had a consultant return eight months
after the study ended for a more subjective follow-up look.
Of the client measures gathered, only one—number of deten-
tions—produced usable data. The AC A analysis of the data, how-
ever, was badly flawed. They failed to control for significant
population differences on race and seriousness of offense, while
controlling instead for nonsignificant differences on a widely used
offender classification scale. Since ACA found little to report, the
data concerning clients need not be examined here.
When examining the staff data, the research team found that
staff morale and satisfaction at Okeechobee were lower than at
Dozier. There was a smaller questionnaire return rate at Okee-
chobee (62 percent vs. 82 percent), and the average length of
employment was much longer at Dozier, so selection factors may
Issues of Quality 139

have been operating. However, on about half the survey questions,

there were significant differences between Okeechobee and Dozier
staff.
Okeechobee staff more often'.
—saw the environment as threatening to clients and staff;
—saw their work as temporary employment and "just a job";
—were dissatisfied with their jobs and work environment.
They less often:
—saw their institution as well organized and well run;
—felt that they were well trained or effective;
—were optimistic about the future of their charges.
In terms of management, the researchers found significant im-
provements in efficiency of purchasing and greatly increased lati-
tude in personnel procedures. However, Eckerd had serious
problems in running Okeechobee that resulted from initial per-
sonnel policies that later had to be reversed.
When it first took over the facility, Eckerd reduced the staff
from 224.5 to 183 positions. Cottage counselors lived with their
charges 24 hours a day, but their work week was redefined from
40 hours (5 x 8) to 80 hours (5 x 16). This (along with other
factors) caused great dissatisfaction and enormous turnover among
the counselors.
Staff turnover at Okeechobee was very high to begin with, about
87 percent the year before the contract began. During the year of
transition, turnover shot up to 167 percent (on an annual basis),
then fell back to 118 percent the next year. 67 Meanwhile, turnover
at Dozier was very low during all three years (about 10 to 12
percent).
In response to this unrest, Eckerd contributed an extra $236,000
above the contract to bring the staff size back to its precontract
level and reduce the work week. The new work week ranged from
40 hours for detention unit counselors to 56 hours for cottage
counselors (16 hours a day for 7 out of 14 days). At this point,
new Okeechobee cottage counselors, after 90 days, were earning
$12,000 per year for a 56-hour week. This was the same hourly
rate ($4.12) as new Dozier counselors, who received $8,560 per
year for a 40-hour week. In addition, Okeechobee counselors re-
140 PRIVATE PRISONS

ceived free room and board, which Dozier counselors did not.
Nonetheless, staff morale and turnover remained much worse at
Okeechobee than at Dozier. Their total compensation advantage
was apparently not worth the onerous work schedule. Later, work
schedules for cottage counselors were reduced to 8 1/2 hours a
day, 45 hours a week, and all staff were paid overtime beyond 40
hours a week (the initial policy after the takeover provided no
overtime pay rate).
Thus, Eckerd's attempt to cut labor costs by reducing staff size
and increasing work hours was clearly a failure, which soon had
to be reversed. While the year of reduced budget for staff allowed
Eckerd to shift funds into making substantial and permanent im-
provements in the facility's physical plant, those improvements
came at a high cost in terms of staff morale. As the AC A study
concluded:

The Eckerd Foundation's assumption that they could "site-adapt"

their Wilderness Camp program to fit Okeechobee, was unrealistic.
This idea, coupled with the loss of experienced state staff and an
immediate reduction in total number of personnel, left the facility
both understaffed and for the most part in the hands of inexperi-
enced personnel. The concomitant increase in work-hours led to an
even higher rate of personnel turnover and contributed to further
lowered staff morale. 68

There must have been unrest among the residents as well as the
staff, judging from Okeechobee's high escape rate. On the other
hand, escapes were unusually common at Okeechobee prior to the
changeover. The escape rate was 25.6, 27.6, and 25.6 per 100
inmates in the years before, during, and after transfer from gov-
ernment to private hands. This is essentially no change, in spite
of the extremely high turnover among the cottage counselors and
the other problems of a difficult transition year.69 At Dozier, the
escape rates for those three years were 10.6, 8.0, and 6.3 per 100.
Management of the education program at Okeechobee was
made more efficient by the contracting arrangement. Previously,
the Florida Department of Education contracted the education
programs at Okeechobee and Dozier to local school systems (a
community college and the county school board, respectively).
Issues of Quality 141

When Eckerd contracted to run Okeechobee, it also applied for,

and won, the contract for its education program. This resulted in
several efficiencies: educational and institutional policies and dis-
ciplinary actions could be coordinated; educational and other staff
could overlap and supplement each other's activities; and overhead
costs authorized at 20 percent of the education contract could be
largely reinvested in the school. At Dozier, the full amount au-
thorized in the contract for overhead costs was used for that pur-
pose by the contracting county school board. At Okeechobee,
however, only one-sixth of the funds authorized for overhead was
used for that purpose; the remainder was used to pay other ex-
penses at the school. Moreover, after the first year, Eckerd used
some of its education contract funds to pay for telephone, utilities,
and maintenance accounted for by the education program. At
Dozier, the school board paid for none of these costs out of its
contract.
The ACA researchers tallied up an inventory of 36 positive
results and 22 negative results of the Eckerd contract. The list is
too long and detailed to present fully, but a sample from each set
is paraphrased here.

Positive results included:

1. Greater flexibility in personnel procedures.
2. Greater flexibility in purchasing and capitalization.
3. Improved training program.
4. New equipment and refurbishment of all occupied cottages.
5. The education program (which must be contracted by the
state) was brought under the same administration as the rest
of the institution.
6. The Foundation lobbies the legislature for program and cap-
ital funds.
7. Improved security and fire safety.
8. Improved food.
9. Improved clothing for clients.
10. Increased supply of maintenance materials.
1). Quicker identification and resolution of problems.
12. Faster renovation and construction.
142 PRIVATE PRISONS

Negative results included:

1. An initial work schedule for counselors that was too de-
manding, resulting in lowered morale and greatly increased
turnover.
2. A requirement of one to five years employment with the
Foundation before transfer of state pension credit to the
Foundation retirement plan.
3. Difficulty in changing staff attitudes and philosophy.
4. Insufficient maintenance staff.
5. Duplicative accounting requirements.
6. Department of Health and Rehabilitative Services perceived
a loss of information and control.
7. Loss of coordination with the rest of the state j uvenile system.
Several positive results that are omitted above concerned em-
ployee benefits or perquisites. They are worth a summary mention
for two reasons. First, they mitigate somewhat the employee dis-
satisfaction over hours and working conditions that occurred dur-
ing the transfer period. Second, they differ from the negative
experiences of some other contracts, in which employee benefits
constitute a sore point. At Okeechobee, under Eckerd, employees
received: a pension plan at least as good as the state offered;
$10,000 free life insurance; free dental care; free housing, meals,
and clothing; tuition for part-time college study; certain profes-
sional and business expenses for key staff; and Foundation-
sponsored social events.
It should also be noted that among the negative "results" men-
tioned in the report were: short client stay; inappropriate referrals
to the facility; and no control over population size. Since these are
determined by the state, it does not seem accurate to describe
them as the "results of contracting."
The ACA study noted in an afterword that a newly changed
administration at Okeechobee chalked up a large number of re-
markably positive achievements following the first draft of the
ACA report.70 Employee turnover was cut by 75 percent following
a return to normal working hours. Staff size was increased and the
facility's population declined, resulting in greater client control
and a more relaxed atmosphere. The student council met regularly
with the superintendent, a grievance procedure was created, and
Issues of Quality 143

the need for disciplinary segregation declined. Work release and

aftercare programs were developed. Sanitation and maintenance
improved. Staff training was strengthened. Improvements were
noted in the classification system, community relations, and many
other areas.

The Silverdale Survey

While a researcher for the American Bar Foundation, Samuel Jan

Brakel, a law professor at Northern Illinois University, studied the
quality of CCA's program at Silverdale, the Hamilton County
Penal Farm, largely from the perspective of the inmates. 71 By
means of a questionnaire, Brakel asked a sample of 20 inmates 16
questions covering the conditions of their confinement, programs
and services, due process procedures (discipline, grievance, legal
access, and release procedures), and relations with the outside
world.
At the aggregate level, the total responses (20 x 16 = 320) broke
down as follows: 157 positive, 67 ambivalent, and 96 negative.
The areas where CCA was most highly rated by inmates in-
cluded: physical improvement, upkeep, and cleanliness; staff
competence and character; work assignments; chaplain (also coun-
selor) services; requests and grievances; correspondence and tele-
phone; and outside contacts.
Areas with a rough balance of positive and negative ratings were:
safety and security; classification; medical care; food; education;
discipline; and legal access.
Areas with a preponderance of negative ratings were: recrea-
tional facilities and programs; and release procedures. It should
be noted that release decisions and allocation of good time credits
are the responsibility of the county superintendent of corrections
and out of the hands of CCA.
Six of the prisoners were able to compare CCA's management
with prior conditions under county administration, at least on some
of the questions. These inmates provided 28 explicit before-and-
after comparisons: 24 favored CCA and 4 favored the county prior
to CCA.
From prisoners' comments and information gathered indepen-
dently by Brakel, it is evident that much of the before-and-after
144 PRIVATE PRISONS

comparison favoring CCA stems from the fact that conditions un-
der the county had become especially bad. This may be a common
problem in evaluating private prisons. Some of these facilities—
especially the early ones—may be responses to desperate situa-
tions, where a "regression to the mean" effect is possible. On the
other hand, while inheriting a "mess" may make it hard to do
worse, it does not guarantee that a contractor will do better. In-
deed, it makes it harder to do well in absolute terms, because the
contractor also inherits the budget, physical plant, staff, and in-
mates that caused many of the prior problems.
Many inmates at Silverdale were able to make external com-
parisons of their experience under CCA to conditions at the county
jail in downtown Chattanooga or to conditions at the state peni-
tentiary in Nashville. These comparisons totaled 102, with 66 fa-
voring Silverdale, 10 favoring the Chattanooga County Jail, and
26 favoring the state penitentiary. The external comparisons sug-
gest that CCA's improvements are real achievements and not just
regression-to-the-mean effects.
Brakel also cites the contract monitor and former warden as
having identified "in unequivocal terms" five major areas of im-
provement: physical plant, classification, staff treatment of in-
mates, the disciplinary system, and medical services.

A Survey of Contracted Services

It is reasonable to expect that the quality of private sector operation

of entire institutions will correlate with the quality of private sector
delivery of the separate services and programs that make up the
whole. Here there is a longer track record, so we have better
information.
In a recent survey, 52 corrections agencies, representing 38 states
plus the District of Columbia, reported having at least one service
contract with a private vendor. 72 The survey found 32 different
services that were contracted privately, covering "literally every
aspect of institutional operations."73 Further details about benefits
and liabilities were provided by 50 of the agencies. When asked
to identify benefits, 62 percent of the responding agencies men-
tioned better quality of service.74 When agencies were asked to
Issues of Quality 145

state the "single best advantage" of their contracts, half of the

advantages listed related to quality of service and 45 percent related
to cost.75 When asked to identify liabilities, using a checklist of 12
possible problems, the 50 agencies produced a total of 161 com-
plaints. About one-third of these (34 percent) concerned service
quality or delivery. 76
It is too soon to do a formal survey of corrections agencies that
have contracted for the operation of entire facilities; the number
is too small. The closest thing we have to such a review is a report
prepared for the National Institute of Justice by the Council of
State Governments and the Urban Institute. The researchers ob-
tained documents such as contracts, Requests for Proposals
(RFPs), inspection reports, and legislative studies from 22 states.
They also visited privately contracted facilities and interviewed
corporate and government officials. While emphasizing that in-
formation on service quality is very limited, they concluded that
these facilities "are perceived by government agency oversight
officials as being quite satisfactory. We have seen no indication to
date that a government agency has been dissatisfied to any signif-
icant extent with the quality of the service provided."77

Comparison Yes, Double Standard No

Perhaps the major contribution of private prisons is that they will

provide an alternative standard against which to measure public
prisons. This is one potential benefit of private prisons that reform
groups should examine closely. Privatization may force improve-
ments in government operations by defining higher standards and
raising expectations.
However, the existence of an alternative, or comparative, stan-
dard should not lead to the creation of a double standard, in which
the legal requirements for private prisons are higher than for their
public counterparts. It is one thing to anticipate higher quality
from private corrections, or to encourage it with incentives in the
form of renewed contracts, longer renewal periods, or higher fees.
But would be another matter to formally require contracted fa-
cilities to meet higher standards than are required of government
operations. Charles Ring, in what is probably the best-balanced
146 PRIVATE PRISONS

monograph on private prisons, points out what could happen then:

"Plaintiffs' attorneys could argue that state prisons which fail to
meet the minimum requirements imposed on private providers are
in violation of the state's own standards of what constitutes ac-
ceptable conditions and humane care."™ Just as government and
contractor prisons should have the same requirements and pro-
cedures of due process (see Chapter 4), so too should they have
the same legally required minimum standards of care.
One purpose of contracting is to improve the quality of our
prisons, but that purpose will be defeated if demands for excellence
are used as a weapon against the process of contracting itself. That
is what will happen if private prisons are required to meet uniquely
high standards of quality and accountability while at the same time
producing substantial cost savings relative to governmental oper-
ations that are not being held to the same standards. It is reasonable
to ask private prisons to live up to their own claims that they can
do better for less, but the comparison must be fair. The insistence
that they meet higher standards for less than what it costs govern-
ment to meet much lower standards ultimately becomes an im-
possible demand. A major function of competition is to spur all
parties on to higher performance. That purpose would be under-
mined if the private sector were handicapped by the application
of double standards.
One promoter of double standards for contracted versus gov-
ernment-run prisons is Professor Ira Robbins. In the Model Con-
tract and Model Statute that he has presented to the American
Bar Association, Robbins proposes criteria for private prisons that
exceed what is required of public facilities.79 To give just one
example here, Robbins incorporates into his Model Contract 22
standards on staff training developed by the American Correc-
tional Association. The ACA identifies the majority of those stan-
dards as nonmandatory for accreditation, but Robbins upgrades
them, making all 22 mandatory under the Model Contract.80
In my view, a "model" contract or statute would be one that
attempted to balance conflicting interests and values, no one of
which can be maximized without sacrificing other legitimate con-
cerns. I refer here not just to the conflicting special interests of
contractors, contracting agencies, prisoners, and the public, but
also to the conflict between more general interests and values:
Issues of Quality 147

accountability versus flexibility, security versus due process, quality

versus economy, innovation versus regulation or restraint, and so
on. Robbins' Model Contract and Model Statute are not well bal-
anced in this sense.81 The guiding principle running throughout his
proposals is the necessity to maximize one value: accountability.
Now that is a very important value, but Robbins' approach to it
is one-sided. He assumes that accountability is inherently prob-
lematic for private prisons, while government agencies are, by their
very nature, more accountable and considerate of the interests of
inmates and the public. As a result of this perceived difference,
Robbins seeks to impose constraints, liabilities, encumbrances, and
costs on private prisons far beyond those that exist for government
prisons.
To be sure, contractual operations require certain mechanisms
of accountability that are different from those used for noncon-
tractual operations. However, this holds for intergovernmental
contracting as well, but Robbins is silent on the need for model
contracts and statutes to assure accountability when one govern-
mental jurisdiction contracts with another. Robbins requires of
private contractors not just different mechanisms (because they are
contractors) but also higher standards (because they are private)
for accountability, liability, insurance, standards of operation,
monitoring, regulation, and economic efficiency.
When Robbins, in the name of accountability, seeks to impose
on private prisons a set of legally mandated requirements that do
not exist for government prisons, he creates special liberty interests
for privately, confined prisoners, and thus new grounds for legal
actions against their keepers.82 This double standard could make
private prisons so relatively expensive, so specially encumbered,
and so uniquely vulnerable to lawsuits that they cannot survive.83

Conclusion

There has been almost no systematic empirical research comparing

private and government-run prisons in terms of quality. Until such
studies are available, we must rely on the type of anecdotal evi-
dence that has been considered in this chapter. Evidence of this
nature should be evaluated with caution. Government-run prisons
148 PRIVATE PRISONS

have been studied extensively for decades. About all that can be
said of them as a class is that they vary enormously in just about
every area that relates to the concept of quality.
Private prisons, because they are still unusual, are particularly
susceptible to the effects of both positive and negative publicity.
Much of the literature in the first few years was rather compli-
mentary and optimistic. As the novelty has worn off, critical stories
are becoming more common, as is often the case. While neither
puff pieces nor hatchet jobs provide a basis for valid conclusions,
these types of stories will probably predominate for some time to
come. The governmental prison system has survived numerous
scandals, along with endless and scathing denouncements, basically
because it is the only system we've got. Private prisons have never
had the advantage of monopoly, and thus will be much more vul-
nerable to negative publicity. There is little danger that jurisdic-
tions will commit themselves completely to private prisons on the
basis of early positive reports. It is possible, though, that some
jurisdictions might completely rule out contractual arrangements
on the basis of early scandals. No matter how convincing stories
of "success" or "failure" may appear, they should not form the
basis of any but the most narrow and tentative of conclusions.
My own conclusion so far is simply that private prisons will fall
variously within the same range of quality as do those run by
government employees. Some private prisons will be better than
some public prisons, and vice versa. As long as there are at least
some jurisdictions where prisons and jails might be improved by
competition from the private sector, that option ought not to be
categorically ruled out. Moreover, even jurisdictions with rela-
tively high quality prisons ought to remain open to competition.
7
Issues of Quantity

Many of the people who are opposed to private prisons take

this stance because they are opposed to prisons generally. They
fear that cheaper and more efficient prisons will mean more im-
prisonment. Few of these critics object to the privately adminis-
tered, profit-making community correctional programs which
serve as alternatives to prison. 1 Rather, they are concerned that
private prisons will distract attention from these and other alter-
natives.2 It is ACLU policy that "new prison or jail capacity
should be increased only after existing unconstitutional condi-
tions have been remedied, if ever." 3 Moreover, capacity should
not be expanded until all alternatives to imprisonment have
been exhausted.4
Jerome Miller, of the National Center on Institutions and
Alternatives, lobbies hard for alternatives to imprisonment.
Miller has voiced strong doubts about for profit jails for
adults 5 and his organization's newsletter has attacked and
ridiculed the idea of proprietary prisons.6 However, Miller
saw nothing wrong with privately contracting out virtually
all of the Massachusetts juvenile corrections system, a revo-
lutionary change for which he was personally responsible.
The private exercise of state authority to carry out correctional
practices is acceptable to Miller, but only where the aim is
to provide treatment rather than increase the prison popu-
lation.

149
150 PRIVATE PRISONS

"Capacity Drives Use"

If, through greater efficiency, more prisons are built, argue the
critics, then they will be filled simply because they are there. "As
long as you build prisons," a program associate at the Edna
McConnell Clark Foundation declared, "you'll fill them. It's a self-
fulfilling prophecy." 7
For this argument to have weight as an objection to privatization,
it would have to be shown that capacity which is produced, owned,
and operated by private agents on a contingent basis under contract
to a government is as likely to be used by the government "because
it is there" as is capacity produced, owned, and operated by the
government itself. On the contrary, however, the government is
more likely to use up "excess" capacity that it already owns than
it is to purchase "unneeded" capacity from a contractor. Owners
may feel compelled to find uses for their idle properties; renters,
however, do not increase their holdings just because new prop-
erties are on the market. Thus, the argument that capacity drives
use would be a better argument against government ownership of
prisons than against privatization.
Nevertheless, those who object to private prisons speak of "the
solid evidence which supports the position that the criminal justice
system inexorably operates to fill every available cell regardless of
the need [or] wisdom of their use." 8 The most widely cited piece
of this "evidence" was supplied by Abt Associates9 in a time-series
analysis relating state prison populations to changes in capacity.
The researchers concluded that all net increases in prison capacity
will be entirely absorbed within two years. When a team of experts
at Carnegie-Mellon University headed by Alfred Blumstein, an
eminent authority on prison population research,10 examined the
Abt analysis they concluded that it contained a crucially incorrect
calculation and was extremely sensitive to the inclusion of two
extreme data points. In addition, they criticized the statistical
model as overly simplistic because it omitted important variables
and ignored simultaneous (two-way) causal effects. Blumstein and
his associates concluded that existing evidence does not support
the hypothesis that prison capacity directly determines the degree
of prison use. While the Blumstein group emphasized that the
Issues of Quantity 151

evidence does not disprove Abt's hypothesis either, they point out
that during the 1960s, when there was considerable spare capacity,
prison populations declined even though crime was increasing."
Obviously, capacity is not unrelated to the degree of prison use,
but expansion of capacity does not automatically produce a rise in
imprisonment. Likewise, nonexpansion does not automatically
prevent an increase in imprisonment; if it did, there would be no
overcrowding. For over two decades, prison reform groups have
advocated a moratorium on all new prison construction, hoping
that lack of capacity would curtail use. That strategy has backfired,
and increasing numbers of prisoners are paying the price in terms
of overcrowding and deteriorating physical conditions.
Given current levels of crowding, oversupply of prisons seems
like a fanciful concern, but that prospect should be viewed as a
hope, not a danger. Overconstruction would allow us to tear down
many ancient and run-down monstrosities that have been tolerated
up to now only because there are no alternatives. Still, supporters
of privatization, who tend also to ascribe to supply-side theories
of economics, ought not to dismiss completely the warning that
private production of public services can increase the demand for
those services.12 Indeed, both liberals and conservatives sometimes
support particular forms of privatization precisely because they
want to see more rather than less activity in those specific areas.
Ideally, in the area of criminal justice, demand should drive
supply, rather than vice versa. Given the reality of scarce resources,
that ideal can never be achieved fully, but any increase in efficiency
is a step in the right direction. Greater efficiency creates more
room for the system to respond to the requirements of justice and
not just to the reality of scarcity. Whether we opt for more im-
prisonment or for less, our decision ought to be based as much as
possible on a sound philosophy of punishment, not merely on the
availability or absence of beds. We, the consumers of justice, must
see to it that our demand for punishment is relatively inelastic with
respect to supply, so that oversupply, if it does occur, will simply
drive down the price. To imprison someone merely because there
is a bed available clearly would be unjust; equally unjust, however,
is the failure to imprison serious felons because space is lacking.
If both oversupply and undersupply can lead to injustice, we
should, in principle, err on the side of oversupply, although this
152 PRIVATE PRISONS

is not likely to happen for some time to come. Distortions of supply

do not relate symmetrically to distortions of justice. While un-
dersupply leads inevitably to either overcrowding or underimpris-
onment, oversupply does not necessitate overimprisonment. None-
theless, oversupply is also undesirable, if only because it is
wasteful. What is needed, for both justice and efficiency, inflexi-
bility of supply, which the private sector is best equipped to
provide.

Vested Interests and Lobbying

Advocates of privatization generally want to see total government

spending go down. Contracting, however, even if it reduces unit
costs, does not necessarily reduce total spending. This might be
the result of effective lobbying on the part of contractors, who
have a vested interest in selling as much of their product or service
as possible. Critics of private prisons are right to worry about
potential distortions in the justice market as a result of political
influence by parties with vested interests.
Critics go overboard, however, when they imply that contracting
will create special interests where none existed before, and when
they exaggerate the power, influence, and supposedly purely self-
serving character of private businesses. Some critics portray private
prison companies as prepared to hang onto each prisoner for as
long as possible, by denying inmates good time credits and un-
dermining their bids for parole, in order to wring out extra per
diem payments. Those who fear that this vested interest will per-
vade the corrections process ask whether prospective jurors who
own any stock in a prison company could be disqualified from
felony cases.13
Opponents of private prisons apparently believe that contractors
will be able to manipulate legislatures and public opinion in order
to ensure more imprisonment and longer sentences. They speak
of the ability of contractors to "buy unlimited television and news-
paper time and space, reach the highest political levels and unleash
upon America a program to convince the public to lock up more
and more people for longer and longer periods, carefully con-
cealing that their motive is profit." 14 Harmon Wray warns that
Issues of Quantity 153

"the most critical flaw in the privatization movement is that it is

inherently expansionist." Wray cites the Vera Institute's Michael
Smith as noting that "the private sector has an enormous invest-
ment in stimulating demand" and he cites Ken Schoen, Director
of the Justice Program at the Edna McConnell Clark Foundation,
who predicts: "Private operators whose growth depends upon an
expanding prison population may push for ever harsher sentences.
. . . And the taxpayers will finance the profit-makers while double-
locking their doors at night." 15

Special vs. Public Interests

Lobbying is an intrinsic and (within certain limits) positive aspect

of representative government. Whatever government does, and
however it does it—whether directly or through contracted
agents—the policy-making process will be subject to lobbying by
all sorts of groups representing any number of different interests.
These groups are not always motivated by economics. Personal,
moral, philosophical, ideological, or reputational investments can
be at stake, with the same biasing effect as economics and the
same potential conflict with the general public interest. Moreover,
it should be noted that private, for-profit companies are not the
only ones with economic interests; such concerns also affect gov-
ernment agencies and nonprofit organizations.
For example, the National Center on Institutions and Alterna-
tives has a strong philosophical and moral commitment against
incarceration and in favor of alternative sanctions. It also tends to
identify with offenders. Yet the NCIA is the largest private pro-
vider of presentence reports in the country, 16 and in that role it is
not completely free to promote its own values exclusively. Surely
it would lose business as well as credibility if all of its presentence
reports argued against imprisonment. Here, then, is a case where
an economic interest acts to restrain an ideological interest and to
force an agency to focus more on the concerns of the general public.
In short, its economic interest keeps NCIA honest and objective,
in the sense that it is forced to consider values and definitions of
justice other than just its own. This same dynamic can also operate
on a profit-making incarcerative organization.
154 PRIVATE PRISONS

While many Americans denigrate them as "special interest"

groups, lobbyists and political action committees often serve
shared interests and can be a creative force for change. 17 Correc-
tions experts have long bemoaned the relative lack of organized
constituencies for corrections, other than correctional employee
unions. One noted expert, John P. Conrad, has concluded glumly:
"A strong and durable constituency for corrections cannot be
built. . . . The reason is simple. Corrections is unrelated to citizens'
personal interests." 18 Adding new parties with incentives to par-
ticipate in the formulation of social policy, as well as incentives to
innovate and experiment in the details of policy implementation,
offers at least an opportunity to shake up a generally stagnant
enterprise.
Whether industry-sponsored policy changes will be good or bad
will depend on whom you ask. Predictably, they will please some
and displease others. As noted, critics who expect private prisons
to lobby for more imprisonment view that prospect negatively and
say that it is not in the public interest. Most of the general public,
however, would dispute this claim. Those who said that the courts
were not harsh enough rose steadily from 48.9 percent in 1965 to
84.9 percent in 1978.19 From 1980 to 1986, between 82 and 86
percent of Americans advocated stiffer penalties for lawbreakers.20
In a recent national poll, 71 percent of the public said that a jail
or prison term is the most appropriate sanction for a broad range
of crimes, including rape, robbery, assault, burglary, theft, prop-
erty damage, drunk driving, and drug offenses.21
It should not be supposed, however, that private corrections
corporations would lobby only for longer sentences. Contractors
can be expected to push for higher correctional budgets for re-
habilitation, education, job training, work programs, medical ser-
vices, conjugal visiting facilities, better food, and any number of
other improvements in prison environments and programs. The
more these are mandated by the state, the more they can be offered
by private prisons as well. Contractors may lobby for capital outlays
to renovate deteriorating facilities, then bid for the work if it is
authorized. As noted above, one major obstacle to significant
change of any sort in corrections has been the fact that the inter-
ested and active constituency has always been so narrow and weak.
Thus, reform-oriented corrections professionals may find in the
Issues of Quantity 155

private sector an ally as well as a competitor, and should welcome

this expansion of the constituency.

Private vs. Public Lobbies

The private sector will not bring politics and lobbies to a field
where none now exists. Corrections is already a political arena.
Wherever governments appropriate and redistribute resources, in-
terest groups of all sorts attempt to influence the allocation of
those resources. Among the interest groups now actively involved
in correctional issues are: corrections officials and their associa-
tions, correctional employee unions and associations, police and
their organizations, attorneys and their associations, crime com-
missions, funding agencies, ex-offender organizations, the Amer-
ican Civil Liberties Union and its state chapters, and a number of
other prison reform groups.
Criminal law and criminal justice generally, not just corrections,
are political, and strongly influenced in their development by spe-
cial interest groups. This fact is recognized by analysts all across
the political spectrum, from right-wing libertarians to the radical
left. 22 Bruce Benson, an economist at Florida State University,
has summarized the special interest theory of law in six prop-
ositions:
1. Legislators and other public officials supply and enforce laws
demanded by politically powerful special interest groups.
2. Effective interest groups tend to be small relative to the pop-
ulation that might be affected by the government activity in
question, because of the difficulty of organizing and making
decisions when many individuals are involved.
3. Political power may take the form of votes, money, or the
ability to disrupt a politically stable situation with such things
as strikes, violence, and disorder.
4. Interest group members may be very self-interested but they
also may be very well intentioned individuals seeking what
they feel is "good" for the society as a whole (although their
definition of "good" is typically affected to some extent by
their personal circumstances).
156 PRIVATE PRISONS

5. Bureaucrats (i.e., police, prosecuting [sic] judges, prison of-

ficials) constitute interest groups that benefit when laws re-
quiring enforcement are passed.
6. Government responds to interest group pressure by favoring
the most powerful group (or groups). . . [however,] conflict-
ing demands tend to lead to compromise, with no group
completely satisfied or dissatisfied.23
In support of the special interest theory of the development of
law and operation of the justice system, Benson24 cites empirical
research by Richard Berk, Harold Brackman, and Selma Lesser25
on changes in the California Penal Code from 1955 to 1971. The
researchers found that just a few criminal justice lobbies domi-
nated, and often practically dictated, criminal justice legislation.
Prominent among these lobbies were the California Peace Officers
Association, the American Civil Liberties Union, the state Bar
Association, and the Friends Committee on Legislation. Benson
notes that:
[An] important finding of the Berk-Brackman-Lesser study was that
"public opinion" played no identifiable role in Penal Code revision.
The study deemed "inescapable" the conclusion that criminal law
was enacted for the benefit of interest groups rather than for the
public good.2''
The result, and often the aim, of this lobbying by bureaucrats and
nonprofit organizations has been to expand the size, scope, power,
and resources of the criminal justice monolith.
Public employee organizations and unions that promote the cam-
paigns of those who promise them larger budgets, more jobs, and
higher salaries are doing exactly what they say private contractors
will do. Thus it is ironic to hear the President of the American
Federation of Government Employees Council of Prison Locals
(which represents Federal Bureau of Prison workers) declare that:
"For the first time, it is in someone's self-interest to foster and
encourage incarceration. It does not take an accountant to figure
out that they will act in their self-interest." 27
Of the one million members of the AFSCME, 50,000 are cor-
rections employees.28 Organized union labor accounts for some of
the strongest lobbies and most lavishly funded political action com-
mittees in the country. As Morgan Reynolds notes: "Unions can
Issues of Quantity 157

supply formidable organization, campaign money, workers, and

direct influence over some members' votes. Public employees par-
ticipate in elections at substantially higher rates than the general
citizenry does, thereby forming a more potent voting bloc than
their share of the work force might suggest." 29
Correctional workers' unions have used collective bargaining to
restrict the ability of administrators to contract for community-
based programs.30 These agreements preserve the jobs of institu-
tional employees and, as a side effect, keep the populations of the
institutions higher than administrators may desire. Those who com-
plain that private prison businesses, in pursuit of profits, will dis-
courage the development and use of alternatives to prison, have
been silent when organized labor, in pursuit of its own interests,
has done the same. Among the policies that correctional unions
have opposed, are: deinstitutionalization, contracting out for ser-
vices, the use of volunteers, establishment or enhancement of due
process rights for inmates, increased community and family in-
volvement in institutional programs, prisoner furloughs, work or
educational release programs, and a shift of emphasis from custody
to treatment. 31 A study of correctional unions in 17 states found
that

Employee organization lobbying, publicity, lawsuits, and job actions

. . . have often been attempts to counteract progressive correctional
programs such as community-based facilities and to reestablish an
emphasis on custody. Another feature of this campaign is that cor-
rectional unions have advocated longer prison terms and more strin-
gent parole policies—for example, an increase in the minimum term
an inmate must serve before he can become eligible for parole.12

One final example of a public-sector lobby is particularly relevant

to the charge that privatization is inherently expansionistic. As
noted in chapter 5, most referenda on bond issues for prison con-
struction yield a mixed message from the general public. People
want more prisons but are unwilling to pay higher taxes to finance
new long-term debt. Such referenda are often defeated by a well-
funded advertising campaign mounted by organized and powerful
opponents of new prison construction. Supporters were better or-
ganized than usual, however, in a 1981 referendum for a bond
issue to build new prisons and expand the system in New York.
158 PRIVATE PRISONS

There, the biggest spender, shelling out $158,000 to lobby for the
bond issue, was the New York Department of Correctional Ser-
vices.33 The referendum was defeated, but it "had no impact on
the construction plans and timetable of the Department of Cor-
rectional Services," whose Commissioner exhorted his employees
to work instead on the regular appropriations process, and "sup-
port the continued expansion of our state prisons with the appro-
priations in [the next] budget." H
Against this background, to argue that privatization will intro-
duce an element of self-interest not otherwise found in the criminal
justice system is absurd. This is not to deny that private firms will
have concerns that may sometimes conflict with the public interest
(or with other conflicting interests competing to be defined as "the
public interest"). The point is that these conflicts of interest are
not unique to the private or profit-making sector and therefore do
not provide legitimate grounds for excluding that sector from par-
ticipation in the formulation of criminal justice policy.
Private prison companies may very well become important actors
in the field of criminal justice policy, just as public actors are today.
Indeed, merely opening up the field has taken some political mus-
cle. As Harmon Wray reports, one lobbyist for Corrections Cor-
poration of America managed the campaigns of Tennessee
Governor Lamar Alexander and served as his Chief of Staff. Alex-
ander has been strongly supportive of prison privatization. "Others
in the CCA-Alexander circle include CCA stockholders who are
current and former Alexander cabinet officers, CCA administra-
tors who are former state GOP chairpersons, a CCA lobbyist who
was a Democratic state senator, and two prominent public figures
[the governor's wife and the state's Speaker of the House] who in
1985 sold their CCA stock to avoid the appearance of conflict of
interest."35
Competition within the industry can serve to dilute, rather than
concentrate, this political power. For example, one of CCA's com-
petitors, Corrections Associates, Inc. (now called Pricor), hired
Tennessee's Finance Commissioner Hubert McCullough away
from his state office, to become its new chairman and CEO.36
With or without the participation of profit-making enterprises,
it is important to maintain democratic political controls to prevent
any one set of narrowly defined interests from squeezing out all
Issues of Quantity 159

others and from asserting an imprimatur over designation of "the

public interest." We cannot prevent "lobbying" (though it may
not always be called that) by nonprofit organizations, government
agencies, public employee unions, or commercial companies, any
of whose agenda may or may not coincide with the public interest.
However, allowing these groups to compete, both in the provision
of a service and in the public formulation of policy for the provision
of that service, is a better method of protecting the public interest
than is granting a monopoly to one particular service provider.
"Pluralism" is what we call the condition in which the "public
interest" must be sorted out from among competing definitions
and claims. We regulate the competition to preserve the integrity
of the processes by which we determine (1) what services are most
in the public interest and (2) who best can supply them. But it is
competition, even more than regulation, that prevents concentra-
tion of power in either area.

Expansion vs. Flexibility and Diversification

The charge that private corrections companies, to remain profit-

able, will require an ever-expanding prison population, is based
on a simplistic caricature of capitalism that reflects a misconception
of the nature of business under conditions of competition. To be
sure, most profit-makers do attempt to drum up business. On the
whole, however, businesses succeed not by stimulating spurious
demand, but by accurately anticipating both the nature and the
level of real demand. This is true whether demand is rising or
falling, and particularly if it is shifting. The ability to predict and
respond is far more important to a business than any supposed
power to artificially stimulate demand.
The supposed need for full occupancy is often referred to by
commentators on private prisons as the "Hilton Inn mentality." 37
The phrase is an unfortunate one when used by critics, and not
just because of the high quality associated with the Hilton.38 Many
hotels make profits without continuous full occupancy. It is not
crucial that demand exactly meet supply. It is far more important
that suppliers be able to predict demand and have the flexibility
to shift resources and to adjust prices accordingly.
160 PRIVATE PRISONS

Still, it is clear that private facilities will have an interest in at

least stable, if not maximum, occupancy levels. Rather than as-
suming that contractors will respond insidiously, by violating the
rights of inmates or by stampeding the legislature and judiciary
into greater use of prisons, critics should recognize that efficient
use of contracted facilities is also in the public's interest.
One response based on that recognition would be to write con-
tracts that allow a vendor to temporarily fill a contracting agency's
excess space by accepting inmates from another agency or juris-
diction. A higher price could be charged for these "extra" inmates,
with the excess shared between the vendor and the primary con-
tracting agency. This arrangement is part of the contract between
CCA and Bay County, Florida. Likewise, in Santa Fe, New Mex-
ico, CCA runs the county jail under contract to Santa Fe county
but has supplemental contracts to hold prisoners for the city of
Santa Fe, the Federal Bureau of Prisons, the U.S. Marshall's Ser-
vice, and some other counties and cities in New Mexico.39 In Hous-
ton, CCA has a supplemental contract with the Texas Board of
Pardons and Paroles to hold preparole inmates in the facility that
it runs as a detention center under a primary contract to the INS.40
These supplemental contracts benefit all parties by increasing
efficiency. Jail space is a scarce resource, but the problem is more
one of distribution than of aggregate supply. While some jails
are desperately overcrowded, many others, especially small to
medium-sized local facilities, have plenty of empty cells. Every
year, some jails close for lack of need. The flexibility of private
contracts can do much to allocate resources to where they are
needed.
Another lesson to be learned from the Hilton Inn mentality is
one that is far more applicable to a private enterprise than to a
government agency. That lesson is: diversification. Because com-
mercial enterprises survive and prosper by accurately anticipating
and responding to shifts in demand, we should not assume that
correctional corporations will always be motivated to lobby for
expansion of high-security facilities. Such corporations can be ex-
pected to diversify both within and outside of corrections. If they
anticipate or sense a shift in public demand toward community
treatment, electronic monitoring, or other alternatives to incar-
ceration—or, indeed, a decline in total demand for correctional
Issues of Quantity 161

services of any type, through falling crime rates or decriminali-

zation—they will prosper more by shifting resources to other ac-
tivities than by attempting to buck the trend. Right now, there is
a genuine unmet demand for imprisonment. However, if the de-
mand for alternatives to prison increases, commercial companies
should be able to respond rapidly to such a shift. One INS detention
contractor, for example, also provides (and aggressively markets)
electronic monitoring services as an alternative to jail.
To sum up briefly, political and economic pressures to expand
the use of imprisonment are neither inherent in nor unique to the
involvement of private enterprise in corrections. Vested interests
in promoting either imprisonment or alternatives to imprisonment
can be found in both public- and private-sector organizations. The
public interest, whether it lies in greater or lesser use of impris-
onment, is therefore no more naturally allied with the special
interests of government agencies, government managers, and gov-
ernment employees, than it is with those of outside contractors.
8
Issues of Flexibility

A major promise of proprietary prisons is that they will bring with

them the flexibility and innovation typical of entrepreneurial ac-
tivity and commonly found among small businesses. They will have
less of the bureaucracy that tends to preserve the status quo in
government and other large organizations. As enterprises subject
to competition and market discipline, they will be more free to
engage in purchasing and subcontracting without the bureaucratic
regulations that are needed to control government spending. They
will be able to respond more rapidly to the changing conditions
and needs of corrections, and act more quickly to correct mistakes
resulting from inaccurate predictions or faulty policies. They will
add new expertise and specialized skills, and promote creativity
and enthusiasm by bringing in "new blood" and new ideas more
often than is possible under civil service.
Joan Mullen, an expert on privatization of corrections at Abt
Associates, sees enhanced flexibility as an even greater potential
benefit than cost containment:
The notion that private organizations can provide more for less is
undeniably attractive, but probably unrealistic. The greatest prom-
ise of the private sector may instead lie in its capacity to satisfy
objectives that might be difficult if not impossible to achieve in the
public sector—introducing public sector managers to the principles
of competitive business; quickly mobilizing facilities and manpower
to meet immediate needs; rapidly adapting services to changing
market circumstances; experimenting with new practices; or satis-

162
Issues of Flexibility 163
tying special needs with an economy of scale not possible in a single
public sector jurisdiction.'
In short, one of the major strengths claimed for private prisons
is that their greater management flexibility and more rapid speed
of response will promote both minor innovations and major program
changes, whether through initiation, expansion, contraction, or
termination.

Rigidity vs. Clarity

Critics of contracting argue that it is impossible to write a contract

that is as broad and flexible as the mission of a public agency needs
to be. Contractors may be unwilling to go beyond the provisions
of their contracts. Renegotiating and changing contracts is time-
consuming and terminating a contract is often very difficult. Thus,
contracts produce their own form of rigidity, and it will be harder
for the government, under contracting, to order and control mar-
ginal changes. As two sociologists put it: "How can innovation be
expected from a contractor who will not offer one iota more than
the contract calls for and who does not have even the limited
flexibility of the state and national correctional programs?" 2 A
political scientist argues:
Writing and enforcing highly detailed contracts may help to guar-
antee accountability, but it does so only at the expense of admin-
istrative flexibility. . . . As a public corrections official with years of
experience in these matters stated, "Either the contractors will be
allowed to run wild as they did in the old days, or we'll make the
specifications, regulations, and monitoring so rigid that the firms
will become as bureaucratized and inefficient as we are. . . . " 3
This all-or-nothing reasoning, however, is clearly a false dilemma.
Private prisons can have latitude in some respects without being
given free rein in all respects.
A government prison that is run by unionized public employees
may suffer from its own kind of systematized rigidity as a result
of a management-labor contract, rather than a government-
management contract. For example, one study of correctional
unionism reported that: "Correctional administrators interviewed
during our research charge that the provisions requiring that as-
164 PRIVATE PRISONS

signments and promotions be based on seniority have done more

to disrupt the operation of correctional institutions than any other
kind of provision." 4
It is fair to predict that private prison companies will use their
contracts to limit what is demanded of them. Within reason, that
is one of the intended functions of a contract. But what critics see
as rigidity, proponents see as specificity and clarity.
Government agencies can issue grandiose mission statements
without fear of penalty if they fail to live up to all their promises.
A contract, on the other hand, is meant to cut both ways: to specify
as well as to limit what is required. The purpose of a contract is
to facilitate, not to restrict, the accomplishment of certain goals.
Contracts force the government to confront the question of what
are its goals, standards, and criteria of performance. All parties—
contractor, government, and taxpayers—benefit from that process.
Contracts protect the government by specifying in advance the
nature and quality of the service it is to receive. At the same time,
contracts also protect the contractor from unanticipated and un-
reasonable demands.
Contracts will not necessarily protect a vendor from reasonable
demands, even if they were previously unspecified. A more flexible
contractor will have a competitive edge, when it comes time for
contract renewal, over one who uses the contract to behave in a
rigid and truly unreasonable fashion. The government's control
over contract renewal and rebidding will tend to prevent abusively
rigid interpretations of contract provisions by the vendor. To ex-
ercise this control, however, the government will have to be pre-
pared to deal with the problem of terminations and transitions.

Contract Terminations and Transitions

In exchange for the flexibility of contracting, government must be

prepared to minimize the risk of interruptions in service when
contracts change. When a contractor takes over a government
operation or when the government resumes control or shifts it to
another contractor through competitive rebidding, many details
must be anticipated. Capital investments must be protected. Prop-
erty and equipment must be transferred. Continuity of records
Issues of Flexibility 165

must be maintained. The interests of current employees must be

considered. In short, terminations and transfers of contracts, even
on a nonemergency basis, entail components not found under nor-
mal governmental continuity.
The American Federation of State, County, and Municipal Em-
ployees maintains a file of instances where it has been necessary
to terminate contracts for public services because they proved to
be too costly or unsatisfactory/ These examples illustrate the prob-
lems of transition referred to above, but they also show something
else. They indicate that it is feasible to terminate a public/private
contract. In contrast, how feasible would it be to replace or halt
the activities of a government agency staffed by tenured and union-
ized civil servants whose services were found to be unsatisfactory?
It may not always be easy to terminate a contract, but experience
has shown that it is nearly impossible to terminate a government
agency, even one supposedly made mortal by a sunset law.6 Thus,
a contract situation can be seen as preferable if an agency is poorly
managed or staffed. Where contracting, and therefore competi-
tion, is possible, bad management (either public or private) is less
likely to become entrenched, because a surgical solution is at hand.
The Immigration and Naturalization Service has successfully ter-
minated and replaced several of its contracts for detention facili-
ties. Robert Schmidt, formerly Supervisor of Detention Services
for the INS, supervised ten private contracts for six different fa-
cilities before he retired. Of those ten contracts, he describes three
as "total successes," four as "OK," and three as "total failures."
He reports that the INS had little difficulty in terminating the
contracts that were failures and replacing them with new contrac-
tors.7 It should be noted, however, that the INS is a national agency
with multiple facilities, which helps to ease such transitions.
Contractors with multiple contracts can deal with rough tran-
sitions by temporarily reassigning personnel. Corrections Corpo-
ration of America uses experienced facility administrators from its
central offices to serve as interim administrators when setting up
a new contract or changing the administration of an ongoing con-
tract. The Eckerd Foundation used counselors from its other pro-
grams to deal with a rough transition period at Okeechobee, during
which there were strong fears of violence. According to an Amer-
ican Correctional Association study, the Superintendent and these
166 PRIVATE PRISONS

outside personnel "were able to maintain the safety of the insti-

tution through heroic effort and countless hours of overtime." 8
CCA's contract for New Mexico's women's prison provides for
a procedure to ensure the government a smooth transition in case
of contract termination. During transfer of operations back to the
NMCD or to another contractor, CCA would work under super-
vision of the New Mexico Corrections Department for a period of
60 days. All records would transfer to the NMCD.
As a final note, since one purpose of contracting is to allow for
competition in the provision of correctional services, it would be
a mistake to regard potential terminations and management tran-
sitions as necessarily negative. Certainly, they can be disruptive,
but they may be either good or bad, depending on what went
before and what comes after.

Innovation and Risk-Taking

A study of contracting for community correctional services found

that flexibility was a major attraction to public agencies.9 Contracts
allow agencies to experiment with new programs without long-
term commitment of funds or of tenured civil service staff. Vested
interest in these programs does 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, flexibility for the public agency translates into inse-
curity for the private contractor. To survive and succeed, a con-
tractor must 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 costs 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 and willingness to take risks arc often seen as virtues
when associated with small businesses. They can, however, be seen
as mixed blessings when it comes to the incarceration business.
On the one hand, the structure of incentives seems to favor orig-
inality and innovation in the private sector. While failure in cor-
Issues of Flexibility 167

rections can be dramatic and costly for either the government or

a private contractor, there are fewer countervailing rewards for
success in the public sector. Private actors can be motivated eco-
nomically to take innovative risks, but public actors' concern with
political survival induces caution and reinforces the status quo.
Under civil service, the careers of line staff and middle managers
are more or less predetermined, while the positions of top officials
depend largely on avoiding embarrassing mistakes. Steering clear
of bad publicity becomes more important than outshining the com-
petition, especially if there is no competition.
On the other hand, in the coercive environment of prisons,
flexibility and innovation are not necessarily desirable. Cumber-
some governmental procedures for review and approval of changes
in policies and programs are not merely obstructionist. They are
necessary to preserve due process constraints on the power of the
state. It is important not to lose these constraints when state power
is extended through contractual agents.
This is a valid point, though it gets carried too far when it is
argued that private firms are incapable of responsibility in inno-
vation. According to sociologists Jess Maghan and Edward Sa-
garin, "Innovation and flexibility, when proposed by the CCA or
some other such organization, can be motivated only by profit
enhancement, hardly a purpose inspiring to those who are grap-
pling with corrections." '" We should not feel that we have to
choose in an all-or-nothing fashion between due process protec-
tions and the ability to innovate. Particular innovations may be
either good or bad, but a structure that is conducive to innovation
is, on the whole, positive.
Contracting passes some of the risks of innovation from the
government to the private sector. This is usually presented as a
point in favor of contracting. John Donahue, however, sees it as
a drawback. While conceding that "uncertainty about the future
is the whole rationale of a market for risk," Donahue questions
the wisdom of shifting risk from the government to private parties.
His primary argument against doing so is that "governments in
general are better at spreading risks than private companies are,
simply because they encompass more people." n What he means
is that governments are better at absorbing costs when risks turn
out to be bad risks.
168 PRIVATE PRISONS

Why should the ability of government to absorb costs be seen

as an advantage? If our goal is to avoid costs, the best strategy is
to transfer risk to the party most motivated to minimize it, not to
the party best prepared to absorb it. In most cases, it is the party
least able to bear a risk that is best prepared to act quickly in
response to innovations that develop snags or go sour.

Flexibility and the Uncertainty of Social Policy

Flexibility is especially important in the administration of public

policy, where the concentration of decision-making magnifies the
consequences of ignorance, uncertainty, and error. Policies re-
garding 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 ac-
curacy 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 be-
cause it can generally respond more quickly to changes in
information.
A market in corrections would share in the general advantages
of markets over central planning. The advantage most relevant
here is that while central planning magnifies the consequences of
erroneous predictions, competition isolates and minimizes them.
If a state launches a major prison construction plan and hires an
army of civil servants, based on a long-term projected trend that
does not materialize, or that unexpectedly reverses itself after a
few years, the cost will be much greater than if several competing
contractors are responding continuously to projected needs. 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 costs of their inaccurate forecasts. In con-
Issues of Flexibility 169

trast, when the state has a monopoly on the prison business, it can
simply pass on to taxpayers the full cost of its errors, and thus has
less incentive to avoid mistakes in the first place.
Contracting can help respond to changes in prison populations
brought about by changes in legislation. In recent years, these
changes have generally meant increases, but contracting can help
with decreasing correctional needs also. In 1984, the Bureau of
Prisons signed a three-year contract to house about 60 Youth Cor-
rections Act offenders in a private facility in La Honda, California.
Repeal of the Youth Corrections Act that year had made it clear
to the Bureau that its YCA population would disappear through
attrition over the next few years. The Bureau's director later re-
ported to Congress that: "Contracting to house these offenders
gave us the flexibility to handle our population without acquiring
additional permanent spaces. This allowed us to respond to the
YCA population reduction in the most cost-effective way." 12

Flexibility Enhances Justice

The flexibility of private prison contracts may also enhance justice,

at least according to a "just deserts" model. Public concerns over
justice and punishment (usually expressed as "getting tough") have
led to changes in many components of the justice system. Abol-
ishing parole, limiting judicial discretion, banning or restricting
plea bargaining, and other such reforms are intended to curb
abuses and to make punishment more uniform and just. Generally,
though not necessarily, this is seen as replacing inconsistent leni-
ence with more consistent firmness and punitivity. One objection
to these "get tough" policies has been the fear that they will pro-
duce further overcrowding of prisons. Therefore, some criminol-
ogists have urged the development of so-called front-door and
back-door mechanisms that would seem to defeat the purpose of
the antidiscretion reforms. Back-door options include emergency
release mechanisms and ongoing early reentry programs. Front-
door mechanisms include diversion programs and sentencing
guidelines that specifically require judges to take capacity into
account. '3
The search for new mechanisms of diversion and release to pre-
170 PRIVATE PRISONS

vent or reduce 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 based on the jus-
tice model, however, makes just the opposite assumption: prison
flow should respond to the crime rate, which is largely beyond the
control of the state; therefore, prison capacity must be flexible.
Reduction (or increases) in imprisonment under a justice model
should occur because such changes are perceived as improving
justice, not because of limited (or expanded) prison space.
At least at the margins, then, the prison system must be able to
expand and contract in accordance with the shifting demands of
justice. 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, adjust-
able contracts, offer an increased prospect of achieving this mar-
ginal flexibility.
9
Issues of Security

One of the strongest reservations about private contractors re-

lates to their ability to run high security prisons. The for-
mer director of the Federal Bureau of Prisons, for example,
has expressed doubts that private firms could ever run max-
imum security institutions.1 The Council of State Govern-
ments and the Urban Institute recommend that only
minimum security facilities be contracted at this time. 2 Their
rationale for this appears to be based largely on the expecta-
tion that local communities will object to privately run max-
imum security institutions being located in their neighbor-
hoods.3
While private firms have not yet run a maximum security
prison, they are running jails that include all levels of security.
In February 1987, the Bay County (Florida) jail, which is
run by Corrections Corporation of America, had over half a
dozen capital murderers along with other offenders in the
jail's maximum security wing. Also, some private juvenile
institutions are designed for serious, including violent, of-
fenders.
Why should private management be seen as more problem-
atic in a maximum security institution? Probably because, in
most people's minds, the higher the level of security, the
greater the need for coercion, including the use of potentially
deadly force.

171
172 PRIVATE PRISONS

Deadly Force

The right to use deadly force is widely regarded as an exclusive

prerogative of government, but this is a misconception. Legitimacy
in the use of force is determined by conformity to provisions of
law, not by the public or private employment status of the person
who exercises force. Under certain circumstances, the use of force
by private actors is both legal and legitimate.
For example, the power of arrest has been delegated to private
railway police, to humane society agents, and to bail bondsmen.
The arrest powers of bondsmen were established under common
law, while a majority of states empower private railway police by
statute. Such police may carry concealed weapons and have all the
arrest powers of public police. Though nominally appointed by a
public official and sworn into office, they are in fact employed and
supervised by the railroads.4
Under common law, a private citizen has the right to use deadly
force in self-defense, in defense of another, or to prevent the
escape of a felon. In 1985, however, the Supreme Court held, in
Tennessee v. Garner, that deadly force may be used by police
against a fleeing felon only where it is "necessary to prevent the
escape and the officer has probable cause to believe that the suspect
poses a significant threat of death or serious injury to the officer
or others." 5 It is not clear what the implications of this decision
are for prison guards, but it appears that the powers of police, and
probably of corrections officers, are approaching those of ordinary
citizens, rather than diverging from them. The Garner decision is
a recognition of the precept that, whatever the specific rules and
regulations, the general principles justifying the use of force by
state agents should not be significantly different from those jus-
tifying the use of force by civilians. In this connection, an obser-
vation from the report of the Council of State Governments and
the Urban Institute is significant: "In States that subscribe to the
provisions of the Model Penal Code, the proper use of deadly
force by private correctional officers would not require further
legislation. The definitions in the Model Penal Code appear to
sufficiently include private prison guards." 6
There is no moral distinction between governmental and private
Issues of Security 173

actors as individuals that would justify a governmental monopoly

on the use of force. Civil servants are not by nature or necessity
the most responsible, accountable, or controllable wielders of
power. There is, however, a practical rationale for limiting par-
ticular uses of force to small and highly visible subsets of the
population. The rationale is that, by doing so, it is easier to keep
the use of force within the limits specified by law. That is why
powers of arrest are generally linked to certification of some sort,
such as licensing or deputization. The certification does not confer
legitimacy in itself, through incantation or some sort of voodoo
jurisprudence. Rather, it aids the enforcement of laws intended
to restrict the legitimate uses of force. In keeping with this goal,
it is consistent for the state to recognize, allow, or authorize force
by nonstate actors under certain circumstances and within the same
legal guidelines that determine whether the use of force by a state
actor is legitimate.
Also consistent with the practical rationale for limiting and con-
trolling the use of force would be for the state to certify, license,
deputize, or swear in all correctional officers, both public and
private. This would bind them formally to specific codes that au-
thorize and restrict their permissible uses of force and provide
mechanisms for disciplining them if they overstep their authority.
The training and other requirements for such certification should
be the same for public and private correctional officers.
In the state of Florida, all correctional officers, including those
employed by private contractors, must be certified by the state. In
New Mexico, legislation authorizing private prisons designates a
contractor's employees as peace officers while in the performance
of their duties.7 The California Department of Corrections places
one or more of its own officers at each of its privately contracted
prisons for parole violators. These state officers have final authority
over any security issues that may arise, such as escapes or distur-
bances, and during any emergency operation may take over com-
mand from private security officers.8 Officers do not carry guns
inside most prisons and jails, either public or private, and the use
of deadly force against inmates is rare. 9 Some private prisons have
guns available outside, with employees who are authorized to carry
them.
What happens in the case of a riot, or other emergency? Critics
174 PRIVATE PRISONS

often ask this question, then move on to other issues as if it were

unanswerable, or required no answer. The answer, however, is
about the same for a private as for a government-run prison. In
the event of disturbances at most prisons and jails, force is de-
ployed from outside. Major disturbances often require reinforce-
ment from state police or national guard units. This practice would
not be different in a privately run prison.

Routine Security

Experience at most contracted facilities, in terms of routine se-

curity, has been positive. At the Bay County Jail, CCA has as-
signed officers to each floor so that they can observe violations
and response time will be quicker. Under the county, there were
often no officers on the floor, which was a major factor in a lawsuit
filed against the sheriff and the county by an inmate who said the
lack of supervision resulted in his being sexually abused by other
inmates.
When Buckingham Security took over operation of the Butler
County Prison, the first thing they had to do was take back control
from the inmates. Even allowing for the partiality of the source,
a First Year Report by Buckingham is worth a lengthy quote on
this score:
Prior to Buckingham's arrival, the staff remained out of the areas
where inmates lived. They stayed out of the passageways when
inmates were moving through them. During the infrequent outdoor
periods, a staff member watched from an overlooking window. The
myth existed that various areas in the prison belonged to the pris-
oners whenever they were in them. It was thought to be dangerous
to intrude. . . .
Beginning immediately, Buckingham took control of the prison.
All staff, including the warden, mingled with the inmates daily. . ..
The entire prison now belongs to the county and the county em-
ployees govern all of it at all times. All inmates are now assigned
specific bunks in specific cells and the staff knows that each is ac-
counted for. 1 "
Among the security reforms instituted by Buckingham were: rou-
tine physical security checks (bars, locks, doors, windows, walls
Issues of Security 175

and perimeter); a key control system; a professional head-count

system; an organized system for inmate files; a tracking system to
determine proper release dates; a sally-port system for security
doors; and an inmate classification system. The custom of allowing
police to wear their guns inside the prison was ended.
At Silverdale, the private prison run by CCA for Hamilton
County, Tennessee, security was significantly enhanced following
a minor disturbance in August of 1986. In his detailed and objective
case study of this prison, Samuel Jan Brakel describes how security
measures that were already under way were accelerated after this
incident. Guns and other riot gear were placed in a secure area
below the control tower (guns are not worn inside the prison); a
new fence topped with razor wire was built around the men's unit;
a more restrictive checkpoint was set up at the gate; new key
control and inmate pass systems were implemented; staff assign-
ments were reviewed to assure 24-hour coverage of all crucial
stations; visitation, transportation, and shakedown procedures
were tightened; surveillance of cells and dorms was increased; and
emergency evacuation drills and procedures were instituted."

Escapes

There have been escapes from private prisons, but their record in
this regard is no worse than their public counterparts, and often
better. Certainly no private facility has been as lax in its security
as the District of Columbia's Oak Hill Youth Center, a "high
security" detention facility for juveniles in Laurel, Maryland,
where a recent check of the official log showed that 30 percent of
the 197 detainees were missing and listed as escapees,12 or the new
Prince George's County (Maryland) Jail—much celebrated for its
"new generation" design—which 11 times during its first year re-
leased wrong prisoners under mistaken identities. 13
During the first 7 months of operation at the Marion Adjustment
Center, there were 4 walk-aways, 3 of whom were recaptured
within 24 hours. In a comparison state facility, walk-aways aver-
aged 1.5 per week.14 Differential selectivity of admission at these
two facilities, however, makes this comparison questionable. A
before-and-after comparison is more instructive. At the Okeecho-
176 PRIVATE PRISONS

bee School for Boys, the escape rate was 25.6, 27.6, and 25.6 per
100 inmates in the years before, during, and after transfer from
government to private hands. The failure of the escape rate to rise
is significant in light of the extremely high employee turnover that
occurred during the transition year15 (see chapter 6).
Over the first few years of private management at Silverdale,
the number of escapes went down slightly. This was in spite of the
fact that the population was increasing and moving sharply toward
a higher mix of felony cases. During most of 1984 (until October
15), the prison was under county management. There were 42
escapes during 1984, all but 9 of which occurred while the prisoners
were on outside work crews under the control of the sheriff. In
1985, there were 39 escapes—9 from within the compound. In
1986, of 35 total escapes, 7 were from within. During this three-
year period, the population rose to 358 from a base of 250-300,
and the number who were state felons rose from 10 percent to 34
percent. Thus, where more escapes might have been expected,
given the changing nature of the population, there were about the
same or slightly fewer under private management. 16

Strikes

The National Labor Relations Act (NLRA) makes it unlawful for

an employer to interfere with the right of workers to strike. Public
employees are specifically excluded from NLRA coverage and in
most jurisdictions there are laws making it illegal for at least some
categories of civil servants to strike. In a review of legal issues
relating to private prisons, Mary Woolley concludes that private
prison employees do possess the legal right to strike. 17 Woolley
asserts that these workers are covered by the NLRA unless Con-
gress clarifies or amends the Act or the National Labor Relations
Board reverts to an interpretation that was abandoned in 1979.
(A 1975 ruling held that a private fire department was exempt from
the NLRA and had no legally protected right to strike, because it
provided an essential municipal service; this interpretation was
rejected in a 1979 case involving private busing of public school
students.) 18
Issues of Security 177

Public Employee Strikes

Those who fear strikes by correctional workers would be better
advised to oppose public employee unionization than to oppose
prison privatization. Although strikes are generally more common
in the private sector, they are rapidly increasing among public
employees. From 1960 to 1984 there was a tenfold increase in the
number of public sector work stoppages and a thirty fold increase
in number of days idle.11J A state-by-state study of public employee
unionization showed that, in each state, the number of strikes
increased dramatically in the years immediately after enactment
of collective bargaining legislation. In some states, the effect was
astounding. Michigan had only one strike in the seven years prior
to its legislation, compared to 290 strikes in the six years thereafter.
New York jumped from an average of 4.1 strikes a year before,
to 27.75 strikes a year after.20
Strikes and other job actions are illegal for correctional officers
(guards) in all states except Hawaii, and generally they are illegal
for most other correctional employees as well.21 The absence of a
right to strike, however, has not prevented public prison guards
from engaging in many strikes, sickouts, and other job actions.
Under the leadership of the American Federation of State,
County, and Municipal Employees, strikes by correctional workers
have been not merely local, but sometimes regional and even na-
tional affairs. In 1974, under coordination by the AFSCME, prison
guards in Rhode Island and at six penal institutions in Ohio walked
off their jobs.22 In 1979, almost all the guards at New York's 33
correctional facilities went out on a strike that lasted 17 days.
During that time, the strikers were responsible for several incidents
of violence and property damage directed at National Guardsmen
called in by the Governor.23 A study of correctional employee
unionism in 17 states found that correctional officers had engaged,
illegally, in strikes or similar job actions in about half of the re-
search states.24
Contrary to popular impression, prison strikes do not generally
create situations that are difficult to control. Lockdowns are fre-
quent, but inmates tend to stay on good behavior during strikes
even when supervisors attempt to run things on a near normal
178 PRIVATE PRISONS

basis without a full lockdown.25 Thus, the risk of strikes in private

prisons is neither as new nor as serious as most people suppose.
Still, it is a risk to be avoided or minimized as much as possible.
Whether the risk will be higher or lower in a private prison remains
an open question.

Strike Risks under Contracting

Privatization provides an opportunity to deal with the problem of
strikes in a fashion more realistic than simply outlawing them.
Mary Woolley recommends that work stoppages be addressed in
both enabling legislation and contract. For example, in the event
of a strike, state police could be required to perform as guards,
as they do already during emergencies at correctional institutions.
The cost of contingency staffing could be covered by insurance or
performance bonds. 26 Another possibility, not mentioned by Wool-
ley, would be to couple legislation requiring that all correctional
officers—public and private—be certified,with legislation provid-
ing for automatic decertification of officers who participate in a
strike.
The Council of State Governments and the Urban Institute rec-
ommend that private prison contracts should "require sufficient
advance notice of the end of an employment contract period, the
onset of labor difficulties or major grievances that could result in
a work stoppage or slowdown." 27 The contract between CCA and
Hamilton County specifies that "A default shall immediately occur
if any threat to the security of property or persons . . . is caused by
strike (whether by employees or inmates). . . "2g The contract—
like other prison contracts—also specifies procedures for the
county to resume immediate control of the prison in the event of
default or serious disruption, such as a strike, a riot, or bankruptcy.
State police and the National Guard would be the ultimate re-
course in a strike by private correctional officers, as is the case
now when public employees walk off the job. However, the cost
of such intervention could be negotiated and specified in the con-
tract, so that it might or might not distribute differently from how
it currently does. Also, a performance bond can be used to defray
the government's cost if it has to take control of a contracted
facility.
Issues of Security 179

Since a strike or other disruption would allow the government

to terminate a contract, unemployment as the result of a strike
will be a credible threat to private officers. In contrast, such threats
do not often deter strikes in the public sector. Although numerous
strikes took place in Ohio prisons from 1970 to 1975, it was not
until May 1975 that any striking officers were fired. Likewise, when
Walpole State Prison officers were fired for striking in 1973, the
Governor of Massachussetts was pressured into rehiring them, by
an AFSCME threat to declare a strike by all state employees.29
Amnesty for strikers is a universal demand in public employee
strikes and generally (assuming that there has been no violence or
other illegal acts incidental to the strike) they get it.
In government-run prisons, strikes typically last only a few days;
a three-week strike is considered extreme.30 There are several
reasons to believe that labor unrest would be resolved even more
quickly in a proprietary prison. Private management has a greater
and more personal stake in rapidly ending a strike and resolving
the underlying issues. They also have more direct control over
many matters at issue, and where they do not, that fact is clearer
than in a public situation, where a strike is often not merely against
management but against the executive and legislative agencies as
well. Private correctional officers, unlike their public counterparts,
do not have a claim on the support of noncorrectional public em-
ployees, especially given the opposition of public employee unions
to privatization. Even inmates in a private prison might be in a
better position to prevent or resolve a strike. Inmates suffer most
from any strike, but ordinarily they have no power to influence
the outcome.31 As third-party beneficiaries to a contract, however,
they will have a new legal standing from which to petition for a
court injunction against a strike or threaten to sue the company.
Finally, the experience of the federal government is reassuring
on the question of strikes against contractors. In a review of almost
30 years of contracting experience under Circular No. A—76, "Per-
formance of Commercial Activities," the Office of Management
and Budget lists the fear of strikes as one of the common miscon-
ceptions about contracting:

Available evidence has not borne out the fear that contracted ser-
vices would be more susceptible to employee strikes than it they
180 PRIVATE PRISONS

were retained in-house. There have been two cases of strikes by

contract employees and in both cases contingency plans proved
effective. The Circular requires that contractors be held accountable
in the case of strikes and that contingency plans be included in
contracts.32
In sum, there is nothing special about the government's ability
to provide security in prisons. Force can be exercised legitimately,
legally, responsibly, and effectively by private actors subject to the
same rules of law as public employees. Experience with security
in private facilities, including jails with maximum security wings,
has been fairly positive so far. And the threat to security posed
by a labor strike is at least as great if not greater under govern-
mental as compared to private management.
10
Issues of Liability

Critics of privatization warn that governments will not escape li-

ability by contracting the administration of their prisons, as some
advocates supposedly claim. In contrast, some opponents of pri-
vate prisons seem to imply that government can escape liability
(through sovereign immunity) if only it stays away from contract-
ing. Lawyer and privatization critic J. Michael Keating summarizes
the liability picture as follows:
What seems reasonably clear is the likelihood that private jail and
prison operators will be subject to all of those statutes fashioned to
render state agencies liable for their misconduct, such as the Civil
Rights Act, while remaining ineligible for the benefits derived from
those statutes and common law doctrines formulated to preclude or
limit the liability of public bodies, such as state tort liability statutes
and the doctrine of sovereign immunity. 1
From context, it is clear that Keating regards this legal vulnerability
of private prisons as an argument against them, but it is not at all
clear why this should be so.
Since critics of private prisons are particularly concerned about
prisoners' rights and the accountability of those who exercise
power, they should regard the increased liability of private prisons
as an argument in favor of contracting. Prisoners in private facilities
gain by the addition of an extra layer of liability, accountability,
and responsibility for their rights and welfare. They are certainly
not worse off in this regard than they would be under total gov-
ernment management. It is therefore ironic for prisoners' rights

181
182 PRIVATE PRISONS

groups to raise liability concerns in opposition to contracting. One

would think that these litigious groups would be reassured, not
alarmed, by the legal vulnerability of private prisons.
So far, there has been at least one case in which an inexperienced
private detention vendor was sued, made a monetary settlement,
and went out of the detention business.2 The government was also
sued in this case, but only for declaratory relief, not monetary
damages. When the government is sued for damages, it passes the
cost on to taxpayers; no government agency yet has been litigated
into bankruptcy. This may be a plus in terms of stability, but it is
a minus in terms of accountability.

Civil Rights and Wrongs

On at least one liability issue, there seems to be universal agree-

ment: with or without privatization, government remains ulti-
mately responsible for the constitutional rights of citizens,
including prisoners. Government may not shirk its constitutional
duties, nor can it avoid liability. Careful legal scholars holding
opposite views on the general wisdom and propriety of private
prisons are nonetheless in agreement that both government and
contractors definitely can be held responsible under the Civil
Rights Act (42 U.S.C. Section 1983) for violations of constitutional
rights by private prison contractors.3 Private prisons will qualify
as Section 1983 defendants because they will be acting "under color
of law" and thus "state action" will be present. Therefore, both
private prisons and their contracting units of government can be
sued under the provisions of this Section.4
The most pertinent case so far has been Medina v. O'Neill,5 the
facts of which were as follows:
In 1981, a vessel arrived in the Port of Houston carrying 26
stowaways, and the Immigration and Naturalization Service was
alerted. Ordinarily, the INS orders that stowaways be detained on
board the same vessel for immediate return transportation. In this
case, however, the ship lacked facilities for detention, so 20 aliens
were placed in a local jail, while 6 were placed with a private
security firm, in a 12 foot by 20 foot cell designed to hold no more
than a half-dozen people. The next day, 10 of the aliens in the
Issues of Liability 183

local jail were transferred to the custody of the private firm and
placed in the cell that was already filled to capacity. A day later,
the aliens attempted to escape and a private guard accidentally
killed one of them with a shotgun.
The federal district court ruled that the plaintiffs were entitled
to sue the INS for violation of constitutional and statutory rights.
The ruling had three significant parts. First, the actions of the
private party were held to constitute "state action." Second, the
conditions of confinement were held to violate constitutional stan-
dards of due process. Third, the INS was held to be responsible
for ensuring that the detention met constitutional standards.
In 1988, the Fifth Circuit Court of Appeals vacated and reversed
two of the three parts of the district court ruling. The Appeals
Court held that "(1) INS officials had no statutory duty to provide
appropriate detention facilities for excludable aliens, and (2) [the]
aliens' due process rights were not violated." 6 Significantly, how-
ever, the Appeals Court did not negate the district court's ruling
that the contractor's conduct constituted "state action." That rul-
ing was allowed to stand despite the fact that excludable aliens
have practically no rights under the U.S. Constitution, and even
though the connection between the INS and the private firm in
the Medina case was very weak. All costs for the detention and
transportation of stowaways must be borne by the carrier on which
they arrive. 7 Though it may designate someone other than the
carrier to provide detention, the INS itself does not have to fi-
nancially contract for that service in the case of excludable aliens.
Compared to this, any correctional facility privately managed un-
der a direct, formal contract to the government is operating even
more clearly "under the color of law." In sum, there seems to be
little doubt that violations of constitutional rights by private prison
staff constitutes "state action" under Section 1983,8 thus allowing
plaintiffs to sue either the government or the private provider.
Professor Ira Robbins both prefaced and concluded a lengthy
discussion of the liability issue with statements that one argument
in favor of prison privatization has been the claim that it will reduce
or eliminate government liability. 9 It is this claim, he says, that
makes the issue of liability so significant in the debate over private
prisons. Robbins' documentation on this point, however, is a state-
ment by an opponent of contracting, not a supporter.10 In point
184 PRIVATE PRISONS

of fact, it is virtually impossible to find any supporters of privati-

zation who claim that contracting can actually immunize govern-
ment against legal responsibility (as opposed to lowering risks by
providing better services and insuring or indemnifying government
against those risks that remain). Charles Thomas calls the argu-
ment about liability a "straw man" and reports that "no repre-
sentative of any private corporation has even one time in any
document I have ever found—and I've reviewed hundreds of
them—ever made that allegation." " My own review of the lit-
erature is consistent with that report.12
The existence of government liability is a nonissue, because all
parties agree that it is there. What remains to be seen is whether
that liability results in higher or lower legal costs to the government
under private as compared to public operation of prisons. In any
case, the liability argument can be turned in either direction. Which
is stronger as an argument for contracting: the claim that the gov-
ernment can thereby escape from liability, or the claim that it
cannot do so? To argue that private prisons will not reduce the
liability of either government or contractors seems to me to be a
very strong argument for privatization, because it emphasizes that
accountability and protection of prisoners' rights will not be com-
promised. Indeed, the legal recourse of prisoners may well be
increased.
In testimony to the President's Commission on Privatization,
Professor Robbins has noted that prison contracts will create two
categories of third-party beneficiaries to the contract: prisoners
and the public. This expands prisoners' rights by giving them the
potential benefit of a contract cause of action, which they would
not have had before. In addition, this new cause of action will
generally have a longer statute of limitations than a suit under
Section 1983 or an ordinary tort action.13
To nail down this unique form of liability, 14 Robbins has rec-
ommended that clauses be inserted in both statutes and contracts
explicitly declaring that prisoners and the public are intended to
be third-party beneficiaries to written agreements between the gov-
ernment and private contractors.15 That would give standing to
enforce the contract not only to the government agency but to
prisoners and to any member of the general public. The ACLU,
or any other group that now sues prisons on behalf of inmates,
Issues of Liability 185

would no longer need inmate plaintiffs; as part of the public, it

could sue private prisons on behalf of itself.
To broaden the scope of claims that could be brought by third
parties, Robbins places in his Model Contract a clause in which
the private provider agrees "to provide inmates with proper care,
treatment, rehabilitation, and reformation." ' 6 As goals of impris-
onment, rehabilitation and reformation are not only controversial,
but almost certainly impossible to achieve fully. When offenders
released from private prisons are convicted of new crimes, will
members of the public (and the newly convicted offenders them-
selves!) be able to sue the contractor for damages because of this
"breach of contract"?

Sovereign Immunity

Another liability question is whether private jailers should be al-

lowed to protect themselves against monetary damages by asserting
the defense of sovereign immunity.
In principle, there is a basic distinction between "policy-making,
planning, or discretionary decision-making activities, which remain
fully protected by the sovereign immunity defense, and operational
or ministerial activities, which are not immune." " In practice,
however, the distinction between making and implementing policy
is often fuzzy. To minimize or avoid legal challenges on delegation
grounds, this distinction should be clarified as carefully as possible.
In some states, legislation authorizing prison privatization in-
cludes a statutory prohibition against the defense of sovereign
immunity. In Florida, legislation permitting correctional privati-
zation at the state and local levels specifies that private companies
are liable for damages in tort with respect to the care and custody
of prisoners under their supervision; the Tennessee Private Prison
Contracting Act denies the defense of sovereign immunity to pri-
vate contractors or their insurers.' 8
Most states have waived sovereign immunity either generally or
substantially.19 In those states, liability for the tortuous acts of
correctional employees will exist whether those acts are committed
by government or by contract employees. On the other hand, many
states waive immunity only if the government purchases liability
186 PRIVATE PRISONS

insurance, and then only up to the amount of coverage.20 Thus,

even where sovereign immunity is waived, the waiver is generally
limited, and liability is capped. Given the large insurance coverage
required of private prisons, and the historically modest size of
damages awarded in correctional lawsuits (both discussed below),
a legislated cap on the liability of private prisons may not be nec-
essary at this time.21
Serious questions can be raised about the advisability of impos-
ing a legal double standard, in which state actors are protected by
sovereign immunity while private actors performing the same func-
tion are not. The major argument for sovereign immunity is that
public officials need that protection in order to perform their jobs
effectively. The major argument for denying sovereign immunity
to contractors is that they need to be held responsible.
The implicit theories behind such a double standard are that:
(1) government actors can be trusted while private actors can-
not; (2) government actors respond best to carrots while private
actors respond best to sticks; and (3) immunity allows people to
act according to their true characters, which for public actors
means altruistically but for private actors means selfishly. But it
simply won't do to assume that public and private actors have
different human natures or respond to incentives in radically
different ways. Indeed, many people change from a public-sec-
tor job to a private-sector job without any simultaneous change
in character.
The doctrine of sovereign immunity was originally based on the
notion that because the King ruled by Divine Right, he could do
no wrong. This notion has now been replaced by one that is only
slightly less questionable: that "public servants" are also touched
by grace, because they are disinterested and benevolent and act
only for the public good; hence, they should be granted immunity
so that they can act boldly on their presumed commitment to
maximize social welfare.
Qualified (or "good faith") immunity for government actors also
rests on the proposition that it would be unfair to prosecute persons
who are legally required to exercise discretionary power in their
jobs. However, government officials and bureaucrats apply for
those jobs and assume the concomitant obligations voluntarily.
Issues of Liability 187

Some are attracted to such positions precisely because they want

to exercise discretionary power. Others accept the power as part
of a job that they want largely for personal gain (satisfaction,
salary, job security). Few are driven by self-sacrifice. Except per-
haps for military draftees, government employees are not legally
required to hold their jobs in the first place, or to remain in them.
Therefore, a civil servant who by virtue of his job is required to
exercise discretionary power is in no different position from a
contractor, who by virtue of his contract may be required to ex-
ercise discretionary power.
It may be that some jobs—like police officer and prison war-
den—are so fraught with discretionary power and potential liti-
gation that the public interest requires a grant of immunity as
one of the incentives needed to fill these posts. Where this is
true, I see no reason why it would be less true for contractual
police or wardens than for those who are direct government
employees. Qualified immunity is a device to make a risky job
more attractive (i.e., more profitable) to a government em-
ployee. The alternative would be to pay civil servants enough to
enable them to buy liability insurance. Exactly the same logic
applies to contractors: they too need an incentive, in the form
of either qualified immunity or fees high enough to pay for ex-
tra insurance.

Actual Risk vs. Legal Exposure

As discussed above, prisoners in private facilities have at least as

many avenues of civil redress as do their fellows in government-
run prisons. Total liability, in other words, is not decreased.
Spreading the liability neither immunizes the government nor re-
lieves it of its ultimate responsibility for the rights and welfare of
prisoners.
However, it is still possible that government, by contracting,
could reduce the actual risks (and therefore costs) for which it
remains liable, as opposed to reducing its legal exposure. That is,
the financial damages likely to be suffered by government as a
result of its legal liabilities can be cut back. Lowering liability risks
188 PRIVATE PRISONS

not only avoids financial damages, it reduces litigation and insur-

ance costs as well.
Liability risks can be reduced by:
1. Running prisons better, and thus avoiding lawsuits.
2. Achieving certification, which greatly enhances the defense
against lawsuits.
3. Carrying adequate and cost-effective insurance.
4. Agreements in which the contractor defends the government
in court and indemnifies it against legal damages.
5. Developing extensive legal expertise and resources, both for
preventing and for fighting lawsuits.
6. Settling quickly out of court, which is easier for private firms
than for public entities.
Private prison contractors today are reducing risks by all of these
methods.
Corrections Corporation of America believes that litigation, and
thus liability exposure, can be reduced by taking a proactive, rather
than a reactive approach. For example, they invite the ACLU to
inspect their facilities. As they put it, "That's good free advice;
and they're the ones most likely to sue us." 22 When they won the
Texas contracts, CCA took a positive step toward assuring that
they would be able to comply with the massive and comprehensive
court orders governing corrections in that state. They retained as
counsel a former Texas Department of Corrections attorney inti-
mately familiar with the Ruiz decision.
At several of their facilities, including Laredo, Bay County, and
Silverdale, CCA is required to supply legal assistance to inmates
beyond the mere provision of a law library. To represent inmates
in lawsuits against them, CCA retains private counsel under con-
tracts that enable the lawyers to act like public defenders, immune
to any control by CCA. One benefit to CCA is that they hear
about a potential problem before a suit is filed and therefore can
resolve it more easily. To represent CCA when it is sued, the
company hires other outside counsel; the job of their own legal
staff is mainly to keep them from being sued in the first place.23
Prior to its contract with CCA, Bay County, Florida, was under
court order to reduce overcrowding in its jail and faced a lawsuit
for the death of a prisoner. The uninsured county commissioners
Issues of Liability 189

were named personally as defendants. 24 Nine months after CCA's

takeover, three prior lawsuits against the county over jail condi-
tions "were dropped in the wake of substantial improvements in
conditions." 25
Changes in procedures instituted by Buckingham Security when
it took over operation of the county jail and prison at Butler
County, Pennsylvania, were designed to take back control of the
facility from the inmates. Those changes should also make less
likely any serious injuries or harmful incidents and reduce legal
liability that might result from a lack of proper procedures in the
event of a significant incident. One measure of the effect of these
changes (which were detailed in Chapter 9) is that prior to Buck-
ingham's takeover, emergency trips to the hospital had been av-
eraging four or five a week, but after the takeover they were rare. 26
The liability experience of private prisons has not been all pos-
itive, however. Corrections Corporation of America recently re-
solved a $100 million lawsuit filed against the company and 11
other defendants in Hamilton County, Tennessee. The suit, which
alleged that a female inmate died as a result of inadequate medical
treatment, was settled out of court for $100,000.27 As will be noted
below, this amount is not unusual for a medical malpractice case.

The Modest Size of Correctional Damages

It is too soon to tell whether private prisons will be sued less often,
less successfully, or for lower amounts than government prisons.
In the final analysis, the effect of privatization on liability costs
will depend more on operational differences than on legal differ-
ences. It is worth noting, however, that while the volume and total
cost of litigation by inmates is enormous, 28 and the costs of outside
insurance and self-insurance are high, a recent national survey by
Contact Center, Inc., indicates that monetary damages and settle-
ments in corrections tend to be rather modest.29
In the Contact Center study, the Federal Bureau of Prisons and
33 states returned surveys reporting their costs for inmate lawsuits
in 1983 and 1984. In compensatory damages, punitive damages,
settlements, and court-awarded attorney fees, these 34 jurisdic-
tions paid a total of $5,920,922 over the two-year period.30 Using
adult inmate population figures for these systems on June 30,1984,
190 PRIVATE PRISONS

combined payouts come to a per capita, per diem value of just

four cents.31
Though a big award could have a serious effect on a small county,
it is evident that for large systems, liability costs resulting from
privatization would have to be several times what they are now
before they would substantially affect costs on a per unit basis.
And for successful suits to multiply, the total volume of litigation
would have to explode, since only a minuscule proportion of inmate
lawsuits result in damages or settlements.32

Insurance and Indemnification

Although liability risks in corrections are generally low, the cost

of insuring against them can be high. In 1986, CCA spent about
$940,000 on premiums for $5 million in general liability insurance,
which did not cover officers of the corporation. No claims were
filed against CCA or its officers during that year. By 1987, CCA
had established a self-insurance plan covering both the corporation
and its officers. For this plan, $5 million from their initial public
stock offering was placed in an interest-bearing trust fund. An
outside firm was paid about $150,000 a year to service or administer
claims. CCA projected savings of about $800,000 on insurance
premiums for 1987.33
At one time, CCA was required by its contract with Hamilton
County, Tennessee, to carry $25 million of insurance, and by its
contract with Bay County, Florida, to carry $15 million. At first,
the company did obtain $25 million coverage, but later they found
it impossible to find an insurer that would underwrite a policy that
large. The company felt that its assets ($30 million in lines of credit
and $12 million in net worth) were adequate to cover risks beyond
a $5 million insurance plan.34
Other corrections companies also carry large amounts of insur-
ance. Behavioral Systems Southwest carries $5 million.35 Pricor,
which manages four correctional facilities, carries $2 million in
general liability insurance.36 Wackenhut Corrections Corporation
typically provides $5 million in coverage and can provide more on
large contracts. In addition, WCC has behind it the enormous
Issues of Liability 191

financial strength of its parent, the Wackenhut Corporation, with

assets of $300 million and revenues of over $400 million in 1988.37
In addition to carrying high levels of insurance, private prison
companies typically offer contracts in which they indemnify the
government against the total cost of any harm resulting from the
operation of their facilities. In Bay County, as elsewhere, CCA
promises to defend the county in court against any legal actions
arising out of the operation of the jail. The contract contains the
following indemnification provision:
CCA shall save and hold harmless and indemnify the COUNTY,
the members of the Board of County Commissioners, county em-
ployees, and agents, including attorneys, and their respective legal
representatives, heirs and beneficiaries, whether acting in their of-
ficial or individual capacity, and shall pay all judgments rendered
against any or all of them for any and all loss or damage of whatever
kind against any and all liability, claims, and cost, of whatsoever
kind and nature for physical or personal injury and any other kind
of injury, including specifically deprivation of civil rights, and for
loss or damage to any property occurring in connection with or in
any way incident to or arising out of the occupancy, use, service,
operation or performance by CCA, its agents, employees or rep-
resentatives of any of the provisions or agreements contained in this
Contract, including any Appendices, for which the COUNTY, the
members of Board of County Commissioners, county employees
and agents, attorneys, or other persons, as noted hereinabove,
whether acting in their official or individual capacity, who may
become legally liable resulting in whole or in part from the acts,
errors, or omissions of CCA, or any officer, agent, representative,
or employee thereof, and for which CCA shall pay all judgments
which may be rendered against the COUNTY, members of Board
of County Commissioners, county employees and agents, including
attorneys, and other persons as noted hereinabove, whether in their
individual or official capacity.38
As formidable as this sounds, CCA is financially able to shoulder
the risk, and may even be overinsured against it. In its own survey
of monetary damages in correctional negligence cases, the largest
award CCA could find was a medical malpractice judgment for
$750,000, which went to the family of an inmate who died in a
prison hospital.39 This study dovetails with the Contact Center
192 PRIVATE PRISONS

finding that nearly all the large (six-figure) awards went to victims
of medical malpractice.40 Health care is the most frequently con-
tracted component of corrections, representing about 23 percent
of all correctional service contracts.41 Apart from this troublesome
area, which correctional officials are often glad to contract out,
liability risks in corrections are no greater than those that occur
in many other enterprises, both public and private.
Ira Robbins has proposed that every private prison contractor
should be required to carry insurance in the amount of $25 million
per occurrence.42 Such a requirement would slap a big "Sue Me"
sign on the back of every privately employed warden, an open
invitation to lawyers in search of deep pockets. Many states place
dollar limits on their own liability. In Virginia, for example, this
limit is $25,000.43 The law does allow higher awards where they
are covered by insurance, but it is hard to see what incentive a
state agency would have to voluntarily carry insurance greater than
$25,000. Thus, in Virginia, under Robbins' proposal that private
prisons must carry $25 million in liability insurance, the incentive
to sue a private prison would be 1,000 times as strong as the
incentive to sue an uninsured state prison.
One has to wonder if Robbins, a foe of private prisons, really
just wants to see them sued out of existence. On the other hand,
Robbins is a sincere champion of prisoners' rights, and the private
companies would be insured, so perhaps he just wants to see that
prisoners' interests are well protected. But if that is the case, why
not insist at the same time that states also be required to set their
own liability limits at $25 million per occurrence? Actually, over-
insurance is not a good idea for either type of prison. In addition
to inviting opportunistic lawsuits, overinsurance may create a
"moral hazard" by weakening the insured's incentive to guard
against risk.
Overinsuring only the private prisons will create other problems
also. Since contractors will pass their insurance costs on to the
state in the form of higher fees, a massive insurance requirement
could price them out of the market. In theory, governments could
be required to calculate what it would cost them to carry that much
insurance themselves, when comparing their own costs against con-
tractors' bids. In practice, however, that is unlikely to happen if
the government is not actually required to have that much insur-
Issues of Liability 193

ance itself. This would become another hidden cost that presents
an inaccurate picture of the differences between public and private
corrections.
Briefly, privatization does not offer government any easy escape
from its responsibility and liability for imprisonment of offenders.
What it does offer is the prospect of sharing that liability, buffering
the government through indemnification, and possibly reducing
the number of lawsuits through improved management.
11
Issues of Accountability
and Monitoring

Critics claim that contracting reduces accountability because pri-

vate actors are insulated from the public and not subject to the
same political controls as are government actors. Also, the critics
charge, contracting diffuses responsibility; government and private
actors can each blame the other when something goes wrong.
Further, contracting may encourage the government to neglect or
avoid its ultimate responsibility for prisons; supervision may
slacken.
This chapter will focus on the issues of accountability and mon-
itoring as they apply to both public and private prisons. The first
step is to recognize that accountability can take many different
forms.

Administrative Accountability

Accountability does not come easily under any sort of system.

Blame avoidance is endemic to both public and private bureau-
cracies. While a major purpose of any contract is to identify areas
of responsibility, critics of contracting are right to point out that
this does not guarantee accountability. Contracting could even
confuse and diffuse responsibility and accountability. The state
may attempt to avoid responsibility by pointing its finger at the

194
Issues of Accountability and Monitoring 195

contractor; the contractor in turn, may hide behind the contract

by insisting that its responsibilities are limited to those that are
explicitly spelled out on paper.
On the other hand, the process of contracting encourages the
government to define the purposes of imprisonment, the respon-
sibilities of prison managers, and measures of performance by
which those managers will be judged. These definitions and mea-
sures can then be applied to governmental as well as to private
prisons.
To limit a vendor's ability to hide behind a contract, contracts
can contain provisions incorporating detailed and compre-
hensive specifications or standards that are spelled out in state
statutes, corrections department policies, consent decrees,
American Correctional Association standards, and other author-
itative sources. Even without such provisos, those contractors
who are willing to live up to the spirit of a contract will have an
edge over competitors who use their contracts as shields to pro-
tect themselves against all unanticipated demands, including
those that are reasonable.

Legal Accountability

Governments are held accountable not only through the politi-

cal process but, more importantly, through the rule of law. This
is especially the case in the area of corrections. It is primarily
the legal system that protects due process and other rights of
prisoners. Legal accountability on the part of proprietary pris-
ons can be no less than that of government prisons and will
probably be greater, since the private sector enjoys no rights of
sovereign immunity.
The importance of the rule of law was discussed at length in
Chapter 4. The issue of liability was examined in Chapter 10. Here,
it may be instructive to compare the accountability of contractors
to that of judges who have been appointed for life. Judges and
wardens, of course, perform very different functions, but both
exercise considerable power over other citizens. Compared to
judges, the problem of imposing legal and political accountability
on wardens, either public or private, is much more simple.
196 PRIVATE PRISONS

Life-tenured judges are appointed by an elected representative

of the people. Contractors are selected through a politically and
legally regulated process either by elected representatives or by
officials who in turn are accountable to elected representatives.
Once appointed, tenured judges are accountable in three ways,
each with a parallel for contractors.
First, even life-tenured judges are subject to the rule of law.
They are not free simply to rule as they see fit, without legal
constraint. Contractors are even more subject to legal restrictions
since, unlike judges, who often have certain absolute immunities,
contractors do not enjoy even limited immunities. They are fully
vulnerable to prosecution in criminal courts, and to lawsuits in civil
courts.
Second, the decisions of all but nine judges—those on the U.S.
Supreme Court—are subject to appellate review. Likewise, cor-
rectional contracts can be written so that all contractor decisions
affecting the rights or liberty of prisoners are subject to review by
governmental and judicial agents. Even if there is no such con-
tractual provision, it is becoming increasingly clear that correc-
tional contractors act "under color of law" such that their conduct
constitutes "state action" under 42 U.S. Code Section 1983. This
means that prisoners, or others acting on their behalf, can sue not
only the contractors but the contracting government agency, and
officials thereof, for any actions that violate constitutional rights
(see "Civil Rights and Wrongs" in Chapter 10).
Third, even tenured judges are ultimately accountable through
articles of impeachment. In a similar vein, the ultimate mechanism
of accountability for contractors is termination or nonrenewal of
the contract. The difference is that judicial impeachment is a rare
event, while breaking or failing to renew a contract is common-
place. Unlike impeachment, termination and nonrenewal are not
reserved for only the most extreme cases of misbehavior or
nonperformance.
In short, there are effective legal remedies for malpractice or
abuse of power by private prison contractors. Private wardens are
merely a special, and not a difficult, case within the broader prob-
lem of how to apply the rule of law to the men and women who
operate the legal system.
Issues of Accountability and Monitoring 197

Economic Accountability

Economic competition is a powerful mechanism of accountability

and discipline. Vendors who are subject to competition are not
only accountable to the government, through their contracts; they
are accountable also to other parties who have interests that some-
times conflict with, but mostly parallel or derive from, those of
the government. Chief among these other parties are competitors,
insurers, and the capital market, i.e., investors.
Competitors hold each other accountable on standards of cost
and quality set by the purchaser of their services. Less effective
or more costly contractors will lose business to rival firms. Com-
petition, comparison, information, accountability: each follows
naturally from the one before. In competing for contracts, vendors
provide comparative information about themselves and each other
(including information about government-run prisons). Such
knowledge is essential to accountability.
Insurers provide independent evaluations of quality in the form
of risk assessments. Their premiums impose discipline by punishing
or rewarding high or low risk. A company with a high-risk profile
may be unable to obtain any insurance at all.
The independence and discrimination of investors and knowl-
edgeable investment advisors adds a powerful form of "supervi-
sion" in the private sector to supplement the direct supervision
and regulation required by the state. Contractors who are suc-
cessful because they run well-managed and profitable businesses
will be able to attract investors. Newly invested capital, in turn,
can be used by such a business to improve its services even further.
But wouldn't potential investors just concentrate on a company's
bottom line (profits) and neglect or even discourage attention to
quality of service? This is not likely. Investors have a stake in
the reputation and the future of their company, not just in im-
mediate or short-term profits. Stock prices anticipate the future.
A private prison corporation that is headed for scandal, lawsuits,
prosecution, or uninsurability as a result of mismanagement, will
see its stock begin to fall even before it actually begins to lose
contracts. Moreover, if employees are also stockholders, this dis-
198 PRIVATE PRISONS

tributes the supervision motive to where it will do the most good.

In severe cases of mismanagement, investors can force reform from
within or from without, through takeovers if necessary.
Citizens who invest in a private prison company are risking their
money; they have an incentive to consider whether the company's
prisons are needed and will be well built, well managed, and in
good favor with government and taxpayers for a long time to come.
In contrast, those who buy general government bonds don't have
to be so careful. They know that the government guarantees them
a fixed return, regardless of the success or failure of the project
they are capitalizing.
Critics sometimes claim that private prison companies, by their
very nature, and particularly because they are accountable to stock-
holders, will have to put private goals and interests first and the
concerns of the public second. If this is the case with private pris-
ons, it will make them unique among commercial enterprises. It
is true that a profit-making company must sooner or later make
profits if it wishes to stay in business. It is also true that businesses
are answerable to the interests of their owners and investors. Nei-
ther of these facts, however, places private interests above all
others. Would we say of other commercial enterprises that they
must place their stockholders ahead of their customers? Or that a
company is accountable to its Board of Directors before it is ac-
countable to the law?
Economic controls do not displace political controls, but they
can operate more quickly and allow finer adjustments. As Joseph
Kalt points out, the "political 'marketplace'. . . meets relatively
rarely; when it does voters are presented with a bundle of nu-
merous and durable choices that cannot be marginally altered.
Moreover, political competition is plagued by high transaction
costs; the costs of organizing and promoting changes in the bundle
of policies offered by the government are substantial." ' In con-
trast, renegotiating and changing a contract is quicker and more
discriminating than reelecting a new administration. If wholesale
change is called for, however, that too can occur through contract
termination, without the long interval required between elections.
For proprietary prisons, market mechanisms of supervision, dis-
cipline, and accountability add to those of the political and legal
Issues of Accountability and Monitoring 199

systems. Economic accountability supplements, more than it con-

flicts with, political and legal accountability.

Political Accountability

Since prisons carry out public policies, they of course must some-
how be held accountable to the public. Where policies are codified
into law, mechanisms of legal accountability will serve. Where
policy is administratively created or interpreted, however, it is
desirable that there be mechanisms for public input, public scru-
tiny, and public control. These mechanisms may be characterized
broadly as "political" and can take many forms.
Direct popular election is the most obvious form, and it is clear
that an elected jailer is politically accountable, at least in principle.2
In practice, however, direct election is not necessary for effective
political accountability. If it were, only those jails that happen to
be administered by an elected sheriff would be politically account-
able. Other jails, and all prisons, are run by administrators who
are appointed or hired, not elected.
The political accountability of a contractor is like that of an
appointed, rather than an elected, official. Appointees and con-
tractors are only indirectly accountable to the electorate, but this
does not make them less responsible. One study compared the
fiscal accountability of cities having a mayor/council structure to
that of cities with a council/manager structure. It found that in
mayor/council cities the actual cost of refuse collection was 41
percent higher than what was shown in their official budgets, com-
pared to 22 percent higher for the council/manager cities.3 Since
the first step toward accountability is the provision of accurate
information, it is significant that informational accountability was
greater under the system with a hired professional manager—which
resembles a contractual arrangement.
Of the many millions of public functionaries and workers re-
sponsible for the execution of public policy, only a tiny proportion
are directly accountable to the electorate. All the rest can be said
to be accountable only in the sense that somewhere up a line of
command, or up a network of crisscrossing chains of supervision,
200 PRIVATE PRISONS

there lies an elected official. But the workers and supervisors in-
volved in running a prison under contract, by virtue of the fact
that they are all simultaneously liable to the threat of termination,
can make a good case that they are more accountable politically
than are most government employees. Only the highest govern-
ment officials and administrators are politically accountable by
virtue of election or appointment. 4 Working down the ranks from
middle management to lower management to line staff, public
employees become progressively less politically accountable. Most
government employees enjoy civil service protections that make
them virtually immune to being fired. They may be reassigned,
but they are not likely to find themselves involuntarily unem-
ployed. In one year, for instance, "only 300 of the 2.8 million
federal employees reportedly were dismissed or terminated for
incompetence." 5 Public employees at state and local levels of
government are also very difficult to discipline. In an extreme
example of political and legal nonaccountability, when government
correctional employees go on strike—illegal to begin with—they
often continue their job actions even in defiance of court orders.6
At the federal level, the judicial branch of government is in-
sulated from direct electoral accountability and the legislative
branch seems to be getting more so. Members of the House of
Representatives are rarely ousted from their posts. In the 1986
elections, 98.4 percent of House incumbents seeking reelection
were returned to office.7 The executive branch, to be sure, is more
susceptible to electoral influence, at least at the top. However, to
the extent that most federal employees who work in executive
agencies are entrenched in civil service jobs and insulated from
the political process, the accountability of the executive branch at
the highest level may not translate into accountability at the op-
erational level.
If accountability does, indeed, become progressively weaker as
one moves down a chain of command to the level of those who
actually deliver public services, contracting can help shorten that
chain and strengthen the linkage by making everyone involved
more directly vulnerable, and therefore more accountable, to of-
ficials and administrators at the highest levels. In the case of pris-
ons, contracting can also be used to hold government, as well as
vendors, accountable. Where poor management has become en-
Issues of Accountability and Monitoring 201

trenched and resistant to reform, contracting provides a surgical

solution. In Bay County, Florida, part of the motivation for con-
tracting the jail was a belief on the part of some county officials
that the sheriff was too powerful and too difficult to control.8 One
official said he believed that sheriffs, most of whom are reelected
continuously, should not be elected officers at all, but professional
employees hired by the county commission. The commission itself
is elected, is responsible for the budget, and is given statutory
authority to incarcerate offenders in a jail or prison. The official
argued that sheriffs derive much of their political power, in effect,
from the points of their guns. Citizens are intimidated by sheriffs
in a way that they are not by county commissioners. Sheriffs control
large budgets and extensive job patronage in addition to the power
inherent in their discretionary arrest authority. "People are ab-
solutely scared to death to disagree with the sheriff, no matter
who he is," said this county official, who cited acts of intimidation
against himself and other functionaries.
Bay County knows too well that simply leaving the jail in the
hands of an elected sheriff is no easy solution to the problem of
accountability. A jury awarded $10,000 damages against the sheriff
in a lawsuit brought by an inmate complaining of inadequate med-
ical care when the jail was under the sheriff's administration.9 The
sheriff was also charged with sexual harassment in another lawsuit
brought by several female employees. In addition, the sheriff's
department was the subject of internal and external investigations
regarding money that was missing from the department's evidence
locker: $12,600 in 1983 and $2,270 in 1987. The money was never
accounted for. 10

Community Accountability

The only time that a prison or jail is directly accountable to an

identifiable segment of the public is at birth. Before a new lockup
can be constructed, it must be located in a community willing to
accept it. Public fears about safety and effects on property values
must be overcome, but despite often strong and widespread resis-
tance, it is not impossible to convince at least some communities
to accept prisons in their backyards. This is because prisons bring
202 PRIVATE PRISONS

jobs to the unemployed and unskilled in a depressed area. They

buy food, fuel, and other items, often locally.
Because they must rely on persuasion and offers of benefits
rather than governmental power to overcome community resis-
tance and fears, private prison companies may be forced to be
more accountable to the public than is the state. Private firms do
not have the power of eminent domain. When Arbor, Inc.,
wanted to open a work release facility in Chicago, it created a
community board of advisors and hired locals to help renovate an
abandoned building. In contrast, when the state proposed a similar
facility in the same area without consulting the community, local
protests aborted the plan."
Site planning also played a role in the final decision by the
Kentucky Corrections Cabinet to award a contract to the U.S.
Corrections Corporation for the Marion Adjustment Center. The
state's first choice of contractor was unable to find a community
willing to accept its proposed facility, while USCC won approval
for its location from Marion County's Fiscal Court. l2 Some Marion
County residents, however, remained opposed and instituted law-
suits, which were not successful in removing the prison. A study
for the Commonwealth of Virginia describes the results of USCC's
public relations efforts:
The "NO PRISON" signs which grace the residents' front yards are
gradually disappearing and only a small group of 3-4 hard-core foes
remain. U.S.C.C., actively working on better public relations in the
community, held an open house the last week of July and invited
all the townspeople (over 90% attended). Tours of the facility were
given and questions answered. The residents complimented the
owners and director on their quality staff and were amazed when
told that the tour guides were not staff but rather (college-educated)
inmates. Two mothers made a specific point of thanking U.S.C.C.
for the employment opportunities presented to their newly em-
ployed sons. This is noteworthy since unemployment is a persistent
problem in the small farm community. 13

Visibility of Prisons

Prisons are more open to public inspection than most people think.
Nonetheless, except during periods of crisis, they tend to have low
Issues of Accountability and Monitoring 203

visibility to the broader public. Private facilities are currently sub-

ject to a great deal of publicity. Most of this is due to their novelty,
but even after the novelty wears off they are likely to remain
controversial and to draw at least some attention indefinitely. In
addition, having a mix of governmental and contractual operations
invites comparisons which make each type more visible than it
would be alone.
Even people with strong objections to private prisons credit them
with drawing attention to problems and issues that are common
to all prisons and worthy of public examination. A vocal critic of
private prisons, Michael Keating, Special Master of Rhode Island's
state facilities, nonetheless notes that the use of private providers
"opens up the process to outsiders," and exposes facility operations
to public view. 14 The contract for the Okeechobee School for Boys
specifies that failure of the Eckerd Foundation to allow public
access to records shall be grounds for termination of the contract.15
Experience in other parts of the criminal justice system supports
the idea that contracting increases visibility and critical evaluation.
For example, privately prepared presentence investigation reports
(PSIs) are subject to much closer scrutiny by more parties than
are PSIs prepared by government investigators. The most impor-
tant part of a PSI is the section where a disposition is recom-
mended. As Herbert Hoelter, director of the National Center on
Institutions and Alternatives, points out:
In many state and federal jurisdictions, the recommendation section
of the PSI, the section which demands the most accountability, is
not disclosed. The opposite is the case with the private report, where
the recommendations and relevant rationale for them are subject
to full disclosure. Any responsible private report must demand a
higher standard as a result of its disclosure and exposure."1

Prison Constituencies

Most citizens have little incentive to monitor the management of

prisons. As individual voters they can have little effect on prison
policies, and still less on prison personnel. Most citizens will never
serve time. Apart from dramatic events, like riots or escapes, they
have no interest in routine information about local jails. Hence,
there is no general public constituency for prisons.17
204 PRIVATE PRISONS

Prisoners, of course, are interested in prisons, and in recent

years they have gained considerable legal power to promote their
interests. However, they have almost no political power. Prison
reform groups, like the American Civil Liberties Union, through
its National Prison Project, have some political influence, but most
of their efforts are legal, rather than political.
The most politically effective constituency for corrections is cor-
rectional officials themselves, who influence their executive and
legislative overseers not through the ballot box, but through con-
trol over information. It is well known that bureaucracies attempt
to control the flow of information so as to advance or protect their
own interests. Various measures, such as the Freedom of Infor-
mation Act, limit the ability of public bureaucracies to control
information completely. However, when the agency that generates
the bulk of routine information about itself also stores that infor-
mation and controls its dissemination, accountability is compro-
mised. Under contracting, there is at least some independence of
interest between the agencies that generate information and those
that then receive and control it. In addition, with on-site moni-
toring, there will also be an independent primary source of
information.

Monitoring

All parties, including vendors, agree that monitoring is important

for private prisons. Private prisons need to be monitored partly
because they are contracted and partly because they are prisons.
As contractual operations, they need to be monitored for com-
pliance with contractual provisions. As prisons, they need to be
monitored for performance. Clearly, compliance and performance
are over-lapping concerns, but it is worth noting them separately
to emphasize their mutual importance. Moreover, as will be argued
below, monitoring for performance—as distinct from monitoring
for contract compliance—should not be viewed as a need that is
special to contracted prisons only.
In a report for the National Institute of Justice, the Council of
State Governments and the Urban Institute reviewed prison man-
agement contracts from the early 1980s and the Requests for Pro-
Issues of Accountability and Monitoring 205

posals (RFPs) issued by government agencies that shaped those

contracts. They found that both the RFPs and the subsequent
contracts were, overall, quite general.18 As a guide for future con-
tracts, the report recommended a monitoring process with the
following components:
1. Statistical summaries of reported unusual incidents, such as
escapes, deaths, major injuries or illnesses, assaults, distur-
bances, staff use of force, and major disciplinary actions (i.e.,
loss of good time, lockdowns, or solitary confinement).
2. Surveys of inmates regarding programs and conditions.
3. On-site inspections using standardized evaluation forms that
focus on actual conditions and behavior, not just written pro-
cedures. These should be conducted by outsiders at least
annually and by on-site or local monitors on a continuous
basis.
4. Timely feedback to contractors, so they can adjust their prac-
tices, and to government officials, so they can make informed
decisions on contract renewal.
The report also recommended that, for comparison, the same mon-
itoring procedures be applied to publicly operated facilities.
Though government may be more inclined to monitor con-
tractors than to monitor itself, it should not be assumed that
any monitoring will occur automatically. The government needs
to provide a specific mechanism for monitoring. Since contract-
ing does relieve bureaucrats of many daily headaches, they
may be tempted to treat it as a quick fix and neglect their duty
to oversee.
Something like this apparently happened for a while in Flor-
ida, with the Okeechobee School for Boys. The Department of
Health and Rehabilitative Services, which was responsible for
monitoring the contract at Okeechobee, requested but did not
receive authorization for a full-time monitor, so the task was as-
signed to a staff member with other responsibilities in West
Palm Beach. However, after the lack of monitoring was criti-
cized in the first draft of an evaluation study by an American
Correctional Association research team, the Department of
Health and Rehabilitative Services initiated a regular, formal
auditing process.20 In its final draft, the ACA study noted that,
206 PRIVATE PRISONS

in addition to the HRS response, the private administration at

Okeechobee responded to the initial ACA report by moving "in
a very positive direction." 21 The ACA's own monitoring found
the contractor to be in compliance on 91 percent of the items in
its contract.22 The lesson here seems to be that both govern-
ment and contractors benefit from outside scrutiny as well as
from a system of monitoring.
Monitoring has, in fact, been a regular feature of most recent
prison and jail contracts. At the Marion Adjustment Center (a
preparole facility), an on-site state employee (a parole officer)
monitors the contract, approves inmate furloughs, and gives fi-
nal approval or disapproval to all good-time determinations
made by private employees.23 The situation is much the same at
the Bay County Jail and Annex, except that the monitor is a
county employee who also has an office and duties downtown.
At Silverdale, the county prison and women's jail at Hamilton
County, Tennessee, monitoring is provided by the county's Su-
perintendent of Corrections, who was the warden at Silverdale
before the Corrections Corporation of America assumed man-
agement. The superintendent handles all release and good time
decisions and makes all work assignments for prisoners doing
county work outside the walls. He also serves as liaison be-
tween the prison and the county commission, the courts, the pa-
role board, the probation board, and the state department of
correction.24 Hidden Valley Ranch was monitored daily by a
federal official when it was under contract to the Bureau of
Prisons. At the Butler County Jail, run by Buckingham Security
Limited, the contract is monitored by a county employee and
disciplinary protocol is outlined in the contract.25
Reports26 that the Immigration and Naturalization Service had
to hire 12 employees to provide 24-hour monitoring at their con-
tracted Houston detention facility arc not accurate. Essentially,
the contractor rents space to an INS staff of about a dozen at that
site. Most of those people spend most of their time on work for
the INS that is independent of the contract. Only one is designated
officially as the contract monitor and even he spends only about
30 to 40 percent of his time on monitoring duties. An evening
monitor spends 20 to 30 percent of his time on activities related
to monitoring.27
Issues of Accountability and Monitoring 207

Informal Monitoring
In addition to formally designated monitors, prison contracts can
be monitored in many other ways. External observers and watch-
dogs, like the media and the ACLU, are at least as interested
in private as in governmental prisons, if not more so. Internal
"monitoring" will be provided by prison inmates, who are veteran
whistle-blowers and will take legal actions against private wardens
as energetically as they do against the government. Significant
input from prisoners could be very useful in contract renewal de-
cisions. While it would not be wise to give inmates any formal
power over the choice of their keepers, some kind of mechanism
for inmate evaluation of their treatment would help to reinforce
accountability.28 Critics who fear neglect or abuse of inmates as a
result of a cost-cutting ethic in private prison management might
be reassured by this provision for evaluation by "insiders." 29
Insurance companies provide a form of monitoring for private
prisons that is often lacking for the government. Insurers have a
vested interest in gathering valid and objective data about areas
of performance that relate to legal liability costs. The premiums
they set for different vendors or types of vendor will give an in-
dependent assessment of risk and, thereby, of quality. This is anal-
ogous to the use of household fire insurance premiums to evaluate
fire departments, or the use of hospital liability premiums to eval-
uate hospitals.30

Sauce for the Gander

It should be emphasized that monitoring is as much a benefit as
it is a burden to the corrections system. Monitoring adds another
level of supervision to an activity that needs as much of that as it
can get. More importantly, it brings a new element of independence
to the system of checks and balances controlling an awesome ex-
ercise of domestic power: the deprivation of human freedom. As
noted in Chapter 4, independent review is vital to due process.
Here, the point is that independent monitoring promotes objec-
tivity and rigor in the overall supervision of a prison. It is easier
to be consistent when imposing standards on outsiders than when
enforcing them on ourselves or our colleagues.31
208 PRIVATE PRISONS

T. Don Hutto, a former commissioner of corrections in two

states, and now the President of CCA International, a division of
Corrections Corporation of America, observes that: "As a director
of corrections, I did a better job of monitoring and evaluating
private sector contracts than I did of monitoring and evaluating
my own operations. I also did a better job of monitoring and
evaluating the jails, which I did not have responsibility for oper-
ating. Through the contracting process, government can be more
objective about the goals it wants to reach." 32
In any case, whether it is seen as a burden or a benefit, both
sides of the private prison debate agree that monitoring is very
important. Indeed, monitoring and other mechanisms of insuring
accountability are so important in the prison context that they
should not be seen as special to contractual operations. It is nec-
essary and desirable to monitor the operation of correctional fa-
cilities no matter who is running them.
Government-run facilities are routinely inspected, audited, reg-
ulated, supervised and monitored not just by correctional agencies
but by other agencies, and sometimes other branches and other
levels, of government. Beyond these routine forms of monitoring,
there are others. One third of jails surveyed in 1986 were under
a court order.33 At least 60 jails are supervised by a court-appointed
jail master.34 In 1984, 31 states and the District of Columbia had
at least one major prison operating under a court order or a consent
decree. For six of the states and the District of Columbia, the
entire prison system was under a court order or a consent decree.35
In 1986, 14 states had court-appointed overseers, such as masters
or compliance monitors, for their prisons.36 Court orders and con-
sent decrees are analogous to contracts, and jail or prison masters
are like contract monitors.37
Judicial monitoring of government-run prisons can be extremely
expensive. Monitoring and enforcing a court order may require
a special master with a staff of attorneys and investigators and
an annual budget running up to three-quarters of a million dol-
lars.38 Moreover, such masters commonly serve long and indefinite
terms on any single case.39 Even when a consent decree is super-
vised directly by a judge, without a special master, that also en-
tails a cost.
As a digression, it would be interesting to apply to special mas-
Issues of Accountability and Monitoring 209

ters, who administer or oversee court-ordered prison reforms,

some of the concerns raised by critics of private prisons about
issues of delegation and accountability. Who monitors the
monitors?
When judges issue broad and detailed orders specifying exactly
how prisons and prison systems must be run, they can be accused
of eroding the principle of separation of powers. The broader the
judges' concerns, the more they encroach on the policy preroga-
tives of the legislature. The more detailed the judicial orders, the
more they encroach on the administrative realm of the executive
branch.
When judges appoint special masters to supervise the enactment
of their orders, they thereby delegate power and authority to pri-
vate individuals. By virtue of this delegation, prison masters may
be considered officers of the court, but the structure of their re-
lation to the court is functionally equivalent to a contract, and
many of the arguments raised against delegation to prison con-
tractors can be raised with equal validity (or invalidity) against
special masters.
There is no legislative mandate for the delegation. The guide-
lines for the masters' interpretations may be vague. The account-
ability of special masters is low.40 With large budgets and
substantial fees, special masters could be accused of having a vested
interest in producing continuous findings of violations, an accu-
sation parallel to the charges of conflict of interest leveled against
profit-motivated contractors.
In reply to this last charge, it may be claimed that the profes-
sional ethics of special masters will inhibit them from finding vi-
olations just to feather their own nests. I tend to agree, but I think
the defense applies just as well to the ethics of professional prison
contractors. More significantly, however, I think that both con-
tractors and special masters are kept honest not just by their ethics,
but by the fact that they have a material interest in their reputations
for honesty and integrity.
To make my point clear, I am not arguing that the practice of
appointing prison masters is either unwise or unconstitutional. I
am saying that objections to delegations of authority over matters
of prison administration, based either on some nondelegation doc-
trine or on concerns about accountability, have no more categorical
210 PRIVATE PRISONS

validity when applied to prison contractors than they have when

applied to special masters.
To return to my primary theme, monitoring should not be seen
as a new burden created by contracting. Rather, private prisons
serve to focus attention on an important question we should ask
of all prisons: How best can we monitor, regulate, and evaluate
them?
The Council of State Governments and the Urban Institute, as
noted above, have suggested what monitoring and evaluation
ought to encompass. More important than the specific ingredients
of monitoring, however, is a recognition of the importance of the
process, and of the fact that it is just as important for public as
for private prisons.
The requirements and procedures for monitoring prisons ought
to be similar for both government and contracted prisons. At least
two states—Massachusetts and Pennsylvania—have implicitly rec-
ognized this principle. They have developed standardized moni-
toring systems, to be applied to both state-run and contracted
facilities for juveniles. In both states, procedures call for exami-
nation of institutional records, site visits by an outside team, and
interviews with staff. In addition, the Massachusetts procedures
include interviews with inmates. 41
While monitoring is necessary for all prisons, it is not sufficient.
Monitoring has not, by itself, saved governmental prisons and jails
from poor management or physical deterioration. What is re-
quired, beyond supervision, is motivation. Court monitoring has
included threats of heavy fines, or total shutdowns, to spur reform
of governmental prison systems. This has produced some good
results, but it is clearly a meat-ax approach, suitable only for ex-
treme cases. In contrast, one of the major advantages of private
contracting is the opportunity to systematically structure incentives
so that performance will respond to feedback on a regular and
routine basis.42
12
Issues of Corruption

Wherever large sums of money and great discretionary power come

together, especially if accountability and control are weak, there
will be a risk of corruption. Corruption has been a problem in
prison contracting historically and it is a problem in other types
of government contracting today. It is therefore quite reasonable
to be concerned about the possibility of corruption in correctional
privatization.
Writing for the American Federation of State, County, and Mu-
nicipal Employees, John Hanrahan asserts:
In recent years, there have been scores of publicized cases of payoffs
and kickbacks in connection with state and local governmental con-
tracting; of price-fixing and bid-rigging; of major contracts being
given to cronies and campaign contributors of public officials; of
contractors' conflicts of interests, and of contracts going to com-
panies with links to organized crime. 1
Certainly, these are real problems, with potential for subverting
the contracting process. Prison contracts, just like other types of
government contracts, carry with them temptations and opportu-
nities for corruption.
To use the possibility of corruption as an argument against con-
tracting per se, however, is illogical. It is fallacious to imply that
corruption-related problems are uniquely inherent in private con-
tracting, or would necessarily diminish if contracting were not al-
lowed. Political corruption is a corollary of government, not just
of government contracting.

211
212 PRIVATE PRISONS

Corruption as a Corollary of Government

People who spend other people's money are always tempted to

find ways to keep some of the money or to spend it on themselves.
This trait is not limited to contractors and their governmental
collaborators. Public administrators who do not use contractors
have other ways of cheating. They may wastefully expand their
budgets and activities, thereby increasing their salaries, their per-
quisites, their status, and their power. They can cheat by building
and protecting sinecures for themselves, thereby lowering the ratio
of work to reward. They may pad their payrolls, their offices, and
their staffs. In addition to their own cheating, they may counte-
nance the cheating of other administrators because it doesn't cost
them anything directly, and it serves as camouflage.
Payroll padding, nepotism, cronyism, patronage, bribery, pay-
offs, featherbedding, dishonest budget inflation, conflicts of inter-
est, misuse of public funds, links to organized crime, and many
other kinds of corruption are known to occur within public em-
ployee unions and within governmental units that provide services
directly, rather than through contracts. Thus, the potential occur-
rence of any of these (for example, the involvement of organized
crime) is no more legitimate as an argument against contracting
public services per se than it would be as an argument against the
existence of government, or of unions. 2
Bernard McCarthy has studied the many ways in which correc-
tional officials and employees sometimes abuse their discretionary
power for personal gain in the form of money, drugs, sex, or other
goods and services.3 He identifies three categories of correctional
corruption. Misfeasance includes granting special privileges or pref-
erential treatment, selective use of authorized rewards and pun-
ishments, the sale of paroles or other releases, and misuse or
misappropriation of state resources for personal purposes. Mal-
feasance includes theft, embezzlement, traffic in contraband, ex-
tortion or exploitation of inmates or their families, protection
rackets, assisting escapes, or engaging in criminal conspiracies.
Nonfeasance includes overlooking inmate violations of rules or
criminal activity, and failure to report or to stop corrupt acts by
fellow workers.
Issues of Corruption 213

All forms of corruption, whether misfeasance, malfeasance, or

nonfeasance, involve the abuse of public power to pursue private
ends. Corruption is not caused by private ownership of agencies
that provide public services. Nor would corruption be prevented
if a law were passed requiring that all state-mandated services be
produced directly by government employees. In the Soviet Union,
for example, the state controls the production and distribution of
virtually all goods and services. Not coincidentally, corruption is
pervasive.
It is ironic that some critics of private prisons are fond of quoting
Dostoevsky—that the degree of a nation's civilization can be seen
in the way it treats it prisoners—and wondering aloud what Dos-
toevsky would think of private prisons.4 The clear implication is
that based on this country's use of private prisons, Dostoevsky
would draw some negative conclusions about our civilization, but
in comparison to what? Let's consider one striking comparison
that the great Russian novelist would surely make if he had not
died in 1881.
If Dostoevsky had been alive in the Soviet Union during the last
70 years he would have been witness to one of the most brutal
and lawless prison systems in history. He'd have seen political
prisoners jammed shoulder to shoulder into airless cells and box-
cars and shipped to punitive slave camps where they were worked,
starved, and frozen to death. If he visited contemporary American
prisons, including private prisons, Dostoevsky would probably be
impressed by the civil and human rights protections, the food and
medical care, the standards of decency, even the space, he would
generally find there, at least in comparison to the Soviet Gulag.5
It would indeed say something about our civilization, but nothing
that would discourage private sector involvement in the running
of prisons.
What this comparison would tell the author of Crime and Pun-
ishment, himself the victim of repression and harsh punishment,
is that it is not state and governmental control that guarantees
good prisons. It is the rule of law, an institution that is strongest
in capitalist and mixed economies with their free markets, property
rights, and libertarian traditions, and weakest in socialist systems
where the management of nearly everything is defined as the ex-
clusive prerogative of the state.
214 PRIVATE PRISONS

In short, private sector involvement in the exercise of power is

not the cause of corruption nor is government monopoly the cure.

The "Lessons" of History

Michael Keating is representative of other private prison oppo-

nents in the picture he paints of the early history of correctional
contracting:
For much of the Nineteenth Century while correctional facilities,
especially in England, were nominally in the hands of government,
they were actually under the control of keepers or petty tradesmen,
who were in effect private contractors rather than salaried employ-
ees. Although they were required to submit accounts to supervising
courts, only mass escapes or gross corruption threatened their ten-
ure. Having once obtained their appointments or "contracts"
through judicial patronage, these early correctional entrepreneurs
were able to settle down to a lifetime of profitable extortion. Every-
thing in the facility was for sale; even release required the payment
of a fee to your friendly keeper."
Although contracting for prison labor is quite different from
contracting for institutional management, it is portrayed by critics
as indicating something basic about the character of private en-
terprise involvement with prisons. Historically, contracting for
prison labor—while found in all parts of the country—was most
common in the South and is often associated with the culture of
slavery and the economics of Reconstruction. Some profess to see
a continuation of this tradition in the fact that private prisons today
are concentrated in the South. A more likely explanation of this
phenomenon is the fact that organized union opposition is weaker
there. Harmon Wray, however, sees a deeper historical pattern
running from slavery, convict leasing systems, and chain gangs, to
contemporary private prison management and the employment of
prisoners by modern private industry. He observes darkly: "Con-
stitutionally, slavery is legal as punishment for crime, and our
Southern prison populations are, of course, overwhelmingly poor
and disproportionately black and Hispanic." 7
Other critics imply that privatization is both racist and a tool of
class oppression:
Issues of Corruption 215
Corporate America is upper-class white. Only a few of its hirelings
are minority people. Privatization places in the hands of the haves
a tool to exploit and further enrich themselves at the expense of
the have-nots.8
John Dilulio describes a bleak history of for-profit corrections
from the pre-Civil War years to the post-World War II years. His
summary judgment is measured but critical:
It is highly unlikely that the ugliest features of this history will repeat
themselves. Increased external monitoring aside, the corrections
profession has grown well beyond the days when such situations
were tolerated or encouraged. But the record does teach that prisons
and profits can be a most unhealthy mix.'
In his review, however, Dilulio himself does some unhealthy
mixing by lumping together three different types of profit-related
activity: private management of institutions, leasing of convict la-
bor, and profit-seeking behavior by public prison officials. Much
of the corruption and abuse found in the last two areas originated
within government, rather than the private sector. Many state-
run prisons attempted either to turn a profit through their own
activities or to profit by leasing involuntary convict labor to outside
employers. These things were not done just by private prison man-
agers. Thus it is not accurate to attribute them specifically to pri-
vatization, as though they were not also common in the absence
of privatization.
Dilulio has supported his historical observations, in part, by
citing two of the earliest and best-known observers of American
penitentiaries, Alexis de Tocqueville and Gustave de Beaumont. 10
While their writings do contain information useful to critics of
private prisons, it should be noted that Tocqueville and Beaumont
provide a two-edged comparison of private and governmental
prison management. They were critical of the exploitation of pris-
oners' labor both by contractors and by governmental prison
administrators. They noted that this exploitation was minimized
when there was a balance between governmental and private
management:
It appeared to us, that the evil which we have thus pointed out, has
been generally avoided in the new penitentiaries in the United
States. In these establishments, neither the system of entire domestic
216 PRIVATE PRISONS

[governmental] management, nor that by contract, have been ex-

clusively adopted."
These scholars concluded that corruption and exploitation were
equally possible under either private contracting or internal public
management. The cause in both cases was some form of monopoly
and the cure was some form of competition to avoid consolidation
of power. For example, contracts were rotated among contractors.
Also, different activities and industries were contracted to different
parties so that no one proprietor had too much power.
The history of private prison management, like the history of
convict leasing, contains many grim examples of corruption, prof-
iteering, and abuse of prisoners. It must be remembered, however,
that most of this took place at a time when corruption was also
much more prevalent in government-run prisons and in the crim-
inal justice system generally. Some states ran their prisons as profit-
making enterprises with as much ruthlessness and exploitation
without the aid of private contractors as others did with them.
Wardens and sheriffs had considerable discretion and autonomy,
and ran their institutions like feudal fiefdoms, as illustrated by this
example from the 1920s:
In one county the cost of feeding a prisoner was eight cents a day
while the sheriff received forty-five. In many counties, the sheriff
is permitted either directly or through concessionaires, to sell special
articles of food, tobacco, or other so-called luxuries, to prisoners.
He is thus permitted to starve them to the point where they or their
friends purchase food to supplement the daily ration. He thus enjoys
the extraordinary privilege of reaping a profit not only from star-
vation but from the relief of starvation. 12

Malcolm Braly, author of On the Yard, recalls in his memoirs

a variation on this theme that he observed during a stint in the
Nevada State Prison:
[Everything in the kitchen was for sale and everyone who worked
there sold food. The convict politicians [powerful inmates] bought
control over most of the meat, butter, eggs, milk—the good stuff—
and the mainline [most convicts] got whatever was left over.13
At the New Mexico State Penitentiary, prior to this country's
bloodiest prison riot in 1980, there prevailed an extensive pattern
Issues of Corruption 217

of theft, extortion, graft, embezzlement, and other forms of cor-

ruption among employees and officials of the corrections depart-
ment. Officials helped themselves to "fill-it-up" privileges at the
state gasoline pumps, free automobile repair, free dry cleaning,
free haircuts at the prison barber, and free dental and medical
services for the whole family, including discount medication and
injections, at the pen infirmary. Officials also helped themselves
to food from the pen kitchen, purchasing eggs and meat at fifty
cents to the dollar. Whole truckloads of meat being shipped to the
penitentiary from the prison farm at Las Lunas would disappear
on a regular basis.14
In some counties today, the sheriffs are allowed to keep for
themselves any money not spent on food for prisoners. When the
Wisconsin legislature tried to end this practice in that state recently,
the bill was vetoed by the governor. The legislature is now trying
again, amidst reports that some sheriffs have squeezed out meal
profits equal to their salaries. One sheriff in 1988 was discovered
to have fed prisoners baked goods that had been donated to the
county as food for goats.15
Nearly all the abuses found by historians in private prisons can
be found also in the history of governmental prisons—including
abuses related directly to the profit motive.16 Corrections Today
recently reprinted an article first published in 1945 by E. R. Cass,
a leading figure in correctional history, in which Cass decried the
exploitation and corruption of jails run for profit on the fee sys-
tem.17 What was most objectionable about these jails, according
to Cass, was not that they made a profit, but that the profit was
based on abuse of power. Those who set the fees, imposed the
fees, and enforced their collection, kept the money for themselves.
Under the fee system, the sheriff was "a public official whose chief
interest is to increase the population of the jail, and thus add to
his fees." 18 Some of these sheriffs would release offenders early
but continue to collect fees until the nominal expiration of the
sentence. Significantly, the jails disparaged by Cass were run en-
tirely by the government. The private sector played no part in
them.
Almost all the arguments from history presented by critics of
private prisons suffer from a common fallacy: they fail to make
contemporaneous comparisons. Instead, they compare the worst
218 PRIVATE PRISONS

aspects of private prisons from the past with the features of modern
prisons (or worse, with idealized, not actual, versions of mod-
ern prisons). Of the possible comparisons between historical,
modern, public, and private prisons, critics focus almost exclusively
on the one analogy that puts private prisons in the worst light.
Never do they compare modern private prisons with public prisons
of earlier eras. That would be equally valid (or invalid) but un-
fortunately (from their perspective) it would make private prisons
look far superior. It would also make more obvious the fallacy of
noncontemporaneous comparison.
At a minimum, a valid historical analysis would have to compare
private prisons to public prisons within the same political and legal
jurisdictions at the same point in time. 19 However, even if such
proper comparisons show private prisons to be worse than their
public counterparts during some earlier era, it is questionable
whether such differences would still apply in the socially, politi-
cally, and (most important) legally different world that exists today.

The Revolution in Prisoners' Rights

At a time when the prison environment was more generally cor-

rupt, contracting often represented an extension and application
of that corruption, but it was not the cause of it. In today's political
and legal environment, especially with the firmly established rev-
olution in prisoners' rights, such extreme and flagrant abuse of
power is very unlikely. William C. Collins, formerly a Senior As-
sistant Attorney General for the State of Washington and a spe-
cialist in correctional law, cautions against comparisons that ignore
the enormous changes of the last two decades:
While the private operation of jails has some historic precedent,
legal and management issues in jail operation have changed so
dramatically in the last 10 to 20 years, especially with the growth
of inmate rights and court involvement, that contracting issues or
problems from the 19th century have little relevance in the waning
decades of the 20th century. 20
Important as the prisoners' rights revolution and heightened
judicial oversight of prison management have been, however, they
Issues of Corruption 219

do not mean that corruption is no longer a problem in corrections.

And it is likely to remain a problem as long as officials and workers
at all levels of prison administration have extensive discretionary
power and relatively low visibility.

Controlling Corruption through Law and the Market

Just as the private sector is not the source of all corruption, neither
is state monopoly of correctional operations a solution to corrup-
tion. Abuses of inmates under the nineteenth-century practice of
leasing out their labor were as much abuses by the state as by
private firms. It was not so much the state as the law that finally
ended those abuses, and it is the law, not the state, that protects
against abuses and violations of prisoners' rights in contemporary
prisons. The way to further guarantee the rights of prisoners is not
to insist that prisons remain in the hands of government employees,
but to maintain the rule of law.
The independence of the judiciary allows it to oversee and reg-
ulate the government's prisons. Prisoners' rights are thus pro-
tected, and the power of their keepers constrained, by the checks
and balances inherent in the distribution and separation of powers.
Contracting, when it operates properly, extends the concept of
separation, and constructive tension, between agents of adminis-
tration and agents of oversight. When there is corruption and
collusion between contractors and government officials, the reg-
ulatory function of contracting breaks down, but it is the collusion,
not the contracting, that is the problem.
Co-optation is a subtle form of collusion. In private prisons, it
is possible that an on-site governmental monitor could become co-
opted by the contractor, even if only through friendships. The
Council of State Governments and the Urban Institute suggest that
this possibility "can be alleviated by periodically changing moni-
tors, by proper training, and by continued interaction between
State home-office personnel and the monitor." 21
The "revolving door" syndrome, in which government pur-
chasers move directly into jobs with private vendors, can produce
another subtle form of collusion, which could be called "antici-
patory collusion." As Joan Mullen points out, however, there are
220 PRIVATE PRISONS

established methods of dealing with this problem, including "con-

flict-of-interest provisions attached to public employment, openly
competitive procurement procedures, and broadly composed con-
tractor selection committees." 22
Problems of corruption in public-private contracting are rooted
in departures from conditions of genuine competition. Govern-
ment monopolization, by outlawing contracting, would not solve
these problems. With honest government, by definition, there
would be no corruption in the delivery of public services. But
governments are not automatically honest; they must be kept that
way. Open competition among contractual agents of government
is one effective method of keeping them honest. "The answer,"
in the words of Robert W. Poole, Jr., "is to have rational, open
bidding procedures and objective selection standards—and to
make sure that they are adhered to. This can be done by requiring
that all such rules, procedures and criteria be matters of public
record and by holding bid openings and other important decision-
making sessions in public." 23
13
Issues of Dependence

Critics worry that some vendors will obtain contracts by "low-

balling," that is, they will make unrealistically low bids. After the
contract goes into effect, the government will gradually lose much
of its capacity to resume the operation itself, and high capitalization
costs will prevent new competitors from entering the field. Once
the government becomes dependent on a contractor, that con-
tractor will be free to jack up prices. Worse, the contractor may
go bankrupt, leaving the government without any correctional
capacity. Worst of all, contracting may devolve into exclusive
franchises that simply replace public monopolies with private
monopolies.
All of these are realistic possibilities against which we should
seek safeguards. None of them, however, is so unavoidable as to
justify a moratorium on private prisons.

Low hailing

A common objection to contracting of all sorts is the danger of

"lowballing." In lowballing, an unrealistically low bid is used to
win an initial contract. Losses are then recovered through cost
overruns or inflated subsequent contracts. If competitors cannot
quickly enter the market and if the government would incur high
costs in resuming the operation itself, the existing contractor can
raise its price gradually but continually.

221
222 PRIVATE PRISONS

A contractor's greatest leverage over the government may be

as much psychological as financial. When a prison or jail is con-
tracted out, it relieves public officials of many worries and head-
aches. To take back a facility, or even to recontract it to a new
firm, would be a hassle. The emotional cost of that hassle would
be borne by the bureaucrats assigned to handle the necessary trans-
actions and paperwork, whereas the financial cost of renegotiat-
ing the same contract at a higher price can be passed along to
taxpayers.
The strategy for a contractor seeking to hook the government,
then, would be to raise prices through a series of modest increases,
each one being too small to be worth the bother of a new bidding
process. Obviously, there must be some limit, but it could well be
higher than what the price would have been under either fair
competition (no lowballing) or continued government monopoly.
Is there a solution to this problem?
First, it should be noted that taxpayers can be "lowballed" by
public agencies as well as by private contractors. A public agency
can start a program with a low budget or as a "pilot project" and
then increase the budget after the program is entrenched and a
constituency strongly interested in its continuation has been es-
tablished. Public agencies also produce "cost overruns" when they
exceed their budgets or inflate the costs of construction and fi-
nancing by dragging out the time to completion.
The caveat that private companies may raise their rates in the
future implies that public agencies, in contrast, will not. The fact
is, however, that since World War II the price of goods and services
has increased much more rapidly in the public sector than in the
private sector,1 partially because contractors are often required to
index their fee increases to the Consumer Price Index.
Experience in contracting for other public services has shown
that while some lowballing probably does occur, it can be con-
trolled. Moreover, it is not so extensive as to have made contracting
more costly, on average, than direct public delivery. A nationwide
study of garbage collection sponsored by the National Science
Foundation demonstrated significant savings through contracting.
Costs of municipal collection were shown to be 29 to 37 percent
greater than contractor prices.2 This comprehensive study presum-
ably included some firms that had been lowballers and now were
Issues of Dependence 223

overchargers, it" that practice really is endemic to private contract-

ing. However, these inflated fees, if any, were not enough to out-
weigh the strong savings provided by contracting overall.
Recent experience with prison contracting does not support the
fears of lowballing. Joan Mullen reports that the INS, which has
the most extensive experience in contracting for custodial con-
finement, has encountered increasing, not decreasing, competition
for its business. INS contracts are renegotiated annually and must
go to the lowest bidder. "This requirement, plus an INS history
of early contracting with low-cost nonprofits, appears to provide
little opportunity for the provider to include substantial cost in-
creases in the contract." 3
Financial reports by Corrections Corporation of America for
its early years do show annual losses to the corporation overall,
but these are not the result of lowballing. Each of CCA's facili-
ties is profitable, and the company's total return from opera-
tions is positive.4 Costs of developing a new market, however,
have produced high central office costs, which can be expected
to diminish, as a percentage of total costs, as the field matures.
Now that Wackenhut, another corporation that is national in
scope, has joined CCA in market development, those costs
should spread more widely and have less effect on particular
contracts.

"Cost Overruns"
One alleged example of lowballing is an oft-cited New York Times
story that referred to what it headlined as a $200,000 "cost over-
run" at Silverdale, the privately operated prison in Hamilton
County, Tennessee.5 Prior to the contract, the county ran the
facility at a cost of $24 per inmate for an average population of
243. After the contract began, vigorous enforcement of laws
against drunken driving raised the average population to more than
300. The contractor did not charge any more than the $21 per
inmate agreed to in the contract, but the county was sending far
more prisoners than anticipated to the facility. While it is true that
the county thus spent more money on Silverdale than it had
planned, it is very misleading to label the excess expense a "cost
overrun." 6
224 PRIVATE PRISONS

A more accurate term, suggested by Charles Ring, is "popula-

tion overrun."
Some Hamilton County officials were upset because they had
accepted a bid that turned out to have been higher than necessary.
But this is not a valid criticism of private contracting generally, or
even of this particular contract. After all, if the average population
had declined, the county would have garnered unexpected benefits
from the contract. Would the county then have volunteered to pay
a fee higher than contracted, to compensate the company for its
lower than projected revenues?
Contracts based on specific prison population predictions are
inherently risky. This risk can be distributed by negotiating con-
tracts that specify lower or higher fees depending on whether proj-
ected volume rises or falls. However, neither party to a contract
can avoid all such risks without paying some sort of premium to
the other party. To judge a contract only after the fact, without
taking into account how events might easily have been different,
is like pronouncing your insurance policy a waste of money because
you did not have an accident. No one can predict the future per-
fectly, but competitive bidding and contractually set prices provide
a motive for all parties to predict as well as they can. Bidding by
private contractors generally tends to keep prices both as low and
as accurate as possible even though unanticipated events may later
show the winning bid to have been, in retrospect, either too high
or too low.
Those who cite the $200,000 cost "overrun" story generally miss
the most significant part of the tale. When it was clear that the
large number of drunk driving offenders was likely to be a per-
manent factor, the contractor and the county worked out a new
price agreement that provided for a much lower per diem payment
for those particular offenders. 7 The cost of the contract is rene-
gotiated annually in any case, but the county's Superintendent of
Corrections was able to arrange an informal adjustment even be-
fore the contract was reworked. 8

Defenses against Lowballing

As illustrated by the adjustment just discussed, contractors have
a vested interest in accommodating, not exploiting, their govern-
Issues of Dependence 225

mental customers. Hard bargaining may be part of the contracting

game, but so is flexibility. The best contracts are those that are
mutually beneficial. While one party may sometimes gain tem-
porarily at the expense of the other, a lopsided contract is unstable
and not in either party's long-term interest.
A contract with a sliding scale can protect the government
against a population overrun, or even work in its favor. At the
Bay County jail complex in 1988, the contract with Corrections
Corporation of America was set at $31.01 per day for the first 300
inmates, about $10 lower for the next 20, and only $7.88 for every
inmate over 320. When the population reached 400 (a bit over the
designed capacity), both CCA and the county were surprised, but
the county was delighted to be paying such a low fee for one-fifth
of its detainees. 9
Public agencies can guard against lowballing by evaluating pro-
posed budgets for their realism and reliability, rather than follow-
ing a rigid rule specifying that contracts must go only to the lowest
bidder. The lowest bid is not necessarily the most realistic bid.
Also, regular renegotiation or renewal of contracts, with at least
the potential for competition through open bidding, can make
lowballing a costly strategy. No private company can raise its fees
very high above a reasonable profit margin without inviting ex-
posure and competition. Competing contractors have the infor-
mation, motivation, and organizational resources to control each
others' prices to a much greater degree than taxpayers are able to
control government costs.

Consolidation and Market Entry

Ironically, one concern about privatization of corrections is that it

will not be competitive enough. Capital costs are said to be so high
as to restrict market entry, thus allowing early entrants to use their
capital advantage to squeeze out would-be competitors with pred-
atory pricing. Further, big corporations may use political power
as well as economic power to control the market. Critics fear that
the private sector in corrections will become "a privileged group
of large, monolithic service providers concerned more about profit
than performance, creativity and compassion." 10 They worry
226 PRIVATE PRISONS

about "the small, independent social service agencies. Hundreds

of them, particularly those supported by federal grants, have dis-
appeared while the private sector's bigger corporations—Eclectic,
Magdala, the Volunteers of America and others—survive and
prosper." n
Some critics even foresee the possibility of dependence at a
national level, with unsettling effects on prisoners: "An American
gulag archipelago, in which the prisons are under the jurisdiction
of private corporations working together, or a single corporation
that has established a dominant if not monopolistic position, could
easily result in the transfer of prisoners from one area of the coun-
try to another..." 12
Market dominance and consolidation are not bad in themselves;
it depends on whether they result from economic or political pro-
cesses. Critics are right to be wary of political distortions of the
contracting process. Even lobbying, a legal form of political influ-
ence, can be objectionable, but only if it departs from the model
of open competition based on relevant criteria established by agen-
cies accountable to the public. What must be avoided is unfair
competition, not successful competition. It may be true that small,
independent, nonprofit agencies supported by federal grants will
not be able to survive in a competitive market. If so, it will not
be because profit-making and competition are somehow incom-
patible with "performance, creativity and compassion." Instead,
such shakeouts are likely to make the industry more stable, more
reliable, and higher in quality.
Critics often cite the experience of defense contracting and
predict that the Pentagon's $500 hammer will be matched by
the Corrections Department's $500 million slammer. The anal-
ogy is not a very good one. Research, product development,
and capitalization costs are not anywhere near as high for cor-
rections as for the defense industry. Nonetheless, the compari-
son to defense contracting might still be useful because the
lessons that have been learned there in recent years may have
application to contracting for prisons. The 1984 Competition in
Contracting Act was partly a result of scandals in defense con-
tracting and was designed to reduce expensive sole-source and
cost-plus contracting. Success in achieving this goal has been re-
ported as "substantial." u Some of the techniques used to keep
Issues of Dependence 227

defense contractors competitive would presumably work for

prison contractors as well.
By the same token, lessons from defense contracting about the
limits of competition may also apply to the prison industry. These
would include lessons about not pushing competition purely for
the sake of competition; about limiting the number of contracts in
order to maintain economies of scale and to avoid overcapacity;
and about the shortsightedness of buying only from the lowest
bidder.
There may be an analogy between developing and producing
a small number of expensive weapons, such as Stealth bombers
or MX missiles, and a U.S. state that decides—as an experi-
ment—to authorize the construction and operation of one, and
only one, maximum security prison. With such a poor ratio of
high investment risk to low potential return, it is unlikely that
this state will attract much competition or realize low prices.
However, capital costs for prison contractors, unlike those for
many defense contractors, are not necessarily so high as to restrict
market entry; they will vary a great deal by size and type of facility
and will depend on whether the contract calls for new construction,
renovation or conversion of other property, or takeover of existing
governmental facilities. Start-up costs for single, especially low-
security, facilities are well within reach of small businesses or
groups of investors. As a new corporation, CCA was able to site,
finance, build, and open a 350-bed prison within 7 months, for $5
million. The U.S. Corrections Corporation, founded by two men
with an initial investment of $1.9 million, opened its first facility
at a seminary purchased for $695,000. If this is all it takes to enter
the market, it is well within the resources of numerous potential
competitors.
Market entry is likely to get easier, not harder. When corrections
is no longer an exclusively government-operated monopoly; when
enabling legislation has been passed and other legal uncertainties
resolved; when jurisdictions have in place mechanisms for con-
tracting and monitoring; when conventions and standards for
prison contracts begin to develop; when investors gain confidence
in the industry—as all these things occur, it will become progres-
sively easier for new companies, including small ones, to enter the
market and compete.
228 PRIVATE PRISONS

Facility Ownership and Government Dependency

on Contractors

If a jurisdiction's prisons are not only managed, but also owned

by a private corporation, will this make the government so de-
pendent that it loses effective control over the company and the
prison?
James Gentry suggests that the physical plant of prisons should
be owned by the government, because that would allow for shorter
contracts and for easier transfer to subsequent bidders.14 While he
recognizes that private ownership facilitates speed, savings, and
innovation in prison construction, Gentry thinks these could be
achieved under state ownership by design competition and lease-
purchase agreements separate from management contracts.
Separating management and ownership does have some advan-
tages, but it also has disadvantages. One problem is that design
and operation are intimately related. Efficiency may be lost if a
management company is neither allowed nor given any incentive
to invest in the design and construction of the facilities it will
operate.
The argument against private ownership of prisons assumes that
the physical assets of a prison are so expensive, specialized, and
nonredeployable that the market will be resistant to new entrants
and firms will require long contracts to recover their investments. '5
These long contracts will lock governments in to single suppliers.
An alternative to a long contract would be a periodically renewable
contract with provisions specifying the terms for immediate gov-
ernment purchase of the facility if and when the contract is ter-
minated. It might be objected that this would give the vendor
something to hold over the government, to deter it from termi-
nating the contract. However, if the government wishes to take
over the facility itself, it will be no worse off than if it had built
and financed the prison in the first place, and if it switches to a
new contractor it can have the new contractor take over the same
financing role as the previous contractor.
On the matter of dependency, it may well be asked: who has
whom over the barrel? Even where a state or county contracts
Issues of Dependence 229

away most of its own correctional capacity to a single private ven-

dor, does that put the government at the mercy of the vendor?
Clearly, the dependence works both ways. There is only one cus-
tomer, in a given jurisdiction, for correctional services. The vendor
may own the buildings, but either they will be difficult to convert
to other uses (which puts the government in a good bargaining
position) or they will not (which means that the government could
also acquire and convert other buildings). Either way, private own-
ership of the prison does not leave the government in an untenable
position.
Further, the company does not "own" its human capital. By
canceling a contract, the government instantly creates a pool of
surplus labor having just the characteristics now needed by the
government, or another contractor, as the new employer.

The Threat of Bankruptcy or Default

It is not inevitable that the government will become dependent on

contractors, but it can happen. To the degree that dependency
does occur, the government will have to worry about that con-
tractor going bankrupt. Even if the company never does go bank-
rupt, it can use the mere threat of bankruptcy to gain concessions
from the government, at least to the extent that the government
is in fact dependent on the contractor.
Thomas Coughlin, Commissioner of the New York Department
of Correctional Services, supports his concern over this issue by
citing a case in a related area. In 1978, the New York Department
of Mental Retardation contracted with a private, nonprofit agency,
the United Cerebral Palsy Association of New York, to run a
substantial part of a large facility in New York City. Three years
later, the company was $17 million in debt and filed for bank-
ruptcy. 16 The state resumed operation of the facility.
The most worrisome aspect of bankruptcy or default is the pros-
pect that a jurisdiction might find itself suddenly without the ca-
pacity to hold and care for its prisoners. Until there are a number
of regionally or nationally viable correctional contractors prepared
to enter the market in a particular jurisdiction, it would be wise
for the government to retain some capacity of its own. This is
230 PRIVATE PRISONS

possible even in a jurisdiction so small that it has only one jail or

prison. Such a jurisdiction would need to retain supervisory ca-
pacity for monitoring purposes, in any case.
What disappears when a contractor goes broke is the financial
structure and highest level of management, both of which the
government will still have. A bankrupt prison company leaves its
line staff and most of its middle management behind, ready for a
new employer. The plant and equipment can be purchased by the
government from the failed company (or its creditors) at liqui-
dation prices with money from insurance carried by the former
vendor as part of the contract. Or another company could step in
and take over a failing competitor—or purchase its assets—and
the government could switch the contract to the new company.
I do not mean to play down the fact that bankruptcy would cause
serious problems. However, it need not leave the government
empty-handed, or force it to empty its prisons or jails.
Since bankruptcy is a worst-case scenario for private prisons,
for the sake of perspective it ought to be compared to worst-case
scenarios for government prisons. While a government is unlikely
to go bankrupt, some prison systems have experienced interrup-
tions of services comparable to what might occur in a bankruptcy.
Courts have ordered the immediate closure of public prisons and
threatened to close down whole systems. Fiscally strapped and
debt-ridden state and local governments have been forced to re-
lease prisoners for purely budgetary reasons. Frustrated sheriffs,
transporting prisoners from overflowing jails, have left them
shackled to the gates of overflowing prisons unwilling to accept
them. 17 Convicts have been housed in warehouses, quonset huts,
tents, gymnasiums, trailers, schools, boats, and other makeshift
accommodations.
Clearly, many public prison systems are already in a state of
crisis and disruption, for which private prison contracts offer some
relief. In such systems, bankruptcy or other failures by a contractor
would mean little more than a return to the status quo ante.

Some Recommendations

To protect against defaults or bankruptcies, the Council of State

Governments and the Urban Institute make several recommen-
Issues of Dependence 231

dations to contracting jurisdictions. During bidding, and periodi-

cally throughout the contract, companies should be judged on
financial soundness and stability. Contracts should specify who is
to pay the costs created by termination (this could vary by whether
termination is for cause or otherwise). Performance bonds (equal,
for example, to the cost of one year's contract) should be consid-
ered. There should be a contingency plan to cover staffing, place-
ment of inmates, and control of the facility during emergency
transitions. 18 State officials in Kentucky required in their Requests
for Proposals that prison contractors be prepared to post a per-
formance bond equal to 70 percent of the annual value of the
contract. 19 In its Santa Fe contract, CCA posted a performance
bond, with a $325,000 certificate of deposit as collateral.20

Contracts and the Expansion of Government

While the general concept of privatization is often attractive to

conservatives and libertarians, because it implies a shrinking of
government, the specific mechanism of contracting has received
support from some liberals for just the opposite reason. These
supporters hope that contracts, by operating more efficiently, will
be able to provide more government services in spite of limited
tax revenues. Whether this is seen as a danger or an advantage,
it must be recognized that contracting can be used to extend, as
well as to reduce, the scope of government.
A major cause of government growth is the existence of special
interest groups that encourage government spending in those areas
that most directly benefit the group and its clients. Paul Starr argues
convincingly that "the most common privatization proposals, such
as contracting out and vouchers, would hardly diminish the domain
of 'special' interests." 21 He warns against the creation of "an
enlarged class of private contractors and other providers dependent
on public money." 22 Starr correctly notes that those who produce
public services—whether as public employees or as private con-
tractors—have a vested interest in increasing government expen-
ditures on those services.
Liberals have been blamed by conservatives for promoting the
growth of a parasitic "New Class"—defined by its relationship to
the means of public production—that thrives on the growth of
232 PRIVATE PRISONS

government. Unlike labor and capital—the two sides of the old

class structure—the New Class does not produce wealth but par-
ticipates in (and benefits from) its governmental transfer. The New
Class consists of the interest groups and lobbyists who call for
government services and programs, the politicians who order them,
the functionaries and professionals who administer the programs
and deliver the services, and the originally targeted individuals and
organizations who receive what is left at the end of this long pipe-
line (or trough, as the cynics refer to it). Now conservatives them-
selves may be vulnerable to charges that they have simply added
private contractors to the ranks of the New Class.
Contracting, by itself, is not going to limit the scope of govern-
ment, but it can restrain the cost. It is not a way of restricting the
demand for government service, but a method of controlling the
price. To do so, however, contracting must be competitive. It does
no good to substitute private monopolies for public ones. This is
one danger of privatization to which even its advocates should be
especially sensitive. While designed to inject characteristics of the
private sector into operations of the public sector, contracting
could have the reverse effect. Characteristics of government could
be extended through the use of private enterprise. We might end
up with the worst, rather than the best, of both worlds.
Something akin to noncompetitive contracting occurs when
government agencies create "off-budget enterprises" (OBEs).23
OBEs, according to Hans Sennholz, are governmentally created
and managed enterprises, whose spending and borrowing are not
recorded on any budget. Familiar examples would be airport, hous-
ing, parking, sewer, water, and other authorities, but there are
many other types as well. OBEs make it harder to control the
growth and cost of government. Indeed, OBEs have arisen in
response to attempts to statutorily or constitutionally restrict gov-
ernment spending and debt, which is why most of them are found
at the state and local, rather than federal, level.24 OBEs allow
governments
to spend and borrow without constraint, to dispense patronage with-
out civil service restrictions, and to bestow favors and benefits on
special groups. An OBE is an anomaly of organization: a govern-
ment entity unfettered by many of the statutory constraints appli-
cable to government, a corporation without stockholders but with
Issues of Dependence 233
a board of directors consisting of politicians or their appointees, a
non-profit business that competes with business or is protected from
competition as an unregulated monopoly. 25
There are important differences, however, between municipal
authorities or other OBEs and private contractors supplying public
services. Contracting reveals more expenses than it hides, and the
fees paid to contractors are on-budget. Contractors are not inde-
pendent authorities; they are accountable to their government
monitors who are, in turn, accountable to the public. In contrast
to contractors, "OBEs pay no taxes or license fees, post no per-
formance bonds, face little paperwork and regulatory tape that
strangle individual enterprise." 26

Exclusive Franchises

While correctional contracting may not create off-budget enter-

prises, it may in other ways contribute to what Sennholz is warning
against: expansion of the power of government. In particular, a
contract that took the form of an exclusive franchise would corrupt
the process of competition into just another form of governmental
monopoly. The State of Tennessee flirted briefly with this when it
took seriously a bold proposal by the Corrections Corporation of
America to lease the entire operation of the Tennessee Depart-
ment of Correction.
The distinction between an exclusive franchise and market dom-
inance is important. With an exclusive franchise, there is formal
protection against competition; in market dominance, there is not.
If one company runs all the prisons or jails for a single jurisdiction,
that does not by itself create either a monopoly or a dependency
situation. As long as there is no exclusive franchise, then the po-
tential for competition will still exist.
Competition occurs through time as well as across space. Even
with a long-term contract, a competitive environment can be main-
tained through provisions for periodic renewal, review, and pos-
sible termination. Nonetheless, wherever possible, government
should strive to preserve actual, as well as potential, competition.
It could do this by continuing to run some facilities itself, thereby
retaining its own capacity to compete. Or it could divide its facilities
234 PRIVATE PRISONS

among different contractors and hold open the possibility of re-

distribution of facilities at a later point. A classic example of com-
petition between contractors and the government, over time,
comes from a field other than corrections. In Phoenix, the city
public works department, by outbidding private competitors, suc-
cessfully regained one sector of the city's garbage collection that
it had lost to a private contractor a few years earlier.
In correctional contracting, Texas, California, and the Immi-
gration and Naturalization Service, as large systems, have had no
trouble maintaining contracts for secure confinement across mul-
tiple providers. Competition that is preserved in large systems like
these, or across jurisdictions nationwide, is competition that is
available for smaller systems as well. Corrections Corporation of
America and Wackenhut Corrections Corporation currently op-
erate facilities in only a handful of states. However, they are al-
ready large enough that they could provide competition in any
state. The absence of private prisons and jails in most states is due
to such factors as strong union opposition, absence of enabling
legislation, and caution or lack of interest on the part of correc-
tional and other governmental authorities. It is not because the
nature of corrections makes it impossible for competition to de-
velop rapidly in response to real demand.

Dependence vs. Competition

Objections that government may become dependent on particular

contractors, that capital costs will restrict market entry, that mar-
kets will be consolidated or distorted by a few powerful corpora-
tions, that corruption may undermine open competition, or that
exclusive franchises may occur, raise a number of legitimate con-
cerns. However, to present these objections, not as warnings of
dangers to avoid, but as arguments against the very concept of
privatization, is self-contradictory. The objections rest on an un-
derlying recognition that competition is desirable. They thus im-
plicitly reaffirm that which they seem to deny. With the market
model accepted as a standard of judgment, an imperfect, distorted,
or manipulated market is obviously undesirable, something to
avoid. But preserving a nonmarket, government monopoly of ser-
Issues of Dependence 235

vices would not avoid or solve the problem of imperfect com-

petition.
The AFSCME warns governments against becoming addicted
to contractors, but what they offer in its place is mandated de-
pendence on organized public employees. If dependence is a real
problem, will that problem be solved if there are no private ven-
dors? Is it not also a form of dependence when a service can be
supplied only by government employees, especially when they are
organized into unions that control the labor market? Private prison
companies can help free some governments from dependence on
public employees and their unions.
In sum, as an objection to private prisons, the forecast of gov-
ernment dependence on contractors has a self-defeating character.
To object that the private supply of a public service may not be
sufficiently competitive is not a very good argument for public
monopoly. The more essential a service, the greater the need for
a diversity of contingent sources of supply.
14
Private Prisons and the
Privatization of Punishment

The Future of Private Prisons

The future of privatization in one form or another in corrections

seems fairly secure. Contracting of services and of nonsecure facili-
ties is already a permanent feature of corrections. At this point,
contracting for the management of secure facilities is still unfamiliar
and controversial, but it gives every indication of gaining wide-
spread acceptance. The forces behind its emergence are still in place
and growing stronger. Crowding in existing prisons is increasing,
along with judicial pressure to do something about it. The judicial
solution—closing institutions or capping them, and fining jurisdic-
tions that, do not respond to court orders—only serves to increase
population pressures at other institutions and financial pressures on
government. While crime rates have declined slightly in recent
years, public demand for imprisonment has not. Polls show that the
public believes in longer sentences than are now being served and is
upset over early release as a response to crowding. At the same
time, the public often rejects the issuance of bonds to build new
prisons on governmentally borrowed money. Proprietary prisons
offer a way out of this dilemma, at an affordable price. As a result,
they are likely to be viewed as a viable option by increasing numbers
of federal, state, and local governments.
That privatization is a viable option has been argued and docu-

236
Private Prisons and the Privatization of Punishment 237

merited at length in this volume. I have examined here every known

criticism of private prisons and jails, and I do not dismiss any of them
lightly. However, I have been unable to find any criticisms of private
facilities that cannot be matched by equally serious and analogous
criticisms of noncontracted facilities. Virtually all potential prob-
lems facing private prisons have close counterparts among the prob-
lems troubling prisons run directly by the government. All prisons,
both public and private, face challenges in the areas of authority, le-
gitimacy, procedural justice, accountability, liability, cost, security,
safety, corruptibility, and so on. They face these challenges primar-
ily because of the nature of their mission, not because of their incor-
poration as public or private entities.
Some students of bureaucracy argue that all organizations are
public; that they vary only in their mix of economic and political
authority and in the degrees to which they exercise, and are con-
strained by, each type of authority. 1 Public choice theorists make
a parallel argument in the opposite direction: that public actors
and organizations respond to the same basic incentives of self-
interest as do those in the private sector.2 From either of these
perspectives, "public" and "private" prisons are more alike than
they are different.
Still, they are different, and these differences should be ex-
plored, experimented with, and exploited. The goal of running
prisons that are safe, secure, humane, efficient, and just, is too
important to reserve to the government. If that goal can be served
better by private companies, they should be allowed a chance to
prove it. If it is best served by the government, then the govern-
ment, too, should be required to demonstrate that fact empirically,
not merely announce it by edict.

The Privatization of Punishment

At the beginning of this book, I suggested that private operation

of prisons and jails can be seen as an extreme test of the limits of
privatization, because the administration of criminal justice, and
especially of punishment, is widely regarded as a core function of
government and the exclusive prerogative of the state. If the penal
function can be privately performed, what function cannot be?
238 PRIVATE PRISONS

However, while I have entered into what might be regarded as an

extreme area of privatization, I have also explored, so far, only
the most limited form of privatization: namely, contracting.
To say that contracting is a limited form of privatization is not to
minimize its significance. Indeed, the significance of contracting can
be seen in the number and variety of arguments that have been raised
against it.3 By examining these arguments in the case of incarcera-
tion, and finding them lacking, I have challenged their validity in
other areas as well. Nonetheless, as a method of privatization, con-
tracting poses little threat to the sovereignty or scope of the state.
Contracting does not deny government's responsibility to provide or
arrange for a particular public service; it only rejects a government
monopoly over the immediate production of that service.
A purist might argue that the term "privatization" ought to be re-
served for the total transfer of a function from the government to the
private sector. In relative terms, privatization could be said to occur
only to the degree that the government formally divests itself of re-
sponsibility for and authority over an activity. Privatization of im-
prisonment in this purist sense of the term is not an issue for practical
policy consideration at this time, or for the foreseeable future. It
might be worthwhile, however, to address the issue here as a theo-
retical question. Perhaps we can put the current policy debate over
"private prisons" in perspective if we try to imagine what autono-
mous private prisons—as opposed to contractually managed pris-
ons—might be like and how they might be justified. I cannot do
justice in one chapter to the concept of a completely private criminal
justice system.4 Moreover, I shall ignore two-thirds of that system
(police and courts) and look only at one type of post-conviction sanc-
tion: imprisonment. The full privatization of all justice functions
would require a more thoroughly libertarian society than the one we
have now, but we can try to envision what private prisons might be
like in such a society. To do so, we need to start with some basic prin-
ciples of libertarian philosophy.

Libertarian Views of the State

While a rationale for privatization can be found within a variety

of nonsocialist ideologies, the philosophy of governance most fa-
Private Prisons and the Privatization of Punishment 239

vorable to it is libertarianism. I use the term "libertarian" here in

a very broad sense, to encompass all political philosophies that
give priority to the liberty and rights of individuals over the welfare
and rights of society or the state. I include the ideas of classical
liberals,5 minimal-state libertarians, 6 and some individualist anar-
chists.7
For all their many differences, there are a few common axioms
among the ideas I lump together here as libertarian. These are:
1. Individual rights are natural, inalienable, and supreme.
2. The most fundamental of these are the rights to life, liberty,
and property.
3. No individual or group may rightfully initiate the use or threat
of coercive force against anyone else.
4. Within certain limits, individuals have the right to respond
with force to the initiation or threat of force by others.
5. The state has no rights or legitimate powers not originally
held by individuals, and therefore no unique claim to the
legitimate use of force.
6. The proper function of government (if any) is to enforce and
protect individual rights under the rule of law.
The challenge raised by anarchist libertarians is: given the ax-
ioms listed above, how can government in the form of a state exist
at all without violating individuals' natural rights? The true an-
archist's answer is: it can't.
Robert Nozick, in Anarchy, State and Utopia, answers the an-
archists with an "invisible hand" theory of the state, which he
offers as conceptually possible, rather than as historically accurate.
Nozick reasons that in a hypothetical, Lockean state of nature,
private protection agencies would arise spontaneously, charging a
fee to protect individuals against the violation of their rights by
others. These private agencies would play the roles now played by
governmental police, courts, jails, and prisons. By the nature of
its business, however, one of these protection agencies would, over
time, become dominant in any given geographic area, and thereby
take on the characteristics of a state. It would have a de facto
monopoly over the use of force, even if it did not claim this as an
exclusive right de jure. 8
Individualist anarchists like Murray Rothbard, 9 "anarchocapi-
240 PRIVATE PRISONS

talists" like David Friedman, 10 and some other libertarian legal

analysts' 1 believe, contra Nozick, that it is possible for multiple
and competing protection agencies to coexist within the same area.
Market mechanisms, they claim, would keep the competition from
being violent; thus, the existence of even a minimal state is neither
necessary nor desirable.
While most arguments about the feasibility of a stateless society
are theoretical, some libertarians use historical arguments. Viewed
historically, the notion that punishing criminals is the exclusive
prerogative of the state appears to be an invention of the state.
That is, its origins are not ancient, but coincident with the devel-
opment of modern nation-states. Prior to that development, re-
sponses to offensive behavior or injury primarily took the form,
not of vengeance or unrestricted violence, but of restitution and
compensation. As described by Randy Barnett:
The image of state criminal punishment arising from a bloody
Hobbesian jungle is pure myth. Monetary payments had replaced
violence as the means of dispute settlement and functioned well for
over 600 years. It was only through the violent conquest of England,
Ireland, and other parts of Europe that state criminal punishment
was reluctantly accepted.17
As a historical example of a "stateless" society, Rothbard cites
1,000 years of ancient Celtic Ireland.13
Unfortunately, there is no modern example of a stateless, com-
pletely libertarian society, and it seems unlikely that libertarian
ideals can ever be fully realized, any more than other ideals. None-
theless, discussions of ideal types (like the "free market" or the
"state of nature") have long been used to clarify philosophical
principles that may then be applied to the real world. It is not
necessary to resolve here the question of whether a more fully
libertarian society should take the form of a "minimal state" or
whether it ever could or should be completely "stateless." Either
way, attempts to limit the scope of government could include pro-
visions for the existence of relatively autonomous private prisons.14

Libertarian Principles of Crime and Punishment

To define criminal punishment as the exclusive prerogative of the

state is to accept a collectivistic (whether statist or communitarian)
Private Prisons and the Privatization of Punishment 241

concept of crime and punishment. In the collectivist view, all

crimes, whatever individual victims they may have, are also crimes
against some collectivity (the community, society, the state). Thus,
crimes are prosecuted, and punishment imposed, primarily on be-
half of the collectivity. Individual victims may even be excluded
from the process entirely, unless their testimony is crucial, in which
case victims may be compelled to testify, along with other wit-
nesses.
Because it is designed to serve the collectivity, the criminal jus-
tice system must be driven and directed primarily by representa-
tives of the collectivity. However, to the extent that such issues as
delegation, representation, and accountability are resolved (as I
have argued in preceding chapters they can be), a public/private
contract for the operation of a prison or jail is not incompatible
with the collectivistic view of crime and punishment. In contrast,
the idea of autonomous private prisons requires an individualistic
concept of crime and punishment.
From an individualistic and libertarian perspective, crime is seen
as a violation of a particular individual's rights. These rights are
natural, not to be created or removed at will by society. If a system
is created to protect these rights, it is still individuals who must
move that system. Whether this means that only victims have a
right to punish criminals is a matter of debate.
Some libertarians believe that punishment is the exclusive right
of victims. For example, Rothbard asserts:
In the libertarian society, there are only two parties to a dispute or
action at law: the victim, or plaintiff, and the alleged criminal, or
defendant. It is the plaintiff that presses charges in the courts against
the wrongdoer. In a libertarian world, there would be no crimes
against an ill-defined "society," and therefore no such person as a
"district attorney" who decides on a charge and then presses those
charges against an alleged criminal. 15

Murder poses a special problem for libertarians which they at-

tempt to resolve by referring to the victim "or his heirs." But this
just creates more problems. If a murder victim dies intestate,
who—if anyone—would have the right to punish the murderer? 16
The phrase "or his heirs" implies that the right to punish is
transferable. If so, why should it be transferable only to heirs?
Rothbard allows that a criminal might be permitted to buy his way
242 PRIVATE PRISONS

out of punishment. 17 Presumably, then, a victim could sell his right

to punish to the highest bidder. Criminals would be protected from
wealthy zealots (or sadists) by the principle of proportionality,
which "imposes the maximum limit on punishment that may be
inflicted before the punisher himself becomes a criminal aggres-
sor." 1X Within this limit, punishment will be imposed only when
there is some specific individual or group, beginning with the vic-
tim, that wants to see a criminal punished more than the offender
is willing—or able—to escape punishment. 19
Defining punishment as a transferable right raises many ques-
tions. The biggest problem with such a view is that it ignores the
moral dimension of punishment. Even if we accept the idea that
all rights (including the right to our own bodies and lives) are
ultimately property rights, this does not mean that every violation
of rights should be reduced to economic terms. A crime is not just
a harm, which arguably can be translated into an economic cost.
It is also a wrong, which cannot be so reduced.
One central thrust of libcrtarianism is its emphasis on individual
rights. Libertarian concepts of justice are normative and rights-
based, not utilitarian. It is contrary to libertarian principles to treat
individuals coercively as the involuntary means to some end. A
libertarian justification of punishment must therefore be a nonu-
tilitarian one.

The Justification of Punishment in Libertarian Society

Since the publication of John Rawls's A Theory of Justice (1971)

and Robert Nozick's Anarchy, State and Utopia (1974), there has
been a greatly increased emphasis on individual rights among phi-
losophers and political theorists, with a corresponding decline in
the influence of utilitarianism. 20 Utilitarian theories of criminal
justice are collectivistic, viewing society as the victim of all criminal
acts and justifying coercive treatment of offenders as the means
to a societal end. The end is crime control and the means include
rehabilitation, deterrence, and incapacitation. Rights-based the-
ories of criminal justice, in contrast, place the rights of individuals
above the interests of social groups and emphasize restitution or
retribution rather than crime control as the purpose of criminal
sanctions. 21
Private Prisons and the Privatization of Punishment 243

Andrew von Hirsch's Doing Justice, which appeared soon after

the books by Rawls and Nozick, reflected a shift among crimi-
nologists and penologists away from utilitarian conceptions of crim-
inal justice.22 Doing Justice sets forth what has come to be called
the "justice model." Following Kant, this model calls for penal
sanctions on moral grounds, as the "just deserts" for criminally
blameworthy conduct.23
Deterrence, incapacitation, and rehabilitation as purposes of
imprisonment are contrary to libertarian precepts. First, they are
utilitarian strategies that place a premium on collective welfare,
rather than individual rights. 24 Second, they are not merely reactive
to past behavior, but aim primarily at molding future conduct. The
offender is coerced not because of what he has done but because
of what he (or others!) might do. This has the effect, if not the
intent, of a rationalization for the initiation of force.
Punishment cannot be justified by a libertarian on the grounds
that it is also deserved; only on the grounds that it is deserved.
Otherwise, the justification is either superfluous or it involves a
degree of coercion beyond that which is called for by considerations
of justice alone. In the latter case, it is an attempt to justify the
initiation of force, which violates a fundamental axiom of liber-
tarianism.

Rights, Duties, and Punishment

How does a nonutilitarian, retributive view of punishment fit in
with libertarian principles of individual rights?
To believe in rights is to believe in duties. Those are alternative
statements of the same concept. To believe in duties is to accept,
implicitly but of logical necessity, the corollary of punishment.
When we say that people have a duty to refrain from violating the
rights of others, we are saying that there must be some negative
sanction if they fail to meet that duty. Duties are given meaning
by the consequences that attach to their nonfulfillment. The mean-
ing of a duty, like that of any other norm, must be socially con-
structed through the attachment of sanctions to behavior. A norm
(a rule, a law, a duty, a right) that had no sanction attached to its
violation would be empty and without meaning.
Recognizing that rights and duties are socially constructed
norms, libertarians understand the importance of sanctions in the
244 PRIVATE PRISONS

creation and maintenance of such norms.25 Thus, the maintenance

of the basic libertarian proscription against force and fraud requires
that acts of force and fraud be punished. The recognition that
rights must be socially defined through punishments, however,
does not mean that these sanctions may only be imposed by col-
lectivities, or in the name of the collectivity.
The belief that a particular social rule is based on a natural
right—or should be treated in this fashion—implies that every
individual has a right to participate in the social definition and
maintenance of that rule. This means that every individual can
help enforce the rule and punish those who violate it, since pun-
ishment is essential to the definition and preservation of the rule.
Thus, the natural right of individuals to be free from force and
fraud gives citizens license to punish acts of force and fraud.
Another way of expressing this idea is to say that sanctions serve
to uphold values. Failure to reward ethical behavior and to punish
misdeeds undermines the values of right and wrong. Persistent
failure to punish theft, for example, would erode and eventually
destroy the value we place on property rights.26 Thus, legal sanc-
tions are integral to legal norms and values. It is in this sense that
we may say that punishment is necessary to preserve the "integrity"
of the law.27 This refers not just to integrity in the sense of probity,
but also to integrity in the sense of completeness or unity.

The Sanction of Imprisonment in a

Libertarian Society

Some libertarians argue for restitution as the only valid function

of the criminal sanction. Randy Barnett and John Hagel believe
that all the rights relevant to criminal justice can be divided without
residue into two categories: victim's rights and offender's rights. 28
No third parties, including abstract third parties like society or the
state, have rights in their scheme. Only the victim has a right to
restitution and may use force against the offender for this, and
only this, purpose. The offender, in turn, has a right to propor-
tionality in the amount of force used against him for purposes of
restitution. "In short, the criminal's rights pick up at the exact
point that the victim's rights leave off." 29
Other libertarians, however, reject a purely restitutionary view
Private Prisons and the Privatization of Punishment 245

of criminal justice. Libertarian philosopher John Hospers, for ex-

ample, objects that restitution, by itself, is morally deficient be-
cause it does not take desert into account.30 Restitution rests on
theories of strict liability, while retribution requires an understand-
ing of the intent, and therefore of the culpability, of the offender.
Richard Epstein, a law professor, cautions against combining crim-
inal trials and civil actions into a criminal proceeding—of rolling
up restitution and punishment into one ball. To do so, Epstein
asserts, blurs important distinctions, confounds issues, and creates
more problems than it solves.31 Defining criminal justice in purely
restitutionary terms ignores the important moral distinction be-
tween a crime and a tort. Restitution concentrates on repairing a
harm (the essential element of a tort), not on punishing a wrong
(the essential element of a crime). The wrong done by a crime
extends beyond the harm done to the immediate victim. Without
reifying "society" or "the state," it can be said that crimes threaten
the conditions, and undermine the values, necessary for free and
civil association among all individuals. A violation of the rights of
one individual is thus in some measure a threat to the rights of all
individuals. Failure to punish the violation will weaken and un-
dermine the rights of all, and will erode the values that define those
rights.
This is what gives others, besides the immediate victim, the right
to punish criminals. It also explains why restitution, though it may
be essential to some other types of justice, is not central to criminal
justice. It is independent, just as the filing of a tort is independent
of the prosecution of a crime. Let torts be used to compensate
victims for harms. Punishment of crimes is still needed to uphold
the values served by criminal justice.
Given a retributive theory of punishment, another question re-
mains. Is there any special reason, consistent with libertarian prin-
ciples, why it should take the form of imprisonment? In a
libertarian society, the purpose of law is to enhance individual
liberty and to secure the natural rights of individuals, provided
they do not interfere with the rights of others. The function of
government, under law, is to maintain the minimal conditions nec-
essary for free and voluntary interaction among human beings. In
broad terms, the law prohibits initiation of force or fraud; beyond
that, it leaves people alone.
As libertarian philosopher J. Roger Lee points out, only one
246 PRIVATE PRISONS

form of legitimate punishment follows directly from this emphasis

on maintaining the conditions for free and voluntary association.32
That punishment is imprisonment: the removal of the civil liberty
of free association. While individual liberty is a natural right, the
civil liberty of free association with others is not. Rather, it is a
conditional privilege that must be earned by refraining from the
introduction of force or fraud in our interactions with others. Im-
prisonment, in Lee's view, is justified as a withdrawal of this liberty
from those who violate that minimal condition of civil association.
It is permissible to remove this liberty from those who, by the use
of force or fraud, limit the liberty of others. Lee notes that, while
society may also require criminals to make restitution and to work
off or pay for the cost of their confinement, those are not punish-
ments. They are civil remedies, not criminal sanctions, even though
they may be combined with the sanction of imprisonment.
A libertarian, or rights-based, jurisprudence, leading to a jus-
tification of imprisonment, can be summarized as follows:
1. Law is an expression, and a definition, of natural rights.
2. A crime is a willful and intentional violation of an individual's
natural rights.
3. In response to crime, penal (punitive) sanctions are called
for, to uphold the right that has been violated.
4. Imprisonment, the removal of the right to civil association,
is an appropriate punishment for crime because it defines,
expresses, and upholds the norms of civil association, which
crime violates and undermines.
Having developed an argument that imprisonment would be
justified as a sanction for crimes in a libertarian society, the ques-
tion remains: what would a libertarian prison look like?

The Libertarian Prison

J. Roger Lee and Laurin Wollan, Jr., in an article that is both

visionary and realistic, have described the rationale and general
design for a libertarian prison. My discussion throughout this sec-
tion relies heavily on their excellent article.33 Lee and Wollan
define a libertarian prison as "one which is itself based, in its
Private Prisons and the Privatization of Punishment 247

workings, on such principles as freedom from coercion, maximi-

zation of autonomy, and individual enterprise." It may or may not
be run by a business corporation. It is, however, conceived and
organized along the lines of a "free market enterprise." 34
The only official (governmental, societal) purpose of a libertar-
ian prison is to suspend the civil privilege of free association in
society at large.
And that is the extent of the permissibility. Nothing more is sanc-
tioned. It is not, for example, permissible to kill criminals, to torture
them, to deny them avenues for sex or for other pleasures, to make
them contemplate their evil ways, to retrain them, to mold them,
etc. No sanction for any such behavior comes out of the fact that
the criminal has breached the trust which is ordinarily placed in
people.'5

Because it is only the civil liberty of the offender that is sus-

pended, others (for example, his family) would be free to visit, do
business, or even to live with him inside the prison. With the crucial
exception of those special restrictions required to maintain the
sheer fact of imprisonment, prisoners would be subject to no reg-
ulations other than those that apply also to citizens on the outside.
Work is a fundamental human (and humanizing) activity. In a
libertarian prison, inmates would have the right to work, and to
earn as much as they can. This would not be "make work," but
productive work filling real needs as identified by a free market.
Inmates could be entrepreneurs, creating their own jobs and own-
ing their own businesses, or they could work for others. This would
neither reward crime nor take jobs away from noncriminals. In-
mates would not simply be "given" jobs. They would compete for
them or create them. Inmate income would derive from wealth
created by their work, not from tax money.
Inmates would not be able to keep all the money they earned.
Like all the rest of us, they would have bills to pay: for food,
shelter, clothing, and so forth. These bills would not necessarily
come from the government. Many costs would be privatized, by
transferring activities from the public to the private sector.
One bill, however, would be special: for restitution. Restitution
is consistent with libertarian principles, and a desirable goal, even
though it is independent of the purpose of imprisonment. It can
248 PRIVATE PRISONS

occur before, during, or after imprisonment, and it should occur;

but it is not directly a penal matter.
Labor may be required in a libertarian prison both to make
restitution and to pay for the cost of imprisonment. By their crimes,
offenders obligate society to punish them. In declaring an act to
be criminal, society makes a threat, or promise, to punish anyone
who commits that act. Punishing criminals, then, is more a mater
of society fulfilling an obligation than of offenders paying their
debts.36 It is not by committing a crime that offenders incur a debt
to society, but by forcing society to punish them. Thus, they cannot
pay their debts simply by being punished, because their punishment
is itself an expense for which they are responsible. As the willful
and blameworthy cause of an expense, an offender should be held
responsible for the cost of that expense.
Notwithstanding the universal obligation to pay one's bills, con-
victs in a libertarian prison would not be forced to work through
physical coercion (violence or threat thereof). They would not be
enslaved, indentured, nor leased like chattel. Offenders with (le-
gally acquired) wealth or independent means could support them-
selves and make restitution without working, if they chose to.
Those who either refused or were unable to support themselves,
and had no wealth to be garnisheed for this purpose, would meet
the same fate as those in a similar position on the outside.37
Crimes committed inside prison would be punished by new pros-
ecution, thereby adding to the sentence, as is already the case. In
most current prisons, however, the longer the sentence the weaker
any possible disincentive to crime inside prison. The limiting case
(in a state without capital punishment) is the "lifer," whose pun-
ishment cannot be increased. In a prison with an internal economy,
on the other hand, the longer an inmate's sentence, the greater
his stake in the stability of his local socioeconomic system. While
sentence time cannot compound indefinitely, fines and restitution
payments can.
The economy of a libertarian prison should flourish, like any
free economy. Participation by inmates in such an economy gives
them a vested interest in social order. The pursuit of voluntary
cooperative exchanges in prison may or may not have long-term
rehabilitative effects on inmates, but just as important would be
the institutional effect. In a laissez-faire prison the source of inmate
Private Prisons and the Privatization of Punishment 249

power would shift toward an economic basis and away from brute
force and intimidation. This new informal mechanism of control
would be less in conflict with the formal mechanisms, thus easing
the task of governance.

Entitlements of Condition in Libertarian Prisons

Libertarian prisons would help clarify many of the moral dilemmas
of imprisonment. One of these is the problem of special entitlement
of prisoners to benefits that are not legally guaranteed to citizens
on the outside. A government that believes in rehabilitation and
in humane treatment may require that prisoners be provided with
free counseling, job training, recreation facilities and programs,
good food, comfortable climate control, clean clothes and linen,
health care including medical, dental, and psychiatric treatment,
prescription drugs and eyeglasses, and a safe, clean, and sanitary
environment. These and other benefits may be required by law,
administrative policy, or court order. A court order in New Mexico
even mandates that prisoners be given free tobacco. Most citizens
outside of prison are not legally entitled to receive benefits like
these without charge. Not even the basic necessities of life such
as food, clothing, shelter, and emergency medical care are uni-
versal entitlements in our society. Why should they routinely be
provided for prisoners?
The answer, which is quite cogent in a nonlibertarian prison, is
that prisoners are not free to fend for themselves. If we deprive
prisoners of this freedom, we are thereupon obligated to provide
for them—and we must debate endlessly the question of what, and
how much, we must provide. In a libertarian prison, prisoners
would not be deprived of either the opportunity or the obligation
to earn things for themselves. Of course, the opportunities might
not be as great as they are on the outside, so perhaps the obligation
ought to be relaxed somewhat also.
When we assume coercive custody of convicted offenders, we
must also accept some responsibility for their welfare. However,
this responsibility does not have to extend to more than the basic
necessities for survival: food, clothing, shelter, and emergency
medical care. Beyond that, neither the courts nor other branches
of government should impose either a floor or a ceiling on the
250 PRIVATE PRISONS

living conditions that prisoners are able to create for themselves.

Considerable inequality of conditions would no doubt result, but
equality of condition is not a goal of libertarian society. Convicts
in a libertarian prison would have to obey rules, and those rules
should be applied and enforced equally, but within those param-
eters prisoners should be as free as possible to achieve the varying,
and therefore unequal, lifestyles they would be able to pursue on
the outside.

Contemporary Precursors to Libertarian Prisons

The libertarian prison envisioned by Lee and Wollan may seem

like a remote prospect, at least in pure form. It is not entirely
fanciful, however. Apparently unknown to Lee and Wollan, in
1973 John Price, an anthropology professor, published an article
titled "Political Enterprise in a Prison: The Free Market Economy
of La Mesa Penitenciaria." 38 That prison, in Mexico, matched
Lee and Wollan's ideal type description of a libertarian prison in
a remarkable number of particulars. The following is from Price's
abstract:
Conjugal visits by the prisoner's spouse, as well as long and very
open family visits, are regularly permitted. In these contacts quan-
tities of food, clothing, and money are allowed to be given to the
prisoners. A market system then develops out of the barter of these
goods, the purchase of bedding and special sleeping quarters, and
the operation of stores, restaurants, and small manufacturing firms.
The resulting prison society is thus similar to the social realities
outside the prison, and it draws the inmates into daily economic
decisions. . . . While La Mesa Penitenciaria has very serious prob-
lems, such as its heroin trade, for most of its inmates it is a humane
prison. y>

For a faint approximation of a libertarian prison in this country,

consider a short period in the recent history of the Maine State
Prison (MSP).40 For about 40 years, this maximum security prison
at Thomaston has allowed inmates to produce wooden craft items
and novelties that are sold at a store run by the prison on nearby
U.S. Route 1, a major tourist thoroughfare. For many years, the
Private Prisons and the Privatization of Punishment 251

crafts trade was limited and subsidiary to a state-run industry pro-

gram that produced goods for state use. 41
In 1976, a new warden was hired who instituted changes that
allowed the inmates greater economic liberty. As a result, Maine's
novelty sales program blossomed rapidly into what was undoubt-
edly the most economically successful inmate crafts program in the
country. What the new warden did was to appoint a supervisory
novelty committee, composed predominantly of inmates, and to
significantly raise the upper limits on value, variety, and volume
of output allowed per inmate. The revenue cap, for example, went
from $5,000 a year in 1976 to $15,000 in 1978.42 To circumvent
even these limits, more enterprising businessmen took on partners,
buying their quotas in exchange for a share of the profits.43
In contrast to most prison industry programs, which employ only
a small fraction of the population, 44 the crafts industry at MSP, at
its peak, involved from half to three-quarters of the inmates. In
1979, the prison store had gross sales of over half a million dollars.45
To keep the store open longer hours in the summer, inmates them-
selves paid the overtime salaries of the state-employed store
managers. Some inmates were said to earn over $38,000 a year,
with claims as high as $100,000 for the most successful "novelty
kings." 46
An unusual aspect of the MSP program was that it allowed
inmates to form businesses and employ other inmates. This pro-
moted labor specialization, managerial skills, and productivity.
The inmate economy was also allowed to have a legal and trans-
ferable currency, in the form of canteen coupons.47 This currency
allowed the development of a secondary economy within the
prison, offering such services as a barbershop, laundry, and tele-
vision rentals.
The novelty committee levied a tax on sales at the prison store.
The revenue was used to buy lounge furniture, athletic equipment,
movie rentals, and other amenities that the state could not or would
not provide. 4X The committee also issued patents for new product
patterns designed by inmates. Although inmates were restricted
in the number of patents that they could own personally, these
were effectively transferable property rights. When patent holders
left prison, they would sell their patterns to others.49 One inmate
controlled 50 patterns, only 10 of which could be in his own name. 50
252 PRIVATE PRISONS

In short, what developed at the Maine State Prison was not just
an inmate work program, but an environment that allowed pris-
oners to become entrepreneurs.
Unfortunately, the laissez-faire atmosphere at the prison may
have extended beyond purely economic matters. In 1980, four
years into this experiment in inmate capitalism, a team of outside
experts called in by the Bureau of Corrections concluded that the
staff had lost control of the prison to the convicts, and recom-
mended a complete lockdown. The warden resigned rather than
follow this directive, but the Director of Corrections stepped in
and confined all prisoners to their cells for two and a half months.
Following a thorough search of the prison, 50 dump-truck loads
of materials and possessions were confiscated and hauled away.
Knives and tools useable as weapons were found, but no guns or
ammunition. 51
Following the lockdown, the prison was reorganized and many
new restrictions were placed on the crafts industry. The cap on
gross income was cut from $15,000 to $8,000, canteen coupons
were declared nontransferable, and procedural restrictions were
placed on inter-inmate employment. 52 Some commentators view
the lockdown and its aftermath as an overreaction. For example,
Jeffrey Shedd asserts that:
In the final analysis, Maine State Prison was locked down because
it didn't fit into the correctional experts' picture of prison life. [This
is a picture in which] there is no place for ambitious and talented
individuals finding a way around bureaucratic restrictions on their
activities, for prison workers' wages being determined other than
by administrative fiat, for some inmates benefitting from others'
desires for haircuts, laundry services, loans, or anything else. In
short, the MSP lockdown occurred, not because authorities did not
have control over the prison, but because they did not have control
over the economic lives of the inmates. 5 '
This statement may be too strong. Even if MSP did not have a
record of violence as bad as many other maximum security prisons,
there was ample and independent substantiation of the investi-
gators' charge that inmates had gained, and staff had lost, too
much power. Whether loss of control by management can be
blamed on having a strong and relatively unrestricted inmate econ-
omy, however, is another matter.
Private Prisons and the Privatization of Punishment 253

Inmate economies of one sort or another are inescapable.54 If

legitimate occupations and legal currency are not available, hus-
tling and illegal trafficking in goods and services (drugs, alcohol,
cigarettes, food, protection, sex, gambling, and so on) will abound.
Some of the most out-of-control prisons, with severe problems of
mismanagement, staff corruption, and domination of some inmates
by others, have been prisons with little or no inmate industry. 55
In 1980, the crafts program at the State Prison of Southern
Michigan at Jackson, with 685 inmates participating, sold only
$5,887 worth of goods at its prison store, about one-tenth the level
of sales the previous year at Maine State Prison (where the total
population was only about 360).56 The crafts program at Southern
Michigan is tightly controlled. No inmate can produce more than
one type of item at a time, or order more than a fixed amount of
raw materials a month, or have more than $200 worth of goods
for sale in the prison store at one time. The limit on ordering
materials means that no inmate may profit by acting as a wholesaler
to other inmates. And no inmate may employ any other inmate.
But controlling the economy at the State Prison of Southern
Michigan has not controlled violence. In 1981, there was a major
riot there. A comparison of the Michigan prison system with that
of Texas, which has long had a self-sustaining (agricultural) prison
industry, found much lower rates of disorder in the latter. 57 Re-
cently, the Michigan prison system was the subject of a federal
investigation of drug-related corruption among guards, with
charges that officials may have protected inmate crime rings. In-
mates were reported to be smuggling drugs and guns into prisons,
where a black market in drugs was thriving. A deputy warden at
Jackson State Prison was arrested on charges of accepting a $10,000
bribe from a leader of the drug ring. One inmate testified that 85
to 90 percent of inmates at the prison sold marijuana. Another
said he made $5,000 a month dealing dope and another boasted
of earning $15,000 a week from the trade.58
In any prison, some inmates will exercise power over other in-
mates. When a major source of this power comes from legitimate
economic activities, it is easier for authorities to supervise and to
influence its exercise. Moreover, those whose power comes from
legal activities and is economic rather than violence-based will have
a greater stake in maintaining stability and a peaceful order. Thus,
254 PRIVATE PRISONS

there is no reason why the positive features of a self-sustaining

inmate economy need to be sacrificed in the name of control.
Inmate self-employment, employment by other inmates, legal cur-
rency, and high degrees of economic freedom and opportunity are
not incompatible with good prison management. While restrictions
on the ownership or location of certain kinds of tools, raw mate-
rials, or products does make sense in terms of safety and security,
income and production limits do not.
The experiment in inmate capitalism at the Maine State Prison
was a far cry from the ideal of a libertarian prison. Further, any
attempt to create a libertarian environment inside prisons today
would probably encounter problems at least a great as those that
occurred at MSP. The experience at Maine, however, does at least
suggest the possibility of constructing a relatively free economy
even in an otherwise very unfrec environment.

Contracting Revisited: A (Comparatively)

Modest Proposal

I do not know if a completely libertarian society, with autonomous,

self-supporting, and completely private prisons will ever be pos-
sible, though I do think the idea is philosophically defensible. What
1 have sketched here, admittedly superficially, is more of a vision
than a currently feasible proposal. Next to that vision, however,
the contracting of prisons and jails (as opposed to their complete
privatization) seems downright traditional and nowhere near as
radical a departure from current practices as critics seem to believe.
Contracting is a reform, not a revolution, and as such, it deserves
at least to be given a try.
Contractual operation of correctional facilities offers many po-
tential benefits. I will recount only a few of the more salient ones
here.
1. Perhaps the greatest advantage of contracting in any context
is that it makes the true costs of a service highly visible, allowing
them to be analyzed, compared, and minimized. This consequence
of contracting can benefit even jurisdictions that do not themselves
contract. They may observe the process elsewhere, or carefully
consider contracting some of their own operations but decide
Private Prisons and the Privatization of Punishment 255

against it. Either way, contracting provides competitive price and

product information that government can use to compare and eval-
uate its own operations, and see if they could be made more ef-
ficient. By revealing otherwise hidden costs, contracting enables
legislators and taxpayers to see where money is going and to decide
more rationally whether a program ought to be continued, ex-
panded, revamped, or discarded.
2. Where a jurisdiction does choose to privatize some of its
facilities, contracting may enable new prisons to be financed, sited,
and constructed more quickly and cheaply than is possible under
customary procedures for governmental construction. While this
process typically takes government two to five years, it has gen-
erally required just six months to a year for private prisons. Fur-
ther, private firms are more apt to design for efficient operation,
carrying these savings forward into future years.
3. In part because of its greater speed, but also because of
greater freedom in matters ranging from personnel to purchasing,
contracting allows greater flexibility, which promotes innovation,
experimentation, and other changes in programs, including ex-
pansion, contraction, and termination. Government can take pro-
grammatic risks through a contractor that it might avoid itself, for
fear that the changes could not easily be undone.
4. Contracting adds new expertise and specialized skills. This
may be particularly true for smaller counties and cities. Businesses
commonly hire outside consultants to help them reorganize and
develop new programs. In a similar fashion, larger correctional
contractors can use their most experienced staff from the central
office to establish a new facility under contract, then turn it over
to a more permanent local staff. 59 The experience and resources
of the central office, which may exceed those of a smaller city or
county by far, remain available throughout the life of the contract.
5. Contracting reduces the tendency toward bureaucratic self-
perpetuation and helps limit the size of government. While it is
true that contractors have a vested interest in encouraging greater
government spending on that which they provide, this impetus
toward growth may be easier to control than is the government's
own seemingly insatiable craving for internal expansion.
6. Contracting increases accountability because market mech-
anisms of control are added to those of the political process. In
256 PRIVATE PRISONS

the electoral competitions that underlie political accountability,

decisions are made by a large number of voters, each of whom
has only limited information and, with only a minuscule margin
of influence, little incentive to gather more. In contracting, com-
petitive bidders are motivated to supply relevant information to a
small number of politically accountable decision-makers. If it is
reasonable at all to suppose that a diffuse public can hold political
actors accountable for their own actions and decisions, then it is
even more reasonable to suppose that those actors in turn, as a
small and well-informed decision-making body, can hold contrac-
tors accountable for theirs.
7. Private prisons will tend to be highly visible, in contrast to
state prisons which, at least historically, have been ignored by the
public and given (until recently) "hands-off" treatment by the
courts. Though the attitude may be ill-informed, there does seem
to be greater public suspicion toward those who would wield power
"for profit" than toward those who would wield it for other rea-
sons, or for its own sake. Vigilance over those who run prisons is
always a gain, though the standards of supervision, and of per-
formance, ought to be the same for both public and private actors.
8. Contracting promotes the development and use of objective
performance measures. When the government spends taxpayers'
money to provide services, it has little natural incentive to measure
objectively the quality of its own performance. In contrast, it is in
the nature of a contract to specify service requirements, and to
some extent broader goals as well. All correctional contracts in-
clude provisions for monitoring and have language that at least
implies the need for some measurement of performance. Unfor-
tunately, we do not have validated, reliable, and standard measures
of performance for the operation of prisons, but there has been a
lot more discussion of the need for them in connection with private
prisons than there was when the government's monopoly was un-
questioned. Accreditation by the American Correctional Associ-
ation may not be fully satisfactory as a measure of performance,
but the fact that it is often required for contracted prisons and
jails, while remaining rare among those run by the government,60
suggests that contracting does encourage the measurement of
performance.
9. By creating an alternative, contracting encourages compar-
Private Prisons and the Privatization of Punishment 257

ative evaluations; this raises standards for the government as well

as for private contractors. In spite of the lack of objective mea-
sures, discussed above, there has always been plenty of criticism
of government performance in running prisons and jails. There
has not, however, been very effective public pressure for reform.
Most criticism of government corrections has been based on ab-
solute rather than relative standards. "Good enough for govern-
ment work" is an attitude that is tolerated by those who ultimately
pay for government work, but only when they can see no realistic
alternative. If the people standing in line at the Department of
Motor Vehicles had a clear vision of a commercial alternative,
public pressures for reform might be as strong in this area as they
are, say, for the postal system. By offering the public a visible
choice, privatization may be a force for correctional reform.
10. Where reform is greatly needed, but public management
has become entrenched and resistant to change, contracting can
provide a surgical solution. It is certainly easier to replace a bad
contractor with a better one than to replace an entire government
agency or operation with another governmental one. The replace-
ment of a government operation with a private one may fall some-
where in between. Where it becomes necessary or desirable to
thoroughly reform a particular prison, or even a small system,
contracting can produce wholesale and sudden change.
Beyond the advantages that I recap above, I have presented in
this book many other arguments that can be made in favor of the
private operation of prisons, jails, and other correctional or con-
finement facilities, and I have attempted to answer all the objec-
tions and criticisms of which I am aware. In the end, however, the
best case for private prisons does not rest on a priori claims that
they will be inevitably or necessarily superior. The strongest case
rests simply on a plea for open-minded comparison, and for eval-
uation against real alternatives rather than against absolute and
ideal standards. The private players ask only for a chance to prove
themselves in competition on a level field. We should ask no less
of the government's team.
This page intentionally left blank
Notes

INTRODUCTION

I.E. S. Savas, Privatizing the Public Sector: How to Shrink Government

(Chatham, NJ: Chatham House Publishers, 1982), p. 1.
2. Seymour Martin Lipset and William Schneider, The Confidence Gap:
Business, Labor, and Government in the Public Mind (New York: Free
Press, 1983).
3. Ronald A. Cass, "Privatization: Politics, Law and Theory," Mar-
quette Law Review 71(1988): 449-523, at 499.

CHAPTER 1

1. This section draws heavily on Charles H. Logan and SharlaP. Rausch,

"Punish and Profit: The Emergence of Private Enterprise Prisons," Justice
Quarterly 2 (1985): 303-305. Adapted with permission of the Academy
of Criminal Justice Sciences.
2. U.S. Department of Justice, Prisoners in 1987. Bulletin (Washington,
DC: Bureau of Justice Statistics, April 1988).
3. U.S. Department of Justice, Prisoners in 1983. Bulletin (Washington,
DC: Bureau of Justice Statistics, 1984).
4. Roy H. Reynolds [Note], "The Role of Special Masters in Federal
Judicial Supervision of State Prisons: The Need for Limitations," Amer-
ican Criminal Law Review 26 (Fall 1988): 491-511, at 491 (citing infor-
mation from the ACLU's National Prison Project).
5. If construction is paid for by a 20-year bond at 10 percent interest,
the real cost will triple the original figure. To these interest costs must be

259
260 Notes to Chapter 1
added an allowance for overruns. In a survey of 15 states, cost overruns
on prison construction averaged 39 percent above the initial budget. Over-
runs would include the effects of inflation during the time from bidding
to completion; hidden costs would include such things as architect and
agency fees, construction supervision, equipment, and insurance. After
calculating these costs, a proposed Connecticut prison reported to cost
$50,000 per bed would actually have cost $62,000 per bed, almost 25
percent more. See Bruce Cory and Stephern Gettinger, Time to Build?
The Realities of Prison Construction (New York: Edna McConnell Clark
Foundation, 1984), p. 16.
6. Gail Funke, "Who's Buried in Grant's Tomb? Economics and Cor-
rections for the Eighties and Beyond" (Alexandria, VA: Institute for
Economic and Policy Studies, 1983), p. 3.
7. Edwin W. Zedlewski, "The Economics of Disincarceration," Na-
tional Institute of Justice Research in Brief (Washington, DC: Department
of Justice, March 1984).
8. U.S. Department of Justice, Report to the Nation on Crime and
Justice: The Data (Washington, DC: Bureau of Justice Statistics, 1983),
p. 93.
9. Cory and Gettinger, Time to Build!, p. 17.
10. Rex Reed and David W. Holm, "The Monopoly Economics of
Juvenile Custody: Could Private Competition Keep Costs Down?" In-
dependence Issue Paper No. 1588 (July 29, 1988):!.
11. U.S. Department of Justice, BJS Data Report, 1988. (Washington,
DC: Bureau of Justice Statistics, April 1989).
12. Tennessee Journal 14, No.47 (November 21, 1988):2-3.
13. Robert Poole makes several points in rebuttal to the broad char-
acterization of contracting out as "destroying jobs." First, it is in the
interest of labor, as well as taxpayers and consumers, that workers be
employed as efficiently as possible. Second, a contract can be made to
specify that displaced government workers be given first preference for
contracted jobs. This is the federal government's official policy on con-
tracting out. Third, both government and nongovernment workers are
given a stake in higher productivity by the existence of competing firms;
they may even want to form their own firms through employee stock
ownership plans. This is the case with at least one of the private prison
contractors, Behavioral Systems Southwest. Finally, it should be remem-
bered that the purpose of both government agencies and private con-
tractors is not to provide jobs but to provide services. See Robert W.
Poole, Jr., "Objections to Privatization," Policy Review 24 (1983): 105-
119.
Notes to Chapter 2 261

14. Ralph de Toledano, Let Our Cities Burn (New Rochelle, NY: Ar-
lington House, 1975), p. 49.
15. Warren I. Cikins, "Privatization of the American Prison System:
An Idea Whose Time Has Come?" Notre Dame Journal of Law, Ethics
and Public Policy 2 (1986): 455.
16. Todd R. Clear and George F. Cole, American Corrections (Mon-
terey, CA: Brooks/Cole, 1986), p. 306.
17. Kevin Krajick, "Prisons for Profit: The Private Alternative," State
Legislatures 10 (1984): 14.
18. American Civil Liberties Union, Policy Guide, Policy #243, Board
Minutes, April 12-13, 1986.
19. Ira P. Robbins, "Privatization of Corrections: Defining the Issues,"
Judicature 69 (1986): 326. The primary constitutional issue referred to in
the ABA resolution is the question of private delegation of state authority.
Ironically, the ABA, which as a private organization accredits law schools
and thereby determines who can sit for bar exams, has survived repeated
attempts to challenge this private delegation of the state's licensing au-
thority to the ABA.
20. Martin Tolchin, "Bar Group Urges Halt in Use of Privately Run
Jails," New York Times, February 12, 1986.
21. Testimony of Thomas Beasley, Chairman, Corrections Corporation
of America, to President's Commission on Privatization, Washington, DC,
December 22, 1987.

CHAPTER 2

1. Though the term "corrections" became standard during the heyday

of rehabilitation, it is still used even by those who regard punishment as
the primary purpose of criminal sanctions. "Penal" facility is likewise
generic although more closely associated with a punitive orientation. Some
would object to calling pretrial detention or juvenile court placements
"penal" and therefore "prisons" because officially they are not punitive.
However, in this book, such distinctions will not often be important and
the term "prison" will frequently include jails and secure juvenile facilities.
2. Any combination of the elements—private ownership, private op-
eration, private management—is possible. There has even been one pri-
vate company that managed a jail owned by the county and staffed by
county employees, thus combining governmental operation and ownership
with private management.
3. U.S. Department of Justice, Children in Custody. 1982/83 Census
262 Notes to Chapter 2
of Juvenile Detention and Correctional Facilities (Washington, DC: Bureau
of Justice Statistic, September 1986), p. 3.
4. Michael A. Kroll, "Prisons for Profit," Progressive, September 1984,
p. 22.
5. The community-based contractors are all nonprofit agencies, by state
regulation. So far, the secure facility contractors have also been nonprofit,
but "DYS officials assert. . . that there would be little opposition to al-
lowing for-profit contractors." Joan Mullen, Kent John Chabotar, and
Deborah M. Carrow, The Privatization of Corrections (Washington, DC:
National Institute of Justice, February 1985), p. 62.
6. U.S. Department of Justice, Children in Custody 1982/83 Census.
7. Unpublished figures supplied by the Bureau of Justice Statistics for
1985 indicate 34,080 juveniles being held in 1,996 private institutions.
8. U.S. Department of Justice, Children in Custody 1982/83 Census.
9. U.S. Department of Justice, Children in Custody: Advance Report
on the 1982 Census of Private Juvenile Facilities, including Comparisons
with Public Facilities (Washington, DC: Bureau of Justice Statistics, March
1984).
10. Data from handwritten and unpublished tables at BJS.
11. Mullen et al, Privatization, pp. 56-58.
12. Kevin Krajick, "Prisons for Profit: The Private Alternative," State
Legislatures 10 (1984): 10.
13. Matthew J. Bronick, "The Federal Bureau of Prisons' Experience
with Privatization" (Washington, DC: Federal Bureau of Prisons, June
1989), Tables 3 and 4.
14. U.S. Department of Justice, Children in Custody 1982183 Census.
15. All data for this paragraph are from handwritten and unpublished
tables at BJS.
16. Kevin Krajick, "Punishment for Profit," Across the Board (1984):
23.
17. James O. Finckenauer, Juvenile Delinquency and Corrections: The
Gap Between Theory and Practice (Orlando, FL: Academic Press, 1984),
p. 182.
18. Ibid., p. 178.
19. Krajick, "Punishment for Profit," p. 25.
20. Mullen et al., Privatization, p. 65.
21. Ibid., pp. 62-63.
22. Krajick, "Punishment for Profit."
23. Mullen ct al, Privatization, p. 65.
24. Phone conversation with William R. Key, Juvenile Court, Memphis,
July 6, 1987.
Notes to Chapter 2 263
25. Los Angeles Times May 29, 1986.
26. Ibid.
27. San Jose Mercury News, March 15, 1985.
28. Behavioral Systems Southwest, "Prospectus" (Pomona, CA: Be-
havioral Systems Southwest, 1985).
29. Money, May 1986, p. 32.
30. Mullen et al., Privatization, p. 67.
31. Behavioral Systems Southwest, "Prospectus."
32. Mullen et al., Privatization, p. 67.
33. Ibid.
34. Peter Young, The Prison Cell (London: Adam Smith Institute, p. 8.
35. Philadelphia Inquirer, April 4, 1984.
36. Money, May 1986: 32.
37. Corrections Corporation of America press release, August 29, 1989.
38. Krajick, "Prisons for Profit," p. 11.
39. Corrections Corporation of America, "Second Quarter and Six
Month Earnings Announced," July 25, 1989.
40. Corrections Corporation of America, Annual Report 1986 (Nash-
ville, TN: Corrections Corporation of America, 1986), p. 24.
41. Hartford Courant, January 3, 1984.
42. T. Don Hutto and G. E. Vick, "Designing the Private Correctional
Facility," Corrections Today, April 1984, p. 85.
43. Mullen et al., Privatization, p. 68.
44. Ibid., pp. 67-68.
45. National Criminal Justice Reference Service, "Privatization Pro-
gram Search" (Rockville, MD: National Criminal Justice Reference Ser-
vice, 1985, computer printout).
46. Conversation with Robert Schmidt, Immigration and Naturalization
Service, May 21, 1987.
47. American Correctional Association, Director of Juvenile and Adult
Correctional Departments, Institutions, Agencies, and Paroling Authorities
(College Park. MD: American Correctional Association, 1987), p. 509.
48. Mullen et al., Privatization, pp. 68-69.
49. Commonwealth of Virginia, "Study of Correctional Privatization"
(Richmond, VA: Secretary of Transportation and Public Safety, 1986),
pp. 66, 71; Chicago Tribune, May 19, 1985.
50. News release from Wackenhut, June 12, 1989.
51. Robert Schmidt, telephone interview, May 21, 1987.
52. Telephone interview with Bob Greene, compliance manager at
Wackenhut Corrections Corporation, June 30, 1989. Information in the
rest of this paragraph is from Bob Greene.
264 Notes to Chapter 2
53. Actually, this facility, a former college, was opened without any
fences or armed guards, so its security at that point consisted mainly of
supervision.
54. Another multi-security county jail was taken over by private man-
agement on October 1, 1985: the Butler County (Pennsylvania) Prison
run by Buckingham Security Limited.
55. Bruce Cory, "From Rhetoric to Reality: Privatization Put to the
Test," Corrections Compendium, May 1986, p. 11.
56. [Louisville] Courier-Journal, May 19, 1986.
57. Ibid.
58. Charles Ring, Contracting for the Operation of Private Prisons: Pros
and Cons (College Park, MD: American Correctional Association, 1987),
p. 33.
59. Courier-Journal, May 19, 1986.
60. Commonwealth of Virginia, "Correctional Privatization," pp. 57-
58.
61. Commonwealth of Virginia, "Correctional Privatization," pp. 65,
80; Courier-Journal, May 19, 1986.
62. Ring, Contracting, p. 33.
63. Commonwealth of Virginia, "Correctional Privatization," p. 59.
64. Of these beds, 158 are kept available for prisoners of the U.S.
Marshall.
65. As described earlier, Hidden Valley was previously under contract
to the federal Bureau of Prisons as a training camp for Youth Corrections
Act offenders. The state contract began the same year the federal contract
ended. Criminal Justice Newsletter, June 16, 1986.
66. Ibid.
67. Site visit by author and interviews with county officials.
68. Cory, "Rhetoric to Reality," p. 13.
69. October 22, 1987, communication from David Myers, CCA Vice
President, Facility Operations.
70. New York Times, May 21, 1985.
71. Samuel Jan Brakel, "Amended Proposal on 'Privatizing' Correc-
tions," (American Bar Foundation, September 1986) p. 20.
72. Ibid., pp. 19-20.
73. National Criminal Justice Association, "Private Sector Involvement
in Financing and Managing Correctional Facilities" (Washington, DC:
National Criminal Justice Association, April 1987), p. 18.
74. The changes at Silverdale are in the following areas: security, med-
ical services, recreation, classification, programs, case management, food
services, maintenance, training and personnel, and chaplaincy. Brakel,
"Amended Proposal," p. 16.
Notes to Chapter 2 265

75. National Criminal Justice Association, "Private Sector Involve-

ment," p. 18.
76. Ibid., p. 18.
77. Communication from David Myers, CCA, October 22, 1987.
78. National Criminal Justice Association, "Private Sector Involve-
ment," p. 18.
79. The New Mexican, June 21, 1986.
80. Commonwealth of Virginia, "Correctional Privatization," p. 72.
Also, David Myers, communication, October 22, 1987.
81. Corrections Corporation of America, press release, June 26, 1987.
82. Corrections Corporation of America, 1986 Annual Report, p. 24.
83. New Mexican, June 21, 1986.
84."New Mexico Stat. Ann. 33-3-1 through 33-3-29 (1984). A 1985
statue also authorizes contracting at the state level: New Mexico Stat.
Ann. 33-3-17 (1985).
85. All information on the New Mexico statute presented in this par-
agraph was taken from Ring, Contracting for the Operation of Private
Prisons, Chapter 3.
86. St. Paul Pioneer Press, March 3, 1985.
87. Ibid.
88. Ibid.
89. New York Times, February 17, 1985.
90. National Criminal Justice Association, Exhibit 2.
91. Buckingham Security Limited, Private Prison Management: First
Year Report 1985-1986, Butler County, Pennsylvania (Lewisburg, PA:
Buckingham Security Ltd., 1986), p. 4.
92. Hartford Courant, April 1, 1984.
93. Commonwealth of Virginia, "Correctional Privatization," p. 73.
94. Buckingham Security Limited, Private Prison Mangement, p. 5.
95. National Criminal Justice Association, "Private Sector Involve-
ment," p. 20.
96. Buckingham Security Limited, Private Prison Management, p. 6.
97. Ibid., pp. 6-7.
98. Ibid., p. 6.
99. Ibid., p. 3.
100. Ibid., p. 1.
101. Ibid., p. 2.
102. Phone conversations on: April 11, 1988, with Joseph Fenton, then
at Buckingham Security; May 12,1989, with Robert Thompson, Chairman
of Butler County Commission; May 12, 1989, with Rick Jalotty, then
Warden of Butler County Prison; May 12, 1989 with Charles Fenton,
formerly warden of the prison under Buckingham. All of these sources
266 Notes to Chapters 3 and 4

described the nonrenewal issue as "very political." Several sources re-

ported that the union shop steward, a leader of the opposition to Buck-
ingham, was a constant source of difficulty while working as a counselor
under Buckingham's administration. When Buckingham eventually fired
him, the incoming county commissioners created a new payroll position
for him as "liaison" between the jail and the commission. After the con-
tract expired, they made him Deputy Warden.

CHAPTER 3

1. This section is adapted with permission from Charles H. Logan,

"Proprietary Prisons," in Lynne Goodstein and Doris L. MacKenzie, eds.,
The American Prison: Issues in Research and Policy (New York: Plenum,
1989).
2. Sources used include the following: Judith C. Hackett, Harry P.
Hatry, Robert B. Levinson, Joan Allen, Keon Chi, and Edward D. Fei-
genbaum, "Contracting for the Operation of Prisons and Jails," National
Institute of Justice Research in Brief (Washington, DC: Department of
Justice, June 1987); Robert B. Levinson, "Okeechobee: An Evaluation
of Privatization in Corrections," Prison Journal 65 (1985): 75-94; Charles
H. Logan and Sharla P. Rausch, "Punish and Profit: The Emergence of
Private Enterprise Prisons," Justice Quarterly 2 (1985): 303-318; Charles
H. Logan, "The Propriety of Proprietary Prisons," Federal Probation 51
(1987): 35-40; Joan Mullen, "Corrections and The Private Sector," Na-
tional Institute of Justice Research in Brief (Washington, DC: Department
of Justice, March 1985); Charles Ring, Contracting for the Operation of
Private Prisons: Pros and Cons (College Park, MD: American Correc-
tional Association, 1987); and E. S. Savas, Privatizing the Public Sector:
How to Shrink Government (Chatham, NJ: Chatham House Publishers,
1987).

CHAPTER 4

1. Jan Elvin, "A Civil Liberties View of Private Prisons," Prison

Journal 65 (1985): 51 (emphasis added).
2. American Civil Liberties Union, Policy Guide, Policy #243, Board
Minutes, April 12-13, 1986.
3. Cited in Newsweek, May 7, 1984.
4. Cited in Kevin Krajick, "Punishment for Profit," Across the Board
21 (March 1984): 27.
5. John J. Dilulio, Jr., "What's Wrong with Private Prisons," Public
Notes to Chapter 4 267

Interest No. 92 (Summer 1988): 66-83; John J. Dilulio, Jr., "Prisons,

Profits and the Public Good: The Privatization of Corrections," Research
Bulletin No. 1 (Huntsville, TX: Sam Houston State University Criminal
Justice Center, 1986). Dilulio and I differ in our beliefs about the propriety
of proprietary prisons. However, on virtually every other aspect of prison
management, I am in total agreement with his views, which are having a
profound effect on today's and tomorrow's thinking about corrections.
Even on the matter at hand, I feel compelled to express appreciation and
respect, before 1 attack. Many other critics of private prisons simply pro-
claim them to be "wrong" for ideological, moral, or "philosophical" rea-
sons, without accepting the philosopher's obligation to examine and
defend that judgment, as Dilulio does.
6. Dilulio, "Prisons, Profits and the Public Good," pp. 4-5.
7. Dilulio, "What's Wrong with Private Prisons," p. 82.
8. Ibid., p. 81.
9. Dilulio, "Prisons, Profits and the Public Good."
10. Dilulio, "What's Wrong with Private Prisons," pp. 78-89.
11. Ibid., p. 83.
12. This and portions of three other sections of this chapter were orig-
inally published in Charles H. Logan, "The Propriety of Proprietary Pris-
ons," Federal Probation 51 (September 1987): 35-40, and reprinted in
Charles H. Logan, "Proprietary Prisons," in Lynne Goodstein and Doris
L. MacKenzie, eds., The American Prison: Issues in Research and Policy
(New York: Plenum, 1989). Adapted with permission.
13. Peter M. Blau, Exchange and Power in Social Life (New York:
John Wiley & Sons, 1967).
14. Ira P. Robbins, "Privatization of Corrections: Defining the Issues,"
Judicature 69 (April-May 1986): 331.
15. Dilulio, "What's Wrong with Private Prisons," p. 79.
16. Michael Walzer, "At McPrison and Burglar King It's. . . Hold the
Justice," New Republic, April 8, 1985, p. 11.
17. I am indebted to Douglas McDonald for suggesting this term.
18. David M. Lawrence, "Private Exercise of Governmental Power,"
Indiana Law Journal 61 (1986): 647-695, at pp. 666-667.
19. Laurin A. Wollan, "Privatization of Criminal Justice," in Proceed-
ings of the 29th Annual Southern Conference on Corrections (Tallahassee
FL: School of Criminology, Florida State University, 1984), pp. 111-124.
20. E. S. Savas, Privatizing the Public Sector: How to Shrink Govern-
ment (Chatham, NJ: Chatham House Publishers, 1987), p. 60.
21. William C. Collins, "Privatization: Some Legal Considerations from
a Neutral Perspective," in Collins: Correctional Law, 1986 (Olympia, WA:
William C. Collins, 1986), pp. 81-93, at p. 85.
22. Ibid., p. 87.
268 Notes to Chapter 4
23. Lawrence, "Private Exercise of Governmental Power," pp. 648-649.
24. Ibid., p. 662.
25. Ronald A. Cass, "Privatization: Politics, Law and Theory" Mar-
quette Law Review 71 (1988): 449-523, at 501. Citations to Ward v. Village
of Monroeville, 409 U.S. 57 (1972); Tumey v. Ohio, 273 U.S. 510 (1927).
26. David N. Wecht [Note], "Breaking the Code of Deference: Judicial
Review of Private Prisons," Yale Law Journal 96 (1987): 815 837.
27. Ibid., p. 834, note 102 and accompanying text.
28. Peter Greenwood, "Private Prisons: Are They Worth a Try?" Cal-
ifornia Lawyer, July/August 1982, pp. 41-42. Greenwood asserts that
these functions account for less than 5 percent of current prison admin-
istration budgets, so it would not burden the state to retain full respon-
sibility for their administration.
29. Ira P. Robbins, The Legal Dimensions of Private Incarceration
(Washington, DC: American Bar Association, 1988), p. 54.
30. Courts may not welcome being drawn directly into the initial decision
process themselves, however. According to a news report, Corrections
Corporation of America "drew the ire of Hamilton County's Sessions
Court judges when the company took inmates downtown to be charged
for what Judge Richard Holcomb said were instances the company should
have handled internally." The court objected to being asked by CCA to
settle internal disputes that state prisons handled by withdrawal of priv-
ileges and denial of good time and other credits after internal adminis-
trative hearings. See Chattanooga Times, August 12, 1986.
31. Timothy L. Fitzharris, The Desirability of a Correctional Ombuds-
man (Berkeley, CA: Institute of Governmental Studies, 1973).
32. Richard A. Waples, "The Privatization of Prisons: The Wrong
Solution for a Real Problem," presented at "A Critical Look at Privati-
zation in Corrections," a conference sponsored by the Indiana Department
of Corrections, Indianapolis, IN, January 29, 1988.
33. John W. Palmer, Constitutional Rights of Prisoners, 2nd ed. (Cin-
cinnati: Anderson Publishing Co., 1977), p. 24.
34. Robbins, Legal Dimensions of Private Incarceration, p. 65.
35. Ibid., p. 449.
36. Ibid., pp. 317-318.
37. Ibid., pp 64-65.
38. Todd R. Clear and George F. Cole, American Corrections (Mon-
terey, CA: Brooks/Cole, 1986), p. 319.
39. Including the ACLU. See William A. Donohue, The Politics of the
American Civil Liberties Union (New Brunswick, NJ: Transaction Books,
1985).
40. Francis A. Allen, The Borderland of Criminal Justice (Chicago:
Notes to Chapter 5 269
University of Chicago Press, 1964); David J. Rothman, The Discovery of
the Asylum (Boston: Little, Brown, 1971); American Friends Service
Committee, Struggle for Justice (New York: Hill & Wang, 1971).
41. Willard Gaylin, Ira Glasser, Steven Marcus, and David J. Rothman,
Doing Good: The Limits of Benevolence (New York: Pantheon Books,
1981).
42. Bruce L. Benson, "Guns for Protection, and Other Private Sector
Responses to the Government's Failure to Control Crime," Journal of
Libertarian Studies 8 (1986): 75-109, at pp. 85-86.
43. A recent note in the Yale Law Journal discusses at length the
advantages of using financial incentives, like bonuses and fines, to monitor
and control private prisons. The author argues that by properly structuring
their incentives, profit-seeking prisons can be powerfully motivated to
maximize not just their profit, but also their effectiveness, as measured
by recidivism. See James Theodore Gentry [Note], "The Panopticon Re-
visited: The Problem of Monitoring Private Prisons," Yale Law Journal
96 (1986): 353-375.
44. David J. Rothman, Conscience and Convenience: The Asylum and
Its Alternatives in Progressive America (Boston: Little, Brown, 1980).

CHAPTER 5

1. Joan Mullen, Kent John Chabotar, and Deborah M. Carrow, The

Privatization of Corrections (Washington, DC: National Institute of Jus-
tice, February 1985), p. 37.
2. Ibid., pp. 40-41.
3. National Institute of Justice, "Corrections and the Private Sector:
A National Forum." Proceedings of a meeting sponsored by the National
Institute of Justice, February 20-22, 1985, Washington, DC, p. 6.
4. Susan M. Duffy, "Breaking into Jail: The Private Sector Starts to
Build and Run Prisons," Barron's, May 14, 1984, pp. 20-22.
5. James B. Jabobs, New Perspectives on Prisons and Imprisonment
(Ithaca, NY: Cornell University Press, 1983), pp. 115-132.
6. Ibid., p. 126.
7. Timothy J. Flanagan and Susan L. Caulfield, "Public Opinion and
Prison Policy: A Review," Prison Journal 64 (Fall/Winter, 1984): 31-46,
at p. 37.
8. Richard Crane, Vice President, Legal Affairs, Corrections Cor-
poration of America, testimony, November 13, 1985, U.S. Congress.
House Committee on the Judiciary. Privatization of Corrections. Hearings
before the Subcommittee on Courts, Civil Liberties, and the Administration
270 Notes to Chapter 5
of Justice, 99th Cong., 1st and 2nd sessions, November 13, 1985, and
March 18, 1986, Serial No. 40 (Washington, DC: U.S. Government Print-
ing Office, 1986), p. 29.
9. Kevin Krajick, "Punishment for Profit" Across the Board 21 (March
1984): 20-27.
10. Ibid.
11. Ibid.
12. Newsweek, May 7, 1984.
13. Hartford Courant, April 1, 1984.
14. T. Don Hutto and G. E. Vick, "Designing the Private Correctional
Facility," Corrections Today (April 1984), p. 85.
15. Crane, Privatization of Corrections, p. 33.
16. Construction itself took 90 days. Telephone interview with Bob
Greene, Wackenhut Corrections Corporation, June 30, 1989.
17. The facility is the Central Texas Parole Violator Facility in San
Antonio, which opened in January 1989. Information from Wackenhut
Corporation, June 14, 1989.
18. George Camp and Camille Camp, "The Real Cost of Corrections:
A Research Report" (South Salem, NY: Criminal Justice Institute, April
1985), p. 3
19. At Hidden Valley Ranch, when run by Electric Communications,
Inc., the cook earned $4 less per hour than he did in the same job at a
nearby San Francisco County boys' ranch. The warden earned one-third
less than he did as warden at the federal prison in Terre Haute, Indiana.
San Jose Mercury News, March 15, 1985.
20. Charles Ring, Contracting for the Operation of Private Prisons: Pros
and Cons (College Park, Md: American Correctional Association, 1987),
pp. 28-29.
21. Erik Larson, "Captive Company," Inc., June 1988, p. 90.
22. Corrections Corporation of America, Company Report, April 3,
1987.
23. Gail Funke, "The Economics of Prison Crowding," Annals, 478
(1985): 88.
24. E. S. Savas, Privatizing the Public Sector: How to Shrink Govern-
ment (Chatham, NJ: Chatham House Publishers, 1982), p. 24.
25. All quotes in this paragraph are from American Correctional As-
sociation, Private Sector Operation of a Correctional Institution (Wash-
ington, DC: U.S. Department of Justice, National Institute of Corrections,
April 1985), p. 48.
26. Buckingham Security Limited, Private Prison Management: First
Year Report 1985-1986, Butler County, Pennsylvania (Lewisburg, PA:
Buckingham Security Ltd., 1986), p. 3.
Notes to Chapter 5 271

27. Ibid., p. 5, with update via personal communication from Charles

Fenton, Warden.
28. Ibid., pp. 4-5.
29. John Hanrahan, "Why Public Services Should Stay Public," Des
Moines Register, March 31, 1983.
30. Kevin Krajick, "Prisons for Profit: The Private Alternative," State
Legislatures 10 (1984): 9-14; Philadelphia Inquirer, April 16, 1984.
31. John D. Donahue, Prisons for Profit: Public Justice, Private Interests
(Washington, DC: Economic Policy Institute, 1988), p. 4.
32. John J. Dilulio, Jr., "Prisons, Profits and the Public Good: The
Privatization of Corrections," Research Bulletin No.l (Huntsville, TX:
Sam Houston State University Criminal Justice Center, 1986), p. 3.
33. Ibid. See also John J.Dilulio, Jr., Governing Prisons: A Comparative
Study of Correctional Management (New York: Free Press, 1987).
34. Peter F. Drucker, Management: Tasks, Responsibilities, Practices
(New York: Harper & Row,1973), pp. 141 ff.
35. Robert W. Poole, Jr., "Objections to Privatization," Policy Review
24 (1983): 106.
36. This also means that they can fully satisfy no one. For a discussion
of this as it pertains to prisons, see Charles R.Tittle, "Prisons and Re-
habilitation: The Inevitability of Disfavor," Social Problems 21 (1974):
385-395.
37. Poole, "Objections to Privatization," p. 107.
38. William L. Megathlin, Dennis D. Murphy, and Robert E. Magnus,
Feasibility of the Establishment of Regional Prisons (Washington, DC:
U.S. Department of Justice, National Institute of Corrections, May 1984)
pp. 45-50.
39. Erwin O. Smigel, "Public Attitudes Toward Stealing as Related to
the Size of the Victim Organization," American Sociological Review
(1956): 320-327.
40. Peter Samuel, "Battling the Budget—Gracefully," Reason 16
(1984): 36.
41. Ibid., p. 38.
42. See Chapter 1.
43. Harry W. Miley, Jr., "Cost Analysis of State vs. Private Correctional
Facilities," (Columbia, SC: University of South Carolina), p. 3.
44. American Correctional Association, Private Sector Operation of a
Correctional Institution p. 47.
45. Information supplied by Bill McGriff, Hamilton County Auditor,
October 27,1987.
46. Samuel, "Battling the Budget," p. 38.
47. Camille G. Camp and George M. Camp, Private Sector Involvement
272 Notes to Chapter 5

in Prison Services and Operations (Washington, DC: U.S. Department of

Justice, National Institute of Corrections, February 1984.)
48. Ibid., p. 5
49. Ibid., p. 7.
50. Ibid., p. 10.
51. Ibid., p. 12.
52. Moreover, the Camps noted that even those agencies not reporting
cost savings "concluded that the operational benefits more than out-
weighed the cost factor." Ibid., p. 10.
53. Camille Camp and George Camp, "Correctional Privatization in
Perspective," Prison Journal 65 (1985): 24.
54. Samuel, "Battling the Budget," p. 39.
55. U.S. Office of Management and Budget, Enhancing Governmental
Production Through Competition: Targeting for Annual Savings of One
Billion Dollars by 1988 (Washington, DC: Office of Federal Procurement
Policy, March 1984).
56. Savas, Privatizing the Public Sector, pp. 89-117.
57. American Correctional Association, Private Sector Operation of a
Correctional Institution, p. 69.
58. Ibid., p. 67.
59. Ibid., p. 52.
60. Ibid., p. 49.
61. Dozier added 17.5 staff positions in 1982-83 and 32 positions the
year after that. During the transition year, Eckerd reduced the staff at
Okeechobee and concentrated on upgrading the physical facilities. The
next year, the Foundation contributed $236,000 of its own money and
returned the staff to above its previous level. Ibid., pp. 15, 48.
62. Judith C. Hackett, Harry P. Hatry, Robert B. Levinson, Joan Allen,
Keon Chi, and Edward D. Feigenbaum, Issues in Contracting for the
Private Operation of Prisons and Jails (Washington, DC: Department of
Justice, October 1987), p. 53.
63. Mullen et al., Privatization of Corrections, p. 65.
64. Philadelphia Inquirer, August 12, 1984.
65. Bruce Cory, "From Rhetoric to Reality: Privatization Put to the
Test," Corrections Compendium, May 1986, p. 12.
66. [Louisville] Courier-Journal, February 16, 1986.
67. Hackett et al., Issues in Contracting, pp. 52-53.
68. Mullen et al., Privatization of Corrections, p. 68.
69. Philadelphia Inquirer April 16, 1984.
70. Crane, Privatization of Corrections, p. 29.
71. Samuel Greengard, "Making Crime Pay," Barrister Magazine 13
(1986): 15.
Notes to Chapter 5 273
72. Money, May 1986, p. 32.
73. John W.Moore, "Paying for Punishment" National Journal No. 11
(March 14, 1987): 616.
74. Santa Fe New Mexican, June 21, 1986.
75. New Mexican, August 16, 1987.
76. Florida Times-Union, November 24, 1985.
77. Ibid. The news account does not explain the discrepancy between
the 12 percent differential for the per diem figures and the 22 percent
differential for total next year's costs. It may be that the per diem figures
referred only to the current jail facility, while the total next year's costs
included a Jail Annex, planned for construction.
78. Interview with John Hutt, Bay County Commissioner, February 25,
1987.
79. Larson, "Captive Company," p. 90.
80. The fee is $28 for work release inmates.
81. Hackett et al., Issues in Contracting, p. 53.
82. All figures are from the Cleveland Plain Dealer, October 1, 1984.
83. Philadelphia Inquirer, April 16, 1984.
84. Lee Kravitz, "Tough Times for Private Prisons," Venture (May
1986): 56; Buckingham Security Limited, Private Prison Management, p. 4
85. Pennsylvania Legislative Budget and Finance Committee, Report
on a Study of Issues Related to the Potential Operation of Private Prisons
in Pennsylvania (Harrisburg, PA: Pennsylvania Legislative Budget and
Finance Committee, 1985), p. 63.
86. Massachusetts Legislative Research Council, Report Relative to Pris-
ons for Profit, July 31,1986, pp. 80-81. See also Samuel Jan Brakel, "Pri-
vatization and Corrections," Federal Privatization Project Issue Paper No.
7 (Santa Monica, CA: Reason Foundation, January 1989), pp. 10-11.
87. Norman A.Carlson, Director, Bureau of Prisons, testimony, March
18,1986, U.S. Congress. House Committee on the Judiciary. Privatization
of Corrections. Hearings before the Subcommittee on Courts, Civil Lib-
erties, and the Administration of Justice, 99th Cong., 1st and 2nd sessions,
November 13, 1985, and March 18, 1986, Serial No. 40 (Washington, DC:
U.S. Government Printing Office, 1986), p. 133.
88. Los Angeles Times, March 29, 1985.
89. Fresno Bee, May 8, 1987.
90. For example, would the new facilities require financing, construc-
tion, and purchase or rental of land, which existing state prisons do not?
91. Hackett et al., Issues in Contracting, p. 6.
92. Before-and-after comparisons might seem, at first glance, to avoid
some of the problems of cross-facility comparisons. Region and location
remain the same, as does (usually) the inmate population. However, many
274 Notes to Chapter 5
other things relevant to cost may change—indeed, it is frequently one of
the goals of privatization to bring about these changes. The private com-
pany may renovate or build to increase capacity. It may introduce new
programs. It is likely to be required to seek accreditation and thus to meet
standards not previously met. Monitoring, which may be included in the
cost of the contract, adds a dimension that makes the operation different
than before. In short, even a before-and-after analysis does not compare
the "same" facility under two forms of management, although it comes
closer to it than a cross-facility comparison does.
93. E. S. Savas, "Municipal Monopolies Versus Competition in De-
livering Urban Services," pp. 473-500, in W. D.Hawley and D. Rogers,
eds., Improving the Quality of Urban Management (Beverly Hills, CA:
Sage, 1974).
94. E. S. Savas, "How Much Do Government Services Really Cost?"
Urban Affairs Quarterly 15 (1979): 23-42.
95. Ibid., p. 31, citing P. Kemper and J. M. Quigley, The Economics
of Refuse Collection (Cambridge: Ballinger, 1976).
96. Ibid., p. 34.
97. Hidden costs are not the same as overexpenditures or unauthorized
expenditures. Such expenditures also will not appear in a budget, which
is a prospective authorization, but they will show up in account books at
year's end (perhaps in a disguised form, if they are unauthorized). Hidden
costs, however, appear in neither the budget nor the report of expenditures
of the particular agency or institution in question.
98. This would parallel the 30 percent underestimate found in studies
of governmental budget figures for refuse collection.
99. George Camp and Camille Camp, "The Real Cost of Corrections:
A Research Report" (South Salem, NY: Criminal Justice Institute, 1985).
100. Ibid., p. 3.
101. The Correction Department gave $26,000 per inmate as the direct
operating expense for its 11 jails in Fiscal Year 1984. When the Correc-
tional Association included such costs as fringe benefits, debt financing,
and interagency costs, they estimated the cost to be $40,000. See William
G. Blair, "Inmate Cost is Put at $40,000 a Year," New York Times,
December 27, 1984.
102. Camp and Camp, "Real Cost of Corrections," p. 3.
103. Cited in Morgan O. Reynolds, Power and Privilege: Labor Unions
in America (New York: Universe Books, 1984), p. 194.
104. President Jimmy Carter's Commission on Pension Policy concluded
that "If the government pension system were subject to the same funding
requirements as private plans, the cost in 1980 would be 79.8% of payroll."
Cited in Reynolds, Power and Privilege, p. 194.
Notes to Chapter 5 275
105. American Correctional Association, Private Sector Operation of a
Correctional Institution, p. 91.
106. Christine Bowditch and Ronald S. Everett, "Private Prisons: Prob-
lems within the Solution," Justice Quarterly 4 (1987): 441-453, at p. 449.
107. While "bankruptcy" may not be the right term, units of government
do face the threat of financial collapse and preventing that from happening
does carry a cost.
108. American Federation of State, County, and Municipal Employees,
"Contracting Out in Local Government." Unpublished paper, March
1984.
109. Hackett, et al., Issues in Contracting, p. 52.
110. Government may create these hidden costs also, by paying less
than subsistence wages or by underfunding pension plans.
111. All data in the following analysis were supplied by Bill McGriff,
Hamilton County Auditor, in written and telephonic communications.
The analysis is available as a separate report. See Charles H. Logan and
Bill W. McGriff, "Comparing Costs of Public and Private Prisons: A Case
Study, "/I National Institute of Justice Research in Action (Washington,
DC: Department of Justice, 1989). It also appears in Charles H. Logan,
"Proprietary Prisons," in Lynne Goodstein and Doris L. MacKenzie, eds.,
The American Prison: Issues in Research and Policy (New York: Plenum,
1989). Adapted here by permission.
112. William D. Berry and David Lowery, "The Growing Cost of Gov-
ernment: A Test of Two Explanations," Social Science Quarterly 65
(1984): 735-749.
113. This is particularly common at the state level, but it happens at
county and federal levels as well. Fringe benefits were included by Ham-
ilton County in its prison budget, but are listed here on the right as a
reminder that they are commonly taken from other budgets.
114. This experience is consistent with the research by Savas, cited
earlier, showing that public administrators generally underestimate the
cost of noncontracted refuse collection.
115. CCA is heavily insured and indemnifies the county against potential
costs of litigation and legal damages.
116. After CCA took over, it subcontracted to a group of local doctors
to visit the facility twice a week, In 1987, after the time of the auditor's
report, the county health department hired a full-time physician, who now
sees patients at the jail and prison as well as at the health department.
CCA pays for that physician's services to the Penal Farm inmates (their
contract with the private doctors having expired).
117. Hamilton County has four divisions and four administrators: fi-
nance, public works, health services, and human services.
276 Notes to Chapter 6

118. General fund obligations were split off by the auditor from the
total county obligations. General funds paid for the county's general gov-
ernment administration. Constitutional officers like the sheriff, registrar
of deeds, tax collector, and clerk of court are separate legal entities,
although they are part of the county's overall budget. While in many
counties the sheriff handles both the jail and the prison out of his budget,
in Hamilton County the Penal Farm was always under human services,
and thus part of general administration.
119. Corrections Compendium, November 1986. This source did not
indicate whether this includes capital costs.
120. U.S. Department of Justice, 1984 Census of State Adult Correc-
tional Facilities (Washington, DC: Bureau of Justice Statistics, August
1987), Tables 18 and 31 (combining data).
121. Another aspect of the methodology understates the savings in
percentage terms, though not in dollar amounts. Certain costs (items 12-
15 in Table 5.2) are identified in Table 5.3 as "continuing, noncontracted
county costs." These costs are paid by the county directly and would be
the same whether the prison were contracted or not. These costs were
included in Table 5.2 (and therefore in Tables 5.3 and 5.4) in order to
show as complete an accounting of the county's total correctional costs
as possible. In Table 5.4, these costs were included on both sides of the
comparison, but they might just as well have been subtracted from both
sides. Had that been done, the percentage differences between contracting
and direct county operation, shown in Table 5.4, would have been 4.3,
3.3, and 9.1 percent instead of 3.8, 3.0, and 8.1 percent. When this effect
and the salary adjustment effect are combined, the new figures on per-
centage savings for the three years are 4.3, 7.9, and 13.1.
122. Besides monitoring, the superintendent now has time for additional
duties that he did not have before the contract. For example, he supervises
a new county program of electronic monitoring as an alternative to im-
prisonment for misdemeanants. Some of the time he would previously
have spent as warden is now available for this sort of expansion of the
county's total correctional program.
123. For a detailed discussion and evaluation of operational changes at
the prison, see Samuel Jan Brakel, "Prison Management, Private Enter-
prise Style: The Inmates' Evaluation," New England Journal on Criminal
and Civil Confinement 14 (Summer 1988): 175-244.
124. Mullen et al., Privatization of Corrections, p. 81.

CHAPTER 6

1. Fern Shen, "Investors Hope to Spring Profit from Private Prison,"

Hartford Courant, April 1, 1984.
Notes to Chapter 6 277
2. John D. Donahue, Prisons for Profit: Public Justice, Private Interests
(Washington, DC: Economic Policy Institute, 1988), p. 14.
3. Quoted in Philadelphia Inquirer, February 1, 1986.
4. Dave Kelly, President, Council of Prison Locals, American Fed-
eration of Government Employees, statement entered into the record,
U.S. Congress. House Committee on the Judiciary. Privatization of Cor-
rections. Hearings before the Subcommittee on Courts, Civil Liberties, and
the Administration of Justice, 99th Cong., 1st and 2nd sessions, November
13, 1985, and March 18, 1986, Serial No. 40 (Washington, DC: U.S.
Government Printing Office, 1986) p. 18.
5. Charles W. Thomas, Lonn Lanza-Kaduce, Linda S. Calvert Hanson,
and Kathleen A. Duffy, The Privatization of American Corrections
(Gainesville, FL: Center for Studies in Criminology and Law, University
of Florida, June 15, 1988), p. 209^210.
6. Todd R. Clear and George F. Cole, American Corrections (Mon-
terey, CA: Brooks/Cole, 1986), p. 532.
7. Joan Mullen, Kent John Chabotar, and Deborah M. Carrow, The
Privatization of Corrections (Washington, DC: National Institute of Jus-
tice, February 1985), p. 68.
8. Judith C. Hackett, Harry P. Hatry, Robert B. Levinson, Joan Allen,
Keon Chi, and Edward D. Feigenbaum, Issues in Contracting for the
Private Operation of Prisons and Jails (Washington, DC: Department of
Justice, October 1987), p. 25.
9. Los Angeles Time, March 29, 1985.
10. Cleveland Plain Dealer, October 1, 1984.
11. James O. Finckenauer, Juvenile Delinquency and Corrections: The
Gap Between Theory and Practice (Orlando, FL: Academic Press, 1984),
p. 181.
12. Ernest van den Haag, "Prisons Cost Too Much Because They Are
Too Secure," Corrections Magazine, April 1980, pp. 39-43.
13. Hackett et al., Issues in Contracting, p. 22.
14. John J. Dilulio, Jr., "Prisons, Profits and the Public Good:
The Privatization of Corrections," Research Bulletin No. 1 (Huntsville,
TX: Sam Houston State University Criminal Justice Center, 1986),
p. 4.
15. Ibid., p. 2.
16. Hackett et al., Issues in Contracting, p. 25.
17. Los Angeles Times, March 29, 1985.
18. This phrase, or one like it, is typical of the charge. See: Jan Elvin,
"Private Prison Plans Dropped by Buckingham," Journal: The National
Prison Project, Winter 1985, p. 11; Jody Levine, "Private Prison Planned
on Toxic Waste Site," Journal: The National Prison Project, Fall 1985,
pp. 10-11; Ira Robbins, "Privatization of Corrections: Defining the Is-
278 Notes to Chapter 6
sues," Judicature 69 (1986): 327; John D. Donahue, Prisons for Profit:
Public Justice, Private Interests (Washington, DC: Economic Policy In-
stitute, 1988), p. 7.
19. Some of these details are included in one of the ACLU reports,
but not in a way that detracts from its overall "horror story" tone. See
Levine, "Private Prison Planned," pp. 10-11.
20. In the public sector, they could look at the state of Massachusetts,
which located its Deer Island House of Correction on the same site as a
sewage treatment plant. When plans were made to relocate the prison,
those plans were made conditional on the ability to locate a garbage
incinerator (a waste-to energy facility) at the same site as the prison. See
Boston Globe, December 17, 1986, pp. 1,13.
21. 491 F. Supp. 1026 (1980).
22. See Donahue, Prisons for Profit p. 7; Levine, "Private Prison
Planned," p. 10; Craig Becker and Amy Dru Stanley, "Incarceration Inc.:
The Downside of Private Prisons," Nation, June 15, 1985, p. 729;
J. Michael Keating, Jr., Seeking Profit in Punishment: The Private Man-
agement of Correctional Institutions (American Federation of State,
County, and Municipal Employees, 1985), p. 15.
23. Hackett et al., Issues in Contracting, pp. 40-41.
24. Ibid., p. 41.
25. Corrections Corporation of America, Company Report, April 3,
1987, p. 10.
26. Federal Bureau of Prisons, 1986 Annual Report (Washington, DC:
Federal Bureau of Prisons), p. 11.
27. Telephone interview with Peggy Wilson, CCA, July 3, 1989.
28. Kevin Archer, "Off with Their Overhead," Policy Review, Fall 1989,
p. 75.
29. Telephone interview with Bob Greene, Wackenhut Corrections
Corporation, June 30, 1989.
30. Lee Kravitz, "Tough Times for Private Prisons," Venture, May 1986,
p. 60.
31. Ibid.
32. Dilulio, "Prisons, Profits and the Public Good," p. 3.
33. Stephen Gettinger, "Accreditation on Trial," Corrections Magazine,
February 1982, pp. 6-21, 51-55.
34. Accreditation has to be pursued, however, and a high rate of ac-
creditation in a particular jurisdiction may be as much a reflection of
political support for the pursuit as it is a reflection of quality. When Florida
successfully won accreditation for all of its penal institutions, some experts
objected to the impression created that Florida's prisons were somehow
exemplary (see Gettinger, Ibid.). Florida was not so much special in the
Notes to Chapter 6 279
quality of its prisons as in its system-wide commitment to apply for ac-
creditation, an expensive and time consuming process.
35. Corrections Corporation of American, Company Report, p. 11.
36. Federal Bureau of Prisons, 1986 Annual Report, p. 8.
37. New York Times, February 19, 1985.
38. A somewhat dated reference on this would be: Department of
Justice, National Institute of Law Enforcement and Criminal Justice, The
National Manpower Survey of the Criminal Justice System: Executive Sum-
mary (Washington, DC,: U.S, Government Printing Office, circa 1976),
pp. 11-13. In that survey, the U.S. Department of Justice reported that
nearly half of all juvenile correctional agencies provided no formal entry
training to line staff; where they did provide training, it averaged about
30 hours in length, with less than one-fourth in excess of 40 hours. For
staff in adult corrections, over half of the training programs were less than
100 hours. Subsequent training on the job was also rare: the Manpower
Survey found that less than 10 percent of state correctional officers had
attended in-service courses. Moreover, supervisors were not much better
trained than line staff: only one-tenth of surveyed correctional agencies
required that supervisors receive training in supervision either before or
after appointment. Training levels of state corrections officers have gen-
erally increased considerably since the time of that survey.
39. Richard Crane, Vice President, Legal Affairs, Corrections Cor-
poration of America, testimony, November 13, 1985, U.S. Congress.
House Committee on the Judiciary. Privatization of Corrections. Hearings
before the Subcommittee on Courts, Civil Liberties, and the Administration
of Justice, 99th Cong., 1st and 2nd sessions, November 13, 1985, and March
18, 1986, Serial No. 40 (Washington, DC: U.S. Government Printing
Office, 1986), p. 30.
40. American Correctional Association, Private Sector Operation of a
Correctional Institution (Washington, D.C.: U.S. Department of Justice,
National Institute of Corrections, April 1985), p. 12.
41. National Criminal Justice Association, "Private Sector Involvement
in Financing and Managing Correctional Facilities" Washington, DC: Na-
tional Criminal Justice Association, April 1987), p. 17.
42. Ibid.
43. John J. Dilulio, Jr., "Prison Discipline and Prison Reform," Public
Interest No. 89 (Fall 1987): 73.
44. Medina v. O'Neill, 589 F. supp. 1028, 1031 (1984).
45. Who will train whom? In a survey of correctional agencies, Camille
and George Camp found that 43 percent contracted with private agencies
for staff training, placing that among the ten most frequently contracted
services. See Camille G. Carnp and George M. Camp, Private Sector
280 Notes to Chapter 6
Involvement in Prison Services and Operations (Washington, D.C.: U.S.
Department of Justice, National Institute of Justice, February 1984), p. 6.
46. George M. Camp and Camille Graham Camp, Corrections Yearbook
1987 (South Salem, NY: Criminal Justice Institute, 1987), p. 42.
47. Dilulio, "Prison Discipline and Prison Reform," p. 89.
48. Clear and Cole, American Corrections, p. 302.
49. U.S. Department of Justice, Law Enforcement Assistance Admin-
istration, Corrections: National Manpower Survey of the Criminal Justice
System, vol. 3 (Washington, DC: U.S. Government Printing Office, 1978),
p. 45.
50. American Correctional Association, 1987 Directory of Juvenile and
Adult Correctional Departments, Institutions, Agencies, and Paroling Au-
thorities (College Park, MD: American Correctional Association, 1987),
pp. xxx-xxxi.
51. Crane, testimony, p. 29.
52. Samuel Jan Brakel, "Prison Management, Private Enterprise Style:
The Inmates' Evaluation," New England Journal on Criminal and Civil
Confinement 14 (Summer 1988): 175-244, at 195.
53. Crane, testimony, p. 38.
54. Samuel Greengard, "Making Crime Pay," Barrister Magazine 13
(1986): 14.
55. Ibid.
56. Becker and Stanley, "Incarceration Inc," p. 729.
57. Chattanooga Times, August 12, 1986.
58. Telephone interview with Bill McGriff, Hamilton County Auditor,
August 14, 1987.
59. League of Women Voters, "CCA—First Year Report Card" (Bay
County, FL: videotape, October 1986).
60. Kevin Krajick, "Punishment for Profit" Across the Board 21 (March
1984), pp. 25-27.
61. Ibid., p. 25.
62. Ibid.
63. Philadelphia Inquirer, August 12, 1984.
64. Ibid.
65. Krajick, "Punishment for Profit," p. 26.
66. American Correctional Association, Private Sector Operation of a
Correctional Institution.
67. These figures are derived from Tables 8.1 and 8.2 of the AC A
Report (Ibid). I extrapolated a weighted average turnover for all staff at
Okeechobee from the ACA figures given separately for "counselors" and
"other staff" in Table 8.2. The approximate ratio of these two categories
was estimated from the data in Table 8.1.
Notes to Chapter 6 281

68. Ibid., p. 92.

69. Ibid., p. 58.
70. Ibid., pp. 99-101.
71. All figures in this section are from Brakel, "Prison Management,
Private Enterprise Style."
72. Camp and Camp, Private Sector Involvement in Prison Services and
Operation.
73. Ibid., p. 5.
74. Ibid., pp. 9-10. The range of benefits mentioned included: improved
administrative operations (more efficient operation, reduced training re-
quirements, better accountability, better use of space); cost savings (fewer
staff, lower costs): and improved services and conditions (better quality,
unique service provided, decreased liability through better conditions).
75. The rest were miscellaneous. Ibid.
76. The rest distributed as follows: 39 percent process (supervision,
bidding, red tape); 22 percent contracting relationships (quality control,
payment, union or other labor problems); 5 percent cost-effectiveness.
Ibid., pp. 11-13.
77. Hackett et al., Issues in Contracting, p. 51.
78. Charles Ring, Contracting for the Operation of Private Prisons: Pros
and Cons (College Park, MD: American Correctional Association, 1987),
p. 17.
79. The criticism that Robbins' report to the ABA creates a double
standard has been leveled by others besides myself. See "Privatization
Backers Criticize ABA for Issuing Negative Report," Criminal Justice
Newsletter 19, No. 24 (December 15, 1988): 3-5.
80. Ira P. Robbins, The Legal Dimensions of Private Incarceration
(Washington, DC: American Bar Association, 1988), pp. 271-279, es-
pecially note 727.
81. Nor does Robbins' report leave much room for later balance or
compromise, at least with respect to the statutory provisions. Note in this
connection that Robbins insists that his Model Statutes "must not be
compromised" and "should not be negotiable." Robbins, Legal Dimen-
sions of Private Incarceration, pp. 414-415.
82. See the statements by Linda Cooper and Charles Logan in "Pri-
vatization Backers Criticize ABA," p. 4.
83. This implication was not lost on at least one member of the American
Bar Association committee that was proposing a formal resolution and
report based on Robbins' report to the ABA. In a letter to the committee,
this member, an opponent of private prisons, wrote: "The resolution itself
is correct in urging caution. More importantly, if followed, the report
makes privatization very difficult." The letter made clear the writer's
282 Notes to Chapter 7
complete opposition to private prisons, In the absence of a condemnation
of the concept (which he preferred), he would settle for a report that
made privatization as difficult as possible.

CHAPTER 7

1. Michael A. Kroll, "Prisons for Profit," Progressive, September 1984,

pp. 18-22.
2. Jess Maghan and Edward Sagarin, "The Privatization of Correc-
tions: Seeking to Anticipate the Unanticipated Consequences." Paper
presented to American Society of Criminology, San Diego, CA, Novem-
ber 1985, pp. 44-46: Russ Immarigeon, "Private Prisons, Private Pro-
grams, and their Implications for Reducing Reliance on Imprisonment in
the United States." Prison Journal 65 (1985): 60-74; Christine Bowditch
and Ronald S Everett, "Private Prisons: Problems within the Solution,"
Justice Quarterly 4 (1987): 451.
3. Jan Elvin, "Private Prison Plans Dropped by Buckingham," Journal:
The National Prison Project Winter 1985, p. 50.
4. Ibid.
5. [New York] Newsday, July 30, 1984.
6. Institutions Etc., "If You Think This Sounds Good, Wait'll You
Hear About Discount Gas Chambers," Investigative Newsletter on Insti-
tutions/Alternatives 6 (November 1983): 6-8.
7. Florida Times-Union, September 23, 1984.
8. J. Michael Keating, Jr., Seeking Profit in Punishment: The Private
Management of Correctional Institutions (American Federation of State,
County, and Municipal Employees, 1985), p. 29.
9. Abt Associates, Inc., American Prisons and Jails (Washington, DC:
U.S. Government Printing Office, 1980).
10. Alfred Blumstein, Jacqueline Cohen, and William Gooding, "The
Influence of Capacity on Prison Population: A Critical Review of Some
Recent Evidence," Crime and Delinquency 29 (January 1983): 1-51.
11. Ibid., p. 50.
12. Paul Starr, The Limits of Privatization (Washington, DC: Economic
Policy Institute, 1987).
13. Maghan and Sagarin, "Privatization of Corrections," pp. 60-61.
14. Edward Sagarain and Jess Maghan, "Should States Opt for Private
Prisons?" A debate with Charles Logan in the Hartford Courant, January
12, 1986.
Notes to Chapter 7 283

15. Harmon L. Wray, Jr., "Cells for Sale," Southern Changes 8 Sep-
tember 1986): 6.
16. Herbert J. Hoelter, "Private Presentence Reports: Boon or Boon-
doggle?" Federal Probation 48 (1984): 53.
17. David M. Lawrence, "Private Exercise of Governmental Power,"
Indiana Law Journal 61 (1986): 653.
18. John P. Conrad, "Corrections and its Constituencies," Prison Jour-
nal 64 (Fall/Winter 1984): 47-48.
19. Bruce L. Benson, "Guns for Protection and Other Private Sector
Responses to the Government's Failure to Control Crime," Journal of
Libertarian Studies 8 (1986): 77.
20. U.S. Department of Justice, Sourcebook of Criminal Justice Statis-
tics—1986 (Washington, DC: Bureau of Justice Statistics, 1987), Table
2.11, pp. 86-87.
21. Joseph E. Jacoby and Christopher S. Dunn, "National Survey on
Punishment for Criminal Offenses" (Bowling Green, OH: Bowling Green
State University, 1987).
22. Benson, "Guns for Protection," p. 79, note 22.
23. Ibid., p. 79.
24. Ibid., pp. 81-83.
25. Richard Berk, Harold Brackman, and Selma Lesser, A Measure of
Justice: An Empirical Study of Changes in the California Penal Code,
1955-1971 New York: Academic Press, 1977).
26. Benson, "Guns for Protection," p. 82.
27. Dave Kelly, President, Council of Prison Locals, American Fed-
eration of Government Employees, statement entered into the record,
U.S. Congress. House Committee on the Judiciary. Privatization of Cor-
rections. Hearings before the Subcommittee on Courts, Civil Liberties, and
the Administration of Justice, 99th Cong., 1st and 2nd sessions, November
13, 1985 and March 18, 1986, Serial No. 40 (Washington, DC: U.S.
Government Printing Office, 1986) p. 17. Emphasis shifted from first oc-
currence of "self-interest."
28. Warren I. Citkins, "Privatization of the American Prison System:
An Idea Whose Time Has Come?" Notre Dame Journal of Law, Ethics
and Public Policy 2 (1986): 455.
29. Morgan O. Reynolds, Power and Privilege: Labor Unions in Amer-
ica (New York: Universe Books, 1984), p. 183.
30. John M. Wynne, Jr., Prison Employee Unionism: The Impact on
Correctional Administration and Programs (Washington, DC: U.S. De-
partment of Justice, NILECJ, January 1978), pp. 184-185.
31. Ibid., pp. 227-228.
284 Notes to Chapter 8

32. Ibid., p. 217.1 am grateful to Bruce Benson for drawing my attention

to this reference.
33. James B. Jacobs, New Perspectives on Prisons and Imprisonment
(Ithaca, NY: Cornell University Press, 1983), p. 118.
34. Ibid., pp. 130-131.
35. Wray, p. 4.
36. Tennessean, September 11, 1985.
37. Joan Mullen, Kent John Chabotar, and Deborah M. Carrow, fThe
Privatization of Corrections (Washington, DC: National Institute of Jus-
tice, Feburary 1985), p. 73.
38. It is the government, not the Hilton Inn, that for decaded has chosen
to cram more "guests" into each "room," rather than to expand its
facilities.
39. Corrections Corporation of America, press release, June 26, 1987.
40. Corrections Corporation of America, press release, July 7, 1987.

CHAPTER 8

1. Joan Mullen, "Corrections and the Private Sector," Prison Journal

65 (1985): 1.
2. Edward Sagarin and Jess Maghan, "Should States Opt for Private
Prisons?" A debate with Charles Logan in the Hartford Courant, January
12, 1986.
3. John J. Dilulio, Jr., "What's Wrong with Private Prisons," Public
Interest No. 92 (Summer 1988): 73.
4. John M. Wynne, Jr., Prison Employee Unionism: The Impact on
Correctional Administration and Programs (Washington, DC: U.S. De-
partment of Justice, NILECJ, January 1978, pp. 170-171.
5. American Federation of State, County, and Municipal Employees,
"Contracting Out in Local Government," unpublished paper, March 1984.
6. Robert Behn, "The False Dawn of the Sunset Laws," Public Interest
49 (Fall 1977): 103-118.
7. Robert J. Schmidt, panel presentation on "Assessments of Private
Corrections," at the National Conference on Alternatives to Jail and
Prison Overcrowding, sponsored by the University of Florida, Orlando,
March 10, 1988.
8. American Correctional Association, Private Sector Operation of a
Correctional Institution (Washington, DC: U.S. Department of Justice,
National Institute of Corrections, April 1985), p. 16.
9. Gene Kassebaum, Billy Wayson, Joseph Seldin, Gail Monkman,
Peter Nelligan, Peter Meyer, and David Takeuchi, Contracting for Cor-
Notes to Chapter 9 285
rectional Services in the Community (Washington, DC: U.S. Department
of Justice, May 1978), Vol. 1, summary.
10. Jess Maghan and Edward Sagarin, "The Privatization of Correc-
tions: Seeking to Anticipate the Unanticipated Consequences." Paper
presented to American Society of Criminology, San Diego, CA, Novem-
ber 1985, p. 53.
11. John D. Donahue, Prisons for Profit: Public Justice. Private Interests
(Washington DC: Economic Policy Institute, 1988), p. 18.
12. U.S. Congress. House Committee on the Judiciary. Privatization
of Corrections. Hearings before the Subcommittee on Courts, Civil Lib-
erties, and the Administration of Justice, 99th Cong., 1st and 2nd sessions,
November 13, 1985, and March 18,1986, Serial No. 40 (Washington, DC:
U.S. Government Printing Office, 1986), p. 140 (statement of Norman
A. Carlson, Director, Federal Bureau of Prisons, submitted with testi-
mony on March 18, 1986).
13. Robert Mathias and Diane Steelman, "Controlling Prison Popu-
lations: An Assessment of Current Mechanisms." Unpublished paper,
National Council on Crime and Delinquency, 1982.

CHAPTER 9

1. Cited in Warren I. Cikins, "Privatization of the American Prison

System: An Idea Whose Time Has Come?" Notre Dame Journal of Law,
Ethics and Public Policy 2 (1986): 462.
2. Judith C. Hackett, Harry P. Hatry, Robert B. Levinson, Joan Allen,
Keon Chi, and Edward D. Feigenbaum, Issues in Contracting for the
Private Operation of Prisons and Jails (Washington, DC: Department of
Justice, October 1987), p. 25. Note, however, that the restriction to min-
imum security facilities, while mentioned in the body of the report, does
not appear in the summary of recommendations.
3. Ibid., p. 25. Of course, there is also community resistance to the
location of prisons when government runs them. While government may
be exempt from some zoning laws and can use force (i.e., eminent domain)
to overcome community opposition, private agencies must rely on per-
suasion. Some communities find the prospect of new jobs and added
property taxes hard to resist.
4. David M. Lawrence, "Private Exercise of Governmental Power,"
Indiana Law Journal 61 (1986): 666, note 77.
5. Connie Mayer, "Legal Issues Surrounding Private Operation of
Prisons," Criminal Law Bulletin 22 (1986): 318.
6. Hackett et al., Issues in Contracting, p. 13.
286 Notes to Chapter 9
1. Massachusetts Legislative Research Council, Report Relative to Pris-
ons for Profit. House Report No. 6225 (Boston: Massachusetts House of
Representatives, July 31, 1986), p. 100.
8. Telephone interview with Bob Greene, Wackenhut, June 30, 1989.
9. Most deadly force in prisons is exercised by inmates, not officers.
A common complaint among corrections officers is that they are sur-
rounded by inmates who are often better armed than their keepers.
10. Buckingham Security Limited, Private Prison Management: First
Year Report 1985-1986, Butler County, Pennsylvania (Lewisburg, PA:
Buckingham Security Ltd., 1986), p. 6.
11. Samuel Jan Brakel, "Prison Management, Private Enterprise Style:
The Inmates' Evaluation," New England Journal on Criminal and Civil
Confinement 14 (Summer 1988): 175-244, at 188.
12. Washington Post, August 9, 1987.
13. Washington Post, February 5, 1988.
14. Commonwealth of Virginia, "Study of Correctional Privatization"
(Richmond, VA: Secretary of Transportation and Public Safety, 1986),
pp. 57-58.
15. At Dozier School for Boys, a "comparison" (but not truly com-
parable) state run facility, the escape rates for those three years were
10.6, 8.0, and 6.3 per 100. American Correctional Association, Private
Sector Operation of a Correctional Institution (Washington, DC: U.S.
Department of Justice, National Institute of Corrections, April 1985),
p. 58.
16. All information for this paragraph is from Brakel, "Prison Man-
agement, Private Enterprise Style," pp. 187-189. Brakel indicates some
slight uncertainty in the escape data. A conflicting set of figures shows 10
escapes from within the prison in 1985, the same number in 1986, and a
drop to 7 in 1987.
17. Mary R. Woolley, "Prisons for Profit: Policy Considerations for
Government Officials," Dickinson Law Review 90 (Winter 1985): 324-
327.
18. Ibid.
19. Morgan O. Reynolds, Power and Privilege: Labor Unions in Amer-
ica (New York: Universe Books, 1984), p. 195.
20. Research cited in Ralph de Toledano, Let Our Cities Burn (New
Rochelle, NY: Arlington Hosue, 1975), pp. 84-86.
21. John M. Wynne, Jr., Prison Employee Unionism: The Impact on
Correctional Administration and Programs (Washington, DC: U.S. De-
partment of Justice, NILECJ, January 1978), pp. 223-224.
22. de Toledano, Let Our Cities Burn, p. 46.
Notes to Chapter JO 287
23. James B. Jacobs, New Perspectives on Prisons and Imprisonment
(Ithaca, NY: Cornell University Press, 1983) p. 142.
24. Wynne, Prison Employee Unionism, p. 202.
25. Ibid., p. 221.
26. Woolley, "Prisons for Profit," pp. 326-327.
27. Hackett et al., Issues in Contracting, p. 11.
28. "Hamilton County, Tennessee, Corrections Facilities Agreement,"
September 20, 1984, section 13 (4), p. 37.
29. Wynne, Prison Employee Unionism, p. 204.
30. Ibid., p. 203.
31. Ibid., p. 223.
32. U.S. Office of Management and Budget, Enhancing Governmental
Productivity Through Competition: Targeting for Annual Savings of One
Billion Dollars by 1988 (Washington, DC: Office of Federal Procurement
Policy, March 1984), p. 11.

C H A P T E R 10

1. J. Michael Keating, Jr., Seeking Profit in Punishment: The Private

Management of Correctional Institutions (American Federation of State,
County, and Municipal Employees, 1985), p. 43.
2. Ira Robbins, "Should Private Firms Run Prisons for Profit?" [New
York] Newsday, March 31, 1985.
3. Charles W. Thomas, Lonn Lanza-Kaduce, Linda S. Calvert Hanson,
and Kathleen A. Duffy, The Privatization of American Corrections
(Gainesville, FL: Center for Studies in Criminology and Law, University
of Florida, June 15, 1988), p. 142; Ira Robbins, The Legal Dimensions of
Private Incarceration (Washington, DC: American Bar Association,
1988), p. 118.
4. Actually, Section 1983 applies only to states and not to the federal
government. However, it is discussed here as though applying to both
levels because it is very similar in logic to so-called Bivens actions against
the federal government. See Thomas, p. 114, note 71.
5. Medina v. O'Neill, 589 F. Supp. 1028 (1984).
6. Medina V. O'Neill, 838 F.2d 800 (5th Cir. 1988) at 800.
1. 589 F. Supp. at 1028.
8. Or in a Bivens-typs action against the federal government.
9. Robbins, Legal Dimensions of Private Incarceration, pp. 72, 118.
10. Ibid., p. 11, note 375.
288 Notes to Chapter 10
11. Quoted in "Privatization Backers Criticize ABA for Issuing Neg-
ative Report," Criminal Justice Newsletter, December 15, 1988, p. 4.
12. In reply to this criticism, Robbins has quoted a statement by Tom
Beasley, chairman of Corrections Corporation of America, promising that
CCA "will protect every political figure associated with the prison system
by means of a multi-million dollar insurance policy which will be sufficient
. . . to finance all litigation, pay all claims and judgments, and in general
insulate those officials from personal liability." (See "Robbins Responds
to Critics of His Report on Privatization," Criminal Justice Newsletter,
February 1, 1989, p. 6.)
This citation, however, does more to refute than to support Robbins'
claim that contractors have promised to reduce or eliminate governmental
liability. The statement by Beasley is a forceful reminder to government
officials that they, personally, remain liable along with their agencies.
CCA does not promise to "immunize" these officials against liability, but
to "insulate" them from the cost of that liability, through indemnification,
insurance, and payment of legal fees. Since CCA will charge government
for the cost of this insulation, it might seem that nothing has been gained,
but that is not the case. While liability for risks remains, it is still possible
to reduce the risks themselves and to reduce the cost of protecting against
the risks. This will be explained shortly.
13. Ira Robbins, Testimony to President's Commission on Privatization,
Washington, DC, December 22, 1987, unpublished transcript, pp. 294-
295.
14. Third-party rights can be created in intergovernmental contracts as
well as public-private contracts. See Robbins, Legal Dimensions of Private
Incarceration, pp. 142-143, note 436.
15. Ibid., pp. 141-149, 418-423.
16. Ibid., p. 149 (emphasis added).
17. Thomas et al., Privatization of American Corrections, p. 93.
18. Ibid., p. 95, note 32.
19. Ibid., pp. 89-90.
20. Ibid., p. 91.
21. However, if private prisons are required to carry insurance at levels
too much higher than the liability of the state, this may create unnecessary
problems. This will be discussed below, in the section on "Insurance and
Indemnification."
22. October 28,1987, interview with Richard Crane, then Vice President
for Legal Affairs at Corrections Corporation of America.
23. Ibid.
24. Bruce Cory, "From Rhetoric to Reality: Privatization Put to the
Test," Corrections Compendium, May 1986, p. 10.
Notes to Chapter 10 289

25. National Criminal Justice Association, "Private Sector Involvement

in Financing and Managing Correctional Facilities" (Washington, DC:
National Criminal Justice Association, April 1987), p. 20.
26. Buckingham Security Limited, Private Prison Management: First
Year Report 1985-1986, Butler County, Pennsylvania (Lewisburg, PA:
Buckingham Security Ltd., 1986), p. 9.
27. See "CCA Settles Lawsuit," press release, October 3, 1988.
28. Roger Hanson reports that prisoner grievances account for 7
percent of all U.S. district court cases and 12 percent of the dockets of
U.S. courts of appeals. Hanson estimates the costs of processing these
cases at well over $100 million a year. By comparison, the total budget
of the entire federal judiciary is $950 million. See Roger A. Hanson,
"What Should Be Done When Prisoners Want to Take the State to
Court?" Judicature 70 (December 1986/1987): 223-227, at 223
and 225.
29. Contact Center, Inc., Inmate Lawsuits: A Report on Inmate Lawsuits
Against State and Federal Correctional Systems Resulting in Monetary
Damages and Settlement (Lincoln, NE: Contact Center, Inc., 1985). I am
grateful to Charles Thomas for drawing my attention to this report.
30. Ibid, p. 14.
31. Ibid. Calculated from figures presented on p. 14.
32. Ibid, p. 1. This survey identified 87 lawsuits lost and 161 settled for
money over a two-year period in 1984 and 1985. In the year ending June
30, 1985, over 22,000 civil rights actions were filed in district and appeals
courts. That does not include the smaller number of tort claims filed in
state courts.
33. Corrections Corporation of America, 1986 Annual Report, p. 31.
34. Chattanooga Times, August 12, 1986.
35. Los Angeles Times, March 29, 1985.
36. Pricor, Prospectus, July 23, 1987.
37. Information supplied by the Wackenhut Corporation.
38. "Bay County Detention Facilities Contract between Corrections
Corporation of America and Bay County, Florida, September 3, 1985,"
pp. 32-33.
39. Thomas W. Beasley, Chairman, CCA. Transcript of testimony to
President's Commission on Privatization, Washington, DC, December 22,
1987, p. 287.
40. Contact Center, Inc., Inmate Lawsuits.
41. Camille G. Camp and George M. Camp, Private Sector Involvement
in Prison Services and Operations (Washington, DC: U.S. Department of
Justice, National Institute of Justice, February 1984), p. 6.
42. Robbins, Legal Dimensions of Private Incarceration, pp. 238-239.
290 Notes to Chapter 11

43. Brian B. Evans, "Private Prisons," Emory Law Journal 36 (Winter

1987): 253-283, at p. 273, note 106.

C H A P T E R 11

1. Joseph P. Kalt, "Public Goods and the Theory of Government,"

Cato Journal 1 (Fall 1981): 565-584.
2. The high reelection rate of sheriffs, in spite of the deplorable con-
dition of many of their jails, casts doubt on their political accountability
in practice, at least on this score.
3. E. S. Savas, "How Much Do Government Services Really Cost?"
Urban Affairs Quarterly 15 (1979): 23-41, at p. 30.
4. In Washington, these officials are regarded as the "summer help"
by the lower echelons of civil service bureaucrats, who form a kind of
permanent government.
5. Morgan O. Reynolds, Power and Privilege: Labor Unions in Amer-
ica (New York: Universe Books, 1984), p. 181.
6. John M. Wynne, Jr., Prison Employee Unionism: The Impact on
Correctional Administration and Programs (Washington, DC: U.S. De-
partment of Justice, NILECJ, January 1978), p. 203.
7. Wall Street Journal, July 22, 1987, p. 20.
8. Site visit by author and interviews with county officials.
9. News-Herald, Panama City, August 6, 1987.
10. News-Herald, Panama City, June 26, 1987.
11. Judy S. Grant and Diane Carol Bast, Corrections and the Private
Sector: A Guide for Public Officials (Chicago: Heartland Institute, 1986),
p. 13.
12. [Louisville] Courier-Journal, June 9, 1986.
13. Commonwealth of Virginia, "Study of Correctional Privatization"
(Richmond, VA: Secretary of Transportation and Public Safety, 1986),
p. 58.
14. Cited in Joan Mullen, Kent John Chabotar, and Deborah M. Car-
row, The Privatization of Corrections (Washington, DC: National Institute
of Justice, February 1985), pp. 73-74.
15. American Correctional Association, Private Sector Operation of a
Correctional Institution (Washington, DC: U.S. Department of Justice,
National Institute of Corrections, April 1985), p. 11.
16. Herbert J. Hoelter, "The Private Presentence Report: Issues for
Consideration," Prison Journal 65 (1985): 57. Emphasis in original.
Notes to Chapter 11 291

17. John P. Conrad, "Corrections and its Constituencies," Prison Jour-

nal 64 (Fall/Winter 1984): 47-48.
18. Judith C. Hackett, Harry P. Hatry, Robert B. Levinson, Joan Allen,
Keon Chi, and Edward D. Feigenbaum, Issues in Contracting for the
Private Operation of Prisons and Jails (Washington, DC: National Institute
of Justice, October 1987), p. 43.
19. Ibid., pp. 50-51.
20. American Correctional Association, Private Sector Operation of a
Correctional Institution (Washington, DC: U.S. Department of Justice,
National Institute of Corrections, April 1985), p. 79, note 4 and accom-
panying text.
21. Ibid., p. 99.
22. Ibid., p. 78.
23. National Criminal Justice Association, "Private Sector Involvement
in Financing and Managing Correctional Facilities" (Washington, DC:
National Criminal Justice Association, April 1987), p. 17.
24. Interview with Floyd Fuller, Superintendent of Corrections, Ham-
ilton County, Tennessee, October 27, 1987.
25. National Criminal Justice Association, p. 20.
26. National Criminal Justice Association, "Private Sector Involve-
ment," p. 10; J. Michael Keating, Jr., Seeking Profit in Punishment: The
Private Management of Correctional Institutions (American Federation of
State, County, and Municipal Employees, 1985), p. 46.
27. Interview with Robert Schmidt, Supervisor of Detention Services,
Immigration and Naturalization Service, May, 21, 1987.
28. At the Volunteers of America Regional Correction Center, a con-
tracted jail for women serving Ramsey County, Minnesota, inmates are
asked to complete an evaluation questionnaire at the time of their release.
The questionnaire asks about safety, food, physical facilities, and program
results. The results are available to county officials and are used by the
contractor to make changes. One other function of the questionnaire is
to allow departing inmates to vent their feelings. (Telephone interview
with Bill Nelson, VOA, Roseville, MN, October 15, 1987.)
29. Inmate evaluations of one county prison (Silverdale) were described
in chapter 6, "Issues of Quality." See Samuel Jan Brake!, "Prison Man-
agement, Private Enterprise Style: The Inmates' Evaluation," New Eng-
land Journal on Criminal and Civil Confinement 14 (Summer 1988): 175-
244.
30. See Samuel Jan Brakel, " 'Privatization' in Corrections: Radical
Prison Chic or Mainstream Americana?" New England Journal on Crim-
inal and Civil Confinement 14 (Winter 1988): 1-39, at 27.
292 Notes to Chapter 11
31. The Massachusetts Department of Mental Health was recently
quoted as deploring the worsening conditions of state institutions, with
the comment, "If private psychiatric hospitals we license did this, we'd
close them down" (National Review, September 16, 1988, p. 11).
32. Interview with T. Don Hutto, Nashville, October 28, 1987.
33. See Corrections Compendium, November 1986, p. 14.
34. Estimate by Howard Messing, NIJ Reports, July/August 1987. Some
unpublished figures provided by Messing, however, indicate that, of jails
under court order, 24 percent had a Master appointed, which would imply
far more than 60 such jails.
35. U.S. Department of Justice, Crime and Justice Facts, 1985 (Wash-
ington, DC: Bureau of Justice Statistics), p. 26.
36. John J. Dilulio, Jr., Governing Prisons: A Comparative Study of
Correctional Management (NY: Free Press, 1987), p. 246.
37. William C. Collins, "Privatization: Some Legal Considerations from
a Neutral Perspective," in Collins: Correctional Law, 1986 (Olympia, WA:
William C. Collins), p. 93.
38. Samuel Jan Brakel, " 'Mastering' the Legal Access Rights of Prison
Inmates," New England Journal on Criminal and Civil Confinement 12
(Winter 1986): 1-69, at pp. 8-9. Also see Samuel Jan Brakel, "Prison
Reform Litigation: Has the Revolution Gone Too Far?" Judicature 70
(June/July 1986): 5-6, 64-65, at p. 64.
The state of New Mexico, with a relatively small prisoner population
but operating under a consent decree, recently reported payments of $1.8
million to support the activities of a court-appointed monitor, his support
staff, and his consulting experts and investigators. When the cost of state
attorneys and inmate legal expenses are included, the direct costs of the
consent decree come to over $4 million. These figures are given in an
annual report, but they are probably cumulative over the four years since
the special master's appointment, rather than annual costs. See The An-
nual Report of the New Mexico Corrections Department for the Seventy-
Fifth Fiscal Year, July 1, 1986, through June 30, 1987 (Santa Fe: New
Mexico Corrections Department), p. 16.
39. Lawyers have been known to pursue appointment as a special master
in order to retire from all other practice with a source on long-term income.
40. Like contractors, masters are not elected but chosen. The manner
of their choosing, however, differs significantly. Contractors are chosen
by or subject to final approval by officials who are either elected or po-
litically appointed and are thus vulnerable to political removal, while
special masters are chosen by federal judges with lifetime tenure. Whereas
contractors are chosen in a process that is public, competitive, and usually
guided by preannounced criteria that are at least subject to legislative
Notes to Chapter 12 293
determination, special masters are chosen by judges in an informal and
subjective fashion from a limited field of insiders.
41. Hackett et al., Issues in Contracting, pp. 49-50.
42. For an elaborate discussion of monitoring and structured incentives
for private prisons, see James Theodore Gentry, "The Panopticon Re-
visited: The Problem of Monitoring Private Prisons," Yale Law Journal
96 (1986): 353-375.

C H A P T E R 12

1. John Hanrahan, "Why Public Services Should Stay Public." Des

Moines Register, March 31, 1983.
2. If it were, it would be a stronger argument against unions than
against contracting. A Rand study of racketeering in New York garbage
collection found that "unions have been essential to racketeer control.
That is to say, only where a corrupt union has been available, or created,
have racketeers been able to establish an industrywide influence." See
Peter Reuter, The Value of a Bad Reputation: Cartels, Criminals, and
Barriers to Entry [Report No. P-6835] (Santa Monica, CA: The Rand
Corporation, December 1982).
3. Bernard J. McCarthy, "Keeping an Eye on the Keeper: Prison
Corruption and its Control," Prison Journal 64 (Fall/Winter 1984): 113-
125.
4. Ira Robbins, "Privatization of Corrections: Defining the Issues,"
Judicature 69 (1986): 331.
5. Professor Robbins' faith in the superiority of governmental over
private management of prisons is all the more remarkable coming from
a man who, a few years earlier, likened America's prison system to that
described in Aleksandr Solzhenitsyn's Gulag Archipelago. While Robbins
saw many similarities between the two penal systems, he acknowledged
no fundamental differences, admitting only to differences of degree. From
a perspective so negative toward America's governmentally run prisons,
to anticipate that privately run prisons would be worse is to imagine great
horrors indeed. See Ira P. Robbins, "Beyond Freedom and Dignity: Alek-
sandr Solzhenitsyn and the American Gulag," Michigan Law Review 78
(1980): 763-789.
6. J. Michael Keating, "Thoughts about Prisons for Profit," in Amer-
ican Federation of State, County, and Municipal Employees, Does Crime
Pay! (Washington, DC: American Federation of State, County, and Mu-
nicipal Employees, 1985), p. 17.
294 Notes to Chapter 12
1. Harmon L. Wray, Jr., "Cells for Sale," Southern Changes 8 (Sep-
tember 1986): 5.
8. Edward Sagarin and Jess Maghan, "Should States Opt for Private
Prisons?" A debate with Charles Logan in the Hartford Courant, January
12, 1986.
9. John Dilulio, Jr., "Prisons, Profits and Public Good: The Privati-
zation of Corrections," Research Bulletin No. 1 (Huntsville, TX: Sam
Houston State University Criminal Justice Center, 1986), p. 3.
10. John J. Dilulio, Jr., "What's Wrong with Private Prisons," Public
Interest No. 92 (Summer 1988): 66-83, at pp. 71-72.
11. Gustave de Beaumont and Alexis de Tocqueville, On the Peniten-
tiary System in the United States and its Application in France (Carbondale,
IL: Southern Illinois University Press, 1964), p. 68.
12. D. Smith, Police Systems in the United States, 2nd ed. (New York:
Harper & Row, 1960), p. 143. Cited in Gilbert Geis, "The Privatization
of Prisons: Panacea or Placebo?" in Barry J. Carroll, Ralph W. Conant,
and Thomas A. Easton, eds., Private Means—Public Ends: Private Busi-
ness in Social Service Delivery (New York: Praeger, 1987), pp. 76-97, at
pp. 81-82.
13. Malcolm Braly, False Starts: A Memoir of San Quentin and Other
Prisons (Boston: Little, Brown, 1976).
14. Roger Morris, The Devil's Butcher Shop: The New Mexico Prison
Uprising (New York: Franklin Watts, 1983) pp. 143-144.
15. Milwaukee Journal, February 1, 1989.
16. One possible exception to this generalization is that private con-
tractors in the nineteenth century frequently went bankrupt. This could
be seen as an abuse in itself (avoidance of responsibility), or it could be
seen as the termination of other abuses. Unlike contractors, states that
historically could not or would not raise the revenues necessary to run
humane, or even viable, institutions did not go bankrupt. They merely
reduced expenditures and worsened conditions further, or turned the
problem over to private contractors. The issue of bankruptcy among con-
temporary contractors is discussed in the next chapter.
17. E. R. Cass, "Jails for Profit," Corrections Today 50 (October
1988): 84, 86.
18. Ibid., p. 86.
19. Ideally, there would also be a random or broadly representative
selection of time periods, jurisdictions, and prisons.
20. William C. Collins, "Privatization: Some Legal Considerations from
a Neutral Perspective," in Collins: Correctional Law,1986 (Olympia, WA:
William C. Collins), p. 81.
21. Judith C. Hackett, Harry P. Hatry, Robert B. Levinson, Joan Allen,
Notes to Chapter 13 295

Keon Chi, and Edward D. Feigenbaum, Issues in Contracting for the

Private Operation of Prisons and Jails (Washington, DC: Department of
Justice, October 1987, p. 49.
22. Joan Mullen, Kent John Chabotar, and Deborah M. Carrow, The
Privatization of Corrections (Washington, DC: National Institute of Jus-
tice, February 1985), p. 73.
23. Robert W. Poole, Jr., "Objections to Privatization," Policy Review
24(1983): 114.

C H A P T E R 13

1. William D. Berry and David Lowery, "The Growing Cost of Gov-

ernment: A Test of Two Explanations," Social Science Quarterly (Sep-
tember 1984): 735-749.
2. E. S. Savas, Privatizing the Public Sector: How to Shrink Govern-
ment (Chatham, NJ: Chatham House Publishers, 1982), p. 93.
3. Joan Mullen, Kent John Chabotar, and Debroah M. Carrow, The
Privatization of Corrections (Washington, DC: National Institute of Jus-
tice, February 1985), p. 68.
4. Thomas E. Burke, "Research: Corrections Corporation of America
Company Report," Boston, April 3, 1987, p. 2. See also Corrections
Corporation of America quarterly and annual reports.
5. Martin Tolchin, "Privately Operated Prison in Tennessee Reports
$200,000 in Cost Overruns," New York Times, May 21, 1985, p. A14.
See also Russ Immarigeon, "Private Prisons, Private Programs, and their
Implications for Reducing Reliance on Imprisonment in the United
States," Prison Journal 65 (1985): 67; Charles Ring, Contracting for the
Operation of Private Prisons: Pros and Cons (College Park, MD: Amer-
ican Correctional Association, 1987), pp. 33-34; John D. Donahue, Pris-
ons for Profit: Public Justice, Private Interests (Washington, DC: Economic
Policy Institute, 1988).
6. Charles Ring, Contracting for the Operation of Private Prisons,
p. 34.
7. National Criminal Justice Association, "Private Sector Involvement
in Financing and Managing Correctional Facilities" (Washington, DC:
National Criminal Justice Association, April 1987), p. 18; Judith C. Hack-
ett, Harry P. Hatry, Robert B. Levinson, Joan Allen, Keon Chi, and
Edward D. Feigenbaum, Issues in Contracting for the Private Operation
of Prisons and Jails (Washington, DC: Department of Justice, October
1987), p. 53.
296 Notes to Chapter 14
8. Telephone interview with Bill McGriff, Hamilton County Auditor,
November 11, 1987.
9. Erik Larson, "Captive Company," Inc., June 1988, p. 90.
10. Philip B. Taft, Jr., "Private Vendors, Part II: Survival of the Fittest,"
Corrections Magazine 9 (February 1983). p. 43.
11. Ibid., p. 42
12. Jess Maghan and Edward Saragin, "The Privatization of Correc-
tions: Seeking to Anticipate the Unanticipated Consequences." Paper
presented to American Society of Criminology, San Diego, CA, Novem-
ber 1985, pp. 43-44.
13. Insight, December 7, 1987, pp. 34-36.
14. James Theodore Gentry, "The Panopticon Revisited: The Problem
of Monitoring Private Prisons," Yale Law Journal 96 (1986): 370.
15. Ibid., p. 369.
16. Thomas Coughlin, "The New York Experience," in American Fed-
eration of State, County, and Municipal Employees, Does Crime Pay?
(Washington, DC: American Federation of State, County, and Municipal
Employees, 1985), p. 32.
17. U.S. News and World Report, December 23, 1985, p. 39.
18. Hackett, et al., Issues in Contracting, p. 40
19. [Louisville] Couier-Journal, March 19, 1985.
20. Corrections Corporation of America, 1986 Annual Report, p. 24.
21. Paul Starr, The Limits of Privatization (Washington, DC: Economic
Policy Institute, 1987), p. 5.
22. Ibid., p. 6.
23. Hans F. Sennholz, "Privatizing Federal Programs," Freeman, June
1987, pp. 223-228.
24. Ibid., p. 405.
25. Ibid., pp. 404-405.
26. Ibid., p. 409.

C H A P T E R 14

1. Barry Bozeman, All Organizations Are Public: Bridging Public and

Private Organizational Theories (San Francisco: Jossey-Bass, 1987),
chap. 6.
2. James M. Buchanan and Robert D. Tollison, eds., The Theory of
Public Choice—// (Ann Arbor, MI: University of Michigan Press, 1984).
3. American Federation of State, County, and Municipal Employees,
Notes to Chapter 14 297

Passing The Bucks (Washington, DC: American Federation of State,

County, and Municipal Employees, 1983).
4. For discussion of this concept see: George Smith, "Justice Entre-
preneurship in a Free Market," Journal of Libertarian Studies 3 (Winter
1979): 405-426; Randy Barnett, "Pursuing Justice in a Free Society: Part
One—Power vs. Liberty," Criminal Justice Ethics 4 (Summer/Fall 1985;
50-72; Randy Barnett, "Pursuing Justice in a Free Society: Part Two—
Crime Prevention and the Legal Order," Criminal Justice Ethics 5 (Winter/
Summer 1986): 30-53; Bruce L. Benson, The Enterprise of Law: Avoiding
Anarchy and Leviathan (San Francisco: Pacific Research Institute on Pub-
lic Policy, 1990); Murray N. Rothbard, For a New Liberty (New York:
Collier Macmillian, 1973): 215-241.
5. For example, John Locke, Adam Smith, John Stuart Mill, and
Thomas Jefferson.
6. For example, Frederic Bastiat, Herbert Spencer, Ludwig von Mises,
Friedrich Hayek, John Hospers, and Robert Nozick.
7. For example, Lysander Spooner, Benjamin Tucker, and Murry
Rothbard.
8. Robert Nozick, Anarchy, State and Utopia (New York: Basic Books,
1974), p. 109. Most social scientists, following Max Weber, regard the de
jure claim of monopoly over the legitimate use of force as a defining
characteristic of states as they exist in the real world. Nozick, however,
accepts the hypothetical case of a purely de facto monopoly as constituting
a minimal state. (See Nozick, pp. 117-118. Actually, the monopoly of
force is used by Nozick to define an "ultraminimal" state. The "minimal"
state requires another feature [taxes, basically] that is not relevant here.)
Nozick argues, further, that there is an advantage to having only one such
agency, in that this will preempt the fewest individuals in the exercise of
their right to punish those who violate the rights of others.
In Chapter 4,1 argued that even if individuals have some kind of contract
with the state, in a Lockean sense, whereby they transfer to the state their
individual right to punish criminals, it may still be legitimate for a gov-
ernment agency to then "subcontract" with private firms to exercise this
authority on behalf of the state. In this chapter, I ask what would happen
if individuals bypassed the government middleman and contracted directly
with private companies to enforce their rights and to punish those who
transgressed against them.
9. Rothbard, For a New Liberty, pp. 215-241.
10. David D. Friedman, The Machinery of Freedom: Guide to Radical
Capitalism (New Rochelle, NY: Arlington House, 1978).
11. See note 4, supra.
298 Notes to Chapter 14
12. Randy Barnett, "Restitution: A New Paradigm of Criminal Justice,"
in Randy Barnett and John Hagel III, eds., Assessing the Criminal: Res-
titution, Retribution, and the Legal Process (Cambridge. MA: Ballinger,
1977), pp. 349-383, at p. 354.
13. Rothbard, For a New Liberty, pp. 231-234.
14. "Relatively autonomous" here means "not subject to a high degree
of direct government regulation." Prisons in a libertarian society would
certainly, however, be strongly constrained by the rule of law.
15. Rothbard, For a New Liberty, p. 259.
16. Barnett, however, proposes that a victim's right to compensation
would become unowned property, available to be claimed, or "home-
steaded," by any entrepreneur. (See Barnett, "Restitution," p. 368.) Bar-
nett argues only for restitution and against punishment, but the
homesteading argument could apply as well to a punishment right.
17. Murray N. Rothbard, "Punishment and Proportionality," in Barnett
and Hagel, eds., Assessing the Criminal, pp. 259-270, at p. 260.
18. Ibid., p. 259.
19. A market in punishment could develop, in which brokers would
seek to profit by buying the right to punish as cheaply as possible from
victims selling it as dearly as possible to offenders or others.
20. Randy Barnett and John Hagel III, "Assessing the Criminal: Res-
titution, Retribution and the Legal Process," in Barnett and Hagel (eds.),
Assessing the Criminal, pp. 1-31.
21. Ibid.
22. Andrew von Hirsch, Doing Justice: The Choice of Punishments (New
York: Hill and Wang, 1976).
23. Von Hirsch's rationale for the justice model, however, is not purely
retributivist. Unlike Kant, von Hirsch see deservedness only as necessary,
but not sufficient, to justify punishment. There is supposedly a "coun-
tervailing moral consideration"—specifically, "the principle of not delib-
erately causing human suffering where it can possibly be avoided" (von
Hirsch, Doing Justice, p. 553). Accepting this principle, von Hirsch argues
that for punishment to be justified, it must also be shown to have a
deterrent effect. A utilitarian element has been added.
Von Hirsch's compromise is internally inconsistent, and thus is weaker
than a purely retributivist justification. The principal that punishment for
wrongdoing is deserved, and the principal against all avoidable suffering,
are logically incompatible. To say that some suffering (i.e., punishment)
may sometimes be deserved is to say that we do not believe that all
avoidable infliction of pain should be avoided. The justice model is
stronger when the utilitarian requirement of deterrence is dropped.
Notes to Chapter 14 299
24. Rehabilitation might be proposed as an individual good in itself,
rather than as a method of crime control. This, however, would be coercive
state paternalism, which is also incompatible with libertarian principles.
For a few exceptions that do not include criminals, see John Hospers.
"Libertarianism and Legal Paternalism," in Tibor R. Machan, eds., The
Libertarian Reader (Totowa, NJ: Rowman and Littlefield, 1982), pp. 135-
144.
25. To say that rights and their corollary duties must be "socially con-
structed" does not conflict with libertarian postulates of natural law, at
least not for secular libertarians. Nonreligious libertarians recognize that
to speak of a "natural right" is to use a verbal shorthand. They do not
believe that a natural right is literally given by God or by nature in some
teleological sense. They understand that a right, like all mental constructs,
is humanly defined. When they speak of something as a "natural" right,
they are saying that it must be treated as absolute and inalienable, as ifii
were given by nature or by God.
26. This view—that sanctions express and uphold values—does much
to explain the role of mercy as a counterpart of desert. Justice and mercy
are not necessarily in conflict. They may be reconciled by noting their
relationship to the value we seek to uphold. In certain cases (e.g., a
starving person steals a small amount of food), punishment, though de-
served, might seem so unfair that its imposition would not serve to uphold
the value at stake (here, property rights). Rather, such punishment would
discredit and thereby weaken that value. To "temper justice with mercy"
in such a case (e.g., by suspending a sentence) is not so much a comprise
between conflicting values as it is a formulation designed to uphold as
effectively as possible the relevant legal value.
27. J. Roger Lee, "The Arrest and Punishment of Criminals: Justifi-
cation and Limitations," in Machan, Libertarian Reader, pp. 86-97, at
p. 95.
28. Barnett and Hagel, "Assessing the Criminal." p. 15.
29. Ibid., p. 13.
30. John Hospers, "Retribution: The Ethics of Punishment," in Barnett
and Hagel, Assessing the Criminal, pp. 181-209, at pp. 204-206.
31. Richard A. Epstein. "Crime and Tort: Old Wine in Old Bottles,"
in Barnett and Hagel, Assessing the Criminal, pp. 231-257.
32. Lee, "The Arrest and Punishment of Criminals."
33. J. Roger Lee and Laurin A. Wollan, Jr., "The Libertarian Prison:
Principles of Lassiez-Faire Incarceration." Prison Journal 65 (Autumn/
Winter 1985): 108-121.
34. Ibid., p. 108.
300 Notes to Chapter 14
35. Lee, "The Arrest and Punishment of Criminals," p. 91.
36. Ernest van den Haag, Punishing Criminals (New York: Basic Books,
1975), p. 15.
37. I leave this fate unspecified. Public welfare? Private charity? Star-
vation? The answer is irrelevant here: only let it be the same inside as
outside.
38. John Price, "Political Enterprise in a Prison: The Free Market
Economy of La Mesa Penitenciaria," Crime and Delinquency 19 (April
1973): 218-227.
39. Ibid., p. 218.
40. The description here draws heavily from Edgar May, "Maine: Was
Inmate Capitalism Out of Control?" Corrections Magazine 1 (May 1981):
17-23; and Jeffrey Shedd, "Making Good(s) Behind Bars," Reason
(March 1982): 23-32.
41. May, "Maine," p. 19.
42. Shedd, "Making Good(s)," p. 25.
43. Ibid., p. 26.
44. In federal prisons, about one-third of prisoners participate in prison
industry programs, but in state prisons it is less than 10 percent.
45. May, "Maine," p. 18.
46. Shedd, "Making Good(s)," p. 26.
47. Ibid., p. 25.
48. May, "Maine," p. 20.
49. Shedd, "Making Good(s)," p. 27.
50. May, "Maine," p. 20.
51. Ibid., p. 18.
52. Shedd, "Making Good(s)," p. 29.
53. Ibid., p. 32 (emphasis in original).
54. David B. Kalinich, The Inmate Economy (Lexington, MA: D.C.
Heath, 1980).
55. Roger Morris, The Devil's Butcher Shop (New York: Franklin
Watts, 1983).
56. Kevin Krajick, "People in Prison Do Not Run Businesses," Cor-
rections Magazine, February 1981, pp. 22-23.
57. See John J. Dilulio, Jr., Governing Prisons: A Comparative Study
of Correctional Management (New York: Free Press, 1987).
58. Detroit News, April 23, 1989.
59. A jurisdiction might even contract with a private company to es-
tablish a new facility or to take over an existing one for a specified period
of time, with the understanding that at the end of this period the facility
would revert to direct government control. The contract could include
training of government employees and managers by the contractors. A
Notes to Chapter 14 301
jurisdiction considering this arrangement would have to expect it to be
more costly, at least in the short run, than a contract that was either long-
term or indefinitely renewable subject to rebidding.
60. Outside of the federal system and a few states that pursue accre-
ditation vigorously.
Selected Bibliography

American Correctional Association. 1985. Private Sector Operation of a

Correctional Institution (Washington, DC: U.S. Department of Jus-
tice, National Institute of Corrections).
Anderson, Patrick, Charles R. Davoli, and Laura J. Moriarty. 1985. "Pri-
vate Corrections: Feast or Fiasco?" Prison Journal 65(2): 32-41.
Auerbach, Barbara J., George E. Sexton, Franklin C. Farrow, and Robert
H. Lawson. 1988. Work in American Prisons: The Private Sector
Gets Involved (Washington, DC: National Institute of Justice).
Bacas, Harvey. 1984. "When Prisons and Profits Go Together." Nation's
Business (October): 62R-63R.
Barnett, Randy. 1986. "Pursuing Justice in a Free Society: Part Two—
Crime Prevention and the Legal Order." Criminal Justice Ethics 5
(Winter/Spring): 30-53.
1985. "Pursuing Justice in a Free Society: Part One—Power vs.
Liberty." Criminal Justice Ethics 4 (Summer/Fall): 50-72.
1979. "Justice Entrepreneurship in a Free Market: Comment."
Journal of Libertarian Studies 3 (Winter): 439-451.
Bast, Diane Carol. 1986. "In Defense of Private Prisons" (Chicago: Heart-
land Institute).
Becker, Craig, and Amy Dru Stanley. 1985. "Incarceration Inc.: The
Downside of Private Prisons." Nation (June 15): 728-730.
Benson, Bruce L. 1990. The Enterprise of Law: Avoiding Anarchy and
Leviathan (San Francisco: Pacific Research Institute on Public
Policy).
Borna, Shaneen. 1986. "Free Enterprise Goes to Prison." British Journal
of Criminology 26(4): 321-334.

302
Selected Bibliography 303
Bowditch, Christine, and Ronald S. Everett. 1987. "Private Prisons: Prob-
lems within the Solution." Justice Quarterly 4: 441-453.
Brakel, Samuel Jan. 1989. "Privatization and Corrections." Federal Pri-
vatization Project Issue Paper No. 7 (Santa Monica, CA: Reason
Foundation).
1988. "Prison Management, Private Enterprise Style: The In-
mates' Evaluation." New England Journal on Criminal and Civil
Confinement 14(2): 175-244.
1988. " 'Privatization' in Corrections: Radical Prison Chic or
Mainstream Americana?" New England Journal on Criminal and
Civil Confinement 14 (Winter): 1-39.
Camp, Camille, and George Camp. 1985. "Correctional Privatization in
Perspective." Prison Journal 65(2): 14-31.
Cikins, Warren I. 1986. "Privatization of the American Prison System:
An Idea Whose Time Has Come?" Notre Dame Journal of Law,
Ethics and Public Policy 2(2): 445-464.
Collins, William C. 1986. "Privatization: Some Legal Considerations from
a Neutral Perspective." In Collins: Correctional Law, 1986 (Olym-
pia, WA: William C. Collins), pp. 81-93.
1985. "Contracting for Correctional Services: Some Legal Con-
siderations" (Washington, DC: National Institute of Corrections).
Cory, Bruce. 1986. "From Rhetoric to Reality: Privatization Put to the
Test." Corrections Compendium (May): 1, 10-14.
Council of State Governments and Urban Institute. 1987. Issues in Con-
tracting for the Private Operation of Prisons and Jails. Executive
Summary and Final Report (Washington, DC: Department of Jus-
tice, National Institute of Justice).
Cullen, Francis T., Jr. 1986. "The Privatization of Treatment: Prison
Reform in the 1980's." Federal Probation 50(1): 8-16.
Dilulio, John J., Jr. 1988. "What's Wrong with Private Prisons," Public
Interest No. 92 (Summer): 66-83.
Donahue, John D. 1988. Prisons for Profit: Public Justice, Private Interests
(Washington, DC: Economic Policy Institute).
Duffy, Susan M. 1984. "Breaking into Jail: The Private Sector Starts to
Build and Run Prisons." Barron's (May 14): 20-22.
Dunham, Douglas W. [Note]. 1986. "Inmates' Rights and the Privatization
of Prisons." Columbia Law Review 86: 1475-1504.
Ellison, W. James. 1987. "Privatization of Corrections: A Critique and
Analysis of Contemporary Views." Cumberland Law Review
17(3:): 683-730.
Elvin, Jan. 1985. "A Civil Liberties View of Private Prisons." Prison
Journal 65(2): 48-52.
304 Selected Bibliography
Evans, Brian B. 1987. "Private Prisons". [Note]. Emory Law Journal 36
(Winter): 253-283.
Fenton, Joseph. 1985. "A Private Alternative to Public Prisons." Prison
Journal 65(2): 42-47.
Field, Joseph E. [Note]. 1987. "Making Prisons Private: An Improper
Delegation of Governmental Power." Hofstra Law Review 15
(Spring): 649-675.
Fitzgerald, Randall. 1988. When Government Goes Private (New York:
Universe Books), pp. 93-119, "Streamlining Our Justice System."
Gage, Theodore. 1982. "Cops, Inc." Reason (November): 23-28.
Geis, Gilbert. 1987. "The Privatization of Prisons: Panacea or Placebo?"
In Barry J. Carroll, Ralph W. Conant, and Thomas A. Easton,
eds., Private Means—Public Ends: Private Business in Social Service
Delivery (New York: Praeger), pp. 76-97.
Gentry, James Theodore [Note]. 1986. "The Panopticon Revisited: The
Problem of Monitoring Private Prisons." Yale Law Journal 96: 353-
375.
Grant, Judy S. 1986. "Prisons for Profit." Hamline Journal of Public Law
and Policy, 7(1): 123-141.
Grant, Judy S., and Diane Carol Bast. 1986. Corrections and the Private
Sector: A Guide for Public Officials (Chicago: Heartland Institute).
Immarigeon, Russ. 1987. "Privatizing Adult Imprisonment in the U.S.:
A Bibliography." Criminal Justice Abstracts (March): 123-139.
1985. "Private Prisons, Private Programs, and their Implications
for Reducing Reliance on Imprisonment in the United States."
Prison Journal 65(2): 60-74.
Joel, Dana. 1988., "A Guide to Prison Privatization." Backgrounder No.
650 (Washington DC: Heritage Foundation).
Kay, Susan L. 1987. "The Implications of Prison Privatization on the
Conduct of Prisoner Litigation under 42 U.S.C. Section 1983."
Vanderbilt Law Review 40(4): 867-888.
Keating, J. Michael, Jr. 1986. Public Ends and Private Means: Account-
ability among Private Providers of Public Social Services. NIDR
Report No. 2 (Washington, DC: National Institute for Dispute
Resolution).
Krajick, Kevin. 1984. Private Financing and Management of Prisons and
Jails (New York: Edna McConnell Clark Foundation, n.p.)
1984. "Punishment for Profit," Across the Board 21(3): 20-27.
1984. "Prisons for Profit: The Private Alternative." State Legis-
latures 10(4): 9-14.
Kravitz, Lee. 1986. "Tough Times for Private Prisons." Venture (May):
56-60.
Selected Bibliography 305

Larson, Erik. 1988. "Captive Company." Inc. June: 86-92.

Lauter, David. 1985. "The Plunge into Private Justice." National Law
Journal (March 11).
Lawrence, David M. 1986. "Private Exercise of Governmental Power."
Indiana Law Journal 61: 647-695.
Lee, J. Roger, and Laurin A. Wollan, Jr. 1985. "The Libertarian Prison:
Principles of Laissez-Faire Incarceration." Prison Journal 65(2):
108-121.
Liebmann, George. 1975. "Delegation to Private Parties in American
Constitutional Law." Indiana Law Journal 50: 650-719.
Logan, Charles H. 1989. "Proprietary Prisons." In Lynne Goodstein and
Doris L. Mackenzie, eds., The American Prison: Issues in Research
and Policy (New York: Plenum).
1987. "The Propriety of Proprietary Prisons." Federal Probation
51(3): 35-40.
Logan, Charles H., and Bill W. McGriff. 1989. "Comparing Costs of
Public and Private Prisons: A Case Study." National Institute of
Justice Research in Action (Washington, DC: Department of
Justice).
Logan, Charles H. and Sharla P. Rausch. 1985. "Punish and Profit: The
Emergence of Private Enterprise Prisons." Justice Quarterly 2(3):
303-318.
McConville, Sean. 1988. "The Privatization of Penal Services." (Stras-
bourg: Draft Report for the Council of Europe, Eighteenth Cri-
minological Research Conference).
1987. "Aid from Industry? Private Corrections and Prison Crowd-
ing." In Stephen Gottfredson and Sean McConville, eds., Amer-
ica's Correctional Crisis: Prison Populations and Public Policy
(Westport, CT: Greenwood Press), pp. 221-242.
McDonald, Douglas, ed. 1990. Private Prisons and Public Interest. (New
Brunswick, NJ: Rutgers University Press).
May, Edgar. 1981. "Maine: Was Inmate Capitalism Out of Control"?
Corrections Magazine 1 (May): 17-23.
Mayer, Connie. 1986. "Legal Issues Surrounding Private Operation of
Prisons." Criminal Law Bulletin 22(4): 309-325.
Miller, Jerome. 1986. "The Private Prison Industry: Dilemmas and Pro-
posals." Journal of Law, Ethics and Public Policy 2: 465-477.
Mullen, Joan. 1985. "Corrections and the Private Sector." Prison Journal
65(2): 1-13.
1985. "Corrections and the Private Sector," National Institute of
Justice Research in Sn'e/(Washington, DC: Department of Justice).
Mullen, Joan, Kent John Chabotar, and Deborah M. Carrow. 1985. The
306 Selected Bibliography
Privatization of Corrections (Washington, DC: National Institute
of Justice).
National Criminal Justice Association. 1987. "Private Sector Involvement
in Financing and Managing Correctional Facilities" (Washington,
DC: National Criminal Justice Association).
Palumbo, Dennis J. 1986. "Privatization and Corrections Policy." Policy
Studies Review 5(3): 598-605.
Pennsylvania Legislative Budget and Finance Committee. 1985. Report
on a Study of Issues Related to the Potential Operation of Private
Prisons in Pennsylvania (Harrisburg, PA: Pennsylvania Legislative
Budget and Finance Committee).
Ring, Charles. 1987. Contracting for the Operation of Private Prisons:
Pros and Cons (College Park, MD: American Correctional
Association).
Robbins, Ira P. 1988. "The Impact of the Nondelegation Doctrine on
Prison Privatization." UCLA Law Review 35: 911-952.
1988. The Legal Dimensions of Private Incarceration (Washington,
DC: American Bar Association).
1986. "Privatization of Corrections: Defining the Issues." Judi-
cature 69(6): 324-331.
Roper, Brian A. 1986. "Market Forces, Privatization and Prisons: A Polar
Case for Government Policy." International Journal of Social Eco-
nomics 13(1): 77-92.
Savas, E. S. 1987. "Privatization and Prisons." Vanderbilt Law Review
40 (May): 889-899.
Shedd, Jeffrey. 1982. "Making Good(s) Behind Bars." Reason (March):
23-32.
Smith, George. 1979. "Justice Entrepreneurship in a Free Market." Jour-
nal of Libertarian Studies 3 (Winter): 405-426.
Starr, Paul. 1987. The Limits of Privatization (Washington, DC: Economic
Policy Institute).
Steinberg, Sheldon S., J. Michael Keating, and James J. Dahl. 1981.
Potential for Contracted Management in Local Correctional Facil-
ities. Report to National Institute of Corrections (Washington,
D.C.).
Stewart, James K. 1986. "Costly Prisons: Should the Public Monopoly be
Ended"? In Patrick B. McGuigan and Jon S. Pascale eds., Crime
and Punishment in Modern America (Washington, DC: Institute
for Government and Politics of the Free Congress Research and
Education Foundation), pp. 365-387.
Taft, Philip B., Jr. 1983. "Private Vendors, Part II: Survival of the Fittest."
Corrections Magazine 9 (February): 36-43.
Selected Bibliography 307
1982. "Private Vendors, Part I: The Fiscal Crisis in Private Cor-
rections." Corrections Magazine 8 (December): 27-32.
Thomas, Charles W., and Linda S. Calvert Hanson. 1989. "The Impli-
cations of 42 U.S.C. Section 1983 for the Privatization of Prisons."
Florida State University Law Review 16(4): 933-962.
Thomas, Charles W., Lonn Lanza-Kaduce, Linda S. Calvert Hanson, and
Kathleen A. Duffy. 1988. The Privatization of American Correc-
tions (Gainesville: FL: Center for Studies in Criminology and Law,
University of Florida).
Travis, Lawrence F., Ill, Edward J. Latessa, Jr., and Gennaro F. Vito.
1985. "Private Enterprise in Institutional Corrections: A Call for
Caution." Federal Probation 49(4): 11-16.
The Urban Institute. 1989. Comparison of Privately and Publicly Operated
Corrections Facilities in Kentucky and Massachusetts (Washington,
D.C.: Urban Institute).
Virginia, Commonwealth of. 1986. "Study of Correctional Privatization."
(Secretary of Transportation and Public Safety).
Voison, Elizabeth. 1985. "Privatization and Prisons." City and State.
(April): 1,30.
Walzer, Michael. 1985. "At McPrison and Burglar King It's . . . Hold the
Justice." New Republic (April 8): 10-12.
Wecht, David N. [Note], 1987. "Breaking the Code of Deference: Judicial
Review of Private Prisons." Yale Law Journal 96: 815-837.
Wooley, Mary R. 1985. "Prisons for Profit: Policy Considerations
for Government Officials." Dickinson Law Review 90 (Winter)
307-331.
Young, Peter. 1987. The Prison Cell (London: Adam Smith Institute).
This page intentionally left blank
Index

Abt Associates, 150-151, 162 American Civil Liberties Union

Accountability. See also Monitoring (ACLU), 10, 12, 18, 49, 72, 119,
administrative, 194-195 125, 149, 155, 156, 184, 188, 204,
community acceptance as form of, 207
201-202 American Correctional Association
double standard of, 145-147 (ACA), 26, 133, 135
economic forms of, 197-199 accreditation by, 20, 23, 24, 27, 44,
issues and arguments analyzed, 128-130, 256
194-210 standards of, 30, 44, 128-130, 146,
issues and arguments listed, 40, 195
44, 47 study of Okeechobee School for
legal, of public and private actors, Boys, 87, 90-91, 102, 138-143,
195-196 165-166, 205-206
monitoring and, 204-210 Arbor, Inc., 202
political, of officials and contrac- Artesian Oaks, 20
tors, 199-201 Arthur G. Dozier School for Boys,
publicity and, 202-203 90-91, 102, 138-143
of special masters, 208-210 Associated Marine Institutes, 19
Accreditation of facilities, 20, 23, 24, Aurora, Colorado, 22, 24, 80, 129
27, 32, 128-130, 256
AFL-CIO, 11 Baker, California Return-to-Custody
Alabama, private facilities in, 34
Center, 28
Alexander, Lamar, 158
Bankruptcy, 40, 48, 103, 178, 182,
American Federation of State,
221, 229-231. See also Depen-
County, and Municipal Employ-
dence (of government on
ees (AFSCME), 10-12, 35-37,
contractors)
36, 103, 156, 165, 177, 179, 211,
Bardacke, Paul, 93
235
Barnett, Randy, 240, 244
American Bar Association (ABA), Bay County (Florida) Jail and Annex,
10, 12, 58, 67, 146
24, 160

309
310 Index
Bay County Jail (cont.) Classification of inmates, 27, 35, 39,
accreditation at, 129 60, 65, 67, 70-71, 116, 122, 143,
costs and savings at, 81, 94, 225 144, 175
described, 29-30 Cleveland, Texas, 27, 28
inspections at, 135-136 Collins, William C., 60, 218
insurance and indemnification at, Colorado, private facilities in, 22, 24,
190-191 80-81, 129
legal assistance and lawsuits at, Colorado Division of Youth Ser-
188-189, 201 vices, 9
monitoring at, 206 Commission on Accreditation for Cor-
security at, 171, 174 rections, 23, 128
Beasley, Tom, 12, 23 Conrad, John P., 154
Beaumont, Gustave de, 215-216 Constituencies in corrections, 42, 154-
Behavioral Systems Southwest, 21-22, 155, 203-204
83, 93, 134, 190 Construction of prisons and jails
Benson, Bruce, 155-156 cost of, 8-9, 79-81
Berk, Richard, 156 need for new, 7-8
Blumstein, Alfred, 150-151 speed of, by contractors, 79-81
Bond referenda and private financing, Contact Center, Inc., 189-190
76-79, 157-158 Corinthian Corporation, 94-95, 121
Brackman, Harold, 156 Correctional Association of New
Brakel, Samuel Jan, 143-144, 175 York, 101
Braly, Malcolm, 216 Corrections Corporation of America
Bridgeport, Texas, 27 (CCA), 12, 122, 129, 143-144,
Buchanan, James, 74 160, 165, 166, 167, 208, 233, 234
Buckingham Security Limited, 34-37, costs and savings, 79, 81, 93—94,
82, 95, 125-127, 134, 174-175, 105-118, 225
189, 206 credentials of executive officers,
Bureau of Justice Statistics, 14, 15, 17 133, 135
Butler County (Pennsylvania) Prison, facilities described, 19-33 passim
34-37, 82, 95, 174-175, 189, inspection of facilities, 135-136
206 liability, 188-191, 231
monitoring of facilities, 70, 206
political connections of, 158
profits, 223
California, private facilities in, 20-22, security at facilities of, 171, 174-
28, 93, 96, 169, 173 175, 178
California Department of Corrections, speed of construction by, 80, 227
22, 28, 134, 173 - staff training and turnover, 130, 132
Camp, Camille, 88, 100, 101 Corruption
Camp, George, 88, 101 control of, 219-220
Carlson, Norman, 96 as corollary of government, 212-214
Carter v. Carter Coal Co., 61 history of, in corrections, 214-218
Cass, E. R., 133, 135, 217 issues and arguments analyzed,
Cass, Ronald, 4 211-220
Central Texas Parole Violator Facil- issues and arguments listed, 40,
ity, 28 44, 47
Index 311
Cost(s) of imprisonment competition and, 234-235
comparing, problems with, 96-105 exclusive franchises and, 233-234
comparing private and public, 90— issues and arguments analyzed,
98, 105-118 221-235
construction, 8-9, 79-81 issues and arguments listed, 40, 45,
financing, 76-79 48
hidden, 98-104 lowballing and, 221-225
issues and arguments analyzed, 76- market entry costs and, 225-227
118 ownership of facilities and,
issues and arguments listed, 39, 41 — 228-229
42, 45-46 Dilulio, John Jr., 50-52, 56, 83, 123,
operation, 8-9, 81-82 124, 129-130, 132, 215
"overrun" in, at Silverdale Deten- Discipline of inmates, 27, 30, 32, 34,
tion Center, 223-224 44, 60, 65, 66-70, 143, 205
in public sector, 8-10 Dobson, Craig, 24
reasons to expect savings, 83-90 Donahue, John, 83, 119, 167
"rebates" from contracting, 104-105 Dostoevsky, Fyodor, 213
skepticism about savings, 82-83 Double standard, public vs. private, 5
Coughlin, Thomas, 229 of accountability, 145-147
Council of State Governments, The, of liability and immunity, 186-187,
103, 128, 145, 171, 172, 178, 204- 192-193
205, 210, 219, 230-231 of quality, 145-147
Crane, Richard, 133 Drucker, Peter, 83
Cream skimming, charges of, 121-125 Due process, 12, 45, 143
Credentials of private prison execu- classification decisions and,
tives, 133-135 70-71
Cuniff, Mark, 49 contracting contributes to, 41,
63-64
good time or disciplinary decisions
Banner, Inc., 131 and, 66-70
Delegation of authority independent review and, 65-66,
at Butler County Jail, 35 69-70
at CCA-Bay County Jail and An- ombudsmen and, 66
nex, 30 private delegation and, 61-71
at CCA-New Mexico Women's Cor- public and private equality of, 62-
rectional Facility, 27 63, 64-66
due process issues in, 61-71
due process theory of, 61-63
propriety of, 49-75 Eagle Mountain Return-to-Custody
rule of law and, 54 Center, 28
at Santa Fe Jail, 32-33 Eckerd Foundation, Jack and Ruth,
at Silverdale Detention Center, 32 18, 19, 82, 90-91, 122, 136, 137-
sovereignty as limit on, 57-61 143, 165-166, 203
special masters and, 208-210 Eckerd Youth Development Center.
Dependence (of government on See Okeechobee School for Boys
contractors) Eclectic Communications, Inc., 21,
bankruptcy risks and, 229-231 134
312 Index
Economies of scale issues and arguments listed, 39, 43,
interjurisdictional contracting and, 46-47
84-85 justice enhanced by, 152
special needs prisoners and, 124- rigidity as opposed to, 163-164
125 unpredictability of correctional
Edna McConnell Clark Foundation, needs and, 168-169
150, 153 Florida, private facilities in
Efficiency of private sector Bay County Jail and Annex, 24,
in constructing prisons, 79-81 29-30, 94, 160, 171, 174, 188-
disputed, 82-83 191, 201, 206, 225
expectations and evidence, Florida Environmental Institute, 19
83-90 Okeechobee School for Boys, 18-
in operating prisons, 19, 90-92, 96, 102, 122, 130, 135-
81-82 143, 203, 205-206
Eisenbuch, Arthur, 137 Florida Department of Health and
Epstein, Richard, 245 Rehabilitative Services, 19
Escapes from private facilities, 25, Florida Environmental Institute, 19
140, 175-176 Force, use of in private prisons, 172—
Evaluations of private prisons. See 175
Quality (of prison operations) Friedman, David, 240

Federal Bureau of Prisons, 126, 171, Gary White and Associates, 28

189 Gentry, James, 228
accreditation of facilities, 129 Gettinger, Stephen, 129-130
construction time, 80 Good time or gain time allocation,
contract at Hidden Valley Ranch, 152, 205
21, 96, 121, 169, 206 private delegation of, 39, 60, 65,
contracts for community treatment, 66-70
16 at specific private facilities, 27, 30,
offenders in private secure facilities, 32, 33, 34, 143, 206
21-24, 32, 33, 160 Government expansion through con-
Federal Law Enforcement Training tracting, 231-233
Center, 130 Governmental functions
Federal Salary Reform Act, 86 criminal justice, 58-59
Fenton, Charles, 34, 36, 125-127 difficulty of defining, 4
Fenton, Joseph, 34 widely delegated, 58, 59
Financing of prisons, private, 76-79 Grace Commission, 86, 87-88
Finkenauer, James, 137 Greene County (Tennessee) Jail, 34
Flexibility of private prisons, 10, 77, Gursky, Henry, 137
96, 141, 159-161, 162-170
disruption during contract changes Haag, Ernest van den, 122
and,164-166 Hagel, John, 244
innovation and, 162-163, 166-168, Hamilton County (Tennessee) Penal
255 Farm. See Silverdale Detention
issues and arguments analyzed, Center
162-170 Hanrahan, John, 82-83, 211
Index 313

Hennepin County, Minnesota, 94 as hidden cost in public sector, 99-

Hidden Valley Ranch, 21, 28, 96, 134, 100
206 liability risks and, 186-193
Hirsch, Andrew von, 243 premiums as measure of quality and
History "lessons" for private prisons, risk, 197
214-218 in Silverdale cost analysis, 106-108,
Hoelter, Herbert, 203 111
Hospers, John, 245 and strikes, 178
Hospital Corporation of America, 22 International Halfway House Associa-
Houston, Texas, private facilities in, tion, 16
23, 28, 80, 160, 206
Houston Processing Center, 23, 28 Jacobs, James, 79
Houston Reintegration Center, 28 Jails, privately contracted, 29-37
Hutt, John, 94 Justice. See also Due process
Hutto, T. Don, 23, 80, 135, libertarian views of, 240-246
208 and supply of prison space, 151-
152, 169-170
Immigration and Naturalization Ser- Juvenile facilities, 15-20
vice (INS), 13, 120, 121, 129-131,
160, 161, 182-183, 206
construction of private facilities for, Kalt, Joseph, 198
80-81 Keating, J. Michael, 181, 203, 214
contract terminations, 165, 234 Kentucky, private facilities in. See
contracts described, 21-24 Marion Adjustment Center
costs at detention centers, 93, 95, Kentucky Corrections Cabinet, 25, 26,
223 92, 202
Imprisonment, expansion of. See Keohane, Tom, 21, 121
Quantity (amount of Krajick, Kevin, 136-138
imprisonment) Kyle, Texas, 27
Indemnification by contractors, 30,
44, 103, 116, 190-193 LaHonda, California, 21, 169
Inspections of private facilities, 33, Laredo (Texas) Processing Center, 23,
104, 135-136, 188, 205. See also 80, 188
Quality (of prison operations) Larragoite, Patricio Dr., 94
Insurance, 40, 44, 147 Lawrence, David M., 57, 61
and bankruptcy, 230 Lee, J. Roger, 245-247, 250
carried by Behavioral Systems Leo Chessley Center, 28
Southwest, 22 Lesser, Selma, 156
carried by Corrections Corporation Liability of government and private
of America, 22-23, 27, 30, 33, actors, 33, 34, 63, 245
116 constitutional torts under 42 U.S.C.
carried by Eckerd at Okeechobee, Section 1983, 182-185
102 costs at Silverdale, 98-99, 103, 110
carried by U.S. Corrections Corpo- coverage for. See Insurance
ration, 26 double standard of, 147, 186-187,
companies as monitors, 207 192-193
314 Index
Liability (cont.) Market entry costs and market domi-
expansion of inmate rights and, nance, 225-227. See also Depen-
184-185 dence (of government on
implications of Medina v. O'Neill, contractors)
182-183 Massachusetts, private juvenile facili-
issues and arguments analyzed, ties in, 15, 149, 210
181-193 Massachusetts Division of Youth Ser-
issues and arguments listed, 40, vices, 15
44, 47 Massey Burch Investment Group, 22
legal damage awards in corrections, Medina v. O'Neill, 182
189-190 Mesa Verde, California, 28
legal exposure vs. actual risk, 187- Miley, Harry W. Jr., 86
190 Miller, Jerome, 15, 149
sovereign immunity, 185-187 Minnesota, private facilities in, 33, 94
Libertarian Model Contract and Model Statute
arguments for delegation of author- proposed to ABA, 67-70, 146-
ity to imprison, 52-54, 60-61 147, 185
justification of imprisonment, 244- Monitoring, 40, 41, 44, 45, 99, 227,
246 230. See also Accountability
philosophy of American govern- and accountability, 204-210, 219
ment, 3 as cost and benefit of contracting,
prisons, 246-254 103-104, 112, 113, 117, 256
views of crime and punishment, elements of, 204-205
240-246 equally needed in public and pri-
views of the state, 238-240 vate prisons, 145-147, 207-210
Lindholm, Tamara S., 22 and independent review, for due
Live Oak, California, 28 process, 65, 70
Lobbying in corrections, 39, 46, 141, at private facilities, 23, 25, 27, 30,
149, 152-159, 160, 226. See also 31, 35, 205-206
Quantity (amount of Monopoly. See Dependence (of gov-
imprisonment) ernment on contractors)
Locke, John, 52, 53 Moore, Doyle H., 129
Los Angeles County Jail, 93 Mud slinging by critics, 125-127
Lowballing by contractors, 39, 45, Mullen, Joan, 162-163, 219-220, 223
221-225. See also Dependence (of
government on contractors)
National Association of Criminal Jus-
McCarthy, Bernard, 212 tice Planners, 49
McCullough, Hubert, 158 National Center on Institutions and
McFarland, California, 28 Alternatives, 149, 153, 203
McGriff, Bill, 105-117, 135 National Commission on Correctional
Maghan, Jess, 167 Healthcare, 129
Maine State Prison, 250-254 National Institute of Corrections, 24,
Management and Training Corp., 90, 133, 138
20, 28 National Labor Relations Act, 176
Marion Adjustment Center, 24-26, National Sheriffs' Association, 10,
92-93, 121, 130, 175, 202, 206 11,29
Index 315

New Mexico, private facilities in, 32— Picariello v. Fenton, 126

33, 93-94, 95, 135, 160, 166, 231 Pitts, Lavelle, 30, 31
New Mexico Corrections Department, Polls of public on crime and prisons,
27, 166 78-79, 154
New Mexico State Penitentiary, 135, Poole, Robert W. Jr., 84, 220
216 Presser, Stefan, 119
New Mexico Women's Correctional Pricor, Inc., 28, 33-34, 134, 158, 190
Facility, 26-27, 166 Prison Health Services, 129
Nissen, Theodore R., 22 Private facilities, descriptions of,
Nozick, Robert, 53, 239, 240, 242, 13-37
243 adult community facilities, 16-17
adult confinement facilities, 20-37
Oak Hill Youth Center, 175 federal contracts, 21-24
Office of Management and Budget, juvenile institutional facilities,
89, 179 17-20
Okeechobee (Florida) School for juvenile open facilities, 15-16
Boys, 96, 122, 132, 165, 203 local contracts, 29-37
costs compared to Arthur G. Do state contracts, 24-28
zier School, 90-92 Private interests and public interests,
described, 18-19 152-159
escapes at, 175-176 Private prisons
evaluation study of, 137-143 arguments favoring, listed, 41-45,
insurance at, 102 254-257
monitoring at, 205-206 arguments opposing, listed, 45-48
personnel efficiency at, 82 defined, 13-14
purchasing efficiency at, 87 forms of, 14, 20-21
staff training, morale, and turnover future of, 236-237
at, 130, 138-140, 142 issues identified, 38-40
Opposition to private prisons, sources number of, 20
of, 10-12 Privatization
Ownership of private correctional fa- compared to contracting, 4-5, 237-
cilities, 13-14, 17, 18, 23, 24-25, 238, 254
29, 77, 80, 130, 150, 228-229 corrections as extreme form of, 4-5
defined, 3-4
Pasadena Immigration Holding Facil- and faith in government, 3-4
ity, 22 as an option for correctional crisis,
Patterson, Richard M., 36 9-10
Pennsylvania, private facilities in, Profit motives, 39, 44, 45, 50, 55, 83,
11-12 119, 152-153, 197-198
Butler County Prison, 34-37, 82, and corruption, historical examples,
95, 174-175, 189, 206 214-218
Weaversville Intensive Treatment good time decisions and, 66-70
Unit, 17-18, 80, 83, 92, 122, 136- versus motives of government and
138 other actors, 63, 71-75, 84
Performance bonds, 32, 178, 231 Profits reports
Personnel efficiency at private pris- by Behavioral Systems Southwest,
ons, 82 22, 183
316 Index
Profits reports (cont.) at Weaversville Intensive Treatment
by Corrections Corporation of Unit, 136-137
America, 223 Quantity (amount of imprisonment)
by RCA Services, 83 "capacity drives use" theory of,
Propriety 150-152
criticisms on principle of, flexibility vs. expansion effects on,
49-52 159-161
of delegation of imprisonment au- issues and arguments analyzed,
thority, 52-54 149-161
due process and issues of, 61-71 issues and arguments listed, 39, 42-
issues and arguments analyzed, 43, 46
49-75 lobbying effects on, 152-159
issues and arguments listed, 39,
41, 45
Rabinowitz, Sandy, 49
legitimacy of authority as issue of,
Radio Corporation of America
50-57
(RCA), 12, 17-18, 80, 83, 92,
as symbolic issue, 55-57
136-138
Public choice theory, 74
Ramsey County, Minnesota, 33, 94
Punishment
Rawls, John, 242, 243
libertarian views of crime and, 240-
Return-to-custody (RTC) facilities, 28
246
Reynolds, Morgan, 156
privatization of, 237-238
Rights of inmates. See Due process
Ring, Charles, 145-146, 224
Quality (of prison operations)
Robbins, Ira, 55-56, 65, 67-70, 146-
ACA standards and, 128-130
147, 183-185, 192
at Butler County Prison, 35-37
Roseville, Minnesota, 33
corner cutting as threat to, 119-121
Rothbard, Murray, 239-241
cream skimming and effects on,
121-125
credentials as predictor of, 133-135 Sagarin, Edward, 167
double standard for, 145-147 St. Mary's, Kentucky, 24
inspections as measure of, 135-136 San Diego County, California, 12
issues and arguments analyzed, Santa Fe County (New Mexico) De-
119-148 tention Facility, 32-33, 93-94,
issues and arguments listed, 39, 135, 160, 231
42, 46 Saugus, California, 20
at Okeechobee and Dozier Schools Savas, E. S., 81, 97-98
for Boys, 137-143 Savin, Peter, 34
in private facilities reflects quality Schmidt, Robert, 165
of public system, 136-143 Schoen, Ken, 153
at Silverdale Detention Center, Security
143-144 at Bay County Jail and Annex,
staff training and, 130-131 29-30
staff turnover as factor in, 131-133 at Butler County Jail, 35, 36
survey of contracted services, impli- at California return-to-custody facil-
cations for, 144-145 ities, 28
Index 317
at CCA-New Mexico Women's Cor- Sobolevitch, Robert H., 137
rectional Facility, 26 Southwest Detention Facilities, 95
and deadly force, 172-174 Standards and goals clarified by con-
at Eckerd Youth Development tracts, 127-130. See also Accredi-
Center, 19 tation of facilities; American
and escapes from private facilities, Correctional Association (ACA)
175-176 Starr, Paul, 231
at Hidden Valley Ranch, 21 Steinhardt, Barry, 119
at immigration facilities, 22-23 Strikes by corrections officers, 176-
at institutional juvenile facilities, 17 180
issues and arguments analyzed, Supply and demand, imprisonment,
171-180 7-8
issues and arguments listed, 39-40, Sweetwater (Texas) Reintegration
43, 47 Center, 28
at Marion Adjustment Center,
24-25
measures at private facilities, 174- Tennessee, private facilities in
175 Greene County Jail, 34
at open juvenile facilities, 15-16 Hamilton County Penal Farm. See
at Shelby Training Center, 19 Silverdale Detention Center
strikes as a threat to, 176-180 Shelby Training Center, 19-20
at Weaversville Intensive Treatment Tall Trees, 20
Unit, 18 Tennessee v. Garner, 172
Sennholz, Hans, 232-233 Termination of contracts, 164-166
Services, contracting of in prisons, Texas, private facilities in, 23, 27-28,
88-89 80, 160, 188, 234
Shedd, Jeffrey, 252 Texas Board of Pardons and Paroles,
Shelby Training Center, 19-20 27, 160
Silverdale Detention Center (Hamil- Texas Department of Corrections, 27,
ton County Penal Farm) 188
costs at, 105-117, 223-224 Texas pre-parole facilities, 27-28
described, 31-32 Thomas, Charles, 184
due process at, 70 Thompson, Milton, 25
escapes at, 176 Tocqueville, Alexis de, 215
evaluation study of, 143-144 Todd, J. Clifford, 25
inspections of, 135 Training of correctional employees,
insurance at, 190 26, 30, 32, 43, 47, 100, 116, 119
legal assistance and lawsuit at, 188, in private facilities, 130-131, 139,
189 141, 143, 146, 173
monitoring at, 206 Turnover of correctional employees,
purchasing efficiency at, 87 42, 47, 81, 120, 131-133
security at, 175 in private facilities, 31, 36, 82, 132,
staff training at, 130 139-140, 142
staff turnover at, 132 Tuscaloosa (Alabama) Metropolitan
strike clause in contract for, 178 Minimum Security Detention Fa-
Smith, Michael, 153 cility, 34
318 Index
U.S. Corrections Corporation, 24-26, Wackenhut Corrections Corporation,
92-93, 134, 202, 227 33, 133-134, 190-191, 223, 234
U.S. Marshall's Service, 32, 160 immigration detention facilities, 22,
Unions, 72, 82, 95, 155, 163-164, 212 23-24, 80-81, 129
as lobbyists for imprisonment, 156- pre-parole and parole violation fa-
157, 159 cilities, 27-28
opposition to privatization from, Walzer, Michael, 57
10-12, 28, 35-37, 46, 120, 235 Weapons at private facilities, 23, 30,
and strikes at prisons, 176, 180 32, 35
Urban Institute, The, 103, 128, 145, Weaversville (Pennsylvania) Intensive
171, 172, 178, 204-205, 210, 219, Treatment Unit, 17-18, 80, 83,
230-231 92, 122, 136-138
Wecht, David, 63
Wollan, Laurin, 58, 246-247, 250
Venus, Texas, 27 Woolley, Mary, 176, 178
Vera Institute, 153 Wray, Harmon, 152-153, 158, 214
Vick, G. E., 80
Volunteers of America (Regional Youth Corrections Act, 21, 96,
Corrections Center), 33, 94 169